Choosing the Right Path: A Practical Guide to Systematic vs. Narrative Reviews in Ecotoxicology for Research and Policy

Abstract

This article provides a comprehensive, practical guide for researchers, scientists, and drug development professionals on the distinct roles, methodologies, and applications of systematic reviews and narrative reviews within ecotoxicology. It begins by exploring the foundational definitions and core philosophical differences between these two major evidence synthesis approaches, highlighting the transition in the field towards more rigorous, systematic methods [citation:5][citation:7]. The guide details the standardized, multi-step methodology of a systematic review, including protocol development and specific challenges in evaluating ecotoxicological evidence such as exposure assessment [citation:1][citation:5]. It further addresses common pitfalls in conducting both types of reviews and offers strategies for optimizing their quality and reporting [citation:3][citation:4][citation:6]. Finally, the article presents a direct comparative analysis of the validity, transparency, and ultimate utility of each review type for informing research and regulatory decisions, concluding with actionable recommendations for selecting the appropriate synthesis method based on the specific goals of the inquiry [citation:5][citation:7].

Defining the Divide: What Are Systematic and Narrative Reviews in Ecotoxicology?

Within the rigorous domain of ecotoxicology and environmental health research, the synthesis of existing evidence is a cornerstone for advancing knowledge, shaping regulation, and guiding drug development. This process exists on a spectrum, anchored by two foundational approaches: the narrative review and the systematic review. These are not merely different labels but represent distinct philosophies of inquiry with profound implications for the conclusions drawn and the decisions informed. A narrative review is traditionally understood as a scholarly summary that provides interpretation, critique, and a broad overview of a field, often crafted through expert judgment to deepen conceptual understanding [1]. In contrast, a systematic review is defined by a predetermined, structured methodology aimed at minimizing bias. It exhaustively searches for, critically appraises, and synthesizes all available evidence on a narrowly focused question, often employing quantitative meta-analysis [2] [3].

This guide posits that a spurious hierarchy that universally elevates systematic reviews over narrative reviews is counterproductive [1]. The critical task is not to declare one method superior but to achieve definitional clarity and methodological appropriateness. The core thesis is that research synthesis in ecotoxicology demands a deliberate choice of approach based on the specific research question. Systematic reviews provide methodological rigor for focused, data-driven questions (e.g., "What is the pooled effect size of chemical X on reproductive endpoints in model fish species?"), while narrative reviews offer indispensable expert synthesis for exploring complex, evolving concepts, debating mechanisms, or integrating disparate evidence streams [4] [1]. Achieving methodological rigor, therefore, depends on either adhering strictly to the transparent protocols of a systematic review or applying "systematic thinking"—such as explicit methodology reporting and bias consideration—within a narrative framework [4].

Foundational Methodologies: A Comparative Analysis

The fundamental differences between narrative and systematic reviews extend across every phase of the research synthesis process. The table below provides a structured comparison of their core characteristics, objectives, and methodological standards, highlighting their complementary roles in evidence synthesis.

Table 1: Comparative Analysis of Narrative and Systematic Review Methodologies

Aspect	Narrative (Expert) Review	Systematic Review
Primary Objective	To provide a broad, expert-led overview, interpretation, and critique of a field; to deepen conceptual understanding and identify theoretical gaps [2] [1].	To answer a specific, focused research question by minimizing bias via a comprehensive, reproducible summary of all available evidence [2] [3].
Research Question	Often broad, flexible, or multiple questions exploring concepts, debates, or historical trends [2].	Narrowly focused and defined a priori using frameworks (e.g., PECO: Population, Exposure, Comparator, Outcome) [5] [3].
Protocol & Methodology	No mandatory pre-published protocol. Methods may be flexible and adapt during the review, though best practice encourages describing search and selection processes [4] [2].	A detailed, pre-registered protocol is mandatory, specifying all steps from search strategy to synthesis plan before commencement [5] [3].
Search Strategy	May not be exhaustive or fully reproducible; often targets key literature and seminal works known to experts [1].	Exhaustive, documented search across multiple databases/registers to identify all relevant evidence, ensuring reproducibility [2] [3].
Study Selection & Appraisal	Inclusion/exclusion criteria may not be explicit or uniformly applied; quality appraisal is often narrative [1].	Explicit, pre-defined eligibility criteria applied consistently by multiple reviewers; formal risk-of-bias/quality assessment using tools [5] [3].
Data Synthesis	Primarily qualitative, narrative, and thematic. Aims for interpretation, critique, and theory-building [6] [7].	Can be qualitative but often involves quantitative synthesis (meta-analysis) to derive pooled effect estimates with increased statistical power [8] [6].
Key Strength	Contextualization & Insight: Capable of integrating diverse knowledge types, clarifying complex issues, and generating novel hypotheses [1].	Minimized Bias & Precision: Provides the least biased estimate of effect, crucial for definitive conclusions and decision-making (e.g., risk assessment) [2] [3].
Common Application in Ecotoxicology	Exploring the development of a scientific concept (e.g., epigenetic transgenerational inheritance); debating mechanisms of mixture toxicity; summarizing a rapidly evolving field.	Supporting chemical risk assessment by deriving reference values; evaluating the strength of evidence for a specific toxicological endpoint; informing regulatory submissions [3].

The Systematic Review Workflow: A Rigorous Protocol

The methodological power of a systematic review stems from its adherence to a structured, multi-stage protocol. This workflow is designed to maximize transparency, reproducibility, and objectivity. The following diagram and subsequent breakdown detail the standardized sequence of steps, as adapted from PRISMA guidelines and ecotoxicology-specific frameworks [9] [3].

Diagram 1: Systematic Review Workflow for Ecotoxicology (48 characters)

Protocol Development & Problem Formulation

The process begins with a clearly articulated problem. In ecotoxicology, this is often framed as a PECO question (Population, Exposure, Comparator, Outcome), which is critical for defining scope [3]. For example: "In freshwater zebrafish (Danio rerio) (P), what is the effect of chronic exposure to microplastic PS beads (E) compared to unexposed controls (C) on fecundity and embryo mortality (O)?" A detailed protocol is then registered, specifying databases, search strings, eligibility criteria, data items, and the planned risk-of-bias assessment and synthesis methods.

Systematic Search & Study Selection

An information specialist should design and execute a comprehensive search across multiple databases (e.g., PubMed, Web of Science, Scopus, TOXLINE) using controlled vocabulary and keywords. This search must be documented in full [3]. The study selection process follows the PRISMA flow diagram [9]. All retrieved records are initially screened by title/abstract against eligibility criteria by two independent reviewers to minimize error. Potentially relevant full texts are then obtained and assessed in detail. Disagreements are resolved through discussion or a third reviewer.

Data Extraction & Critical Appraisal

A standardized, piloted data extraction form is used to collect relevant details from each included study: author, year, study design (e.g., lab experiment, field study), test organism, exposure regime, endpoint measures, results, and funding source. Concurrently, each study's methodological quality and risk of bias are assessed using discipline-appropriate tools (e.g., SYRCLE's risk-of-bias tool for animal studies, COSMOS-E for observational studies). This step differentiates a systematic review from a simple literature list [5] [3].

Evidence Integration & Synthesis

Qualitative Synthesis: All reviews include this step. Extracted data are organized (often in tables) and analyzed narratively to identify patterns, consistencies, inconsistencies, and relationships between study characteristics and findings [6] [7].

Quantitative Synthesis (Meta-Analysis): When studies are sufficiently homogeneous in PECO and design, a meta-analysis can be performed [8]. This involves calculating a weighted average of study-specific effect sizes (e.g., standardized mean difference, risk ratio). A forest plot is the key graphical output, displaying each study's effect estimate and confidence interval, and the pooled estimate. Statistical heterogeneity is quantified using metrics like I². Subgroup analysis or meta-regression can explore sources of heterogeneity (e.g., differences in exposure duration, species) [8] [7].

Confidence Rating & Reporting

The overall strength or certainty of the synthesized evidence is formally graded using frameworks like GRADE (Grading of Recommendations Assessment, Development and Evaluation). Factors such as risk of bias, inconsistency, indirectness, imprecision, and publication bias are considered to rate confidence as high, moderate, low, or very low [3]. Reporting must adhere to the PRISMA 2020 statement, including the completed flow diagram and checklist, to ensure transparency and completeness [9].

Conducting a high-quality review requires specific tools and resources. The following table details key solutions for the ecotoxicology researcher.

Table 2: Research Reagent Solutions for Evidence Synthesis

Tool/Resource	Function in Review Process	Key Features for Ecotoxicology
PRISMA 2020 Statement & Flow Diagram [9]	Reporting Guideline: Provides a minimum set of items for transparent reporting of systematic reviews and meta-analyses.	The flow diagram is essential for documenting the study selection process from many databases. The checklist ensures all methodological aspects are reported.
Cochrane Handbook for Systematic Reviews	Methodological Guidance: The definitive guide for conducting systematic reviews of health interventions, with principles applicable to toxicology.	Chapters on risk-of-bias assessment, qualitative synthesis, and meta-analysis provide foundational statistical and methodological standards.
SYRCLE's Risk of Bias Tool	Critical Appraisal: Tool designed to assess risk of bias in animal intervention studies, highly relevant for ecotoxicological lab studies.	Evaluates sequence generation, blinding, outcome assessment, incomplete data, and selective reporting specifically in animal-based research.
GRADE Framework	Evidence Grading: System for rating the certainty (quality) of a body of evidence in systematic reviews.	Allows reviewers to communicate how much confidence risk assessors or regulators should place in the synthesized findings [3].
DistillerSR, Rayyan, Covidence	Review Management Software: Web-based platforms to manage the entire review process: importing references, screening, data extraction, and collaboration.	Essential for managing large volumes of search results from multiple databases and enabling blinded screening by multiple reviewers to reduce bias.
R Statistical Software (metafor, meta packages)	Quantitative Synthesis: Open-source environment for conducting comprehensive meta-analysis and creating forest plots.	Enables complex statistical models, meta-regression to explore heterogeneity, and sophisticated assessment of publication bias.
PECO Framework	Problem Formulation: Adaptation of the clinical PICO framework for environmental health. Defines Population, Exposure, Comparator, Outcome.	Crucial for framing a focused, answerable research question in exposure science and toxicology at the protocol stage [3].

Application in Ecotoxicology: Navigating the Choice

The choice between a narrative and systematic approach in ecotoxicology is not a matter of prestige but of strategic fit. The decision pathway below illustrates the key considerations that should guide this choice.

Diagram 2: Decision Pathway for Review Type Selection (59 characters)

Systematic Reviews are the preferred method for regulatory toxicology and quantitative risk assessment. For instance, the Texas Commission on Environmental Quality (TCEQ) employs a formal six-step systematic review framework to develop toxicity factors like reference values and unit risk factors [3]. This ensures the decision-making process is transparent, minimizes bias, and yields a reproducible estimate of hazard potency. They are also ideal for resolving controversies where a definitive, pooled estimate of effect is needed from existing, consistent data.

Narrative Reviews excel in contexts requiring expert synthesis. This includes mapping the historical development of a paradigm (e.g., the endocrine disruption hypothesis), debating the ecological relevance of laboratory biomarkers, or integrating evidence across vastly different levels of biological organization (from molecular initiating events to population-level effects). Their strength lies in providing context, generating novel hypotheses, and offering a critical perspective that strict protocol-driven methods might constrain [1].

Current evidence suggests that the methodological rigor of systematic reviews in environmental health can be inconsistent, with reviews implementing a median of only 6 out of 11 recommended practices [5]. The use of a pre-published protocol, however, is strongly associated with higher rigor (mean score of 7.77 vs. 5.39) [5]. This underscores that the label "systematic" alone is insufficient; adherence to the complete, rigorous protocol is paramount.

The synthesis of ecotoxicological evidence is a critical endeavor that supports science, policy, and public health. The false dichotomy and implied hierarchy between narrative and systematic reviews serve neither rigor nor understanding. As this guide has detailed, each approach has a defined epistemology, methodology, and domain of applicability.

The path forward requires methodological precision. Researchers must consciously select the appropriate synthetic tool based on a clear problem formulation. When a systematic review is warranted, commitment to its full protocol—from registration to GRADE assessment—is non-negotiable. When a narrative review is the fit-for-purpose tool, authors should incorporate elements of "systematic thinking": describing their search methodology, being transparent about selection criteria, and explicitly acknowledging potential biases [4]. By moving beyond labels to embrace core definitions and rigorous execution, ecotoxicology researchers can produce syntheses that are not only authoritative and insightful but also trustworthy and foundational for future research and decision-making.

Historical Context and the Shift Towards Systematic Methods in Toxicology

Historical Foundations of Toxicology

The science of toxicology has evolved from ancient empirical knowledge to a structured scientific discipline. Early civilizations, including the Ancient Egyptians (3000 BC), Greeks, and Romans, utilized plant, animal, and mineral-based poisons for hunting, warfare, and political purposes, developing a rudimentary understanding of dose-response relationships [10]. The systematic study of poisons, however, began with key Renaissance and Enlightenment figures.

The German-Swiss physician Paracelsus (1493-1541) is considered the "Father of Toxicology." He introduced the foundational concept that "All things are poison, and nothing is without poison; only the dose makes a thing not a poison," establishing the dose-response principle central to the field [10] [11]. In the 19th century, Mathieu Orfila (1787-1853) systematized the discipline. As the "Father of Forensic Toxicology," he published the first comprehensive treatise, Traité des Poisons (1814), which classified poisons and developed reliable methods for their detection in human tissues, laying the groundwork for forensic and experimental toxicology [10] [11].

The 19th century saw critical methodological advances driven by forensic needs. The Marsh test (1836) provided a sensitive, specific method for detecting arsenic in biological samples, revolutionizing criminal investigations [10]. Simultaneously, pioneers like François Magendie and Claude Bernard pioneered systematic animal experiments to study the physiological effects of poisons, moving the field from observation to controlled experimentation [10].

The 20th century marked the institutionalization and specialization of toxicology. The establishment of regulatory agencies like the U.S. Food and Drug Administration (FDA) and the Environmental Protection Agency (EPA) created a framework for safety assessment [10]. This period also saw the standardization of toxicity testing protocols, such as the LD50 test for acute toxicity and the Ames test for mutagenicity, alongside the emergence of specialized sub-disciplines including environmental, occupational, and clinical toxicology [10].

Historical Milestones in Toxicology

The Narrative Review Paradigm and Its Limitations

Historically, knowledge synthesis in toxicology, as in clinical research, was dominated by the narrative or expert review [12]. This traditional approach involves an expert authoritatively summarizing a field or addressing a specific question based on a subset of the literature. The process for identifying, selecting, and weighting evidence is typically implicit and non-transparent, relying on the author's expertise and perspective [12].

This paradigm carries significant limitations and risks. The non-systematic selection of evidence increases the potential for selection bias, where studies supporting the author's preconceived ideas are given undue weight [12]. Furthermore, the lack of explicit quality assessment of included studies can perpetuate errors or biases from the primary research into the review's conclusions. The absence of a reproducible search and selection methodology makes it impossible to independently verify or update the review, while the frequent lack of a quantitative synthesis (meta-analysis) limits the objective assessment of the evidence [12]. These shortcomings can lead to conflicting conclusions from the same evidence base, as seen in historical debates over chemicals like Bisphenol A, creating uncertainty for regulators, industry, and the public [12].

While narrative reviews retain value for providing broad expert perspectives or quick overviews when time is limited, their methodological weaknesses render them unsuitable for informing high-stakes regulatory or clinical decisions, where objectivity, transparency, and reproducibility are paramount [12] [2].

The Systematic Review Revolution

Systematic reviews emerged from the Evidence-Based Medicine (EBM) movement as a rigorous response to the limitations of narrative reviews. This methodology was subsequently adopted and adapted for toxicology under the umbrella of Evidence-Based Toxicology (EBT) [12]. A systematic review is defined by its transparent, methodologically rigorous, and reproducible process to summarize all available evidence on a precisely framed question [12].

The core distinction lies in its explicit, pre-defined protocol. Unlike a narrative review, a systematic review begins with a specific, focused research question, often structured using frameworks like PICO (Population, Intervention, Comparator, Outcome) [2]. It employs a comprehensive, documented search strategy across multiple databases to minimize selection bias. Studies are then screened against pre-specified inclusion/exclusion criteria [12]. A critical appraisal of the methodological quality of each included study is conducted using explicit tools. Finally, evidence is synthesized, either qualitatively or quantitatively (via meta-analysis), to arrive at an objective conclusion [12] [2].

Table 1: Comparative Analysis of Narrative and Systematic Reviews

Feature	Narrative (Traditional) Review	Systematic Review
Research Question	Broad, often informal or implicit [12].	Specific, focused, and explicitly stated (e.g., using PICO) [12] [2].
Search Strategy	Not typically specified; often non-comprehensive [12].	Comprehensive, documented search across multiple sources [12].
Study Selection	Implicit, based on author's judgment [12].	Explicit, pre-defined inclusion/exclusion criteria [12].
Quality Assessment	Usually absent or informal [12].	Critical appraisal using explicit, standardized tools [12].
Synthesis	Qualitative summary [12].	Qualitative and/or quantitative (meta-analysis) summary [12].
Transparency & Reproducibility	Low; process is not documented [12].	High; full methodology is reported [12].
Primary Application	Providing broad expert overview, identifying debates or gaps [2].	Informing regulatory, clinical, or policy decisions with high-level evidence [12] [2].

The systematic review workflow is a linear, staged process designed to eliminate bias at each step.

Systematic Review Workflow in Toxicology

Adopting this methodology in toxicology presents unique challenges. Toxicological evidence comes from multiple streams (in vivo, in vitro, in silico, epidemiological), which are difficult to integrate [12]. Reviews often involve multiple species, strains, and complex endpoints, and human data is frequently lacking, requiring extrapolation [12]. To address these, guidance documents have been developed by organizations like the National Toxicology Program's Office of Health Assessment and Translation (OHAT) and the Collaboration for Environmental Evidence (CEE), adapting the core systematic review principles to the specific needs of toxicology [12].

Table 2: Essential Research Reagents and Solutions for Systematic Reviews

Item / Tool Category	Function in Systematic Review	Examples / Notes
Protocol Registries	To pre-register the review plan, enhancing transparency and reducing reporting bias.	PROSPERO, Open Science Framework.
Bibliographic Software	To manage, de-duplicate, and screen large volumes of search results efficiently.	DistillerSR, Rayyan, Covidence, EndNote.
Critical Appraisal Tools	To assess the methodological quality and risk of bias in individual studies.	OHAT Risk of Bias Tool, SYRCLE's tool (animal studies), Cochrane RoB 2.0 (clinical trials).
Data Extraction Tools	To systematically collect quantitative and qualitative data from included studies.	Custom spreadsheets, SRDR+ (Systematic Review Data Repository).
Statistical Analysis Software	To perform meta-analysis and other quantitative syntheses of extracted data.	R (with metafor, meta packages), RevMan, Stata.
Reporting Guidelines	To ensure complete and transparent reporting of the review process and findings.	PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement.

Modern Systematic Approaches: Quantitative Systems Toxicology

The shift towards systematic methods extends beyond literature synthesis to primary research. Quantitative Systems Toxicology (QST) represents a paradigm shift from traditional, observational animal testing towards a predictive, mechanism-based, and human-relevant framework [13]. QST integrates computational (in silico) modeling with targeted in vitro experiments to quantitatively understand how perturbations at the molecular level lead to adverse outcomes at the organism level [13].

QST integrates several core modeling approaches. Quantitative Structure-Activity Relationship (QSAR) and ADMET modeling predict the toxicity and pharmacokinetic properties of chemicals based on their structural and physicochemical descriptors [13]. Network-based modeling uses graph theory to map interactions between genes, proteins, and metabolites, identifying key toxicity pathways and mechanisms [13]. Physiologically Based Pharmacokinetic/Toxicokinetic (PBPK/TK) modeling simulates the absorption, distribution, metabolism, and excretion (ADME) of a compound in a virtual human body, predicting target organ exposure [13].

This approach is applied to develop organ-specific toxicity platforms. For example, the Comprehensive in vitro Pro-Arrhythmia (CiPA) initiative aims to improve prediction of drug-induced cardiac arrhythmia by integrating in vitro ion channel assays with computational models of human cardiomyocyte electrophysiology [13]. Similarly, the DILI-sim Initiative focuses on modeling mechanisms of drug-induced liver injury to better predict human hepatotoxicity risk [13].

The evolution of toxicology from an expertise-driven art to a systematic science reflects its maturation. The historical reliance on narrative synthesis has been progressively supplanted by the rigorous, transparent methodology of systematic reviews for evidence integration and by the predictive, quantitative framework of systems toxicology for hypothesis testing and risk prediction. This dual shift towards systematicity in both evaluating existing evidence and generating new knowledge is fundamental to the future of toxicology. It enhances the reliability, reproducibility, and regulatory utility of toxicological assessments, ultimately enabling better protection of human and environmental health in the face of increasing chemical exposure. For the ecotoxicology researcher, embracing these systematic methods is no longer optional but essential for producing high-impact, decision-grade science.

Ecotoxicology, a discipline critical for environmental and chemical risk assessment, relies heavily on the synthesis of existing research to inform regulation and guide future studies [14]. Within this field, two primary methodologies for evidence synthesis exist: the systematic review and the narrative review. These approaches embody a fundamental trade-off between methodological rigor and adaptable scope, a tension central to advancing reliable science [12].

Systematic reviews are defined by a structured, pre-defined protocol aimed at minimizing bias. They offer maximal transparency and reproducibility, making them the cornerstone for evidence-based decision-making [12] [2]. In contrast, narrative reviews provide a broader, more flexible exploration of a topic, synthesizing knowledge through expert perspective and narrative analysis. They are invaluable for mapping developing fields, identifying theoretical gaps, and providing context [2]. This guide examines the key characteristics of both methodologies within ecotoxicological research, detailing their applications, experimental protocols, and the practical tools required for their execution.

Comparative Analysis: Systematic vs. Narrative Reviews

The choice between a systematic and narrative review shapes the entire research process, from the initial question to the final conclusion. The table below summarizes their fundamental distinctions [12] [2].

Table 1: Core Methodological Distinctions Between Narrative and Systematic Reviews in Ecotoxicology

Feature	Narrative (Traditional) Review	Systematic Review
Research Question	Broad, often exploratory; may cover multiple facets of a topic.	Focused and specific, typically formulated using PICO/PECO frameworks.
Protocol & Pre-registration	Rarely uses a pre-defined, publicly registered protocol.	Requires a detailed, pre-published protocol (e.g., on PROSPERO).
Search Strategy	Not always systematic or explicitly reported; may rely on known literature.	Comprehensive, documented search across multiple databases; strategy is reproducible.
Study Selection	Criteria often implicit and subjective; selection may be non-systematic.	Explicit, pre-defined inclusion/exclusion criteria applied by multiple reviewers.
Quality/Risk of Bias Assessment	Informal or absent; reliance on author's judgment.	Formal critical appraisal using validated tools (e.g., OHAT, SYRCLE).
Data Synthesis	Qualitative, narrative summary.	Structured synthesis (qualitative and/or quantitative meta-analysis).
Time & Resources	Generally requires less time and fewer resources [12].	Typically a long-term project (>1 year) requiring significant resources and a team [12].
Primary Application	Providing overview, context, theory development, and hypothesis generation.	Informing risk assessment, guideline derivation, and regulatory decision-making.

Foundational Research Designs in Ecotoxicology

Underpinning both review types is primary research, which employs various quantitative designs. Understanding these is crucial for appraising studies within a review.

Table 2: Common Quantitative Research Designs in Ecotoxicological Primary Studies

Research Design	Key Characteristics	Ecotoxicology Application Example	Strength for Evidence
Experimental	Random assignment to control and treatment groups; manipulation of independent variable (e.g., chemical concentration).	Laboratory toxicity test (e.g., OECD 203 fish acute toxicity test).	Gold standard for inferring causality; high internal validity [15].
Quasi-Experimental	Non-random group assignment (e.g., by basin, sex, or pre-existing condition); intervention is controlled.	Field study comparing impacted and non-impacted sites that are not randomly assigned.	Stronger than observational designs but prone to confounding variables.
(Causal) Comparative	Ex post facto; compares pre-existing groups to infer potential causes for observed differences.	Comparing biomarker levels in fish from a polluted estuary vs. a reference site.	Useful when experimentation is unethical or impossible; suggests association.
Correlational	Measures relationship between variables without manipulation or group comparison.	Analyzing the correlation between pesticide run-off concentration and invertebrate diversity indices.	Identifies relationships and predictions; cannot establish causation [16].
Descriptive	Observes and describes characteristics, patterns, or prevalence without inferring relationships.	Documenting the species composition and abundance in a habitat post-spill.	Provides essential baseline data and generates hypotheses.

Experimental Protocols for Reliability

The Systematic Review Process

The systematic review follows a strict, sequential protocol to ensure objectivity and reproducibility [12].

Diagram: The 10-Step Systematic Review Workflow [12]

Key Steps Explained:

• Protocol Development (Step 3): The foundation of the review. A detailed protocol defining the research question (using PECO: Population, Exposure, Comparator, Outcome), inclusion/exclusion criteria, search strategy, and analysis plan is registered on a platform like PROSPERO before the review begins [12].
• Systematic Search (Step 4): A comprehensive, documented search across multiple databases (e.g., PubMed, Scopus, Web of Science, TOXLINE) with tailored search strings. Grey literature and unpublished studies are sought to mitigate publication bias [12].
• Critical Appraisal (Step 6): Each included study is evaluated for risk of bias using domain-based tools (e.g., the Office of Health Assessment and Translation (OHAT) tool for animal studies). This assesses internal validity and influences the confidence in the overall evidence [12].
• Data Synthesis (Step 8): Findings are synthesized qualitatively (e.g., descriptive summary tables) and, if feasible, quantitatively via meta-analysis. This statistically combines results from multiple studies to provide an overall effect estimate [12].

Protocol for Primary Aquatic Toxicity Testing

A cornerstone of ecotoxicological evidence is the standardized laboratory test. Key protocol elements to ensure reliability and reproducibility include [17]:

• Test Organism Husbandry: Maintaining organisms under optimal, stable conditions (temperature, photoperiod, water quality) prior to and during testing is critical. Stress from poor husbandry can significantly alter sensitivity to toxicants [17].
• Modulating Factors Control: Factors like temperature, dissolved oxygen, pH, and light regime must be tightly controlled and reported. Even minor variations can modulate biological responses and compromise reproducibility between labs [17].
• Chemical Exposure Characterization: Accurate measurement and reporting of nominal vs. measured exposure concentrations, water chemistry (e.g., hardness, organic carbon), and renewal schedules are essential for interpreting dose-response relationships.
• Blinding & Randomization: Personnel measuring endpoints should be blinded to treatment groups to prevent observer bias. Random allocation of test organisms to exposure chambers is necessary to avoid systematic error.

The Scientist's Toolkit: Essential Research Reagents & Materials

Conducting primary studies or appraising them requires familiarity with key experimental materials.

Table 3: Key Research Reagent Solutions in Aquatic Ecotoxicology

Item Category	Specific Examples	Function & Importance	Considerations for Reliability [17]
Test Organisms	Daphnia magna (water flea), Danio rerio (zebrafish), Oncorhynchus mykiss (rainbow trout).	Model species representing different trophic levels with standardized test guidelines.	Genetic strain, life stage, health status, and acclimation history must be documented and controlled.
Reference Toxicants	Potassium dichromate, Sodium chloride, Copper sulfate.	Used in periodic tests to validate the sensitivity and health of the test organisms.	Regular use confirms laboratory consistency and organism responsiveness over time.
Water Preparation Systems	Reverse Osmosis (RO), Deionization (DI) systems, water quality test kits.	Produces standardized, contaminant-free dilution water for tests.	Consistent water quality (hardness, pH, conductivity) is vital for reproducible chemical bioavailability.
Exposure Apparatus	Glass or chemical-inert plastic beakers, flow-through diluter systems, semi-static renewal setups.	Houses organisms during the exposure period under controlled conditions.	Material must not adsorb the test chemical; system must maintain stable exposure concentrations.
Endpoint Measurement Tools	Dissolved oxygen/pH meters, spectrophotometers, microscopes, qPCR systems for molecular endpoints.	Quantifies apical (survival, growth, reproduction) and sub-organismal (gene expression, enzyme activity) effects.	Calibration, precision, and validation of analytical methods are required for accurate data.

The Narrative Review Process

While more fluid, a rigorous narrative review follows a logical structure to ensure comprehensive coverage and scholarly value [2].

Diagram: The Iterative Process of a Narrative Review

Key Process Elements:

• Iterative Exploration: The process begins with broad reading to define the scope and identify seminal papers, knowledge clusters, and controversies [2].
• Thematic Analysis: Instead of extracting quantitative data, the author identifies, compares, and synthesizes overarching themes, theoretical perspectives, and methodological trends across the literature.
• Argument-Based Structure: The review is organized to build a coherent narrative argument or perspective, rather than to answer a single focused question. It contextualizes findings within the historical and theoretical evolution of the field [2].
• Identification of Gaps: A primary output is the clear articulation of knowledge gaps, contradictory findings, and avenues for future primary research, thereby guiding the scientific community.

The choice between a systematic and narrative review is not one of superiority but of purpose. For ecotoxicologists addressing a specific, answerable question to directly inform risk assessment or regulation—where transparency, protocol, and reproducibility are paramount—the systematic review is the unequivocal standard [14] [12].

Conversely, for exploring the breadth of a complex field, synthesizing heterogeneous evidence types, developing novel hypotheses, or providing a comprehensive, flexible scope for an audience new to the topic, a well-executed narrative review is an indispensable scholarly tool [2]. The most robust ecotoxicological research programs strategically employ both methodologies: using narrative reviews to map the landscape and define critical questions, and systematic reviews to provide definitive, bias-minimized answers to those questions.

Within ecotoxicology and regulatory toxicology, the choice between a systematic review and a narrative review is fundamentally a choice about the primary objective of the evidence synthesis. A systematic review is a formal, protocol-driven methodology designed to minimize bias and support causal inference about a specific question, such as whether a chemical exposure causes an adverse outcome. In contrast, a narrative review offers a broad, expert-led overview of a field, synthesizing established knowledge, identifying trends, and highlighting controversies without aiming for formal causal conclusions [12]. This distinction is critical for research integrity, resource allocation, and the application of evidence in policy and drug development.

Core Methodological Divergence: A Comparative Framework

The fundamental differences between these review types stem from their opposing goals: precision and objectivity for causal inference versus breadth and expert synthesis for overview.

Table 1: Comparative Framework: Systematic Review vs. Narrative Review

Feature	Systematic Review (For Causal Inference)	Narrative Review (For Expert Overview)
Primary Goal	To answer a precise, causal question with minimal bias; to inform risk assessment or regulation [12] [3].	To provide a broad, integrative summary of a field; to educate, contextualize, and identify research gaps [12].
Research Question	Highly specific and structured (e.g., using PECO: Population, Exposure, Comparator, Outcome) [12].	Broad, flexible, and often not explicitly stated [12].
Protocol & A Priori Plan	Mandatory. A detailed, publicly registered protocol defines all methods before the review begins [12] [3].	Not required. The process is iterative and guided by the expert’s judgment.
Literature Search	Comprehensive, systematic, and reproducible across multiple databases. Search strategy is documented [12].	Selective and often not specified. May rely on the expert’s known literature and snowballing [12].
Study Selection & Screening	Explicit, pre-defined inclusion/exclusion criteria applied by multiple reviewers to minimize bias [12].	Implicit selection by the author based on relevance and perceived importance [12].
Critical Appraisal	Formal assessment of the risk of bias and quality of each included study using standardized tools [12] [3].	Informal, variable, or absent. Expert may comment on study strengths/weaknesses without systematic criteria [12].
Evidence Synthesis	Structured, often involving quantitative meta-analysis if data permits. Characterizes confidence in the body of evidence [12].	Qualitative, narrative summary. Weight given to studies is subjective and based on the author’s perspective [12].
Time & Resources	High (often >1 year). Requires expertise in systematic review methodology, literature search, and data analysis [12].	Lower to moderate. Primarily requires deep subject matter expertise [12].
Key Output	A causal conclusion (e.g., "The evidence suggests that exposure X is hazardous to organism Y") with a stated level of confidence.	An informed perspective on the state of the science, its history, key theories, and future directions.

Experimental Protocols for Evidence Synthesis

Systematic Review Protocol for Causal Inference

The following protocol, synthesized from toxicology-specific guidance, details the steps required to produce a review capable of supporting causal inference [12] [3].

Table 2: Systematic Review Protocol for Causal Inference in Ecotoxicology

Step	Primary Objective	Detailed Methodology & Key Tools
1. Problem Formulation & Protocol	To define the causal question precisely and plan the entire review to prevent bias.	Develop a PECO statement (Population, Exposure, Comparator, Outcome). Register the protocol (e.g., on PROSPERO or an institutional registry). Define all subsequent steps in detail [12] [3].
2. Systematic Search	To identify all potentially relevant evidence in an unbiased, reproducible way.	Search multiple databases (e.g., PubMed, Web of Science, TOXLINE). Use controlled vocabulary and free-text terms. Document the full search strategy. Search for grey literature [12].
3. Study Screening & Selection	To apply eligibility criteria objectively, minimizing selection bias.	Use dual-independent screening (title/abstract, then full-text) against pre-defined PECO criteria. Resolve conflicts via consensus or a third reviewer. Document reasons for exclusion [12].
4. Data Extraction	To accurately capture study details and results relevant to the causal question.	Use a standardized, pilot-tested extraction form. Extract study design, sample characteristics, exposure/outcome details, effect estimates, and key covariates. Dual-independent extraction is recommended [12].
5. Risk of Bias / Study Quality Assessment	To evaluate the internal validity of each study and its potential to yield a biased estimate of effect.	Use domain-based tools tailored to toxicology study designs (e.g., OHAT, SYRCLE's RoB tools). Assess domains like randomization, blinding, exposure characterization, and outcome assessment [12] [3].
6. Evidence Integration & Synthesis	To statistically and qualitatively combine evidence to estimate a causal effect.	For quantitative synthesis: perform meta-analysis using appropriate models (fixed/random effects), assess heterogeneity (I² statistic). For qualitative synthesis: use structured summary tables and narrative to explain patterns [12].
7. Rate Confidence in Body of Evidence	To grade the overall strength and certainty of the causal conclusion.	Apply frameworks like GRADE or OHAT to rate confidence based on risk of bias, consistency, directness, precision, and other factors [12].
8. Interpretation & Reporting	To transparently communicate findings, limitations, and implications for decision-making.	Report according to PRISMA guidelines. State the causal conclusion clearly, link it to the confidence rating, and discuss for risk assessment or regulatory application [12] [3].

The narrative review process is less prescriptive and more iterative, leveraging the author’s expertise to map and synthesize a complex field [12] [18].

Table 3: Common Workflow for a Narrative (Expert) Review

Stage	Process Description	Key Considerations for Ecotoxicology
Topic Scoping & Conceptual Mapping	The expert defines the breadth and depth of the review based on the intended audience and purpose.	Decide whether to cover a chemical class, a toxicological endpoint (e.g., endocrine disruption), or a methodological area. Identify core themes and historical context [18].
Exploratory & Iterative Literature Engagement	Searching is non-linear, combining known foundational papers, database searches, and following citation trails.	Depth is prioritized over exhaustive breadth. The expert’s knowledge is crucial for identifying seminal studies and influential but less-cited work [12].
Critical Synthesis & Narrative Development	Information is organized thematically rather than by study. The expert weighs evidence, reconciles contradictions, and develops a coherent story.	Synthesis is interpretative. The author provides perspective on why certain findings are more credible, places controversies in context, and highlights paradigm shifts [12] [19].
Identification of Gaps & Future Directions	Based on the synthesized landscape, the expert identifies unresolved questions and promising new research avenues.	This is a key value of the narrative review, leveraging the expert’s insight to guide the research community toward fruitful areas of inquiry [12].

Visualizing Methodological Pathways

Diagram 1: Systematic Review Process for Causal Inference

Diagram 2: Iterative Workflow of a Narrative Expert Review

The Scientist's Toolkit: Essential Research Reagent Solutions

Table 4: Key Tools and Resources for Conducting Reviews in Ecotoxicology

Tool / Resource Category	Specific Item or Framework	Primary Function in Research
Protocol & Reporting Guidelines	PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [12]; PROSPERO (Protocol Registry)	Ensures transparent, complete reporting of systematic reviews. Provides a platform to register an a priori protocol to reduce bias.
Question Formulation Tools	PECO Framework (Population, Exposure, Comparator, Outcome) [12]	Structures a precise, answerable research question for systematic reviews in toxicology.
Risk of Bias / Quality Assessment	OHAT Risk of Bias Tool [12]; SYRCLE's RoB tool for animal studies; GRADE framework for rating confidence [12]	Provides standardized, domain-based criteria to evaluate methodological quality and internal validity of individual studies and the overall evidence.
Evidence Integration Software	RevMan (Cochrane Collaboration); R packages (metafor, robvis)	Facilitates statistical meta-analysis, creates forest plots, and visualizes risk of bias assessments.
Causal Inference Frameworks	Potential Outcomes (Rubin Causal Model) [20] [21]; Directed Acyclic Graphs (DAGs) [20] [22]	Provides the theoretical and graphical foundation for defining, identifying, and estimating causal effects from observational and experimental data.
Specialized Methodologies	Marginal Structural Models [23]; Mendelian Randomization [23]; Propensity Score Methods [20]	Advanced statistical techniques to control for time-varying confounding, leverage genetic data for causal inference, and balance covariates in observational studies to emulate randomized trials.

The Systematic Review Blueprint: A Step-by-Step Guide for Ecotoxicology

In ecotoxicology and broader toxicological research, the shift towards evidence-based practices has fundamentally changed how knowledge is synthesized and applied in regulatory and decision-making contexts [24]. This movement emphasizes transparency, objectivity, and reproducibility, moving beyond traditional narrative reviews, which are often expert-driven but can lack methodological rigor and are susceptible to selective citation and unreproducible conclusions [24]. The systematic review has emerged as the core evidence-based tool, providing a structured, predefined, and transparent method for summarizing all available evidence on a precisely framed question [24].

The critical first step in this rigorous process is the development of a detailed protocol, the cornerstone of which is a well-structured research question. A clearly framed question defines the review's structure, objectives, and methodology, guiding everything from literature search strategies to inclusion criteria and data synthesis [25] [26]. In ecotoxicology, where research often investigates the adverse effects of chemical exposures on organisms and ecosystems, the PECO framework (Population, Exposure, Comparator, Outcome) has been widely adapted from the clinical PICO framework (Population, Intervention, Comparator, Outcome) to better suit the field's needs [25] [24]. This initial step of protocol and question development is what differentiates a systematic, unbiased synthesis from a narrative summary, setting the stage for a review that can reliably inform risk assessment and public health policy [25] [24].

Conceptual Foundations: PICO, PECO, and the Nature of Review Types

The choice of framework and review methodology is dictated by the research objective. Understanding the distinctions is vital for selecting an appropriate approach.

• PICO vs. PECO: The PICO framework is designed for questions about the effects of an intentional intervention, such as a drug or therapy [26] [27]. In contrast, PECO is tailored for questions about the effects of an exposure, which is often unintentional or environmental, such as to a chemical pollutant [25]. The "Exposure" component in PECO presents specific challenges, requiring careful consideration of dose, duration, timing, and route of exposure, which are central to toxicological risk assessment [25] [24].

• Systematic Review vs. Narrative Review: These represent two fundamentally different approaches to evidence synthesis, as summarized in the table below.

Table 1: Key Differences Between Narrative and Systematic Reviews in Toxicology [24]

Feature	Narrative (Traditional) Review	Systematic Review
Question Formulation	Broad, often not explicitly stated.	Focused, precise, and framed using a structured framework (e.g., PECO).
Protocol & Methods	Rarely pre-specified; methods often implicit.	A detailed, publicly registered protocol is mandatory, specifying all methods in advance.
Search Strategy	Not systematic; potential for selective searching.	Comprehensive, reproducible search across multiple databases to capture all evidence.
Study Selection	Criteria not defined; selection may be subjective.	Explicit, pre-defined inclusion/exclusion criteria applied consistently.
Risk of Bias Assessment	Not routinely performed.	Critical appraisal of individual study validity is a standard step.
Evidence Synthesis	Qualitative summary; may be influenced by author perspective.	Structured synthesis (narrative, tabular, meta-analysis); aims to minimize bias.
Reproducibility	Low.	High, due to explicit reporting of all methods.
Primary Utility	Providing a broad expert overview; generating hypotheses.	Answering a specific question to support decision-making and policy.

A Step-by-Step Guide to Protocol Development and PECO/PICO Formulation

Developing a systematic review protocol is a multi-stage process that begins with a scoping phase and culminates in a formal registration.

Scoping and Preliminary Searches

Before formally framing the question, conduct scoping searches on key databases using simple terms [26] [27]. This helps gauge the volume and nature of existing literature, identifies key papers, and informs whether the question needs to be broadened or narrowed [26].

Defining the PECO/PICO Components with Precision

Each element must be defined with operational detail to guide the review.

• Population (P): The organisms or systems under study. In ecotoxicology, this requires precise specification of species, strain, life stage, sex, and health status (e.g., "Juvenile fathead minnows (Pimephales promelas), laboratory-raised") [25] [24].
• Exposure (E) / Intervention (I): For PECO, define the stressor: the specific chemical (including its form), its dose/concentration, the route of exposure (e.g., waterborne, dietary), duration, and timing [25]. For PICO, define the therapeutic or management intervention.
• Comparator (C): The baseline against which the exposure or intervention is compared. This can be an unexposed control group, a group exposed to a different level of the same stressor, or an alternative intervention [25] [26]. Defining the comparator is particularly challenging in exposure studies [25].
• Outcome (O): The measured endpoints of interest. These should be clinically or ecologically relevant and measurable. Specify primary and secondary outcomes (e.g., Primary: mortality at 96 hours; Secondary: expression of cytochrome P450 1A mRNA) [26] [27].

Operationalizing the PECO: Five Common Scenarios

A key advancement in applying PECO is recognizing that the framework can be operationalized differently depending on the review's context and phase of investigation [25]. The table below outlines five paradigmatic scenarios.

Table 2: Framework for Formulating PECO Questions in Exposure Research [25]

Systematic Review Context	Approach to Exposure & Comparator	Example PECO Question
1. Explore a dose-effect relationship	Explore the shape of the relationship across the observed range of exposure.	In freshwater Daphnia magna, what is the effect of a 1 mg/L incremental increase in waterborne nickel concentration on 48-hour mortality?
2. Compare exposure extremes within studied populations	Use cut-offs (e.g., tertiles, quintiles) defined by the distribution in the identified studies.	In brown trout (Salmo trutta), what is the effect of exposure to the highest quartile of sediment PCB concentration compared to the lowest quartile on reproductive success?
3. Apply a known external exposure standard	Use a cut-off value derived from legislation, guidelines, or other populations.	Among amphibians in agricultural landscapes, what is the effect of ambient atrazine concentrations ≥ 10 μg/L (EPA benchmark) compared to concentrations < 1 μg/L on larval developmental malformations?
4. Identify a protective exposure threshold	Use an existing health-based cut-off to define the comparator.	In soil nematodes (Caenorhabditis elegans), what is the effect of chronic exposure to cadmium concentrations below the known EC₅₀ for reproduction compared to concentrations above it on population growth rate?
5. Evaluate an intervention to reduce exposure	Select comparators based on achievable exposure levels via an intervention.	In honey bee colonies, what is the effect of implementing a pesticide mitigation buffer zone (reducing exposure by an estimated 70%) compared to standard agricultural practice on colony survival?

Documenting the Full Protocol

The framed PECO/PICO question anchors a comprehensive protocol that must detail:

• Background and rationale for the review.
• Explicit inclusion/exclusion criteria derived directly from the PECO/PICO components.
• Detailed search strategy, including databases, search terms, and filters.
• Data extraction and management plans.
• Methodology for risk of bias/study quality assessment.
• Data synthesis methods (e.g., meta-analysis, narrative synthesis).
• A realistic timeframe [26] [27].

Protocol Registration

Before beginning the formal search, register the protocol in a public repository such as PROSPERO or the Open Science Framework [26] [27]. This enhances transparency, reduces duplication of effort, and guards against outcome reporting bias by locking in the planned methods.

Visualizing the Process

The following diagrams illustrate the systematic review protocol development workflow and the logical relationship between PICO and PECO frameworks.

Systematic Review Protocol Development Workflow

Relationship Between PICO and PECO Frameworks

Table 3: Key Research Reagent Solutions for Systematic Review Protocol Development

Tool / Resource Name	Type	Primary Function in Protocol Development
PROSPERO	Protocol Registry	International database for registering systematic review protocols in health-related fields; essential for checking for duplicate reviews and ensuring transparency [26] [27].
Cochrane Handbook	Methodological Guidance	Foundational reference detailing rigorous methods for systematic reviews of interventions; provides the basis for adapting methods to toxicology [25] [24].
Navigation Guide / OHAT Handbook	Field-Specific Guidance	Provides methodology adapted for environmental health and toxicology, offering direct guidance on PECO formulation and evidence integration from animal and human studies [25] [24].
RevMan (Review Manager)	Software	Cochrane's tool for preparing and maintaining systematic reviews; facilitates protocol writing, data extraction, risk of bias tables, and meta-analysis [26] [27].
DistillerSR, EPPI-Reviewer	AI-Assisted Screening Software	Streamlines the title/abstract and full-text screening process defined in the protocol, using machine learning to prioritize records and manage workflow [28].
PRISMA-P Checklist	Reporting Guideline	Checklist of essential items to include in a systematic review or meta-analysis protocol; ensures completeness and adherence to reporting standards.

Application in Ecotoxicology: From Protocol to Practice

Ecotoxicology presents unique challenges that directly influence how the PECO framework is applied during protocol development [24]:

• Multiple Evidence Streams: Protocols must plan for integrating data from diverse study types (e.g., field studies, controlled lab experiments, in vitro assays), each with different strengths and validity concerns [24].
• Species and Model Selection: Defining the "Population" requires justification for the choice of test species (e.g., standard lab models vs. wild species) and consideration of ecological relevance [24].
• Complex Exposures: The "Exposure" may need to account for chemical mixtures, transformation products, or non-chemical stressors, moving beyond single-chemical protocols [24].
• Outcome Harmonization: A wide array of measured endpoints (from molecular to population-level) must be categorized into meaningful health outcomes for synthesis, which should be pre-specified in the protocol [24].

Example: A protocol for a systematic review on neonicotinoid insecticides and aquatic invertebrate health might use PECO as follows: P (Populations: freshwater aquatic invertebrate species), E (Exposure: waterborne concentrations of imidacloprid, clothianidin, or thiamethoxam), C (Comparator: unexposed control or exposure at a no-observed-effect concentration (NOEC)), O (Outcome: primary - immobilization/mortality; secondary - growth, reproduction, behavioral change). The protocol would pre-define the search strategy across environmental science databases, set inclusion criteria for exposure duration and study design, and select a relevant risk of bias tool for ecotoxicological studies.

Future Directions: Living Protocols and Automation

The field is evolving towards living systematic reviews, where the question and protocol are periodically updated as new evidence emerges [28]. This paradigm necessitates "living protocols" that are designed from the outset for iterative updates. Furthermore, the integration of automation and artificial intelligence is beginning to assist in protocol development stages, such as using text mining during scoping to refine PECO components and employing machine learning tools to streamline the screening process mandated by the protocol [28]. These innovations promise to enhance the efficiency and sustainability of the rigorous systematic review process in ecotoxicology.

In ecotoxicology research, the synthesis of existing evidence is fundamental for hazard identification, risk assessment, and policy formulation. The approach to this synthesis—be it a systematic review or a narrative review—dictates the rigor, transparency, and applicability of the findings [2]. This guide details the comprehensive search strategies and literature management techniques essential for both methodologies, framed within the broader context of evidence synthesis in environmental science. Systematic reviews, characterized by explicit, pre-specified protocols, aim to minimize bias and provide reliable evidence to support regulatory decisions [2] [29]. Conversely, narrative reviews offer a flexible, exploratory synthesis of literature, invaluable for mapping broad or emerging fields, such as the environmental fate of novel pollutants or innovative bioremediation techniques [30] [31]. Understanding and implementing the distinct search and management strategies for each type is a critical skill for researchers, scientists, and drug development professionals engaged in environmental health and toxicology.

Foundational Distinctions: Systematic vs. Narrative Review Protocols

The choice between a systematic and narrative review is dictated by the research objective, which in turn defines the entire methodological approach, from search strategy to reporting [2].

Table 1: Core Methodological Distinctions Between Systematic and Narrative Reviews in Ecotoxicology

Aspect	Systematic Review	Narrative (Traditional) Review
Primary Objective	To answer a specific, focused research question using all available evidence.	To provide a broad overview, explore concepts, or summarize existing debates on a topic [2].
Research Question	Highly structured, often using PICO/PECO frameworks.	Broad and flexible, may address multiple related questions [2].
Protocol & Eligibility	Mandatory, pre-registered protocol with strict inclusion/exclusion criteria [32].	No formal protocol; scope evolves based on literature discovered [30].
Search Strategy	Comprehensive, reproducible search across multiple databases; explicitly documented [32].	Selective search; may not be exhaustive or fully reproducible [30].
Study Selection & Appraisal	Formal screening by multiple reviewers; critical risk-of-bias assessment [32].	Selective, often single-reviewer; informal appraisal of study quality [30].
Data Synthesis	Structured qualitative synthesis or quantitative meta-analysis [32].	Narrative, descriptive summary and critique.
Key Applications in Ecotoxicology	Informing chemical risk evaluations (e.g., TSCA) [29], dose-response analysis, guideline derivation.	Exploring emerging contaminants (e.g., TFA) [30], reviewing technological advances (e.g., bioremediation) [31].

As exemplified in recent literature, a narrative review on trifluoroacetic acid (TFA) prioritized recent, high-quality evidence from select databases to summarize its environmental fate and toxicology [30], while a systematic review protocol for TSCA risk evaluations mandates exhaustive searches and formal bias assessments to guide regulatory action [29].

Architecting a Comprehensive Search Strategy

A robust search strategy is the cornerstone of a reliable review. Its complexity and formality differ markedly between review types.

3.1 Formulating the Research Question For systematic reviews, a structured framework is essential. The PICO framework (Population, Intervention, Comparator, Outcome) is most common and can be adapted for ecotoxicology as PECO (Population, Exposure, Comparator, Outcome) [32]. For example:

• P (Population): Freshwater fish species.
• E (Exposure): Chronic exposure to trifluoroacetic acid (TFA).
• C (Comparator): Unexposed control populations.
• O (Outcome): Measured endpoints of hepatic toxicity [30].

This precision informs every subsequent step, including search terms and eligibility criteria [32]. For narrative reviews, the question is broader, such as exploring the role of nanotechnology in the bioremediation of hydrocarbon-polluted soils [31].

3.2 Database Selection and Search Syntax A systematic review requires searching multiple bibliographic databases to capture all relevant literature. Key databases for ecotoxicology include:

• PubMed/MEDLINE: For biomedical and toxicological literature.
• Embase: Strong coverage of pharmacology and environmental chemistry.
• Web of Science and Scopus: For multidisciplinary science and citation tracking.
• Specialized databases (e.g., TOXLINE, GreenFile).

Table 2: Core Databases for Ecotoxicology Literature Reviews [32]

Database	Primary Focus & Coverage	Utility in Ecotoxicology
PubMed/MEDLINE	Life sciences and biomedicine; uses MeSH terms.	Core resource for toxicological, epidemiological, and pathological studies.
Embase	Biomedical pharmacology, toxicology, environmental health.	Excellent for chemical and drug-related ecotoxicity studies.
Scopus	Multidisciplinary scientific abstracts and citations.	Broad coverage for interdisciplinary topics like bioremediation [31].
Web of Science	Multidisciplinary sciences with strong citation network.	Useful for identifying seminal papers and tracking research evolution.

Search syntax should combine keywords and controlled vocabulary (e.g., MeSH in PubMed). Boolean operators (AND, OR, NOT) are used to link concepts. An example search string for TFA in aquatic species might be: ("trifluoroacetic acid" OR TFA) AND (aquatic OR fish OR algae) AND (toxic* OR ecotoxic*). For systematic reviews, this search string is documented verbatim in the protocol and methods [29].

Narrative reviews typically employ less exhaustive searches, often within one or two primary databases, using flexible keyword combinations to explore the literature landscape [30].

3.3 Supplementary Search Techniques To mitigate publication bias, systematic reviews must incorporate supplementary searches [32]:

• Grey Literature: Searching regulatory dossiers (e.g., EPA reports) [29], thesis repositories, and conference proceedings.
• Citation Tracking: Reviewing reference lists of included studies (backward chaining) and papers that cite them (forward chaining).
• Hand-searching key journals in the field.

Diagram 1: Systematic Search Strategy Workflow

Managing the Literature Base: From Screening to Synthesis

Managing the volume of literature retrieved is a critical, multi-stage process that is highly formalized in systematic reviews.

4.1 The Screening Process The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram is the standard for documenting this process [9]. It involves:

• Identification: Records from all searches are pooled, and duplicates are removed using reference managers like Zotero or EndNote [33] [34].
• Screening: Titles and abstracts are screened against eligibility criteria by at least two independent reviewers to minimize bias. Tools like Rayyan or Covidence facilitate blinded screening and conflict resolution [34].
• Eligibility: The full text of potentially relevant studies is assessed for final inclusion [32].

4.2 Data Extraction and Critical Appraisal For included studies, data is systematically extracted into pre-designed forms (e.g., study design, population, exposure, outcomes, results). Concurrently, each study's risk of bias is assessed using standardized tools (e.g., Cochrane Risk of Bias Tool for experimental studies) [32]. This appraisal informs the interpretation and strength of the synthesized evidence. Narrative reviews conduct data extraction and appraisal more informally, focusing on conceptual insights rather than standardized metrics [30].

4.3 Synthesis and Reporting Systematic reviews synthesize findings either narratively (descriptively) or quantitatively via meta-analysis, which statistically combines results from multiple studies to produce an overall effect size [32]. Software like RevMan is specifically designed for preparing Cochrane reviews and conducting meta-analysis [35] [34]. Narrative reviews synthesize information thematically to construct a coherent story or argument about the state of the field [31].

Diagram 2: Literature Screening & Synthesis Funnel

Experimental Protocols in Cited Ecotoxicology Reviews

5.1 Protocol for a Narrative Review: The Case of Trifluoroacetic Acid (TFA) A 2025 narrative review on TFA provides a clear example of a less structured, yet rigorous, methodological approach [30].

• Objective: To critically summarize current knowledge on TFA's properties, exposure pathways, toxicology, and regulation.
• Search Strategy: A literature search was conducted over two months (Dec 2024-Jan 2025) in PubMed and Scopus. Keywords included "trifluoroacetic acid," "ecotoxicology," "human exposure," and "regulation," combined with Boolean operators.
• Study Selection: Inclusion was based on relevance, prioritizing recent studies and those with reliable quantitative data. No formal screening protocol or bias assessment was used.
• Data Management: Sources were managed using Zotero for citation organization [30].
• Synthesis: Findings were synthesized narratively under thematic headings (e.g., environmental fate, ecotoxicological data) to provide an integrated overview.

5.2 Protocol for a Systematic Review: The EPA TSCA Framework The U.S. EPA's draft protocol for systematic reviews under TSCA represents a high-stakes, regulatory-grade methodology [29].

• Objective: To establish a robust, transparent process for selecting and evaluating scientific studies for chemical risk evaluations.
• Protocol: A pre-registered, publicly peer-reviewed protocol defines the entire process.
• Search Strategy: Mandates comprehensive searches across multiple databases, plus grey literature from regulatory agencies. Search strategies must be fully documented.
• Study Selection & Appraisal: Involves dual independent review at all stages. Each study is critically appraised for risk of bias and relevance using established tools.
• Data Synthesis: Evidence is integrated, with a preference for quantitative meta-analysis where possible, to characterize hazard and dose-response [29].

The Scientist's Toolkit: Software and Reagents

Effective literature review and ecotoxicological research rely on specialized digital tools and physical reagents.

Table 3: Digital Tools for Literature Review Management [36] [34]

Tool Name	Category	Primary Function	Best For
Covidence	Systematic Review Manager	Streamlines title/abstract screening, full-text review, risk-of-bias assessment, and data extraction.	Teams conducting high-quality systematic reviews [34].
Rayyan	Systematic Review Manager	Free web-based tool for collaborative screening and selection of studies.	Researchers starting with systematic reviews or working with limited budgets [34].
DistillerSR	Systematic Review Manager	Web-based platform for screening, data extraction, and reporting for complex reviews [2].	Large-scale or regulatory-compliant systematic reviews.
RevMan	Meta-analysis Software	Software for preparing and analyzing Cochrane Reviews; performs meta-analysis [35] [34].	Authors conducting meta-analyses, particularly for Cochrane.
Elicit	AI Research Assistant	Uses language models to find relevant papers, extract data, and summarize findings based on a research question [36].	Early-stage exploration and rapid literature synthesis.
Semantic Scholar	AI-Powered Search Engine	Provides AI-generated summaries of papers and visualizes citation networks [36].	Discovering key papers and understanding research landscapes.

Table 4: Key Research Reagents and Materials in Featured Ecotoxicology Reviews

Reagent/Material	Function in Ecotoxicology Research	Example from Literature
Trifluoroacetic Acid (TFA)	A persistent, water-soluble pollutant studied for its chronic toxicity to aquatic and terrestrial organisms [30].	Used as a target analyte to investigate environmental fate, exposure pathways, and chronic ecotoxicological effects [30].
Hydrocarbon Mixtures (e.g., Crude Oil)	Complex pollutant used to study soil contamination, toxicity to biota, and efficacy of bioremediation strategies [31].	Used in experiments to test the effectiveness of bioaugmentation, biostimulation, and nano-enhanced remediation techniques [31].
Biosurfactants & Molecular Probes	Used to enhance bioavailability of hydrophobic contaminants for microbial degradation or to detect specific genetic activity in bioremediating organisms [31].	Employed in biostimulation experiments to increase microbial degradation rates of petroleum hydrocarbons in soil [31].

The choice between a systematic and narrative review in ecotoxicology dictates a cascade of methodological decisions. A systematic review demands a comprehensive, pre-specified search across multiple sources, rigorous dual-reviewer screening, formal bias assessment, and structured synthesis to answer a focused question—a process essential for regulatory risk assessment [29]. A narrative review employs a more flexible, targeted search and a narrative synthesis to explore broad concepts, map emerging fields, or summarize debates, as demonstrated in reviews of TFA and hydrocarbon bioremediation [30] [31]. Mastery of the respective search strategies and literature management tools—from Boolean syntax and PRISMA flowcharts to platforms like Covidence and Rayyan—empowers researchers to produce syntheses that are fit for purpose, whether that purpose is setting a safe drinking water limit or charting the future of green remediation technology.

This guide details the third step of a systematic review in ecotoxicology: the critical appraisal of individual studies. This process involves a dual assessment of a study's Risk of Bias (internal validity) and its Study Sensitivity (ability to detect a true effect). Rigorous critical appraisal is a foundational element that differentiates a systematic review from a traditional narrative review, transforming a simple literature summary into a transparent, reproducible, and defensible evidence synthesis [12]. Within the broader thesis comparing systematic and narrative reviews, this step is a key differentiator; narrative reviews typically lack a formal, explicit appraisal process, which can obscure the reliability of their conclusions [12] [2].

Core Concepts: Bias, Sensitivity, and Validity

Risk of Bias refers to the potential for systematic error in a study's design, conduct, or analysis that would lead to a consistent deviation from the true effect [37]. Assessing risk of bias is an evaluation of a study's internal validity—the credibility of the link it establishes between an exposure and an outcome [38].

Study Sensitivity is a complementary concept, describing a study's ability to detect a true effect if one exists. An insensitive study, due to factors like poor analytical methods, low statistical power, or inappropriate exposure concentrations, may fail to show a real difference, leading to a false-negative conclusion [39]. It is analogous to the sensitivity of a diagnostic assay.

The relationship between systematic error (bias), random error (precision), and overall validity is illustrated in Figure 1 [37]. It is critical to distinguish these concepts from other quality constructs such as external validity (applicability), reporting completeness, or ethical compliance [37].

Precision vs. Bias in Study Results

Table 1: Distinguishing Systematic Reviews from Narrative Reviews

Feature	Narrative Review	Systematic Review
Research Question	Broad, often not explicitly specified [12] [2].	Specific and focused, using PECO/PICO structure [12] [2].
Search Strategy	Usually not specified or comprehensive [12] [2].	Comprehensive, explicit, and reproducible search across multiple sources [12].
Study Selection	Implicit, based on author judgment [12].	Explicit, pre-defined inclusion/exclusion criteria [12] [2].
Critical Appraisal	Usually absent or informal [12] [2].	Mandatory, structured assessment of Risk of Bias and Sensitivity [39] [37].
Data Synthesis	Qualitative summary [12].	Structured qualitative synthesis; quantitative meta-analysis where possible [12] [2].
Time & Resources	Generally lower [12].	High (often >1 year), requiring specific expertise [12].
Output	Expert opinion, prone to selective citation [12].	Transparent, reproducible evidence synthesis to guide decision-making [12] [2].

Methodologies for Critical Appraisal

The FEAT Principles for Risk of Bias Assessment

A robust risk of bias assessment should adhere to the FEAT principles: be Focused on internal validity, Extensive in covering all key bias domains, Applied to influence the synthesis and conclusions, and Transparent in reporting [37] [40]. The workflow in Figure 2 provides a framework for implementing these principles.

Risk of Bias Assessment Workflow

Protocol for Applying the ROBINS-E Tool

The Risk Of Bias In Non-randomized Studies - of Exposure (ROBINS-E) tool is a state-of-the-art instrument for observational environmental studies [41]. Its application protocol involves:

• Specify the Causal Effect: Clearly define the exposure, outcome, and estimated effect for the study result being assessed [41].
• Answer Signalling Questions: For each of the seven bias domains, answer detailed questions about the study's methods. Domains include bias due to confounding, participant selection, exposure classification, departures from intended exposures, missing data, outcome measurement, and selection of the reported result [41].
• Make Domain Judgements: For each domain, judge the risk of bias (low, some concerns, high, or very high) and the predicted direction of bias (towards or away from the null) [41].
• Reach an Overall Judgement: Synthesize domain-level judgements to assign an overall risk of bias and direction for the study result [41].

Protocol for Species Sensitivity Distribution (SSD) Modeling

SSD modeling is a key ecotoxicological method for deriving protective hazard concentrations (e.g., HC5). A standardized protocol includes:

• Data Compilation: Curate toxicity data (e.g., LC50, NOEC) from a wide range of species and taxonomic groups, often using databases like U.S. EPA ECOTOX [42].
• Data Preparation: Log-transform toxicity values. Integrate both acute and chronic endpoints, applying assessment factors if necessary to normalize data [42].
• Model Fitting: Fit a statistical distribution (e.g., log-logistic, log-normal) to the ordered toxicity data. Advanced approaches may use specialized models for chemical classes like agrochemicals [42] [43].
• HC5 Estimation: Calculate the hazardous concentration for 5% of species (HC5) from the fitted distribution as a key benchmark [42] [43].
• Validation & Application: Validate models with hold-out data. Apply to untested chemicals using quantitative structure-activity relationship (QSAR) properties for prediction [42]. This workflow is shown in Figure 3.

Species Sensitivity Distribution (SSD) Workflow

Comparative Analysis of Tools and Models

Table 2: Key Risk of Bias Tools for Ecotoxicology

Tool Name	Primary Scope	Core Domains/Features	Output	Source
ROBINS-E	Non-randomized studies of exposure (environmental/occupational) [41].	7 domains: Confounding, selection, exposure, missing data, measurement, reporting. Uses signaling questions.	Risk of bias judgement (Low/High) & direction of bias for each domain and overall [41].	[41]
NTP Risk of Bias Tool	Human & animal studies for chemical hazard assessment [38].	Focuses on internal validity. Parallel structure for human and animal evidence streams [38].	Risk of bias rating. Designed to be tailored to specific review questions [38].	[38]
FEAT Framework	General framework for any quantitative environmental systematic review [37] [40].	Not a tool itself, but provides FEAT (Focused, Extensive, Applied, Transparent) principles to guide tool use/development [37].	Ensures the assessment process is fit-for-purpose and robustly conducted [37].	[37] [40]

Table 3: Species Sensitivity Distribution (SSD) Models in Ecotoxicology

Model / Approach	Description	Data Input	Key Output	Application & Notes
Global SSD Model	A generalized model built from a large, curated dataset spanning multiple taxonomic groups [42].	Acute (LC/EC50) and chronic (NOEC/LOEC) toxicity data for many species (~3250 entries) [42].	Predicted HC5 values for untested chemicals; identification of toxicity-driving substructures [42].	Used for initial screening and prioritization of large chemical inventories (e.g., EPA CDR) [42].
Class-Specific SSD Model	Specialized models tailored for high-priority chemical classes (e.g., pesticides, personal care products) [42].	Toxicity data specific to the chemical class and relevant taxonomic groups [42].	More accurate, class-relevant HC5 values; supports targeted risk mitigation [42].	Addresses regulatory needs where generic models may be inadequate [42].
Trait-Based Sensitivity Prediction	Approach using mechanistic traits (toxicokinetic, toxicodynamic, ecological) to predict sensitivity [43].	Species trait data and/or molecular information (e.g., target receptor ortholog sequences) [43].	Insights into drivers of sensitivity; predictions for data-poor species [43].	Aims to move beyond statistical extrapolation to mechanistically informed assessment [43].

The Scientist's Toolkit: Research Reagent Solutions

Table 4: Essential Tools and Resources for Critical Appraisal in Ecotoxicology

Item	Function in Critical Appraisal	Key Features / Notes
ROBINS-E Tool Template	Provides a structured worksheet to consistently assess risk of bias in observational exposure studies [41].	Available as a Word or Excel template; includes signaling questions for seven bias domains [41].
ECOTOX Knowledgebase	Authoritative source for curated single-chemical toxicity data for aquatic and terrestrial organisms [42].	Essential for compiling data for SSD modeling and evaluating the comprehensiveness of a study's tested species [42].
SSD Modeling Software (e.g., OpenTox SSDM)	Platforms for fitting statistical distributions to toxicity data and calculating hazard concentrations (HC5) [42].	Facilitates reproducible SSD development; some platforms offer open-access tools and model sharing [42].
Chemical Prioritization Database (e.g., EPA CDR)	Inventory of industrial chemicals used to apply predictive models for screening and prioritization [42].	Allows researchers to identify high-priority chemicals for further testing or assessment based on predicted hazard [42].
Systematic Review Management Software	Platforms to manage screening, data extraction, and appraisal processes for complex reviews.	Enhances transparency, reproducibility, and collaboration among review team members.

In conclusion, the critical appraisal step—rigorously assessing both risk of bias and study sensitivity—is what allows a systematic review to evaluate the internal validity and detective capability of the evidence base. This structured, transparent process stands in direct contrast to the informal, often opaque evaluation typical of narrative reviews. By applying standardized tools like ROBINS-E and quantitative methods like SSD modeling within frameworks like FEAT, ecotoxicologists can produce syntheses that reliably inform environmental risk assessment and regulation.

Within the broader thesis examining systematic versus narrative reviews in ecotoxicology, the phase of evidence integration and synthesis represents the analytical core where data transforms into insight. Traditional narrative reviews, while valuable for exploratory discussion, often employ implicit, non-reproducible methods for weighting and combining evidence, increasing the risk of selective interpretation and biased conclusions [24]. In contrast, a systematic review mandates an explicit, objective, and protocol-driven synthesis. This phase directly addresses a key weakness of narrative approaches by systematically minimizing bias during the integration of findings from multiple, often heterogeneous, primary studies [24].

In ecotoxicology, synthesis is particularly complex due to multiple evidence streams (e.g., in vivo, in vitro, in silico, and epidemiological data), diverse model organisms, varied exposure regimes, and a multitude of measured endpoints [24]. The synthesis step must therefore employ rigorous, tailored methodologies to integrate this diversity into a coherent and reliable evidence base suitable for informing robust scientific conclusions, chemical risk assessments, and environmental policy [3]. This guide details the advanced technical approaches for both qualitative (narrative) and quantitative (meta-analytic) synthesis, providing researchers with a critical toolkit for conducting high-quality, transparent evidence integration.

Foundational Framework: From Systematic Review to Synthesis

Evidence synthesis is not a standalone activity but the culmination of a meticulous, multi-stage process. The integrity of the synthesis is entirely dependent on the rigor applied in prior steps [44].

• Protocol-Driven Foundation: A pre-registered, detailed protocol defines the synthesis plan a priori, preventing subjective, data-driven analytical choices and ensuring transparency [44].
• Comprehensive Search & Screening: A replicable search across multiple databases and grey literature sources, followed by dual-independent screening against PICOS/PECO criteria, ensures the identified evidence base is comprehensive and relevant [45].
• Critical Appraisal: Before synthesis, each included study must be assessed for risk of bias (e.g., using tools like SYRCLE's RoB for animal studies or ROBINS-I for non-randomized studies). This assessment informs the interpretation of findings and can be used to stratify or weight studies during synthesis [44].
• Data Extraction: Structured extraction of quantitative, qualitative, and descriptive data into pre-defined forms provides the raw material for synthesis. Consistency, verified by dual independent extraction, is paramount [3].

The following workflow diagram illustrates the pivotal position of synthesis within the systematic review process, highlighting its dependence on prior stages and its role in generating final, graded conclusions.

Systematic Review Workflow with Synthesis as Core

Quantitative Synthesis (Meta-Analysis)

Meta-analysis is a statistical technique for combining quantitative results from independent studies to produce a single, pooled estimate of effect (e.g., standardized mean difference, odds ratio) with greater precision and generalizability [44].

Technical Protocol for Meta-Analysis in Ecotoxicology

• Effect Size Calculation: For each study, calculate a comparable effect size (ES) and its variance.

  • Continuous Data (e.g., enzyme activity, growth): Use Hedges' g (standardized mean difference), which includes a correction for small sample bias. Appropriate for comparing mean endpoint values (e.g., body weight, gene expression fold-change) between exposed and control groups [44].
  • Binary Data (e.g., mortality, incidence): Calculate log odds ratios (OR) or risk ratios (RR). Commonly used for apical endpoints like survival or morphological deformity incidence [24].

• Model Selection:

  • Fixed-Effect Model: Assumes all studies estimate a single true ES, and variability is due only to sampling error. Use only if heterogeneity is negligible (I² < 25%).
  • Random-Effects Model: Assumes the true ES varies between studies (due to differing species, exposure conditions, etc.) and estimates the mean of this distribution. This is the default and more conservative choice for most ecotoxicological meta-analyses given inherent biological and methodological diversity [44] [24].

• Heterogeneity Quantification: Statistically assess variability between studies.

  • Cochran's Q-test: Tests the null hypothesis of homogeneity. A significant p-value (<0.10) indicates significant heterogeneity.
  • I² Statistic: Describes the percentage of total variation across studies that is due to heterogeneity rather than chance. I² values of 25%, 50%, and 75% are interpreted as low, moderate, and high heterogeneity, respectively [44].

• Subgroup Analysis & Meta-Regression: To explore sources of heterogeneity, pre-specified subgroups can be analyzed separately (e.g., freshwater vs. marine species, acute vs. chronic exposure). Meta-regression uses study-level covariates (e.g., exposure concentration, organism life stage) as moderators to explain variance in effect sizes [24].

• Sensitivity Analysis & Bias Assessment: Test the robustness of results by repeating the analysis excluding high-risk-of-bias studies or using different statistical models. Assess publication bias visually using funnel plots and statistically using Egger's regression test [44].

Data Presentation for Meta-Analysis

Quantitative data extracted for meta-analysis must be structured, complete, and traceable. The following table outlines essential data categories.

Table 2: Quantitative Data Extraction Framework for Ecotoxicological Meta-Analysis

Data Category	Specific Variables	Description & Purpose
Study Identification	Author, Year, DOI	For referencing and grouping.
Population (P)	Test organism (species, strain, life stage), Habitat (freshwater, marine, terrestrial)	Defines the biological system. Essential for subgroup analysis [44].
Exposure (I/E)	Stressor/Chemical (name, purity), Concentration (nominal, measured), Duration, Route (waterborne, dietary)	Defines the intervention/exposure. Critical for dose-response assessment and grouping [3].
Comparator (C)	Control group description (vehicle, sham), Control group mean & SD (or SE)	Baseline for effect size calculation.
Outcome (O)	Endpoint category (mortality, growth, reproduction, biochemical), Specific measurement (e.g., LOEC, EC₅₀, mean enzyme activity), Units	The measured effect. Must be consistent across studies for pooling [24].
Effect Size Data	Treatment group mean, SD (or SE, CI), Sample size (N)	Required raw data for calculating standardized effect sizes (Hedges' g) [44].
Study Design (S)	Laboratory/Field, Test type (acute/chronic), Replication, Compliance (e.g., OECD guideline #)	Informs risk of bias and suitability for synthesis.

Qualitative Synthesis (Narrative Synthesis)

When quantitative pooling is inappropriate due to conceptual heterogeneity, incompatible outcome measures, or a preponderance of non-quantitative data, a structured narrative synthesis is conducted. This is a systematic, descriptive approach that organizes, explores, and interprets study findings without statistical pooling [44] [46].

Technical Protocol for Narrative Synthesis

• Developing a Preliminary Synthesis: Systematically tabulate and summarize the results of included studies. This goes beyond simple description to organize findings by key dimensions (e.g., by type of stressor, organismal response level, or outcome direction) [46].
• Exploring Relationships in the Data: Analyze patterns across studies. Identify factors that may explain differences in findings, such as study quality, organism characteristics, or exposure context. This step is analogous to investigating heterogeneity in meta-analysis [44].
• Assessing the Robustness of the Synthesis: Critically evaluate the strength of the evidence contributing to the narrative conclusions. This involves explicitly considering the volume, consistency, and risk of bias of the underlying studies [3]. A formal assessment using the GRADE framework for grading the certainty of evidence can be adapted for ecotoxicology [44].
• Presenting the Synthesis: Structure the narrative to tell a clear, evidence-based story. Use visual aids like concept maps to illustrate hypothesised pathways or summary tables to compare study characteristics and findings [46].

Visualizing Mechanistic Evidence: A Signaling Pathway Example

Qualitative synthesis in molecular ecotoxicology often involves integrating evidence on disrupted biological pathways. Visualizing these pathways clarifies complex mechanistic narratives. The following diagram, informed by studies such as those on toxicant-induced signaling disruptions [47], models a generalized stress response pathway.

Generalized Stress Response Signaling Pathway

The Scientist's Toolkit: Essential Reagents for Synthesis

Table 3: Research Reagent Solutions for Evidence Synthesis

Tool/Resource Category	Specific Item	Function in Synthesis
Methodological Guidance	Cochrane Handbook [44], CEE Guidelines [45]	Provides gold-standard protocols for designing and executing systematic reviews and meta-analyses.
Reporting Guidelines	PRISMA 2020 Statement & Checklist [45]	Ensures transparent and complete reporting of the review, including synthesis methods and results.
Statistical Software	R (metafor, meta packages), RevMan, Comprehensive Meta-Analysis	Performs complex meta-analytic calculations, generates forest and funnel plots, and conducts heterogeneity and bias analyses [44].
Qualitative Analysis Software	NVivo, ATLAS.ti, Delve	Aids in managing, coding, and theorizing from qualitative data extracted from studies during narrative synthesis [46].
Risk of Bias Tools	SYRCLE's RoB (animal studies), ROBINS-I (non-randomized studies)	Standardized tools to critically appraise internal study validity, a critical input for interpreting synthesis results [44].
Evidence Grading Framework	GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) [44]	Provides a systematic approach to rate the overall certainty (high, moderate, low, very low) of the synthesized evidence body.
Color Palette Tools	ColorBrewer [48], Viz Palette [48]	Assists in selecting accessible, perceptually uniform color schemes for data visualizations (forest plots, concept maps), adhering to best practices for color contrast and colorblind readability [49] [50].

Application in Ecotoxicology: Synthesis of Current Research

The approaches outlined above are actively applied to address pressing questions in ecotoxicology. For instance, a systematic review on the effects of short-chain per- and polyfluoroalkyl substances (PFAS) on Daphnia magna could employ meta-analysis to pool effect sizes for reproductive endpoints across multiple studies, while using narrative synthesis to integrate findings on less standardized molecular or behavioral responses [47]. Similarly, a review of microplastics toxicity might use meta-regression to explore how particle size and polymer type moderate effects on growth across aquatic species, while qualitatively synthesizing evidence on microbiome alterations [47].

Within the continuum of review methodologies, the structured synthesis phase is what decisively differentiates a systematic review from a narrative one. It replaces expert opinion and selective citation with explicit, auditable, and often quantitative methods for integrating evidence [24]. For ecotoxicology researchers and risk assessors, mastering both qualitative and quantitative synthesis techniques is essential for building a transparent, reproducible, and robust evidence base. This foundation is critical for advancing scientific understanding and informing sound environmental and public health decisions in the face of global contamination challenges [3].

The synthesis of ecotoxicological evidence is fundamentally shaped by the methodological choice between narrative and systematic reviews. This choice directly influences how the field addresses its core challenges: integrating multiple, disparate evidence streams and mitigating pervasive exposure misclassification. Traditional narrative reviews, while valuable for exploring broad concepts and tracking scientific development, typically employ an implicit, non-standardized process for literature identification, selection, and synthesis [2]. This lack of transparency increases the risk of bias, where the selection and weighting of studies may be influenced by the author's preconceptions, making it difficult to reproduce conclusions or assess their validity [12].

In contrast, systematic reviews provide a transparent and methodologically rigorous alternative. They are defined by a pre-specified protocol that includes a clearly formulated research question, a comprehensive and reproducible literature search, explicit criteria for study selection, and a critical appraisal of individual study quality [51] [2]. The primary goal is to minimize reviewer bias and error, thereby producing a more objective and reliable evidence synthesis [12]. This approach is particularly critical in ecotoxicology, where evidence is derived from diverse streams—including epidemiology, wildlife observations, in vivo laboratory studies, in vitro assays, and in silico models—and is frequently compounded by uncertainties in exposure assessment [51] [52]. The systematic review framework offers a structured process to evaluate and integrate these varied data types transparently, a necessity for robust environmental risk assessment (ERA) and regulatory decision-making [51] [53].

Table 1: Core Methodological Differences Between Narrative and Systematic Reviews in Ecotoxicology [12] [2].

Feature	Narrative (Traditional) Review	Systematic Review
Research Question	Broad, often informal or implicitly defined.	Focused, specific, and explicitly defined (e.g., using PICO elements).
Search Strategy	Often not specified; may not be comprehensive.	Comprehensive, reproducible, and documented across multiple databases.
Study Selection	Implicit criteria; selection process not transparent.	Explicit, pre-defined inclusion/exclusion criteria applied consistently.
Quality Assessment	Rarely formal or systematic; based on expert judgment.	Critical appraisal using explicit, validated tools to assess risk of bias.
Data Synthesis	Qualitative summary, often selective.	Structured synthesis (qualitative and/or quantitative, e.g., meta-analysis).
Conclusions	Based on expert interpretation of a possibly non-representative sample.	Based on transparently evaluated and synthesized evidence.
Time & Resources	Typically lower (months).	Typically high (often >1 year).

Foundational Challenges in Ecotoxicological Evidence Synthesis

Ecotoxicology faces inherent complexities that complicate evidence synthesis, irrespective of the review methodology. Two of the most significant are the need to integrate multiple lines of evidence and the frequent misclassification of exposure.

The Imperative for Integrating Multiple Evidence Streams

Ecotoxicological risk assessment is rarely based on a single study type. Determining whether a chemical is an endocrine disruptor, for example, requires evidence of an adverse effect, evidence of endocrine activity, and a plausible link between the two [51]. This necessitates the synthesis of data from mechanistic in vitro studies, controlled in vivo laboratory tests (across multiple species), field-based wildlife observations, and epidemiological studies [51] [53]. Each stream has distinct strengths and weaknesses regarding biological relevance, control of confounding variables, and direct applicability to ecosystem health. Narrative reviews often struggle to weigh these streams objectively, while systematic review frameworks like SYRINA (Systematic Review and Integrated Assessment) are specifically designed to evaluate each stream individually before integrating them into an overall conclusion [51].

The Pervasive Problem of Exposure Misclassification

Accurate exposure assessment is a notorious challenge in environmental science. In wildlife and field studies, true exposure levels are difficult to measure directly and are often estimated with significant error [54]. Laboratory studies, while controlling exposure, frequently use concentrations and durations not representative of environmental conditions, creating an extrapolation problem [52] [54]. Misclassification arises from spatial/temporal sampling errors, the use of imperfect biomarkers, and analytical limitations [54]. This error is not random; it often biases exposure-response relationships toward the null, leading to false-negative conclusions or underestimations of risk [54]. Addressing this requires methodologies that explicitly account for measurement uncertainty.

Methodological Approaches for Evidence Integration

Systematic review methodologies provide structured frameworks to tackle the challenge of multiple evidence streams. A prominent example is the SYRINA framework for endocrine disrupting chemicals (EDCs), which tailors the systematic review process to the specific needs of ecotoxicology [51].

The SYRINA Framework Protocol

The SYRINA framework outlines a seven-step protocol for the systematic review and integrated assessment of EDC studies [51]:

• Formulate the Problem: Define the specific chemical and adverse outcome of interest.
• Develop the Review Protocol: Pre-specify the methods for search, selection, appraisal, and synthesis.
• Identify Relevant Evidence: Execute a comprehensive, documented literature search.
• Evaluate Evidence from Individual Studies: Critically appraise the internal validity (risk of bias) of each included study.
• Summarize and Evaluate Each Stream of Evidence: Synthesize findings and assess strength within mechanistic, in vivo animal, and epidemiological streams separately.
• Integrate Evidence Across All Streams: Use a weight-of-evidence approach to combine stream-specific conclusions.
• Draw Conclusions and Evaluate Uncertainties: State the overall level of evidence and highlight key data gaps [51].

The final integration (Step 6) is critical. It moves beyond listing findings from different streams to evaluating their coherence, consistency, and biological plausibility. For instance, strong evidence of a specific molecular initiating event in vitro, consistent adverse apical outcomes in laboratory fish tests, and observed population-level effects in contaminated field sites would constitute a coherent and compelling body of evidence.

Quantitative Evidence Integration: The Plausibility Database Approach

Innovative quantitative methods are emerging to standardize integration. One approach involves creating a "plausibility database" that translates conclusions from systematic reviews into standardized Levels of Evidence (LoE) for each chemical-outcome pair within different evidence streams (e.g., toxicological, epidemiological) [55]. These stream-specific LoEs are then combined using pre-defined algorithms to generate an overall LoE (e.g., "unlikely," "likely," "very likely") [55]. This transforms qualitative expert judgment into a transparent, reproducible scoring system, facilitating the identification of high-priority hazards and critical research gaps.

Evidence Integration Workflow in Systematic Review

Advanced Methodologies for Addressing Exposure Misclassification

Addressing exposure misclassification requires specialized statistical and modeling approaches that explicitly account for measurement error.

Bayesian Network (BN) Framework for Error-Aware Analysis

Bayesian Networks provide a powerful tool for quantifying the impact of measurement error on exposure-response inferences. A BN is a probabilistic graphical model that represents a set of variables and their conditional dependencies. In the context of exposure assessment, key nodes include True Exposure (TE), Measured Exposure (ME), Accuracy of Exposure Measurement (AcEM), True Response (TR), and Measured Response (MR) [54].

The model incorporates prior knowledge or assumptions about the accuracy of measurements (AcEM, AcRM) and the potential relationship between TE and TR. It then uses Bayes' theorem to update the probabilities of the true states (e.g., "strong relationship," "no relationship") based on the observed, error-prone data (ME, MR) [54]. This allows researchers to:

• Quantify how measurement error biases the estimated strength of an exposure-response relationship.
• Determine the sample sizes required to detect a true effect given a known level of measurement imprecision.
• Evaluate the potential value of improving measurement accuracy in future study designs [54].

Protocol for Implementing a Bayesian Network Analysis

The following protocol outlines the key steps for applying a BN to an exposure-response problem [54]:

• Define Network Structure (Directed Acyclic Graph): Identify all relevant variables (TE, ME, TR, MR, AcEM, AcRM, confounding factors) and specify their causal relationships (e.g., TE influences TR; AcEM influences the difference between TE and ME).
• Parameterize the Model: Define the conditional probability tables for each node. This requires expert judgment or prior data to estimate probabilities (e.g., the likelihood that a "High" true exposure is classified as "Medium" in a measurement with "Low" accuracy).
• Incorporate Observational Data: Input the collected, error-prone data from epidemiological or toxicological studies (ME, MR values) into the corresponding nodes as "evidence."
• Perform Belief Updating: Use inference algorithms (e.g., clustering, importance sampling) to update the probabilities of the unobserved variables (TE, TR, strength of relationship) based on the entered evidence.
• Interpret and Validate: Analyze the posterior probabilities. Conduct sensitivity analyses to see how conclusions change with different prior assumptions or model structures.

Bayesian Network for Exposure-Response with Measurement Error

Application to Complex Scenarios: Chemical Mixtures

Real-world ecotoxicology involves exposure to complex chemical mixtures, which amplifies the challenges of evidence integration and exposure assessment.

Integrating Evidence for Mixture Risk Assessment

The risk assessment of mixtures requires integrating evidence on individual chemicals and their potential interactions. Systematic review principles remain paramount. Key steps include [53]:

• Problem Formulation: Defining the mixture of concern (whole mixture or specific component-based group).
• Evidence Stream Identification: Gathering data on individual component toxicity, mechanisms of action (MoA), and mixture toxicity from in vitro, in vivo, and monitoring studies.
• Evaluation of Interactions: Critically assessing studies for evidence of synergism, antagonism, or additivity, using predefined criteria for study quality [53].

Protocol for Assessing Mixture Toxicity Studies

A robust protocol for evaluating studies on chemical mixture interactions should assess [53]:

• Dose-Relevance: Were the tested concentrations environmentally relevant, or excessively high, potentially triggering non-specific toxic effects?
• Statistical Power: Was the study adequately powered to detect the type of interaction being claimed?
• Model Choice: Was the appropriate additivity model (dose addition for similar MoA, response addition for dissimilar MoA) used as the baseline for identifying deviation (interaction)?
• Confounder Control: In non-experimental studies, were potential confounding factors adequately considered?
• Replicability: Are the experimental conditions and results described with sufficient detail to permit replication?

Table 2: Core Evidence Streams in Ecotoxicology and Their Role in Integrated Assessment [51] [55] [53].

Evidence Stream	Typical Study Designs	Key Strengths	Primary Limitations	Role in Integrated Assessment
Epidemiological (Field Observations)	Cohort, case-control, cross-sectional studies in wildlife or human populations.	High ecological/real-world relevance; identifies direct associations in exposed populations.	Confounding factors; difficult to establish causation; exposure misclassification is common.	Provides critical "anchor" for real-world relevance. Consistent findings increase overall plausibility.
In Vivo Toxicology	Controlled laboratory tests on model organisms (fish, invertebrates, amphibians, etc.).	Establishes causation; controls confounders; allows detailed dose-response analysis.	Extrapolation from lab to field; interspecies differences; often uses high, non-environmental doses.	Provides strongest evidence of causal relationships and quantitative dose-response data.
In Vitro & Mechanistic	Cell-based assays, sub-organismal studies (e.g., receptor binding, -omics).	Elucidates molecular mechanisms; high-throughput; reduces animal use.	Uncertain extrapolation to whole organism; may lack metabolic context.	Establishes biological plausibility and mode of action; supports causal interpretation.
In Silico (QSAR, Modeling)	Quantitative structure-activity relationship models, PBPK models, population models.	Predictive; can fill data gaps; explores scenarios.	Dependent on quality of input data and model validation.	Used for prioritization, screening, and extrapolation under defined uncertainties.

The Scientist's Toolkit: Research Reagent and Material Solutions

Table 3: Key Research Reagent Solutions for Advanced Ecotoxicology Testing [52].

Reagent/Material	Function in Ecotoxicology	Application Example	Consideration for Evidence Synthesis
Passive Sampling Devices (e.g., SPMDs, POCIS)	Integrative sampling of bioavailable fractions of hydrophobic or polar contaminants in water.	Measuring time-weighted average exposure concentrations for comparison with laboratory toxicity thresholds.	Reduces exposure misclassification by providing a more relevant measure of bioavailable contaminant fraction than grab water samples. Data from such methods should be prioritized in systematic reviews.
Fluorescent Vital Dyes (e.g., CFDA-AM, PI)	Staining to discriminate live/dead cells or indicate metabolic activity in in vitro or small-organism assays.	High-throughput cytotoxicity screening in fish cell lines (e.g., RTgill-W1).	Enables rapid, mechanistic toxicity data generation. Systematic reviews should note the specific endpoint (e.g., membrane integrity vs. enzyme activity) when synthesizing in vitro evidence.
Species-Specific Enzyme-Linked Immunosorbent Assay (ELISA) Kits	Quantification of biomarker proteins (e.g., vitellogenin, stress proteins) in hemolymph or tissue homogenates.	Detecting endocrine disruption (vitellogenin induction in male fish) or general stress response.	Provides molecular-level evidence bridging mechanism and apical effect. Kit validation and cross-reactivity should be assessed during study quality appraisal in a review.
Standard Reference Sediments & Waters	Control matrices with certified properties for testing poorly-soluble substances or providing consistent background in tests.	Determining the toxicity of hydrophobic chemicals in standardized sediment tests with benthic invertebrates.	Essential for reliable and reproducible laboratory testing. Reviews should check if studies used appropriate controls to isolate the effect of the contaminant of interest.
Cryopreserved Primary Cell Cultures	Source of metabolically competent cells from relevant target species/organs for in vitro testing.	Using primary hepatocytes from fish to study metabolic activation of pro-toxins.	Provides more physiologically relevant mechanistic data than immortalized cell lines. This increased relevance should be considered when weighting in vitro evidence.
Environmental DNA (eDNA) Extraction & Sequencing Kits	Assessing genetic diversity, population structure, or microbial community composition as an ecotoxicological endpoint.	Monitoring impacts of chemical exposure on genetic variation in a population, a key parameter for ecological resilience [52].	Generates complex data on a highly ecologically relevant endpoint. Systematic reviews need clear protocols for synthesizing this emerging type of genomic evidence.

Navigating Pitfalls and Elevating Quality in Both Review Types

The field of ecotoxicology is dedicated to understanding the impacts of chemical, physical, and biological agents on living organisms within ecosystems, with profound implications for environmental regulation, public health, and drug development. Within this complex and high-stakes domain, the synthesis of existing research through literature reviews is a cornerstone of scientific progress and policy formulation. However, not all review methodologies are created equal. This whitepaper examines the critical methodological divide between narrative reviews and systematic reviews, framing the discussion within the context of ecotoxicology research. The central thesis posits that while narrative reviews offer exploratory utility, their inherent susceptibility to selective citation and lack of transparency systematically undermines the reliability of their conclusions, thereby advocating for the systematic review as the superior standard for robust, actionable evidence synthesis.

The distinction is foundational. A narrative review traditionally provides a broad, qualitative summary of literature on a topic, guided by the author's expertise and perspective without a formalized, pre-specified protocol [2]. In contrast, a systematic review employs an explicit, transparent, and reproducible methodology to identify, appraise, and synthesize all relevant studies on a clearly formulated question, minimizing bias through every stage of the process [2]. This methodological rigor is paramount in ecotoxicology, where research findings directly influence risk assessments, regulatory thresholds for environmental contaminants, and the safety profiling of pharmaceuticals.

Methodological Comparison: Systematic vs. Narrative Review Protocols

The fundamental differences between narrative and systematic reviews are embedded in their respective protocols. These protocols dictate the review's objectivity, reproducibility, and ultimately, its scientific credibility.

The Systematic Review Protocol: A Framework for Transparency

The systematic review process is a linear, pre-defined workflow designed to eliminate arbitrariness. Key environmental health frameworks like the COSTER (Conduct of Systematic Reviews in Toxicology and Environmental Health Research) recommendations provide domain-specific guidance, covering 70 practices across eight performance domains, including protocol registration and management of grey literature [56].

The following diagram illustrates the standard workflow for conducting a systematic review, highlighting its staged and accountable structure.

Table 1: Foundational Characteristics of Narrative and Systematic Reviews

Characteristic	Narrative (Traditional) Review	Systematic Review
Primary Objective	Provide a broad overview or theoretical exploration of a topic [2].	Answer a specific, focused research question using all available evidence [2].
Research Question	Often broad or multiple questions; can evolve during writing.	Narrow, focused, and defined a priori using frameworks (e.g., PICO) [2].
Protocol & Methods	No mandatory protocol; methods are often not pre-specified or reported in detail [2].	Mandatory, publicly registered protocol detailing every methodological step [56] [2].
Literature Search	Not systematically comprehensive; selection may be subjective or based on author familiarity.	Exhaustive, structured search across multiple databases; search strategy is reported [2].
Study Selection	Criteria not explicit; high risk of selection bias based on author perspective.	Explicit, pre-defined eligibility criteria applied consistently (often by multiple reviewers) [2].
Critical Appraisal	Variable, rarely systematic; quality of included studies may not be formally assessed.	Mandatory systematic assessment of risk of bias/quality for each included study [2].
Data Synthesis	Typically qualitative and narrative; may group studies thematically.	Structured synthesis (qualitative or quantitative/meta-analysis); explores heterogeneity [2].
Reporting & Transparency	No standard reporting guideline; methodology often opaque.	Follows standards like PRISMA; high transparency required at all stages [2] [57].
Reproducibility	Low. The process cannot be reliably repeated to achieve the same results.	High. The explicit methods allow for the review to be updated or replicated [57].

The Narrative Review Approach: Flexibility at the Cost of Structure

In stark contrast, the narrative review lacks a formalized workflow. Its design is largely contingent on the author's objectives and preferences, often following a traditional IMRAD (Introduction, Methods, Results, and Discussion) structure but without mandated methodological rigor [2]. The literature search is typically non-systematic, and the selection, appraisal, and synthesis of evidence are not governed by pre-specified, objective rules. This flexibility, while useful for generating hypotheses or exploring nascent fields, creates a fertile ground for the introduction of bias, most notably through selective citation and opaque methodology.

The informal methodology of narrative reviews directly enables their two most critical flaws: selective citation and a fundamental lack of transparency. These flaws are not isolated but interact to create a cascade of bias that distorts the scientific record.

Selective citation refers to the non-objective inclusion or emphasis of references that support a preferred hypothesis while ignoring equally valid contradictory evidence. Data indicates this is a pervasive issue; one audit found approximately 20% of citations in a manuscript contained errors, with selective citation being a prominent category [58]. The consequences propagate through the literature, turning inaccuracies into conventional knowledge and misinforming critical decisions in toxicology and drug development [59].

Table 2: Common Citation Errors and Their Prevalence in Scientific Literature

Error Type	Description	Example from Ecotoxicology Context	Proposed Remediation [58]
Selective Citation	Citing one's own work, colleagues', or smaller supportive studies while ignoring contradictory or more extensive evidence [58].	A review on chemical X's safety only cites industry-funded studies with negative results and omits independent studies showing toxicity.	Conduct systematic literature searches using objective database searches.
Incorrect Source Type	Citing a review (secondary source) for a factual claim instead of the original primary research paper [58].	Claiming "compound Y was first shown to be genotoxic in 1995" but citing a 2010 review instead of the 1995 primary paper.	Always cite the primary source. If inaccessible, use "as cited in" notation.
Unjustified Extrapolation	Misrepresenting or over-extending the conclusions of the cited work to fit the author's argument [58].	A study showing toxicity of chemical Z in zebrafish embryos is cited as evidence of its reproductive toxicity in mammals.	Carefully read the full text to understand the precise study scope and conclusions.
Factual Error	Incorrectly describing the findings, mechanisms, or numerical data (e.g., effect size, prevalence) from the cited work [58].	Misreporting the NOAEL (No Observable Adverse Effect Level) from a key toxicology study, altering the risk assessment.	Download and verify claims against the full text of the primary source.
Reliance on Retracted Work	Citing a paper that has been formally retracted, thereby propagating invalidated information [59].	A review on drug safety cites a retracted paper claiming a lack of adverse effects, lending false credibility to the claim.	Verify the publication status of key references prior to submission.

The diagram below models how a single instance of selective citation or misquotation can initiate a cascade of error through subsequent narrative reviews, leading to the entrenchment of biased or false information.

Lack of Transparency: From Methodological Opacity to Irreproducibility

Lack of transparency is the second core flaw. It encompasses the failure to disclose the methods used to gather, select, appraise, and synthesize evidence. In a meta-review of meta-analyses, major issues were found concerning completely reproducible search procedures and the practical absence of analysis script code sharing [57]. While this study focused on meta-analyses, the problem is far more acute in narrative reviews, which operate without the reproducibility safeguards required of systematic reviews.

This opacity manifests in several ways:

• Unreported search strategies: Readers cannot assess the comprehensiveness of the literature covered.
• Unstated selection criteria: The rationale for including or excluding specific studies is hidden, raising concerns about selection bias.
• Undisclosed conflicts of interest: Author affiliations or funding sources that may influence the review's perspective are not declared [60].

The consequence is irreproducibility. A lack of transparency violates a fundamental tenet of the scientific method, preventing other researchers from verifying findings, updating the review with new evidence, or applying alternative analytical techniques to the same dataset [57].

Mitigation Strategies: Tools and Protocols for Robust Reviews

Addressing the flaws inherent in narrative reviews requires adopting the disciplined practices of systematic methodology. The following experimental protocols and tools provide a pathway for researchers in ecotoxicology and related fields to produce more credible, transparent, and useful evidence syntheses.

Experimental Protocol for a Systematic Review in Ecotoxicology

The following protocol is aligned with the COSTER recommendations for environmental health [56] and the PRISMA reporting guideline.

1. A Priori Protocol Development & Registration:

• Action: Prior to beginning the search, develop a detailed protocol specifying the research question (using PECO: Population, Exposure, Comparator, Outcome), eligibility criteria, search strategy, data items, and synthesis plan.
• Quality Control: Register the protocol on a platform like PROSPERO or the Open Science Framework (OSF). This prevents arbitrary changes during the review process and guards against outcome reporting bias [56].

2. Comprehensive, Documented Search:

• Action: Design search strings with a librarian or information specialist. Search at least two major databases (e.g., PubMed, Scopus, Web of Science) and specialist databases (e.g., TOXLINE). Include grey literature sources (e.g., regulatory reports, dissertations) as appropriate [56].
• Quality Control: Document the full search strategy for each database, including dates searched and limits applied. Use reference management software (EndNote, Zotero) to deduplicate records.

3. Blinded, Duplicate Study Screening & Selection:

• Action: Use screening software (e.g., Rayyan, DistillerSR) to manage the process. At least two reviewers independently screen titles/abstracts and then full texts against the eligibility criteria.
• Quality Control: Calculate inter-rater agreement (e.g., Cohen's kappa). Resolve conflicts through discussion or a third reviewer. Document reasons for exclusion at the full-text stage.

4. Systematic Data Extraction & Critical Appraisal:

• Action: Pilot a standardized data extraction form. Extract study characteristics, exposure/outcome details, and key results. In parallel, perform a critical appraisal of each study's risk of bias using a domain-based tool (e.g., RoB 2 for randomized trials, OHAT or SYRINA for environmental health studies).
• Quality Control: Perform data extraction and appraisal in duplicate. Ensure blinding of reviewers to study authors and journals where possible.

5. Transparent Data Synthesis & Reporting:

• Action: Synthesize extracted data. For quantitative synthesis (meta-analysis), pre-specify the statistical model, heterogeneity measures (I²), and sensitivity analyses. For qualitative synthesis, use a structured summary (e.g., GRADE evidence tables).
• Quality Control: Adhere to PRISMA 2020 guidelines for reporting. Publicly share the extracted data, analysis code, and materials on a repository like OSF or Figshare to enable reproducibility and reanalysis [57].

The Scientist's Toolkit: Research Reagent Solutions for Rigorous Reviews

Table 3: Essential Tools for Conducting Transparent, High-Quality Evidence Syntheses

Tool / Resource	Category	Primary Function	Key Benefit for Mitigating Flaws
PROSPERO Registry	Protocol Registry	Prospective registration of systematic review protocols.	Eliminates selective reporting; ensures methodology is fixed a priori [56].
PRISMA 2020 Statement	Reporting Guideline	Checklist and flow diagram for transparent reporting of systematic reviews.	Enforces complete transparency in methods and findings [2].
Covidence, Rayyan, DistillerSR	Screening Software	Platforms for managing blinded, duplicate study screening and selection.	Reduces human error and bias in the study selection process [2].
EndNote, Zotero, Mendeley	Reference Management	Software to collect, organize, annotate, and cite references.	Prevents technical citation errors; aids in organizing large searches [58] [59].
ROBIS Tool, OHAT Framework	Risk of Bias Assessment	Structured tools to evaluate methodological quality of included studies.	Provides systematic, not selective, appraisal of evidence credibility [56].
R, Python (metafor, meta)	Statistical Analysis	Programming languages/packages for advanced meta-analysis and plotting.	Enforces reproducible, script-based analysis that can be shared publicly [57].
Open Science Framework (OSF)	Data Repository	Free platform to preregister protocols, share data, code, and materials.	Facilitates full transparency, reproducibility, and collaborative updating [57].
COSTER Guidelines	Methodological Standard	Recommendations for conducting systematic reviews in toxicology/env. health [56].	Provides field-specific standards, addressing challenges like grey literature and exposure assessment.

The comparison between narrative and systematic reviews in ecotoxicology is not merely academic; it is a question of scientific integrity and societal impact. While narrative reviews can offer valuable perspective and hypothesis generation, their structural vulnerabilities to selective citation and lack of transparency render them unsuitable as a foundation for evidence-based decision-making in risk assessment, regulation, or drug development. The systematic review, with its explicit, pre-registered, and reproducible protocol, stands as the corrective to these flaws, actively minimizing bias and maximizing transparency.

The future of evidence synthesis in ecotoxicology lies in the widespread adoption and further strengthening of systematic methodologies. This includes:

• Mandatory protocol registration for all systematic reviews intended to inform policy.
• Integration of machine learning tools to assist in screening and data extraction, while maintaining human oversight.
• Development of standardized risk-of-bias tools for complex, non-randomized ecotoxicological studies.
• A cultural shift among researchers, journals, and funders to value methodological rigor and transparency as highly as novel findings.

By embracing these principles, the scientific community can ensure that its synthetic work provides a reliable, unbiased map of the evidence landscape—a map that is crucial for navigating the complex environmental health challenges of the future.

Within the evolving landscape of ecotoxicology research, the choice between systematic reviews and traditional narrative reviews represents a fundamental methodological crossroads. Systematic reviews, characterized by their transparent, protocol-driven, and reproducible methodology, are increasingly advocated as the gold standard for evidence-based decision-making in regulatory and public health contexts [12] [61]. In contrast, narrative reviews offer a broader, expert-driven synthesis valuable for exploring debates and identifying knowledge gaps but are susceptible to selective citation and lack of methodological transparency [12] [2]. This technical guide examines two pivotal challenges that hinder the effective application of systematic reviews in fields like ecotoxicology, where observational data predominates: their inherent resource intensity and their frequent inadequate adaptation to observational study designs. Addressing these challenges is critical for leveraging systematic reviews to synthesize evidence on chemical exposures, ecological risks, and toxicological outcomes, thereby strengthening the scientific foundation for environmental and health policy [12].

The Challenge of Resource Intensity

Conducting a high-quality systematic review is a profoundly resource-intensive endeavor, requiring significant investments of time, specialized expertise, and financial support. This intensity creates a substantial barrier to their widespread adoption, particularly in academic settings without dedicated review funding.

Quantitative Analysis of Time and Cost Commitments

The following table summarizes the comparative resource demands of narrative versus systematic reviews, illustrating the scale of the commitment [12].

Table 1: Resource Comparison: Narrative vs. Systematic Reviews

Resource Dimension	Narrative Review	Systematic Review	Implications for Ecotoxicology
Typical Timeline	Months	>1 year (usually)	Impedes rapid response to emerging contaminants or crises.
Required Expertise	Subject matter expertise (Science)	Science, systematic review methodology, literature searching, data analysis (meta-analysis)	Need for interdisciplinary teams or extensive researcher training.
Estimated Financial Cost	Low	Moderate to High	Often necessitates dedicated grant funding, limiting who can undertake them.
Methodological Transparency	Usually not specified; implicit process.	Explicit, pre-specified protocol (e.g., PRISMA, Cochrane).	Ensures reproducibility, a key need for regulatory acceptance of toxicological evidence [61].

Protocol-Driven Workflow and Labor

The systematic review process is a multi-stage, iterative workflow that demands meticulous documentation and repeated, independent verification. The diagram below outlines this resource-heavy process.

Diagram: The sequential and often dual-review stages of a systematic review contribute significantly to its resource-intensive nature.

Key Resource-Intensive Stages:

• Protocol Development & Registration: A prerequisite involving detailed a priori specification of the PECO/S question (Population, Exposure/Intervention, Comparator, Outcome, Study design), search strategy, and inclusion criteria [62] [63] [61].
• Comprehensive Search: Requires searching multiple databases (e.g., PubMed, Embase, Web of Science, specialized toxicology databases) with tailored, complex search strings, alongside grey literature searches [61].
• Dual Independent Screening & Data Extraction: To minimize error and bias, title/abstract screening, full-text review, and data extraction are typically performed in duplicate, effectively doubling the personnel effort [61].
• Risk of Bias Assessment: Utilizing specialized tools like ROBINS-I (for observational studies) requires training and independent dual assessment [62].
• Complex Data Synthesis: Integrating observational data, which often involves handling adjusted effect estimates and exploring heterogeneity, requires advanced statistical expertise [62] [64].

The Challenge of Inadequate Adaptation to Observational Data

Systematic review methodology was pioneered for randomized controlled trials (RCTs). Its direct application to observational studies—which constitute the majority of evidence in ecotoxicology (e.g., field studies, cohort studies on chemical exposure)—is fraught with methodological misfit [63] [64]. The core challenge is managing bias and confounding, which are not controlled by design in observational studies as they are in RCTs.

Core Methodological Shortcomings

The table below contrasts the idealized RCT-based review process with the realities of reviewing observational studies, highlighting areas of inadequate adaptation.

Table 2: Methodological Gaps in Applying Systematic Reviews to Observational Data

Review Component	Traditional (RCT-focused) Approach	Challenges with Observational Studies	Consequences for Evidence Synthesis
Defining Exposure	Precisely defined intervention (dose, timing).	Exposure (e.g., environmental contaminant) is often heterogeneous, poorly measured, or estimated [62].	High heterogeneity; difficulty pooling studies.
Risk of Bias Tool	Cochrane RoB 2 (for RCTs).	Inappropriate use of RCT-focused tools. Requires tools like ROBINS-I, which assesses confounding, selection bias, and exposure measurement [62].	Bias in primary studies may be misclassified or overlooked.
Data for Synthesis	Pooling of crude (unadjusted) outcome data.	Crude estimates are often invalid due to confounding. The primary interest is in adjusted estimates (controlling for confounders), but methods of adjustment vary widely [62].	Pooling unadjusted estimates yields misleadingly precise, biased results. Pooling adjusted estimates is methodologically complex.
Heterogeneity	Explored via sensitivity/subgroup analysis.	Inherently extreme due to variation in study design, exposure assessment, confounder adjustment, and populations [64].	Often limits the feasibility and meaningfulness of meta-analysis.
Evidence Certainty (GRADE)	Starts at "High" for RCTs.	Starts at "Low" for observational studies, due to residual confounding [62].	Conclusions are automatically tempered, requiring more cautious interpretation.

Critical Adaptation: From PICO to PECOS and Tailored Bias Assessment

A fundamental adaptation is reframing the research question from PICO (Intervention, Comparison) to PECOS (Exposure, Comparator), acknowledging that the factor under study is an observed exposure, not a randomized intervention [62].

The protocol must pre-specify the key confounders and co-exposures relevant to the exposure-outcome relationship (e.g., age, socioeconomic status, smoking, or other chemical exposures in an ecotoxicology study). The risk of bias assessment using ROBINS-I then evaluates whether primary studies adequately measured and controlled for these pre-identified confounders [62]. The following workflow details this adapted assessment phase.

Diagram: A protocol for systematic reviews of observational data must prioritize confounding and the analysis of adjusted effect estimates.

Protocol for Handling Adjusted Effect Estimates in Meta-Analysis

One of the most technically challenging steps is synthesizing adjusted estimates (e.g., odds ratios, hazard ratios). The following is a detailed protocol derived from current methodological recommendations [62] [64].

Objective: To quantitatively synthesize adjusted effect estimates from observational studies while acknowledging heterogeneity in adjustment methods. Procedure:

• A Priori Specification: In the review protocol, define the minimum set of confounders for which adjustment is considered essential for a valid estimate.
• Data Extraction:
  • Extract both the crude/unadjusted and the most fully adjusted effect estimate and its confidence interval from each study.
  • Document all variables included in the adjusted model.
• Categorization: Classify each study's adjusted estimate as:
  • Optimally Adjusted: Adjusts for all pre-specified core confounders.
  • Partially Adjusted: Adjusts for some, but not all, core confounders.
  • Not Adjusted: Provides only crude estimates.
• Analysis Strategy:
  • Primary Analysis: If feasible, perform meta-analysis using only optimally adjusted estimates. This is the preferred but often restrictive approach.
  • Sensitivity & Subgroup Analyses:
    • Compare pooled results from optimally adjusted, partially adjusted, and unadjusted estimates.
    • Perform a meta-regression to explore whether the number or type of confounders adjusted for explains heterogeneity in the effect sizes.
  • Alternative Approach: If too few studies provide optimally adjusted estimates, present a structured narrative synthesis tabulating adjusted estimates and their adjustment factors, explicitly discussing the implausibility of a valid quantitative pool [62].

Successfully navigating the aforementioned challenges requires leveraging a suite of established tools and resources. The following toolkit is essential for researchers conducting systematic reviews in ecotoxicology.

Table 3: Research Reagent Solutions for Systematic Reviews of Observational Data

Tool/Resource Name	Primary Function	Application in Ecotoxicology
PROSPERO Registry	International prospective register for systematic review protocols.	Ensures transparency, reduces duplication of effort, and allows for peer feedback on the review plan early in the process [62].
PECOS Framework	Structured format for framing the review question (Population, Exposure, Comparator, Outcome, Study design).	Critical for correctly defining questions about chemical exposures and environmental health outcomes, moving beyond the RCT-centric PICO [62].
ROBINS-I Tool	Cochrane's "Risk Of Bias In Non-randomised Studies - of Interventions."	The recommended tool for assessing risk of bias in observational studies, with a dedicated domain for confounding [62].
GRADE (Grading of Recommendations, Assessment, Development, and Evaluations)	Framework for rating the overall certainty of a body of evidence.	Explicitly downgrades evidence from observational studies for risk of bias (confounding), ensuring conclusions are appropriately cautious [62] [61].
PRISMA 2020 Statement & Flow Diagram	Evidence-based minimum set of items for reporting systematic reviews.	Ensures complete and transparent reporting, which is vital for reproducibility and for readers to assess the review's strengths and weaknesses [61].
DistillerSR, Rayyan, Covidence	Software platforms for managing the systematic review process.	Facilitates dual-independent screening, data extraction, and conflict resolution, managing the high administrative burden [2].

Advanced Visualization for Communicating Complex Evidence

Given the complexity of synthesizing observational data, advanced data visualization is crucial for clear communication to stakeholders, including policymakers and other scientists [65].

Effective Visualization Strategies:

• Summary of Findings (SoF) Tables: Incorporate GRADE assessments to visually display the certainty of evidence for each key outcome [66].
• Interactive Dashboards: For living systematic reviews or large projects, dashboards allow users to filter data by study type, exposure level, or outcome, exploring the evidence base dynamically [65].
• Traffic-Light Evidence Maps: Tools like the Overview Reporting Map (ORMap) use color-coded cells to represent the direction of effect and the certainty of evidence for multiple interventions or exposures at a glance, making complex overviews accessible [66].
• Forest Plots with Subgrouping: Visually display pooled estimates separately for different study designs (e.g., cohort vs. case-control) or levels of adjustment, graphically illustrating sources of heterogeneity.

Within the field of ecotoxicology, which focuses on understanding the effects of toxic chemicals on populations, communities, and ecosystems [67], the synthesis of existing research is paramount for risk assessment and regulatory decision-making. Two primary methodologies for research synthesis are employed: the narrative review and the systematic review. A narrative review provides a qualitative, summary overview of the literature on a topic, often selected based on the author's expertise and perspective. While valuable for discussing concepts and identifying broad themes, its informal, non-replicable search and selection methods can introduce bias, making it difficult to distinguish a comprehensive summary from a selective one.

In contrast, a systematic review is a structured, protocol-driven, and reproducible methodology designed to minimize bias. It involves a comprehensive search for all relevant studies, explicit and consistently applied criteria for selecting evidence, and a critical appraisal of the validity of the findings of the included studies [68]. The results are often synthesized quantitatively via meta-analysis. For ecotoxicology—a field dealing with complex environmental interactions and regulatory consequences—the transparency, reproducibility, and reduced bias of systematic reviews are critical for generating reliable evidence. Reporting guidelines like PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) and the field-specific COSTER (Recommendations for the conduct of systematic reviews in toxicology and environmental health research) provide the essential framework to ensure these rigorous standards are met and clearly communicated [69] [70].

The push for robust reporting has led to the development of several key guidelines. The core principles and applications of the most relevant ones for ecotoxicology are summarized below.

Table 1: Comparison of Key Reporting Guidelines for Evidence Synthesis

Guideline	Primary Focus & Scope	Core Components	Key Application in Ecotoxicology
PRISMA 2020 [71] [68]	Generic reporting of systematic reviews & meta-analyses, primarily for intervention effects. Minimum set of items for transparent reporting.	27-item checklist, flow diagram (for new/updated reviews), abstract checklist [72].	Provides the foundational reporting standard for any systematic review in the field, ensuring methodological transparency.
PRISMA 2020 Extensions (e.g., ScR, DTA) [71]	Specialized reporting for specific review types (e.g., scoping reviews, diagnostic test accuracy).	Adapted checklists (e.g., 20 items for ScR [73]), often with specialized flow diagrams.	PRISMA-ScR is useful for mapping the breadth of literature on a broad ecotoxicological issue before committing to a full systematic review [73].
COSMIN for OMIs [74] [75]	Conducting and reporting systematic reviews of Outcome Measurement Instruments (OMIs). Provides methodology and reporting standards.	Methodology checklist for study quality, reporting guideline for OMI reviews.	Ensures reliable evaluation and selection of biological or ecological effect measurement tools (e.g., biomarker assays, ecological indices) used in ecotoxicological studies.
PRISMA-COSMIN for OMIs 2024 [74] [75]	Reporting of systematic reviews of OMIs. An extension of PRISMA 2020.	54-item checklist for full reports, 13-item abstract checklist, and a data flow diagram.	Enables transparent reporting of reviews that assess the quality and suitability of measurement instruments, critical for ensuring data reliability in the field.
COSTER 2020 [69] [70]	Conduct and reporting of systematic reviews in toxicology and environmental health. Field-specific guidance.	Recommendations covering all stages of review conduct, from protocol to synthesis, with an emphasis on environmental evidence.	Addresses unique challenges in ecotoxicology (e.g., non-randomized studies, in vivo/in vitro data, ecological hierarchies) and aims to set robust, consensus-based standards [70].

PRISMA 2020 is the cornerstone guideline. It mandates clear reporting of the review's rationale, methods (including search strategy, selection process, and data extraction), results, and discussion [76] [68]. Its iconic flow diagram provides a transparent account of the study selection process, documenting the number of records identified, screened, eligible, and finally included [72].

COSTER complements PRISMA by providing detailed, field-specific guidance on the conduct of reviews, which directly informs rigorous reporting. Developed through a consensus process involving NGOs, academia, industry, and government agencies [70], COSTER addresses the practical challenges of synthesizing environmental health evidence, such as handling diverse study designs (epidemiological, toxicological, ecological) and assessing risk of bias in non-randomized studies.

The recent PRISMA-COSMIN for OMIs 2024 extension demonstrates the evolution of reporting standards. Developed via a rigorous process including a Delphi study with over 100 panelists and pilot testing [74] [75], it provides a detailed checklist for reporting reviews that evaluate measurement instruments. In ecotoxicology, applying this guideline ensures that reviews assessing the validity of biomarkers or ecological assessment tools are themselves reported with maximum clarity and reproducibility.

Experimental Protocols: Methodological Workflows for Review Synthesis

The reliability of a systematic review is rooted in a pre-defined, reproducible protocol. The following table contrasts the general experimental protocol for a guideline-adherent systematic review against the less structured approach of a traditional narrative review.

Table 2: Experimental Protocol for Systematic Review vs. Narrative Review

Protocol Stage	Systematic Review (Adhering to PRISMA/COSTER)	Traditional Narrative Review
1. Protocol Development	Mandatory. A detailed, publicly registered protocol is developed a priori, specifying the research question (PICO/PECO), search strategy, inclusion/exclusion criteria, and analysis plan. This is recommended by COSTER [70] and can be reported using PRISMA-P [71].	Uncommon. The approach is typically not pre-specified or registered, leaving the methodology open to change during the writing process.
2. Search Strategy	Comprehensive & reproducible. Systematic searches across multiple databases (e.g., PubMed, Web of Science, Environmental Sciences specialized databases), trial registers, and grey literature are documented with full search strings [72]. COSTER emphasizes searching for unpublished data and non-English literature where relevant [70].	Selective & non-reproducible. Literature selection is often non-systematic, based on the author's existing knowledge, key articles, and convenience. The search process is rarely documented.
3. Study Selection	Structured & unbiased. A minimum of two reviewers independently screen titles/abstracts and full texts against explicit criteria, with disagreements resolved by consensus or a third reviewer. The process is documented using a PRISMA flow diagram [68] [72].	Unstructured & subjective. Selection is performed by the author(s) without formal screening criteria or independent verification, increasing the risk of selection bias.
4. Data Extraction & Risk of Bias	Standardized & critical. Data are extracted using piloted forms by multiple reviewers. The methodological quality or risk of bias of each included study is assessed using appropriate tools (e.g., COSTER provides guidance for toxicological studies [70]).	Variable. Data extraction is informal and rarely includes a systematic critical appraisal of study quality.
5. Synthesis	Explicit & analytical. Findings are synthesized narratively and, if appropriate, quantitatively via meta-analysis. Heterogeneity among studies is investigated. COSTER provides specific guidance for synthesizing evidence from diverse environmental health studies [70].	Descriptive & integrative. Findings are summarized narratively, often chronologically or thematically, without formal synthesis methods.
6. Reporting	Structured & transparent. The final report strictly follows the relevant reporting guideline (e.g., PRISMA 2020 checklist [71], PRISMA-COSMIN for OMIs [75]) to ensure all essential methodological and results information is disclosed.	Free-form. Reporting follows general academic writing conventions without a standardized structure for methodological transparency.

Visualization of Guideline Structures and Workflows

The following diagrams, created using Graphviz DOT language, illustrate the logical structure of the PRISMA reporting process and the development pathway for a reporting guideline extension.

Diagram 1: PRISMA 2020 Systematic Review Selection Workflow [68] [72]

Diagram 2: Development of a Reporting Guideline Extension (e.g., PRISMA-COSMIN) [74] [75]

The Scientist's Toolkit: Essential Research Reagent Solutions

Conducting a guideline-compliant systematic review requires specific "research reagents" or essential materials. The following toolkit details key resources for ecotoxicology researchers.

Table 3: Research Reagent Solutions for Conducting Systematic Reviews

Toolkit Item	Function & Description	Relevance to Guideline Adherence
PRISMA 2020 Checklist & Flow Diagram [71] [72]	Reporting Template: The 27-item checklist ensures all critical elements of the review are reported. The flow diagram templates (for databases only or databases plus other sources) provide a standardized way to document study selection.	Core tool for complying with the PRISMA 2020 statement. Using the official templates guarantees all required information is addressed and presented clearly.
COSMIN Risk of Bias Checklist [74]	Methodological Quality Assessment: A tool designed to evaluate the methodological quality (risk of bias) of primary studies on the measurement properties of Outcome Measurement Instruments (OMIs).	Essential for the conduct phase of a review of measurement tools, which underpins reliable reporting as guided by the PRISMA-COSMIN extension [75].
COSTER Recommendations [69] [70]	Field-Specific Conduct Guidance: A comprehensive set of recommendations covering the entire systematic review process specifically for toxicology and environmental health, from planning to synthesis.	Provides the methodological foundation to conduct a robust review in ecotoxicology, which in turn enables complete and transparent reporting as per PRISMA. It addresses field-specific challenges not covered by generic guidelines.
Systematic Review Management Software (e.g., Covidence, Rayyan)	Workflow Management: Web-based platforms that facilitate collaborative reference screening, full-text review, data extraction, and quality assessment. Some automatically generate PRISMA flow diagrams [72].	Enables the reproducible and auditable execution of key protocol stages (screening, data extraction), directly supporting the transparent reporting of methods required by PRISMA and COSTER.
Environmental Health Databases (e.g., PubMed, TOXLINE, GreenFile)	Evidence Sources: Specialized bibliographic databases that index the primary literature in toxicology, environmental science, and public health.	Performing a comprehensive, multi-database search as mandated by PRISMA [68] and emphasized by COSTER [70] is impossible without access to and knowledge of these key resources.

The Editor's and Reviewer's Role in Upholding Methodological Standards

The exponential growth of scientific literature, particularly in fields like ecotoxicology, has made evidence synthesis reviews indispensable for researchers, policymakers, and practitioners. Within this landscape, systematic reviews (SRs) and narrative reviews serve distinct but complementary purposes. However, the proliferation of reviews claiming to be "systematic" without adhering to methodological rigor poses a significant threat to scientific integrity and evidence-based decision-making. A survey in ecology and evolutionary biology found only about 16% of SRs referenced any reporting guideline, yet those that did scored significantly higher on quality metrics [44]. In health sciences, an estimated only 3% of published SRs are considered to have good methodological quality and provide usable clinical evidence [77]. This discrepancy highlights a fundamental crisis in scholarly publishing: the misclassification of review types and the publication of methodologically unsound syntheses that can misdirect research, policy, and clinical practice.

Editors and peer reviewers constitute the final, critical gatekeepers in the scientific publication process. Their role transcends basic quality checks; they are custodians of methodological standards who ensure that published syntheses—particularly in specialized fields like ecotoxicology—are transparent, reproducible, and scientifically sound. This technical guide defines the explicit protocols and evaluative frameworks that editors and reviewers must employ to distinguish robust systematic reviews from lesser syntheses, thereby upholding the integrity of evidence-based science.

Foundational Context: Systematic Review vs. Narrative Review in Ecotoxicology

Choosing the appropriate review methodology is foundational to a synthesis's validity. In ecotoxicology, where research spans molecular mechanisms to ecosystem-level effects, this choice dictates the review's scope, depth, and applicability. The table below delineates the core methodological and philosophical differences between these two review types.

Table 1: Comparative Analysis of Systematic Reviews and Narrative Reviews

Characteristic	Systematic Review	Narrative (Traditional) Review
Primary Objective	To answer a specific, focused research question by synthesizing all available evidence in a manner that minimizes bias [2].	To provide a broad overview or summary of a topic, exploring debates, concepts, and theories [2].
Research Question	Clearly defined at the outset using frameworks (e.g., PICO/PICOS) [44].	Can be broad or explore multiple questions; not necessarily pre-specified with explicit criteria [2].
Protocol & Registration	Requires a pre-published or registered protocol (e.g., in PROSPERO). Protocol registration is a key indicator of rigor [78] [77].	No protocol or registration required or expected.
Search Strategy	Comprehensive, exhaustive, and documented with the goal of replicability. Must search multiple databases and other sources to find all relevant studies [44] [77].	May or may not be comprehensive. Methods are often not explicit or reproducible; may rely on a subset of known literature [2] [77].
Study Selection & Data Extraction	Conducted by at least two independent reviewers using pre-defined criteria to minimize error and bias. A reconciliation process for conflicts is mandatory [78].	Typically conducted by a single author or team without explicit, independent verification.
Risk of Bias / Quality Assessment	Mandatory critical appraisal of included studies using validated tools (e.g., RoB 2, ROBINS-I) [44] [78].	May or may not include quality assessment of cited works [77].
Data Synthesis	Can be narrative (qualitative), quantitative (meta-analysis), or both. Analysis is structured and follows the protocol [44].	Typically narrative, conceptual, thematic, or chronological [2] [77].
Reporting Standards	Must adhere to guidelines like PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), with a flow diagram required [44] [78].	No consensus on standard structure, though often follows IMRAD format. No specific reporting guideline [2].

For ecotoxicology, the distinction has practical implications. Leading journals like Ecotoxicology and Environmental Safety emphasize mechanistic understanding and hypothesis-driven research, often requiring studies to link molecular effects to population or community-level outcomes [79]. A systematic review is uniquely suited to test a specific hypothesis about an exposure-effect relationship (e.g., "Does sub-lethal concentration X impair reproductive success in species Y?") by systematically aggregating and appraising experimental evidence. Conversely, a narrative review might expertly chart the historical development of a paradigm, such as the evolution of the Adverse Outcome Pathway (AOP) framework, synthesizing ideas and identifying conceptual gaps without a formal, replicable search protocol.

The Editorial and Reviewer Evaluation Protocol

Editors and reviewers must employ a structured, checklist-driven approach to assess review manuscripts. The following workflow and detailed protocol are designed to identify methodological soundness or deficiency at each stage of the review process.

Diagram 1: Systematic review evaluation workflow for editors and reviewers.

Phase A: Initial Evaluation and Rationale

The first assessment determines if the review justifies its existence and is correctly classified.

• Justification and Novelty: Reviewers must verify that the SR addresses a meaningful gap. If previous SRs exist, the new submission must offer a substantive update, new analysis, or correct a methodological flaw [78]. A redundant review should be discouraged.
• Protocol Registration: A prospective protocol registered on platforms like PROSPERO, INPLASY, or the Open Science Framework (OSF) is a non-negotiable hallmark of a true SR [78] [77]. The reviewer must obtain and compare the submitted manuscript against the registered protocol to check for undisclosed deviations in outcomes or methods.
• Adherence to Reporting Guidelines: The manuscript must include a completed PRISMA 2020 checklist (or MOOSE for observational studies) as a supplementary file [78]. Its absence is a major red flag.

Phase B: In-Depth Methodological Scrutiny

This phase is the core of the quality assessment, where most flaws are found.

• Search Strategy: The methods must detail databases searched (e.g., PubMed, Web of Science, Scopus, Environment Complete), exact search strings with Boolean operators, search dates, and efforts to locate grey literature. A search of only one database is universally inadequate [78] [77]. The strategy must be reproducible.
• Study Selection Process: The manuscript must state that screening of titles/abstracts and full-texts was performed independently by at least two reviewers, with a process for resolving disagreements [78] [77]. A PRISMA flow diagram must accurately document the number of records at each stage and reasons for exclusion [44] [78].
• Data Extraction: Similarly, data extraction should be performed in duplicate to minimize error [78].
• Risk of Bias (RoB) Assessment: The use of a domain-based, validated tool is mandatory. For ecotoxicology, tools like RoB 2 (for randomized trials), ROBINS-I (for non-randomized studies), or domain-specific tools for ecological studies are relevant [44] [78]. The results should inform the synthesis (e.g., sensitivity analyses excluding high-risk studies).

Phase C: Analysis and Synthesis Evaluation

Reviewers must assess whether the analytical approach is appropriate and correctly executed.

• Quantitative Synthesis (Meta-analysis): Check if pooling of data is justified (i.e., sufficient clinical and methodological homogeneity). Reviewers must look for appropriate effect measures (e.g., standardized mean difference, risk ratio), the use of random/fixed effects models, and quantification of heterogeneity (I² statistic) [78].
• Qualitative/Narrative Synthesis: Assess if findings are structured and presented clearly (e.g., tabulated), and that conclusions are logically derived from the data presented.
• Assessment of Certainty of Evidence: For SRs informing decision-making, the use of the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) framework to rate the overall certainty of evidence for key outcomes is a best practice [44] [78].

Phase D: Vigilance for Ethical and Emerging Issues

Modern reviewing requires awareness of new threats to integrity.

• Authorship and AI: Guest, ghost, or fraudulent authorship must be checked. The use of generative AI in manuscript writing must be transparently declared per journal policy [79] [67].
• Paper Mill and Fraud Detection: Reviewers should be alert to signs of paper mill output: unusual author email addresses, generic phrasing, template-like figures, or suspiciously positive results across multiple included studies [78].
• Conflict of Interest: All authors must disclose relevant financial and non-financial conflicts [79] [80].

Visualizing the Systematic Review Process: A Standard for Comparison

For editors and reviewers to effectively audit a submitted SR, they must have a clear mental model of the ideal process. The following diagram maps the essential stages of a rigorously conducted systematic review, serving as a benchmark against which manuscript methods can be compared.

Diagram 2: The 8-stage protocol for a rigorous systematic review.

The Scientist's Toolkit: Essential Materials for Review

Editors and reviewers, as well as authors conducting SRs, require a common set of conceptual and practical tools. The following table catalogs key resources.

Table 2: Essential Toolkit for Conducting and Evaluating Systematic Reviews

Tool Type	Specific Tool / Resource	Primary Function & Relevance
Reporting Guideline	PRISMA 2020 Statement & Checklist [44] [78]	The mandatory reporting standard for systematic reviews. Authors must complete it; reviewers use it for evaluation.
Protocol Registry	PROSPERO, INPLASY, OSF Registries [78]	Public platforms for prospectively registering a review protocol to prevent duplication and bias.
Risk of Bias Tool	RoB 2 (randomized trials), ROBINS-I (non-randomized studies) [44] [78]	Validated instruments for critically appraising the internal validity of included primary studies.
Evidence Grading Framework	GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) [44] [78]	Framework for rating the overall certainty (high, moderate, low, very low) of a body of evidence.
Journal Policy Reference	Ecotoxicology and Environmental Safety Guide for Authors [79], Ecotoxicology Submission Guidelines [67]	Domain-specific guidance on scope, article types, and methodological expectations for ecotoxicology research.
Reviewer Checklist	Consolidated SRMA Evaluation Checklist (Derived from [78])	A unified checklist (see Appendix) for structured, thorough peer review of systematic reviews.

Application in Ecotoxicology: Domain-Specific Considerations

Upholding standards in ecotoxicology requires applying general SR principles to the field's unique challenges. Key considerations include:

• Defining the Research Question (PICOS): Ecotoxicology questions often involve complex exposures and outcomes. A robust PICOS framework is vital [44]. Example: P: Freshwater amphipods (Gammarus pulex); I: Exposure to fluoxetine (antidepressant); C: Untreated controls; O: Mortality, feeding rate, locomotor behavior; S: Laboratory microcosm studies.
• Handling Diverse Study Designs: Ecotoxicology evidence includes lab experiments, mesocosm studies, and field observations. The risk of bias tool must be appropriate (e.g., SYRCLE's RoB tool for animal studies, or adapted tools for ecological data) [81].
• Journal-Specific Enforcement: Leading journals have clear policies. Ecotoxicology notes it will not consider papers dealing only with toxicity testing or pollutant levels without linkage to population effects or specific field situations [67]. Editors must desk-reject submissions that fail this fundamental scope test.
• Intervention by Editors: A workshop with editors of toxicology journals prioritized short-term actions including mandating protocol registration, requiring PRISMA adherence, and providing training for reviewers on SR methodology [81]. These interventions directly improve the quality of published syntheses.

The role of the editor and reviewer in the era of evidence synthesis is fundamentally one of methodological enforcement and education. By wielding structured evaluation protocols, understanding the stark operational differences between review types, and applying domain-specific knowledge, they act as the essential guarantors that published systematic reviews in ecotoxicology are truly systematic. This, in turn, ensures that the foundation of evidence for environmental and health policy is robust, reliable, and worthy of the trust placed in it by the scientific community and society.

---

Appendix: Consolidated Reviewer Checklist for Systematic Review & Meta-Analysis (SRMA) Manuscripts

Based on synthesis of [44] [78] [77]

Section	Item	Check (Y/N/NA)	Notes/Questions for Authors
A. Rationale & Protocol	A1. Clear justification and novelty vs. existing SRs provided.
	A2. Protocol prospectively registered (ID provided).
	A3. PRISMA/MOOSE checklist submitted and followed.
B. Methods	B1. PICOS eligibility criteria explicitly defined.
	B2. Search strategy is comprehensive, multi-database, and fully reproducible.
	B3. PRISMA flow diagram included, documenting screening.
	B4. Dual-independent study selection & data extraction stated.
	B5. Appropriate risk of bias tool used for included studies.
C. Results & Synthesis	C1. Synthesis method (narrative/meta-analysis) appropriate for data.
	C2. If meta-analysis, heterogeneity assessed (I²) and model justified.
	C3. Risk of bias results presented and integrated (e.g., sensitivity analysis).
	C4. Certainty of evidence assessed (e.g., GRADE).
D. Integrity & Ethics	D1. Author contributions transparent; no ghost/guest authorship suspected.
	D2. AI use in writing properly declared (if applicable).
	D3. Conflicts of interest fully disclosed.
	Overall Recommendation:	Accept / Minor Revisions / Major Revisions / Reject

The field of ecotoxicology is undergoing a profound methodological transformation, driven by the convergence of three powerful trends: the automation of research processes, the evolution of dynamic "living" systematic reviews, and the rapid integration of New Approach Methodologies (NAMs). This transformation occurs within a critical scholarly context: the ongoing discourse on the appropriate use of systematic versus narrative review methodologies for synthesizing evidence and guiding policy. Historically, narrative reviews have provided broad, contextual overviews of ecotoxicological fields, ideal for exploring emerging contaminants or mechanisms. In contrast, systematic reviews, with their stringent, protocol-driven methodology, have been the gold standard for answering focused questions on chemical risk, requiring minimization of bias and maximal reproducibility [2] [82].

The advent of NAMs—encompassing in vitro assays, computational models, and high-throughput screening—fundamentally challenges these traditional evidence synthesis paradigms [83] [84]. These methods generate vast, complex datasets that are more human-relevant but also faster and more diverse than traditional animal studies. This data deluge makes manual, narrative synthesis increasingly impractical and highlights the risk of selective citation. Conversely, the rigid, time-consuming nature of conventional systematic reviews struggles to keep pace with the rapid generation of NAMs data, risking obsolescence upon publication [85].

This whitepaper posits that the integration of laboratory and informatics automation is the essential catalyst enabling a new synthesis paradigm. Automation facilitates the continuous data extraction and updating necessary for "living" reviews, creating a responsive feedback loop where NAMs efficiently generate human-relevant safety data, and automated living reviews synthesize this evidence in near real-time to guide testing strategies and regulatory decisions. This integrated framework promises to enhance the translational relevance of ecotoxicology, moving the field toward more predictive and proactive safety assessment.

Foundational Context: Systematic vs. Narrative Review in Ecotoxicology

Choosing between a systematic and a narrative review is a foundational decision that shapes the entire research process, its outcomes, and its impact. The distinction is not merely procedural but philosophical, relating to the core objectives of the synthesis [82] [86].

A narrative (or traditional literature) review aims to provide a comprehensive, critical, and objective analysis of current knowledge on a topic. It is characterized by a flexible, non-standardized approach without a pre-specified protocol. The author surveys literature—often without an exhaustive search—to summarize, interpret, and discuss a broad field, identify trends, debates, and theoretical frameworks, and contextualize new research [2] [85]. Its strength lies in its scope and ability to provide a "big picture" perspective, making it invaluable for exploring novel, interdisciplinary, or poorly defined topics in ecotoxicology, such as the ecological implications of a new class of polymers. However, this flexibility introduces susceptibility to author selection and interpretation bias, making it less suitable for definitive, evidence-based conclusions [82].

A systematic review, in contrast, is defined by its methodological rigor and transparency. It is conducted to answer a specific, focused research question (often formulated using PICO—Population, Intervention, Comparator, Outcome) through a pre-registered protocol. Its hallmark is an exhaustive, reproducible search strategy across multiple databases and grey literature, followed by the systematic screening, critical appraisal, and synthesis of all eligible studies [2] [85]. When applicable, a meta-analysis provides a quantitative summary of effect sizes. This process minimizes bias, maximizes reproducibility, and produces the highest level of evidence to directly inform risk assessment and regulation [82]. The trade-off is a narrow scope, high resource demands, and a timeline that can render the review outdated if the field evolves rapidly during its conduct.

The table below summarizes the key methodological distinctions between these two review types.

Table 1: Comparative Analysis of Narrative and Systematic Review Methodologies

Characteristic	Narrative (Traditional) Review	Systematic Review
Primary Objective	Provide broad overview, context, and theoretical framework; identify trends/gaps [2] [82].	Answer a specific, focused research question with minimal bias [2] [85].
Research Question	Can be broad or general; often evolves [82] [85].	Narrow, specific, and fixed; often uses PICO framework [2] [85].
Protocol & Registration	No pre-specified protocol; not registered [85].	Mandatory pre-registered protocol (e.g., PROSPERO) [85].
Search Strategy	Not necessarily systematic or exhaustive; may be selective [82] [85].	Comprehensive, documented search across multiple databases/grey literature; reproducible [2] [85].
Study Selection	Often subjective; criteria not explicit [82].	Based on explicit, pre-defined inclusion/exclusion criteria; performed in duplicate [2].
Critical Appraisal	Not required; quality may be discussed narratively [85].	Mandatory, rigorous quality/risk-of-bias assessment of included studies [2] [85].
Data Synthesis	Qualitative, narrative summary; thematic analysis [82] [85].	Structured narrative, often tabular; may include quantitative meta-analysis [2] [85].
Timeline & Resources	Weeks to months; manageable for individual researcher [85].	Months to years (avg. ~18 mos); requires a team [85].
Susceptibility to Bias	High (selection, interpretation bias) [82].	Low (methodology designed to minimize bias) [2] [82].
Ideal Application in Ecotoxicology	Exploring emerging fields, developing hypotheses, textbook chapters, contextualizing a new study [82] [86].	Substantiating chemical risk assessments, evaluating intervention efficacy, informing regulatory guidelines [2].

The Rise of New Approach Methodologies (NAMs) in Safety Science

New Approach Methodologies represent a paradigm shift in toxicology and ecotoxicology, moving away from reliance on apical endpoints in whole animal studies toward mechanistically informed, human- and ecologically-relevant models. Driven by ethical, economic, and scientific imperatives—including the desire to reduce animal testing, lower costs, increase throughput, and improve human relevance—NAMs leverage advances in biotechnology, genomics, and computational science [83] [84].

NAMs encompass a broad suite of tools:

• In vitro assays using human or ecological receptor-based cell systems.
• Tissue/organ-on-chip microphysiological systems (MPS) that mimic functional units [83].
• High-content screening and high-throughput transcriptomics.
• Computational (in silico) models, including quantitative structure-activity relationship (QSAR) models and kinetic simulations.
• Alternative animal models like C. elegans for rapid screening [84].

A primary scientific driver is the recognition of species-specific differences that limit the translational predictability of traditional rodent models for human health, and of standard test species for diverse ecosystems [83]. NAMs based on human or conservation-relevant cell and tissue models can provide more direct insight into mode-of-action.

Regulatory acceptance is accelerating. The U.S. FDA Modernization Act 2.0 (2022) explicitly allows for alternatives to animal testing for drug safety [83]. The European Chemicals Agency (ECHA) and the U.S. Environmental Protection Agency (EPA) are actively developing frameworks for integrating NAMs into chemical risk assessment under initiatives like Next Generation Risk Assessment (NGRA) [83].

Table 2: Key Categories of New Approach Methodologies and Their Applications

NAM Category	Description	Example Technologies/Tools	Primary Application in Ecotoxicology
Advanced In Vitro Models	Cell-based systems ranging from 2D cultures to complex 3D organoids.	Induced pluripotent stem cell (iPSC)-derived cells, 3D liver/heart/renal spheroids, primary cell co-cultures [83].	Mechanistic toxicity screening, metabolism studies, organ-specific toxicity.
Microphysiological Systems (MPS) / Organ-on-a-Chip	Microfluidic devices housing living tissues that simulate organ-level physiology and dynamic flow.	Liver-chip, kidney-chip, blood-brain-barrier-chip, multi-organ "body-on-a-chip" systems [83].	Modeling systemic toxicity, ADME (Absorption, Distribution, Metabolism, Excretion), and inter-organ signaling.
High-Throughput Screening (HTS)	Automated testing of chemicals across large batteries of biochemical or cellular assays.	Robotic liquid handling, automated imaging, high-content screening (HCS), ToxCast assay pipeline [83].	Prioritizing chemicals for further testing, identifying biological pathways of concern.
Computational Toxicology (In Silico)	Computer-based models to predict toxicity from chemical structure or biological data.	QSAR models, physiologically based kinetic (PBK) modeling, molecular docking, machine learning classifiers [83].	Hazard prediction for data-poor chemicals, read-across, and risk prioritization.
Omics Technologies	Global profiling of molecular changes (genes, proteins, metabolites) in response to exposure.	Transcriptomics (RNA-seq), proteomics, metabolomics, epigenomics [83].	Uncovering novel biomarkers of effect/ exposure, understanding adverse outcome pathways (AOPs).

The validation and integration of NAMs into decision-making frameworks is an active area of consortia work (e.g., IQ Consortium, EFPIA) [83]. A critical challenge is establishing bioinformatic and computational pipelines to translate the complex, high-dimensional data from NAMs into interpretable evidence for risk assessment. This data complexity directly underscores the need for advanced, automated evidence synthesis methods.

Automation as the Enabling Engine: From the Lab Bench to the Literature

Automation is the critical infrastructural trend that amplifies the potential of both NAMs and modern evidence synthesis. It operates at two interconnected levels: physical laboratory automation and digital informatics automation.

Laboratory Automation: In the context of NAMs, automation transforms testing capacity. Robotic liquid handling systems enable precise, reproducible execution of high-throughput in vitro assays. Automated incubators, imagers, and plate readers facilitate continuous, unattended operation. Integrated end-to-end automated workcells can conduct complex, multi-step protocols—from cell seeding and compound dosing to staining and analysis—dramatically increasing throughput and data consistency while reducing human error and variability [87]. This is essential for generating the robust, high-quality datasets required for regulatory acceptance of NAMs.

Informatics Automation: This is the cornerstone of the "living review" paradigm. Software tools now automate nearly every stage of the systematic review workflow, which is notoriously labor-intensive [2].

• Search & Screening: AI-powered tools can execute complex, pre-defined search strategies across multiple bibliographic databases, deduplicate results, and prioritize references for screening using machine learning based on titles and abstracts.
• Data Extraction: Natural Language Processing (NLP) algorithms can be trained to extract specific data points (e.g., dosage, effect size, species) from full-text PDFs into structured databases.
• Critical Appraisal: Tools can guide reviewers through risk-of-bias checklists, ensuring consistency.
• Continuous Surveillance: Automated systems can monitor for newly published primary studies or preprints relevant to the review question, triggering an update cycle.

This digital automation compresses the systematic review timeline from years to months or weeks, making the concept of a "living systematic review"—a review that is continuously updated as new evidence emerges—a practical reality [85]. The logical and data workflow integrating these automated components is illustrated below.

Diagram 1: Integrated workflow from NAMs to decision-making.

Synthesis: Integrating NAMs, Automation, and Living Reviews into a Cohesive Framework

The true transformative power lies in the intentional integration of these three trends into a cohesive evidence-generation and synthesis ecosystem. The framework functions as a continuous, iterative cycle:

1. NAMs Generate Novel Evidence: Automated, high-throughput NAM platforms test chemicals across mechanistic endpoints, producing data that is more human-relevant and fills knowledge gaps for data-poor substances [83] [84].

2. Automation Curates and Structures Data: Informatics pipelines automatically extract, harmonize, and deposit NAMs data (and traditional study data) into FAIR (Findable, Accessible, Interoperable, Reusable) knowledge bases and structured toxicological databases. This step is critical for making diverse data streams machine-readable for synthesis [87].

3. Living Reviews Synthesize Dynamically: Automated living review platforms continuously monitor these knowledge bases. Using pre-specified, transparent protocols, they integrate new NAMs and traditional evidence as it emerges, updating systematic reviews or meta-analyses in near real-time [85]. This addresses the "speed" challenge of traditional systematic reviews.

4. Synthesis Informs Strategy and Regulation: The updated, evidence-based conclusions directly feed into agile chemical prioritization, refinement of testing strategies (guiding which NAMs to apply next), and responsive regulatory risk assessment. This creates a closed loop where the synthesis of past evidence directly optimizes future evidence generation.

This integrated framework resolves key tensions in the systematic vs. narrative review debate. It preserves the rigor, transparency, and reduced bias of systematic methodology while overcoming its historical slowness and static nature through automation. Simultaneously, it channels the broad, exploratory strengths of narrative reviews into the initial development of review questions and the interpretation of complex, mechanistic NAMs data within a more structured, living evidence context.

The Scientist's Toolkit: Essential Reagents and Platforms

Implementing this integrated approach requires familiarity with a suite of modern tools. Below is a selection of key research reagent solutions and platforms central to NAMs and automated workflows.

Table 3: Key Research Reagent Solutions and Platforms for NAMs and Automated Workflows

Tool Category	Specific Example/Technology	Function in Research Workflow
Advanced Cell Models	Induced Pluripotent Stem Cell (iPSC)-derived cardiomyocytes, hepatocytes, neurons.	Provide a human-relevant, genetically defined, and renewable cell source for in vitro toxicity testing and MPS [83].
Organ-on-a-Chip Systems	Commercially available liver-chip or blood-brain-barrier-chip platforms (e.g., from Emulate, Mimetas).	Microfluidic devices that recreate tissue-tissue interfaces, mechanical cues, and vascular perfusion to model organ-level physiology and toxicity [83].
Multi-Electrode Array (MEA) Systems	Integrated MEA platforms for cardiomyocyte or neuronal network assessment.	Non-invasively measures the electrophysiological activity of cell monolayers or tissues, critical for detecting functional cardiotoxicity or neurotoxicity [83].
High-Throughput Screening Assays	Fluorogenic or luminescent assay kits for cytotoxicity, oxidative stress, mitochondrial function, etc.	Robust, miniaturizable biochemical endpoints compatible with automated liquid handlers and plate readers for screening large chemical libraries [83].
Automated Liquid Handlers	Bench-top robotic pipetting systems (e.g., from Opentrons, Hamilton, Tecan).	Execute precise, reproducible liquid transfers for cell culture, assay setup, and compound dosing, enabling HTS and reducing manual error [87].
Laboratory Information Management System (LIMS)	Cloud-based LIMS platforms (e.g., Benchling, LabVantage).	Centralizes and structures experimental metadata, sample tracking, and results, ensuring data integrity and traceability for automated workflows [87].
Systematic Review Automation Software	AI-assisted screening tools (e.g., ASReview, DistillerSR).	Use machine learning to prioritize references for review, significantly accelerating the title/abstract screening phase of systematic reviews [2].

Detailed Experimental Protocol: A Template for an Automated NAM-Based Screening Study

The following protocol outlines a standardized workflow for conducting an automated, high-throughput in vitro screening study to assess the potential hepatotoxicity of a chemical library, generating data suitable for integration into a living evidence synthesis.

Protocol Title: Automated High-Throughput Screening for Chemical-Induced Hepatotoxicity Using a 3D Human Liver Spheroid Model.

1. Objectives and Research Question (PICO Format):

• Population: 3D human liver spheroids (e.g., HepaRG or iPSC-derived).
• Intervention: Exposure to a library of environmental chemicals (e.g., 1000 compounds).
• Comparator: Vehicle control (e.g., 0.1% DMSO).
• Outcomes: Cell viability (ATP content), intracellular glutathione (GSH) depletion, and albumin secretion as a marker of hepatic function.

2. Materials and Equipment:

• Biological Model: Cryopreserved human liver spheroids or cells for spheroid formation.
• Chemical Library: Pre-formatted in 384-well source plates at 10mM in DMSO.
• Assay Reagents: CellTiter-Glo 3D (ATP), GSH-Glo Glutathione Assay, Human Albumin ELISA kit.
• Core Automated Equipment:
  • Automated liquid handler (e.g., Hamilton Microlab STAR) with a 384-channel head.
  • CO2 incubator with automated stacker.
  • Multimode microplate reader (luminescence, fluorescence).
  • Automated imaging system (optional for morphology).
  • LIMS for protocol orchestration and data capture.

3. Detailed Stepwise Procedure: A. Study Initiation & Plate Layout (Automated): 1. The LIMS uploads the study design and chemical plate map. 2. The liquid handler prepares intermediate dilution plates in culture medium, serially diluting compounds from source plates to create a 6-point concentration range (e.g., 100 µM to 0.3 µM). 3. A 384-well ultra-low attachment microplate is designated as the "assay plate."

4. Data Processing and Analysis (Automated): 1. Raw luminescence/fluorescence data from the plate reader is automatically parsed and uploaded to the LIMS/data analysis platform (e.g., Genedata Screener). 2. Concentration-response curves are automatically fitted for each compound and endpoint. 3. Benchmark Dosing: Potency metrics (e.g., TC50 - concentration causing 50% reduction in ATP) are calculated. Results are benchmarked against internal controls and historical data for quality control. 4. A structured data output (e.g., CSV file with compound ID, TC50, curve fit parameters, assay QC flags) is generated for upload to a public repository or internal knowledge base.

5. Integration with Evidence Synthesis: * The structured results file is deposited in a public toxicogenomics database (e.g., NIH's Chemical Effects in Biological Systems - CEBS) with rich metadata. * A living systematic review on hepatotoxicity mechanisms, configured for automated surveillance, ingests this new dataset via an API call to the database. * The review's evidence profile and/or meta-analysis is automatically updated, refining the overall hazard assessment for the screened chemicals.

The trajectory of ecotoxicology points toward an increasingly automated, data-driven, and integrated future. Key emerging trends include the maturation of microphysiological systems (MPS) to model complex ecological interactions and multi-species dynamics, the application of artificial intelligence not just for data analysis but for experimental design and predictive modeling of ecosystem-level effects, and the formalization of standardized data exchange formats to seamlessly connect NAMs data streams with automated evidence synthesis platforms [83] [88].

The scholarly debate between systematic and narrative reviews will evolve within this new environment. The dichotomy will blur, giving rise to hybrid, technology-enabled review modalities. The foundational principle of systematic review—minimizing bias through transparency and protocol—will remain paramount. However, the execution will be fundamentally augmented by AI and automation, allowing reviews to be both continuously updated (living) and incredibly comprehensive in their handling of diverse data types, from traditional in vivo studies to high-dimensional omics data from NAMs [85].

For researchers and drug development professionals, the imperative is to build interdisciplinary competency. Ecotoxicologists must engage with data scientists, automation engineers, and bioinformaticians. Regulatory scientists must develop frameworks to evaluate the evidentiary weight of integrated NAMs data streams. Ultimately, the convergence of NAMs, automation, and living systematic reviews forms a powerful, virtuous cycle. It promises to accelerate the pace of safety science, enhance the human and ecological relevance of our predictions, and provide decision-makers with a current, rigorous, and actionable evidence base to navigate the complex chemical landscape of the 21st century.

Head-to-Head Comparison: Utility, Validity, and Choosing the Right Tool

Empirical Comparison of Methodological Rigor and Transparency

This whitepaper provides a technical, empirical comparison of the methodological rigor and transparency inherent to systematic reviews versus traditional narrative reviews, contextualized within ecotoxicology research. The analysis draws upon current guidelines, performance metrics, and case studies to demonstrate that systematic reviews employ a structured, protocol-driven process that minimizes bias and enhances reproducibility [12] [2]. In contrast, narrative reviews offer flexible, broad overviews but are characterized by implicit, non-standardized methodologies that increase the risk of selective evidence use and unreproducible conclusions [89] [12]. For ecotoxicology—a field burdened by complex evidence streams, data gaps, and high-stakes regulatory decisions—the adoption of systematic review standards is critical for generating reliable, transparent evidence syntheses [3] [56]. This guide details standardized experimental protocols for conducting systematic reviews, provides visual workflows, and outlines an essential toolkit of research reagents and methodologies to empower researchers and drug development professionals in implementing evidence-based toxicology practices.

Ecotoxicology is tasked with assessing the impact of chemical contaminants on ecosystems and human health, a challenge compounded by multifaceted evidence from field studies, in vivo and in vitro assays, and computational models [90] [91]. Synthesizing this evidence to inform regulation and risk management requires methodologies that are both scientifically robust and transparently documented. Historically, narrative reviews have dominated the field, providing expert-driven summaries of literature [12]. However, their informal approach to literature search, selection, and appraisal introduces risks of bias and limits reproducibility, potentially leading to conflicting risk assessments for the same chemical [12].

Systematic reviews, pioneered in clinical medicine and now advanced in evidence-based toxicology, offer a corrective framework [12] [56]. They are defined by an a priori protocol, comprehensive search strategies, explicit inclusion/exclusion criteria, critical appraisal of study quality, and systematic synthesis [44] [89]. Initiatives like the COSTER recommendations provide tailored guidance for conducting systematic reviews in toxicology and environmental health, addressing domain-specific challenges such as integrating diverse evidence streams and assessing non-randomized studies [56]. This whitepaper empirically compares these two review paradigms, providing ecotoxicology researchers with the technical guidance necessary to conduct syntheses that meet the highest standards of methodological rigor and transparency.

Empirical Comparison of Review Methodologies

The fundamental differences between narrative and systematic reviews can be quantified across several dimensions of methodological rigor. The following table synthesizes criteria from toxicology-specific guidance to provide a direct comparison [12] [2].

Table 1: Empirical Comparison of Narrative Reviews and Systematic Reviews in Ecotoxicology

Comparison Feature	Narrative (Traditional) Review	Systematic Review (with/without Meta-Analysis)	Source of Criteria / Implication
Research Question	Broad, often informally defined or unspecified.	Specific, focused, and structured using frameworks (e.g., PICO, PICOTS).	[44] [12]; A precise question determines all subsequent steps.
Protocol & Registration	Typically not published or pre-registered.	A detailed, publicly registered protocol is mandatory (e.g., via PROSPERO).	[44] [56]; Prevents bias from post-hoc changes in methods.
Search Strategy	Not systematic; sources and search terms often unspecified.	Comprehensive, documented search across multiple databases + grey literature.	[3] [12]; Maximizes recall of relevant evidence, minimizing selection bias.
Study Selection	Criteria implicit and subjective; process not documented.	Explicit, pre-defined eligibility criteria; selection process documented (e.g., PRISMA flow).	[89] [2]; Ensures reproducibility and limits reviewer bias.
Risk of Bias / Quality Assessment	Rarely performed or informal.	Mandatory, using validated tools for each study design (e.g., OHAT, SYRCLE).	[3] [44]; Critical for interpreting the credibility of synthesized findings.
Data Synthesis	Qualitative, narrative summary; often selective.	Structured synthesis (narrative, tabular, or quantitative meta-analysis).	[89] [12]; Systematic synthesis provides a clearer, more complete evidence picture.
Certainty of Evidence	Not formally assessed.	Formally graded using frameworks like GRADE or CERQual.	[44]; Informs users about the confidence they can place in the conclusions.
Time & Resource Commitment	Lower (typically months).	Substantially higher (often >1 year).	[12]; Systematic reviews are resource-intensive but produce higher-quality evidence.
Primary Utility	Background context, hypothesis generation, exploring broad fields.	Informing policy/regulation, clinical/safety guidelines, resolving controversies.	[89] [2]; Systematic reviews are considered the highest level of evidence for decision-making.

The empirical superiority of the systematic review framework for objective decision-making is clear. This is particularly vital in ecotoxicology, where, for example, a narrative review might subjectively highlight studies on a contaminant's toxicity, while a systematic review would transparently identify, appraise, and integrate all relevant evidence—including studies showing no effect—leading to a more balanced and reliable conclusion [12].

Detailed Methodological Protocols for Systematic Reviews in Ecotoxicology

Adhering to a standardized protocol is the cornerstone of a rigorous systematic review. The following steps, synthesized from current guidelines, provide a detailed experimental protocol [3] [44] [12].

Step 1: Problem Formulation and Protocol Development

• Action: Define a specific, structured research question. In ecotoxicology, the PICOTS framework is recommended: Population (e.g., Daphnia magna), Intervention/Exposure (e.g., microplastic particles), Comparator (e.g., control conditions), Outcome (e.g., mortality, reproduction), Timeframe, and Study design [44].
• Protocol: Develop and publicly register a protocol detailing the search strategy, eligibility criteria, risk-of-bias assessment tools, and data synthesis plans. This is a mandatory step per COSTER and PRISMA guidelines [44] [56].

Step 2: Systematic Literature Search and Study Selection

• Search Strategy: Design a comprehensive, reproducible search string using keywords and controlled vocabulary (e.g., MeSH, Emtree). Search multiple databases (e.g., PubMed, Web of Science, Scopus, specialized ecotoxicology databases) and grey literature sources [44] [91].
• Study Selection: Implement a two-phase screening process (title/abstract, then full-text) using pre-defined eligibility criteria. This process should be performed independently by at least two reviewers, with conflicts resolved by consensus or a third reviewer. Document the flow of studies using a PRISMA diagram [89].

Step 3: Data Extraction and Critical Appraisal

• Data Extraction: Use a standardized, piloted form to extract relevant data from included studies (e.g., study design, sample characteristics, exposure details, outcomes, results). Dual extraction is recommended for accuracy [12].
• Risk of Bias (RoB) Assessment: Assess the methodological quality and internal validity of each included study using design-appropriate tools. For animal studies, tools like SYRCLE's RoB tool are used; for in vitro studies, adapted tools are applied. This assessment is critical for evidence integration [44] [12].

Step 4: Evidence Integration, Synthesis, and Grading

• Data Synthesis: Synthesize findings narratively, tabularly, and, if appropriate, quantitatively via meta-analysis. Heterogeneity (e.g., I² statistic) must be assessed and explored [44].
• Certainty Assessment: Rate the overall certainty (or confidence) in the body of evidence for each key outcome using a framework like GRADE, considering RoB, consistency, directness, precision, and publication bias [44].

Diagram 1: Systematic Review Workflow in Ecotoxicology [3] [44] [12].

Case Study in Ecotoxicology: Contrasting Review Approaches

A 2025 systematic review on the "Ecotoxicological implications of environmental contaminants on disease vectors" provides a concrete example of rigorous methodology [90]. It explicitly formulated a research question, conducted a comprehensive systematic search, assessed the quality of included studies, and synthesized evidence to reveal that pollutants can synergistically enhance insecticide resistance in mosquitoes—a finding with direct implications for integrated vector management.

In contrast, a narrative review on the same topic might selectively cite only the most dramatic examples of pollution-induced resistance, lack a formal assessment of study limitations, and present a less nuanced conclusion. This demonstrates how the systematic methodology guards against bias and provides a more reliable evidence base for complex ecological interactions [90] [12].

Diagram 2: Contrasting Workflows: Narrative vs. Systematic Review [89] [12] [2].

Advancing Rigor with New Approach Methodologies (NAMs) and Computational Tools

The transition to New Approach Methodologies (NAMs) in toxicology, including in silico models, creates new data streams for systematic reviews. Quantitative Structure-Activity Relationship (Q)SAR models are increasingly used to fill data gaps for cosmetic and industrial chemicals, especially under animal-testing bans [92] [91]. Systematic reviews evaluating chemical safety must now incorporate criteria for assessing the reliability of such computational tools.

Table 2: Performance of Selected (Q)SAR Models for Environmental Fate Prediction (Adapted from a Comparative Study) [92]

Prediction Endpoint	High-Performing Model(s)	Software Platform	Key Consideration for Review Inclusion
Persistence (Ready Biodegradability)	Ready Biodegradability IRFMN; Leadscope model; BIOWIN	VEGA; Danish QSAR; EPISUITE	Prediction should be within the model's defined Applicability Domain (AD).
Bioaccumulation (Log Kow)	ALogP; ADMETLab 3.0; KOWWIN	VEGA; ADMETLab; EPISUITE	Qualitative classification (e.g., bioaccumulative) is more reliable than precise numeric output.
Bioaccumulation (BCF)	Arnot-Gobas; KNN-Read Across	VEGA	Models should be used in a weight-of-evidence approach with other data.
Mobility (Log Koc)	OPERA; KOCWIN-Log Kow	VEGA	Model performance can vary significantly based on chemical space.

Integrating NAMs into systematic reviews requires adapting the risk of bias assessment to evaluate the model's Applicability Domain, algorithmic transparency, and validation status [92] [91]. This represents an evolving frontier in methodological rigor for ecotoxicology evidence synthesis.

Diagram 3: Conceptual Signaling Pathway: Environmental Contaminants and Vector Resistance [90].

The Scientist's Toolkit: Essential Reagents & Methodologies for Rigorous Reviews

Conducting a high-quality evidence synthesis requires a suite of conceptual and technical tools. The following table details key "research reagent solutions" for ecotoxicology systematic reviews.

Table 3: Essential Toolkit for Conducting Systematic Reviews in Ecotoxicology

Tool Category	Specific Tool / Resource	Function & Purpose in the Review Process	Key Reference
Protocol & Reporting	PRISMA 2020 Checklist & Statement	Ensuring complete, transparent reporting of the systematic review.	[44] [89]
	PROSPERO Registry	International prospective register for systematic review protocols.	[44] [56]
Question Formulation	PICOS / PICOTS Framework	Structuring a focused, answerable research question.	[44]
Search Strategy	Bibliographic Databases (PubMed, Web of Science, Scopus, TOXLINE)	Identifying published literature comprehensively.	[44] [91]
	Grey Literature Sources (Regulatory reports, theses, conference proceedings)	Minimizing publication bias by locating unpublished data.	[56]
Study Appraisal	SYRCLE's Risk of Bias Tool (for animal studies)	Assessing internal validity of in vivo ecotoxicology studies.	[12]
	OHAT / NTP Risk of Bias Tool	Adapted tool for human and animal observational studies in toxicology.	[12] [56]
Evidence Grading	GRADE Framework	Rating the overall certainty (confidence) in a body of evidence.	[44]
Data Synthesis	R packages (metafor, robvis)	Conducting meta-analysis and creating risk-of-bias visualizations.	[44]
NAM Integration	QSAR Model Repositories (VEGA, EPISUITE)	Providing in silico predictions for persistence, bioaccumulation, and toxicity (PBT).	[92]
	OECD QSAR Validation Principles	Criteria for assessing the regulatory acceptability of (Q)SAR models.	[92]

The empirical comparison is unequivocal: systematic reviews provide a demonstrably more rigorous, transparent, and reproducible method for synthesizing ecotoxicological evidence than traditional narrative reviews. While narrative reviews retain value for providing introductory overviews or theoretical exploration, they lack the structured safeguards necessary for informing high-stakes regulatory and public health decisions [89] [12].

The future of credible ecotoxicology research lies in the widespread adoption of systematic review standards, as outlined in COSTER and other guidelines [56]. This necessitates training researchers in advanced review methodologies, developing field-specific risk-of-bias tools, and creating robust processes for integrating diverse data streams, including those from emerging NAMs [92] [91]. By consistently applying these rigorous protocols, the ecotoxicology community can produce syntheses that truly minimize bias, enhance transparency, and provide a reliable foundation for protecting environmental and human health.

1. Introduction: Systematic vs. Narrative Review in Ecotoxicology

Within ecotoxicology, evidence synthesis is critical for risk assessment and regulatory decision-making. The field employs a spectrum of review methodologies, anchored by two distinct paradigms: the systematic review and the narrative review. A systematic review is defined by a structured, protocol-driven process designed to answer a specific research question with minimal bias, using explicit methods for searching, selecting, appraising, and synthesizing evidence [2]. In contrast, a narrative review provides a qualitative summary and critique of literature on a broader topic, offering insight based on reasoned argumentation and expert wisdom without following a strict, standardized methodology [93].

Systematic Evidence Maps (SEMs) and Interpretive Expert Narratives represent advanced outputs derived from these respective paradigms. An SEM is a form of systematic review product that visually catalogs and characterizes the available research on a broad topic, identifying the quantity, distribution, and key features of evidence (e.g., studied populations, exposures, outcomes) without performing a quantitative synthesis of results [94] [89]. Its primary objective is to map the research landscape to reveal evidence clusters and knowledge gaps [95]. An Interpretive Expert Narrative is the culmination of a narrative review, delivering a synthesized, critical argument that contextualizes findings within theoretical frameworks, debates existing knowledge, and proposes new directions based on the author's expertise and interpretive synthesis of the literature [96] [93].

This analysis provides a comparative technical examination of these two outputs, framing them within the broader methodological thesis of systematic versus narrative review. It details their respective experimental protocols, visualizes their workflows, and delineates their specific applications within ecotoxicological research and chemical risk management [95] [97].

2. Comparative Analytical Framework

The core differences between Systematic Evidence Maps and Interpretive Expert Narratives are rooted in their foundational methodologies, as summarized in Table 1.

Table 1: Comparative Framework of Evidence Maps and Expert Narratives

Analytic Dimension	Systematic Evidence Map (SEM)	Interpretive Expert Narrative
Primary Objective	To systematically catalog and characterize the extent, range, and nature of available evidence; to identify gaps and clusters for future research or systematic reviews [94] [95].	To provide a comprehensive, critical overview of a field; to synthesize themes, debate perspectives, and propose new theoretical models or research directions based on expert interpretation [96] [93].
Methodological Foundation	Adheres to a pre-defined, transparent, and reproducible systematic review methodology. It is a form of knowledge synthesis that uses systematic searching, screening, and data extraction [89] [97].	Follows a flexible, non-standardized approach. While it may describe the literature searched, it lacks a mandatory protocol, predefined search strategy, or formal inclusion/exclusion criteria [2] [93].
Research Question Scope	Broad and exploratory. Aims to "map" a research field rather than answer a narrowly focused question on intervention efficacy [94] [95].	Broad and conceptual. Often seeks to clarify complex issues, track historical developments, or integrate knowledge across disciplines [93].
Search Strategy	Comprehensive and systematic across multiple databases. Search strategy is documented and reproducible [97].	Selective and pragmatic. May focus on seminal works, specific databases, or literature known to the expert author, with potential for selection bias [96].
Critical Appraisal	May or may not include formal quality assessment or risk-of-bias evaluation. If included, it is used to characterize the evidence base, not to exclude studies [89] [97].	Typically does not include formal, systematic quality assessment of included studies. Evaluation is narrative and integrated into the discussion [2].
Synthesis Method	Descriptive synthesis focused on categorizing and counting studies based on coded characteristics (e.g., chemical, organism, endpoint). Results are often presented in searchable databases, matrices, or heat maps [94] [89].	Narrative synthesis through reasoned argument. Integrates findings thematically, chronologically, or conceptually to build a coherent story and advance a point of view [96] [93].
Primary Output Format	Interactive databases, visual matrices, heat maps, and systematic reports with tabular summaries [94] [98].	Scholarly article or book chapter presenting a sustained, written argument [93].
Key Strength	Transparency, reproducibility, and comprehensive visual overview of the evidence landscape. Unbiased identification of research gaps [95] [97].	Depth of insight, conceptual innovation, and ability to contextualize findings within broader theoretical and historical frameworks [93].
Key Limitation	Does not provide synthesized answers regarding the magnitude or direction of effects. Resource-intensive to produce [94] [95].	Susceptible to author selection and confirmation bias. Lack of reproducible methods makes it difficult to audit or verify conclusions [2].
Ideal Application in Ecotoxicology	Priority-setting for research funding, identifying needs for future systematic reviews, scoping chemical classes for regulatory re-evaluation, and providing evidence inventories for stakeholders [95] [97].	Introducing a complex field, critiquing methodological trends, proposing novel hypotheses or frameworks, and synthesizing knowledge from disparate sub-fields to identify overarching principles [93].

3. Experimental Protocols

3.1 Protocol for Developing a Systematic Evidence Map (SEM) The following stepwise protocol, adapted from contemporary guidance, outlines the rigorous process for constructing an SEM in ecotoxicology [97].

• Define Scope and Stakeholder Engagement: Collaboratively establish the map's broad boundaries (e.g., "health effects of per- and polyfluoroalkyl substances (PFAS) in aquatic biota") and engage stakeholders (regulators, funders) to ensure relevance.
• Develop an A Priori Protocol: Document the research question(s), search strategy, and eligibility criteria (PECO: Population, Exposure, Comparator, Outcome) in a publicly accessible registry.
• Execute Systematic Search: Conduct comprehensive searches across multiple bibliographic databases (e.g., PubMed, Web of Science, TOXLINE, Scopus). Use controlled vocabularies and free-text terms. Document the search strategy fully.
• Screen Studies Systematically: Use pre-defined inclusion/exclusion criteria to screen titles/abstracts and full texts, typically with two independent reviewers and reconciliation of conflicts.
• Code and Extract Data: Develop a detailed data extraction codebook. Extract descriptive metadata (author, year) and map-specific variables (chemical studied, species, exposure route, measured endpoint, study design) into a structured database.
• Optional Critical Appraisal: If the SEM aims to inform future syntheses, conduct a risk-of-bias assessment to characterize the methodological quality of the evidence base.
• Visualize and Report: Generate visual outputs such as evidence atlases, heat maps (showing volume of research by chemical/outcome pair), or interactive online databases. Prepare a final report detailing methods, results, and a narrative summary of evidence gaps and clusters.

3.2 Protocol for Constructing an Interpretive Expert Narrative The methodology for a narrative review is less prescriptive but follows a recognized scholarly process [2] [93].

• Topic Formulation and Conceptualization: Define a broad, significant question or theme (e.g., "The evolution of biomarker use in environmental monitoring").
• Exploratory and Iterative Literature Search: Conduct searches based on the author's expertise, using known key papers, selective database searches, and citation snowballing. The search is adaptive, evolving as new themes are discovered.
• Critical Analysis and Thematic Organization: Read selected literature critically. Organize materials conceptually (e.g., by theory, methodological approach, historical period) rather than merely descriptively. Identify central debates, conflicts, and trends.
• Argument Development and Narrative Synthesis: Synthesize the literature into a coherent, persuasive narrative. The author interprets findings, weighs contradictory evidence, and constructs an argument that offers new insights, critiques existing paradigms, or proposes future directions.
• Manuscript Preparation: Structure the narrative to guide the reader through the logical argument, using subheadings to reflect thematic organization. Integrate citations to support the narrative arc rather than to inventory all existing literature.

4. Visualization of Methodological Pathways and Outputs

The logical relationship between primary research, different synthesis methodologies, and their outputs within ecotoxicology is depicted below.

Graph 1: Relationship between research synthesis paradigms and their outputs in ecotoxicology.

The workflow for creating a Systematic Evidence Map involves sequential, protocol-driven steps, as shown below.

Graph 2: Sequential workflow for conducting a Systematic Evidence Map (SEM) [97].

5. The Ecotoxicologist's Toolkit: Essential Reagent Solutions

The production of high-quality evidence syntheses in ecotoxicology relies on specific methodological tools and resources, as detailed in Table 2.

Table 2: Key Research Reagent Solutions for Evidence Synthesis

Tool/Resource Category	Specific Examples	Function in Synthesis
Protocol Registries	PROSPERO, Open Science Framework (OSF)	Provides a platform to register and publish an a priori review protocol, enhancing transparency, reducing duplication, and minimizing reporting bias. Essential for systematic reviews and SEMs [97].
Bibliographic Databases	PubMed/MEDLINE, Web of Science, Scopus, TOXLINE, Embase	Repositories for primary research literature. Comprehensive searching across multiple databases is mandatory for systematic methods to minimize selection bias [94] [97].
Reference Management & Screening Software	DistillerSR, Rayyan, Covidence, EndNote	Software platforms designed to manage citations, facilitate dual-independent screening of titles/abstracts and full texts, resolve conflicts, and track decisions through the review process. Crucial for managing the large volume of records in systematic reviews and SEMs [2] [97].
Data Extraction & Coding Tools	Custom spreadsheets (Excel, Google Sheets), Systematic Review Data Repository (SRDR+)	Structured templates for consistent coding and extraction of predefined variables (PECO, study design, results) from included studies. Ensures reliability and reproducibility of the data collection phase [97].
Risk of Bias Assessment Tools	Cochrane RoB 2, SYRCLE's RoB tool, OHAT RoB Tool	Standardized instruments to evaluate the methodological quality and potential for bias within individual primary studies (e.g., selection, performance, detection bias). Used in systematic reviews and optionally in SEMs [97].
Visualization & Mapping Software	EPPI-Reviewer, Tableau, R (ggplot2, network packages), Python	Generates visual outputs such as heat maps, evidence gap maps, and flow diagrams (e.g., PRISMA). Transforms coded data into accessible visual summaries that are the hallmark of an SEM [94] [89] [97].
Reporting Guidelines	PRISMA (Systematic Reviews), PRISMA-ScR (Scoping Reviews), ROSES (Environmental Systematic Reviews)	Checklists and flow diagrams that standardize the reporting of methods and results, ensuring completeness and transparency in published review articles [89] [97].

6. Applications in Ecotoxicology and Chemical Risk Management

The distinct outputs of SEMs and Expert Narratives serve complementary, non-overlapping roles in the scientific and regulatory ecosystem, as illustrated in Table 3.

Table 3: Application Scenarios in Ecotoxicology

Decision-Making Context	Systematic Evidence Map Application	Interpretive Expert Narrative Application
Chemical Prioritization & Agenda Setting	Identifies which chemicals or chemical classes within a large group (e.g., pesticides, plasticizers) have the most/least available toxicological data, visually highlighting critical evidence gaps. Directly informs agencies on where to allocate research funding or regulatory scrutiny [95].	Critiques the historical, social, or economic factors driving research attention toward certain chemicals while neglecting others. Argues for a re-prioritization based on emerging exposure patterns or novel toxicological paradigms.
Informing Targeted Systematic Reviews	Serves as a scoping tool to determine whether enough evidence exists on a specific chemical-outcome link to justify a full, resource-intensive systematic review and meta-analysis. Increases efficiency by preventing unnecessary reviews [95] [97].	Synthesizes the conclusions of multiple existing systematic reviews across a related set of chemicals (acting as an "umbrella review" [89]) to derive higher-order principles about mechanisms of action or susceptibility factors.
Regulatory Assessment & Policy Development	Provides a comprehensive, auditable inventory of evidence relevant to a broad regulatory question (e.g., endocrine disruption potential of bisphenol analogues). Offers transparency to stakeholders about the volume and type of evidence considered, supporting the legitimacy of subsequent decisions [95].	Develops a persuasive narrative to support a proposed regulatory framework or testing strategy. Integrates evidence from toxicology, ecology, and policy science to argue for a precautionary approach or a new risk assessment model.
Knowledge Synthesis for Emerging Contaminants	Rapidly maps the nascent and fragmented literature on a newly identified contaminant. Categorizes study types, model organisms, and endpoints to show what is known and where the most pressing research needs lie [97].	Provides an authoritative, state-of-the-art overview of an emerging field (e.g., nanotoxicology, mixture toxicity). Connects disparate early findings to established theory, proposes a coherent research agenda, and defines key conceptual challenges.
Education & Foundational Understanding	Less commonly used for direct education, but publicly accessible interactive evidence maps can serve as advanced resources for researchers and students to explore the evidence landscape [94].	The primary vehicle for textbooks, handbook chapters, and introductory review articles. Essential for training new scientists by providing a curated, conceptually organized entry point into a complex field [93].

7. Synthesis and Conclusion

In ecotoxicology, the choice between generating an Objective Evidence Map and an Interpretive Expert Narrative is not a matter of hierarchy but of strategic purpose, dictated by the research question and intended use of the output.

The Systematic Evidence Map is an indispensable tool for evidence-based management of the scientific literature itself. It provides a transparent, objective, and comprehensive audit of the research landscape. Its primary value lies in resource allocation—guiding funding, targeting systematic reviews, and setting regulatory priorities by making evidence gaps and clusters visually undeniable [95] [97]. It is a tool for planning and prioritization.

The Interpretive Expert Narrative remains the premier vehicle for knowledge-based advancement of the field. It delivers context, critique, and conceptual innovation by weaving evidence into a compelling argument. It challenges assumptions, integrates knowledge across domains, and proposes new hypotheses, thereby driving scientific discourse and theoretical development [96] [93].

Therefore, within the thesis of systematic versus narrative review, these outputs are not in opposition but exist in a necessary dialectic. Robust ecotoxicology and chemical risk management require both: the systematic map to objectively inventory what is known and unknown, and the expert narrative to interpret what it means and where we should go next. The most rigorous and impactful research agendas will be informed by the clear-eyed gap analysis of the former and guided by the synthetic wisdom of the latter.

In the field of ecotoxicology, the task of synthesizing scientific evidence to inform regulatory decisions is both critical and complex. With approximately 350,000 chemical substances in widespread use and fewer than 1% regulated under international conventions, the gap between scientific discovery and environmental governance is vast [99]. This context places a premium on robust, transparent methods for evidence synthesis. The choice of review methodology—systematic or narrative—directly impacts the efficiency, credibility, and policy relevance of scientific assessments. This whitepaper provides an in-depth technical comparison of these two fundamental approaches, analyzing their respective strengths and limitations across the key dimensions of time, cost, expertise, and influence on policy within ecotoxicology research.

Comparative Analysis of Strengths and Limitations

The following table synthesizes the core characteristics, strengths, and limitations of systematic and narrative reviews, providing a direct comparison across the specified domains.

Table 1: Comparative Analysis of Systematic and Narrative Reviews in Ecotoxicology

Feature	Systematic Review	Narrative (Traditional) Review
Core Definition	A literature review that uses explicit, pre-specified, and reproducible methods to identify, select, critically appraise, and synthesize all relevant studies on a focused question [12] [2].	A scholarly summary that uses flexible, interpretive methods to describe the current state of knowledge on a broader topic, relying heavily on the author’s expertise and narrative analysis [2] [100].
Primary Objective	To answer a specific, well-defined research question (e.g., using PECO/PICO) with minimal bias, often to directly support decision-making [12] [2].	To provide a broad overview, explore theories, identify debates, and contextualize knowledge within a field [2] [100].
Time Requirement	High. Typically requires >1 year to complete due to rigorous, multi-step protocol [12]. Assessing toxicity for a large set of chemicals can span decades; evaluating 10,000 contaminants with traditional methods is estimated to take a century [99].	Variable, but generally lower. Often completed in months, though authoritative reviews by agencies can also take years [12] [100].
Financial Cost	Moderate to High. Costs are significant due to personnel time for comprehensive searching, data extraction, and analysis. Traditional ecotoxicity tests for a single chemical average $118,000 [99].	Generally Low. Primarily requires the time of the expert author(s), with minimal procedural overhead [12] [100].
Required Expertise	Multidisciplinary Team. Requires subject matter experts, systematic review methodologies, information specialists for literature searches, and statisticians for meta-analysis [12].	Subject Matter Expertise. Primarily relies on the deep disciplinary knowledge and interpretive skill of the author(s) [12] [100].
Influence on Policy & Regulation	High, and explicitly designed for this purpose. Considered the gold standard for evidence-based decision-making. Increasingly mandated by agencies like EFSA, US NTP, and EPA for hazard and risk assessment [12] [101] [102].	Indirect and Variable. Useful for identifying knowledge gaps and framing new issues but seen as less objective for direct regulatory action due to methodological transparency issues [102] [100].
Key Strengths	• Minimizes bias via transparent, a priori protocol.• Enhances reproducibility.• Comprehensive search reduces risk of overlooking evidence.• Formal quality assessment of included studies.• Provides quantitative summary (meta-analysis) when possible [12] [2] [102].	• Interpretive depth and capacity for theoretical innovation.• Holistic synthesis across diverse evidence types and disciplines.• Efficient for exploring broad, complex, or emerging topics.• Effectively frames historical context and future directions [2] [100].
Key Limitations	• Resource-intensive (time, cost, expertise).• Methodological rigidity can hinder exploration of heterogeneous evidence.• Less suited for broad, exploratory questions or integrating non-empirical research [12] [100].	• Susceptible to bias (e.g., selection, confirmation) due to lack of explicit methods.• Low transparency and reproducibility.• Potential for selective citation and incomplete evidence base.• Limited critical appraisal of included studies [12] [102] [100].

Detailed Methodological Protocols

Systematic Review Protocol

The conduct of a systematic review in toxicology and ecotoxicology follows a standardized, multi-step workflow designed to ensure rigor and transparency [12] [101].

Diagram Title: Systematic Review Workflow in Toxicology

• Planning and Team Formation: Assemble a team with complementary expertise in the subject matter, systematic review methodology, statistics, and information science [12] [101].
• Formulate the Research Question: Define a precise question using a structured framework. In ecotoxicology, the PECO framework is standard: Population (e.g., Daphnia magna), Exposure (e.g., microplastic concentration), Comparator (e.g., control group), and Outcome (e.g., mortality, reproduction) [101].
• Develop and Register a Protocol: A priori development of a detailed protocol is mandatory. It must specify: the PECO question, search strategy, study eligibility criteria, risk-of-bias assessment methodology, data extraction plan, and synthesis approach. Registration on platforms like PROSPERO ensures transparency and reduces duplication [12].
• Comprehensive Literature Search: Search multiple electronic databases (e.g., PubMed, Web of Science, Embase, specialized toxicology databases) using a pre-defined, peer-reviewed search strategy. The search must be supplemented by reviewing reference lists and, where appropriate, searching grey literature to mitigate publication bias [12] [103].
• Study Selection (Screening): Apply eligibility criteria in a two-stage process (title/abstract, then full-text) using independent, blinded review by at least two investigators. Disagreements are resolved through consensus or a third reviewer [12].
• Critical Appraisal (Risk-of-Bias Assessment): Assess the internal validity of each included study using a domain-based approach tailored to observational or experimental ecotoxicology data. Key domains include selection bias, exposure characterization (a major challenge in environmental studies), confounding, outcome assessment, and selective reporting [101]. This step informs the weight of evidence contributed by each study.
• Data Extraction: Extract predefined data from included studies using standardized, piloted forms. Data typically includes study identifiers, PECO elements, key results (e.g., effect size, variance), and risk-of-bias judgments [12] [103].
• Evidence Synthesis: Synthesize findings qualitatively (descriptive summary, tables) and, if studies are sufficiently homogeneous, quantitatively via meta-analysis. This involves calculating summary effect estimates (e.g., pooled odds ratio, hazard ratio) and exploring heterogeneity using subgroup analysis or meta-regression [12] [103].
• Interpretation and Reporting: Interpret results in the context of the overall strength and limitations of the evidence. Report the review following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, including a flow diagram [12] [103].
• Plan for Updating: Establish a schedule for future review updates as new evidence emerges [12].

Narrative Review Protocol

The methodology for a narrative review is less codified and more flexible, centering on the author's scholarly synthesis.

Diagram Title: Narrative Review Development Process

• Topic Selection and Scoping: The author defines a broad area of interest, often to explore a developing field, summarize a mature one, or propose a new theoretical framework [100].
• Exploratory Literature Search: Unlike a systematic search, the author uses their knowledge of key papers, seminal authors, and major journals to identify relevant literature. Searches may be iterative, following citation trails [2] [100].
• Critical Evaluation and Theme Identification: The author subjectively evaluates the identified literature, discerning major trends, conflicting results, methodological strengths/weaknesses, and conceptual gaps. This process is implicit and driven by the author's judgment [100].
• Narrative Synthesis: The author constructs a coherent story or argument that synthesizes the literature. This involves organizing themes logically, using selected studies as exemplars, and developing conclusions about the state of the field and future directions [2] [100].
• Structured Reporting: While flexible, narrative reviews often follow a standard manuscript structure (Introduction, Methods, Results, and Discussion) or a structure tailored to the narrative's logic [2].

The Scientist's Toolkit: Essential Research Reagents and Solutions

Table 2: Key Research Reagent Solutions for Ecotoxicology Evidence Synthesis

Item	Category	Function in Research
High-Throughput Toxicology (HTT) Platforms	Experimental Technology	Enable rapid, cost-effective screening of thousands of chemicals for bioactivity, generating data to prioritize substances for full systematic review [99].
PECO/PICO Framework	Methodological Tool	Provides a structured template for formulating a precise, answerable research question, which is the critical first step in a systematic review [101].
Risk-of-Bias (RoB) Tools	Methodological Tool	Standardized checklists (e.g., OHAT, SYRCLE for animal studies) used to critically appraise the internal validity of individual studies, a cornerstone of systematic review [12] [101].
Meta-Analysis Software	Data Analysis Tool	Software packages (e.g., R metafor, Stata metan, RevMan) enable the statistical combination of effect estimates from multiple studies to produce a summary quantitative result [103].
Generative AI (GenAI) Models	Emerging Technology	Potential to assist in automating stages of the review process, such as screening titles/abstracts or drafting sections, though human oversight remains essential [104].
Literature Review Management Software	Workflow Tool	Platforms (e.g., DistillerSR, Rayyan, Covidence) support collaborative management of the systematic review workflow, including reference deduplication, blinded screening, and data extraction [2].

Within the pressing context of ecotoxicology—characterized by a vast and growing universe of unregulated environmental contaminants—the choice of evidence synthesis methodology has profound implications. Systematic reviews offer an indispensable, rigorous framework for producing policy-ready evidence on specific hazards, albeit at significant investment of time and resources. Narrative reviews provide intellectual agility and conceptual insight crucial for navigating emerging fields and shaping research agendas. They are not inherently inferior but serve a different primary purpose. The most robust approach to informing both science and policy often involves their strategic, sequential use: narrative reviews to map complex landscapes and identify critical questions, followed by systematic reviews to definitively answer those questions with minimal bias, thereby effectively bridging the gap between scientific discovery and environmental protection [99].

The field of ecotoxicology is dedicated to understanding the impacts of chemical, physical, and biological stressors on living organisms within ecosystems, with profound implications for environmental protection and public health policy [105]. As the volume of primary research grows, the synthesis of evidence through literature reviews has become indispensable. Traditionally, this synthesis has been dominated by the narrative review, an expert-driven summary of a field [12]. However, there is increasing recognition of the need for more transparent, rigorous, and reproducible methods to support high-stakes regulatory and remediation decisions [105] [101]. This has spurred the adoption of the systematic review, a methodology pioneered in clinical medicine and now being adapted for environmental health and toxicology [12] [56].

This guide establishes a decision framework for choosing between these two fundamental review types within ecotoxicological research. It frames the discussion within the broader thesis that while narrative reviews provide essential context and hypothesis generation, systematic reviews offer a superior structure for minimizing bias and producing reliable, actionable evidence for decision-making [12] [2]. The choice is not merely methodological but strategic, impacting the credibility, utility, and resource demands of the review process.

Defining the Review Types: Core Characteristics and Methodologies

A narrative review provides a qualitative, comprehensive summary of a broad research area. It is characterized by its flexibility, relying on the author's expertise to select, interpret, and synthesize literature. The objective is typically to provide context, explore existing debates, identify knowledge gaps, and speculate on future directions [2]. There is no formal, pre-specified protocol; the structure is often organized thematically or chronologically, following conventions like IMRAD (Introduction, Methods, Results, and Discussion) at the author's discretion [2]. The methodology for literature search and selection is rarely explicit or reproducible, making it difficult for readers to assess potential for selective citation or bias [12] [106].

Systematic Reviews: The Structured Evidence Synthesis

A systematic review is defined by its explicit, pre-defined, and reproducible methodology aimed at minimizing bias. It begins with a precisely framed research question (often structured using PECO: Population, Exposure, Comparator, Outcome) [101]. A publicly registered protocol details the plan for comprehensive literature searching, study selection based on strict eligibility criteria, critical appraisal of individual study quality (risk-of-bias assessment), and data synthesis [12] [56]. Synthesis can be qualitative or quantitative (meta-analysis). The entire process is documented with high transparency, allowing for independent verification and update [2]. Its primary aim is to provide the least biased answer to a specific question to directly inform policy or practice [12].

Table 1: Foundational Comparison of Narrative and Systematic Reviews

Feature	Narrative Review	Systematic Review
Primary Objective	Provide broad overview, context, and hypothesis generation [2].	Answer a specific, focused research question with minimal bias [12].
Research Question	Broad, informal, and often not explicitly stated [12].	Narrow, specific, and framed using structures like PECO [101].
Protocol & Methodology	No pre-registered protocol; methods are implicit and flexible [2].	Pre-registered public protocol; methods are explicit, transparent, and reproducible [56].
Literature Search	Not systematic or comprehensively reported; potential for selective citation [12].	Comprehensive, multi-database search with explicit, documented search strategy [12].
Study Selection	Criteria not specified; selection based on author expertise [12].	Defined by explicit, pre-specified eligibility criteria (inclusion/exclusion) [12].
Critical Appraisal	Usually absent or informal [12].	Formal assessment of risk of bias/study quality for each included study [107] [101].
Data Synthesis	Qualitative, narrative summary [12].	Structured qualitative synthesis; may include quantitative meta-analysis [12].
Key Output	Expert opinion, theoretical framework, identification of research gaps.	Evidence-based conclusion on a specific question, often with confidence rating.

Quantitative Comparison: Prevalence, Impact, and Resource Investment

The adoption patterns of different review types offer insight into their perceived value and practical constraints. A survey of five top medical journals found that non-systematic reviews (narrative and narrative reviews with some methodology) constituted approximately 73% of published reviews, while systematic reviews accounted for only 27% [106]. Notably, narrative reviews were more frequently published in journals with higher impact factors (Risk Ratio = 1.114) [106]. However, systematic reviews received a higher median number of citations (43) compared to the combined narrative group (median 35), suggesting they generate more downstream research activity despite being less prevalent in the highest-tier journals [106].

The resource disparity between the approaches is significant. A narrative review is typically completed in months, whereas a full systematic review often requires more than a year [12]. The expertise required also expands: while a narrative review demands deep subject matter knowledge, a systematic review additionally requires skills in systematic review methodology, advanced literature searching, and often specialized data analysis for meta-analysis [12]. Consequently, costs for systematic reviews are moderate to high, compared to the generally low costs of a narrative review [12].

Table 2: Quantitative Comparison of Output and Resource Requirements

Metric	Narrative Review	Systematic Review	Data Source
Typical Timeframe	Months [12].	>1 Year (usually) [12].	Survey of guidance documents [12].
Relative Prevalence in Top Journals	Higher (73% of sample) [106].	Lower (27% of sample) [106].	Survey of 5 major medical journals [106].
Association with Journal Impact Factor	Positive association (published in higher IF journals) [106].	Inverse association in the sampled journals [106].	Regression analysis [106].
Median Citation Count	35 (narrative* group combined) [106].	43 [106].	Citation analysis from journal survey [106].
Expertise Required	Deep subject matter (Science) [12].	Subject matter + SR methodology, searching, data analysis [12].	Guidance primer [12].
Relative Cost	Low [12].	Moderate to High [12].	Guidance primer [12].

Methodological Protocols for Ecotoxicology

Core Experimental Protocol for a Systematic Review

Conducting a systematic review in ecotoxicology involves a multi-stage, iterative process. The following protocol is synthesized from established guidance like the Cochrane Handbook and the COSTER recommendations for environmental health [12] [56].

• Problem Formulation & Protocol Development: Define a specific PECO question. Develop and publicly register a detailed protocol outlining the entire methodology [56] [101].
• Comprehensive Search Strategy: Search multiple academic databases (e.g., PubMed, Web of Science, Scopus, specialized toxicology databases) using a peer-reviewed search strategy. Supplement with grey literature searches (theses, agency reports) [56]. Document all sources and dates.
• Study Screening & Selection: Use pre-defined eligibility criteria (PECO elements, study type) to screen titles/abstracts, then full texts. This is typically performed by two independent reviewers with conflict resolution [12].
• Data Extraction & Risk-of-Bias Assessment: Extract relevant data using standardized forms. Critically appraise each study using a domain-based risk-of-bias tool appropriate for ecotoxicology (e.g., assessing exposure confirmation, blinding, confounding). The focus should be on internal validity and study sensitivity (the ability to detect a true effect) [107] [101].
• Evidence Synthesis & Integration: Synthesize findings qualitatively (e.g., grouping by outcome or organism) and quantitatively via meta-analysis if studies are sufficiently homogeneous. Integrate evidence by considering the direction and magnitude of potential biases, consistency across studies, and biological plausibility [101].
• Reporting: Report following PRISMA guidelines. Discuss limitations, strength of evidence, and implications for research and policy [12].

Key Experimental Considerations in Ecotoxicology

Ecotoxicology presents unique challenges requiring methodological adaptation:

• Exposure Assessment: A major source of potential bias is exposure misclassification. Reviews must critically evaluate how exposure was measured and confirmed (e.g., analytical verification of test concentrations), which is often poorly reported [105] [101].
• Multiple Evidence Streams: Reviews must integrate data from diverse study types (in vivo, in vitro, (Q)SAR, field observations), each requiring specific quality appraisal tools [107].
• Heterogeneity: Variability in test species, endpoints, and exposure scenarios is high. The framework must plan for this heterogeneity and not force inappropriate meta-analysis [12] [101].

Flowchart: Comparative Workflows for Systematic and Narrative Reviews

Table 3: Research Reagent Solutions for Conducting Reviews

Tool / Resource	Function	Relevance to Review Type
PECO Framework	Structures the research question into Population, Exposure, Comparator, Outcome elements [101].	Systematic Review: Essential for defining scope & eligibility. Narrative Review: Can help informally scope topic.
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) Statement & Flow Diagram	An evidence-based minimum set of items for reporting systematic reviews. The flow diagram documents the study selection process [12].	Systematic Review: Mandatory reporting guideline. Narrative Review: Not required, but can improve transparency if used.
Risk-of-Bias (RoB) Assessment Tools	Structured tools to evaluate internal validity of studies (e.g., for in vivo, in vitro, or epidemiological studies) [107] [101].	Systematic Review: Core component for critical appraisal. Narrative Review: Seldom used formally.
Reference Management Software (e.g., EndNote, Zotero)	Manages bibliographic references and PDFs, facilitates de-duplication and screening.	Both: Essential for handling literature.
Systematic Review Management Platforms (e.g., DistillerSR, Rayyan)	Web-based platforms designed to support the entire SR process: screening, data extraction, and RoB assessment with multiple reviewers [2].	Systematic Review: Highly recommended for team-based reviews. Narrative Review: Typically not needed.
Meta-Analysis Software (e.g., R with metafor, Stata)	Statistical software for conducting quantitative synthesis (meta-analysis) of effect estimates from multiple studies.	Systematic Review: Required for quantitative synthesis. Narrative Review: Not applicable.
Protocol Registries (e.g., PROSPERO, Open Science Framework)	Public platforms to register and time-stamp a review protocol before it begins [56].	Systematic Review: Best practice to reduce bias and duplication. Narrative Review: Not standard practice.

Decision Framework: Choosing the Right Review Type for Your Ecotoxicology Research

The choice between a systematic and narrative review should be driven by the research goal, available resources, and the intended use of the output. The following framework guides this decision.

Flowchart: Decision Pathway for Choosing a Review Type

Guiding Principles:

• Choose a Systematic Review when: The objective is to answer a precise question to directly inform policy, regulation, or specific remediation strategies (e.g., "What is the effect concentration (EC50) of chemical X on the reproduction of freshwater cladocerans?"). This is mandatory when a comprehensive, unbiased summary is needed, and resources allow for the rigorous process [12] [101].
• Choose a Narrative Review when: The objective is to explore a broad, complex, or emerging field, provide historical context, synthesize disparate ideas, or formulate novel theoretical frameworks (e.g., "Advances and challenges in the bioremediation of hydrocarbon-polluted soils") [2] [31]. It is also appropriate when time or resources are severely limited, and an expert perspective is the immediate need.
• Hybrid Approaches: A "systematized" narrative review can incorporate some elements of systematic methodology (e.g., explicit search strategy) to improve transparency without meeting the full requirements of a systematic review [106].

Within the evolving landscape of ecotoxicology, both narrative and systematic reviews serve vital but distinct purposes. The narrative review remains the instrument of choice for mapping uncharted territory, educating newcomers to a field, and crafting the theoretical narratives that drive science forward. In contrast, the systematic review is the essential tool for building a robust, trustworthy evidence base for environmental protection decisions, offering unparalleled transparency and rigor.

The central thesis of this framework is that the choice is consequential. Opting for a narrative review when a systematic one is warranted risks perpetuating bias and undermining evidence-based policy. Conversely, demanding a full systematic review for an exploratory, hypothesis-generating question is an inefficient use of resources. By applying the decision criteria outlined—research goal, required certainty, and resource availability—ecotoxicologists can strategically select the review methodology that maximizes the scientific and societal value of their synthesis work, thereby strengthening the entire discipline's foundation for protecting environmental health.

The proliferation of anthropogenic contaminants, such as pesticides and per- and polyfluoroalkyl substances (PFAS), presents a complex challenge for environmental risk assessment and regulatory policy. Navigating the vast, interdisciplinary body of primary literature requires robust, fit-for-purpose evidence synthesis methodologies [108]. The choice between a systematic review and a narrative review fundamentally shapes the questions that can be answered, the transparency of the process, and the strength of the conclusions drawn [109] [108].

Systematic reviews employ predefined, protocol-driven methodologies to minimize bias, comprehensively identify all relevant evidence, and provide reproducible conclusions. They are essential for quantitative meta-analyses and for informing definitive risk assessments and regulatory guidelines [108] [110] [111]. In contrast, narrative reviews offer a more flexible, exploratory synthesis of the literature, useful for identifying research trends, mapping emerging fields (e.g., airborne pesticide fate), and framing complex problems where data are heterogeneous or scarce [109]. This guide examines the application of these complementary approaches through contemporary case studies on pesticides and PFAS, providing researchers with a technical framework for selecting and executing appropriate evidence synthesis strategies.

Methodological Framework: Systematic vs. Narrative Review Approaches

The foundational step in any evidence synthesis is the deliberate selection of a methodological framework aligned with the review's purpose. The table below contrasts the defining characteristics of systematic and narrative reviews in ecotoxicology.

Table 1: Comparison of Systematic Review and Narrative Review Methodologies in Ecotoxicology

Aspect	Systematic Review	Narrative (Traditional) Review
Primary Objective	Answer a specific, focused research question; often for quantitative synthesis (meta-analysis) or definitive hazard identification.	Provide a broad overview, explore concepts, identify trends, or contextualize a field.
Protocol & Registration	Mandatory. A detailed a priori protocol (e.g., on PROSPERO) defines search strategy, inclusion criteria, and analysis plan [108] [112].	Not required. Scope and methodology may evolve during the writing process.
Search Strategy	Comprehensive, reproducible, and documented. Searches multiple databases, grey literature, and uses explicit search strings [108] [111].	Often selective and non-exhaustive. May focus on key journals or seminal papers.
Study Selection & Bias Minimization	Explicit, pre-defined inclusion/exclusion criteria applied by multiple independent reviewers to minimize selection bias [108].	Implicit or subjective criteria applied by the author(s).
Critical Appraisal	Formal quality assessment of included studies using standardized tools (e.g., AMSTAR 2 for reviews, risk-of-bias tools for experiments) [108].	Variable; often informal discussion of study strengths and limitations.
Data Synthesis	Structured synthesis, often quantitative (meta-analysis). Focus on effect sizes, confidence intervals, and exploration of heterogeneity [110].	Qualitative, thematic synthesis. Narrative summary of findings from selected literature.
Output & Reproducibility	High reproducibility. Results in a database of included studies, flow diagram (PRISMA), and transparent reporting [111].	Low reproducibility. Represents the author’s interpretative synthesis of the field.
Typical Application	Informing regulatory decisions, deriving consensus effect thresholds, validating testing strategies (e.g., total PFAS analysis) [111].	Scoping emerging issues (e.g., atmospheric pesticide dispersion), framing research agendas, educational summaries [109].

A systematic evidence map, a subtype of systematic review, is increasingly used to chart the breadth of available literature. A 2024 map of 175 PFAS systematic reviews revealed a strong focus on human health outcomes, with significant gaps in wildlife and environmental compartment research, demonstrating how this method can guide future primary research [108].

Synthesis of Experimental Evidence: Protocols and Data Integration

Synthesizing primary experimental data requires understanding standardized protocols and developing methods for integrating diverse endpoints. The following workflows and data tables are central to this process.

Experimental Workflows for Contaminant Assessment

The pathway from primary research to synthesized evidence involves standardized steps for both field monitoring and laboratory toxicity testing.

Quantitative Synthesis of Pesticide Effects on Non-Target Species

A 2025 meta-analysis of 1,705 studies provides a comprehensive quantitative synthesis of pesticide effects, demonstrating the power of systematic review [110]. The following table summarizes key effect sizes.

Table 2: Synthesis of Pesticide Effects on Non-Target Organisms: Meta-Analysis Results [110]

Taxonomic Group	Pesticide Class	Endpoint	Mean Effect Size (lnRR)	95% Confidence Interval	P-value	Interpretation
Animals	All Classes	Growth	-0.091	[-0.128, -0.055]	<0.001	Significant decrease
		Reproduction	-0.395	[-0.464, -0.325]	<0.001	Significant decrease
	Insecticides	Reproduction	-0.466	[-0.550, -0.382]	<0.001	Largest negative effect
	Fungicides	Behaviour	-0.419	[-0.693, -0.145]	0.003	Significant alteration
Plants	Herbicides	Growth	-0.482	[-0.602, -0.362]	<0.001	Very large decrease
	All Classes	Reproduction	-0.346	[-0.538, -0.155]	<0.001	Significant decrease
Microorganisms	All Classes	Growth	-0.445	[-0.630, -0.260]	<0.001	Significant decrease

lnRR (Log Response Ratio): A negative value indicates a decrease in the endpoint (e.g., growth, reproduction) in treated groups compared to controls. An lnRR of -0.482 for plant growth under herbicides corresponds to an approximately 38% reduction [110].

IntegratingIn VitroandIn SilicoData: The PFAS Case

For emerging contaminants like PFAS with limited in vivo toxicity data, systematic reviews integrate novel data streams. A 2025 study on Great Lakes tributaries used Polar Organic Chemical Integrative Samplers (POCIS) to derive time-weighted water concentrations for 16 PFAS [113]. These were compared to multiple toxicity thresholds:

• Published Water Quality Guidelines: For 9 PFAS.
• Primary Literature Apical Endpoints (from ECOTOX Knowledgebase): For 10 PFAS.
• In vitro High-Throughput Screening (EPA ToxCast): Activity Concentration at Cutoff (ACC) for 14 PFAS [113].

Exposure-Activity Ratios (EARs; exposure concentration/ACC) were summed for chemicals affecting common targets to assess mixture effects, which showed potency increases up to 5.6-fold over individual chemicals [113]. This integrated approach prioritized PFOS, PFOA, PFHxS, PFBS, and PFNA for further investigation.

Case Study Applications: From Environmental Fate to Human Health

Narrative Review: Atmospheric Dispersion of Pesticides

A 2025 narrative review on airborne pesticides in France exemplifies the use of this method to scope a complex environmental fate issue [109]. Without attempting a quantitative meta-analysis, it synthesized knowledge on:

• Emission Pathways: Spray drift, volatilization, and wind erosion.
• Influencing Factors: Physicochemical (vapor pressure), environmental (wind), and operational (droplet size).
• Monitoring Data: Highlighted seasonal peaks and the persistent detection of banned substances like lindane.
• Research Gaps: Identified fragmented modeling approaches and a lack of integrated monitoring frameworks [109].

This synthesis provides a conceptual model essential for designing future systematic research and monitoring programs.

Systematic Review & Meta-Analysis: Pesticide Mixture Toxicity

A 2025 laboratory study systematically testing pesticide mixtures on non-target plants (Sinapis alba, Lepidium sativum) followed a protocol-driven approach [114]. It evaluated six active ingredients (e.g., glyphosate, azoxystrobin) individually and in mixtures at field-realistic concentrations.

Key Protocol Steps [114]:

• Test System: Phytotoxkit microbiotests in controlled climate chambers.
• Exposure: Seeds placed on filter paper impregnated with pesticide solutions.
• Endpoints: Seed germination and root/shoot growth after 3-5 days.
• Mixture Modeling: Observed effects compared to predictions from Additive, Dominance, and Multiplicative models.

Finding: All mixtures exhibited stronger phytotoxic effects than individual components, with the additive model being the best predictor, though observed synergies suggested it may still underestimate risk [114]. This work feeds directly into systematic reviews of mixture effects.

Systematic Evidence Mapping: The PFAS Literature Landscape

A 2024 systematic evidence map analyzed 175 systematic reviews on 35 PFAS [108]. It employed a modified AMSTAR-2 checklist to evaluate methodological quality, revealing that reviews on environmental and wildlife topics were fewer and often less methodologically rigorous than those on human health. This map created an interactive database (https://hi-this-is-lorenzo.shinyapps.io/PFASSEMShiny_App/), a tool that directly enables other researchers to access the synthesized evidence base efficiently [108].

Meta-Analysis: Pesticide Exposure and Thyroid Cancer Risk

A 2025 meta-analysis of 13 epidemiological studies demonstrates systematic review for human health outcomes [115]. It found a positive association between exposure to insecticides, herbicides, and fungicides and thyroid cancer risk, with a higher risk for women. The proposed mechanistic pathway involves endocrine disruption, oxidative stress, and inflammation [115].

The Scientist's Toolkit: Essential Reagents and Methods

Table 3: Key Research Reagent Solutions and Materials for Contaminant Evidence Synthesis

Item / Solution	Function in Research	Application in Evidence Synthesis
Polar Organic Chemical Integrative Sampler (POCIS)	Passive in situ device that accumulates hydrophilic contaminants over time, providing a time-weighted average (TWA) concentration [113].	Generates field exposure data for comparison with toxicity thresholds; essential for calculating Exposure-Activity Ratios (EARs).
Combustion Ion Chromatography (CIC)	Analytical method for determining Total Fluorine (TF) in solid and liquid samples by combustion followed by ion analysis [111].	Critical for "total PFAS" assessments in systematic reviews of environmental contamination, bridging targeted and non-targeted analysis.
Total Oxidizable Precursor (TOP) Assay	Chemical oxidation method that converts unknown PFAS precursors into measurable perfluoroalkyl acids (PFAAs) [111].	Used in reviews to estimate the contribution of precursor compounds to overall PFAS burden, addressing a key data gap.
Phytotoxkit Microbiotest	Standardized plant toxicity test using seeds on filter paper in transparent containers [114].	Provides reproducible phytotoxicity data (germination, growth) for primary studies that are amenable to meta-analysis of plant responses.
EPA ToxCast Database & ACC Values	Repository of in vitro high-throughput screening data. The Activity Concentration at Cutoff (ACC) is a consistent potency metric [113].	Enables screening-level risk assessment for thousands of chemicals, including data-poor PFAS, by calculating EARs for integrative reviews.
ECOTOX Knowledgebase	EPA-curated database summarizing peer-reviewed toxicity studies for aquatic and terrestrial life [113].	Primary source for extracting apical endpoint toxicity values (LC50, EC50) for use in ecological risk assessments within systematic reviews.

Data Integration and Visualization for Decision-Making

Effective synthesis for regulatory or management purposes requires translating complex data into actionable insights. The diagram below illustrates the logic flow for assessing mixture risks based on integrated evidence streams, a central challenge in modern ecotoxicology [113] [114] [110].

Synthesizing evidence on contaminants is not a one-size-fits-all endeavor. The choice of method must be deliberate:

• Use Systematic Reviews to answer specific questions for risk assessment (e.g., "What is the EC50 for compound X on species Y?"), to perform meta-analysis of effect sizes, or to create auditable evidence maps for priority-setting [108] [110] [111].
• Use Narrative Reviews to explore emerging issues, map interdisciplinary connections, frame broad research questions, or summarize the state of knowledge for a non-specialist audience [109].

Best Practices for Researchers:

• Define the Purpose Early: Align the review question with the appropriate synthesis method.
• Embrace Transparency: For systematic reviews, publish protocols and follow PRISMA guidelines. For narrative reviews, clearly state the scope and literature selection rationale.
• Integrate Data Streams: Combine traditional apical endpoint data with novel HTS and in silico data to address data gaps for emerging contaminants [113].
• Account for Complexity: Systematically address mixture effects and interactions, which standard risk assessments often miss [114] [110].
• Leverage Technology: Utilize interactive databases and evidence maps to make synthesized knowledge accessible and actionable for the broader scientific and regulatory community [108].

Conclusion

Systematic and narrative reviews serve fundamentally different, yet complementary, purposes in ecotoxicology. Systematic reviews provide a transparent, methodologically rigorous, and reproducible foundation for causal inference and high-stakes decision-making in research and regulation, though they require significant resources and careful adaptation to the field's unique challenges, such as complex exposure assessment [citation:1][citation:5][citation:7]. Narrative reviews offer valuable, timely expert perspectives to map broad fields, generate hypotheses, and communicate to diverse audiences, but their conclusions are more susceptible to bias due to less formalized methods. The future of evidence synthesis in ecotoxicology lies not in the supremacy of one approach over the other, but in the judicious selection of the appropriate tool based on the specific question, available resources, and intended use of the review. Advancing this field requires wider adoption of reporting standards, continued development of ecotoxicology-specific methodologies, and education on the strengths and limitations of each synthesis type to ensure that the conclusions drawn from scientific evidence are as robust and actionable as possible for protecting environmental and human health [citation:3][citation:4][citation:6].

References

• 1. Time to challenge the spurious hierarchy of systematic over ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6001568/]
• 2. The Difference Between Narrative Review and Systematic ... [https://www.distillersr.com/resources/systematic-literature-reviews/the-difference-between-narrative-review-and-systematic-review]
• 3. Guidelines for performing systematic reviews in the ... [https://www.sciencedirect.com/science/article/pii/S0273230017303100]
• 4. Policies and Guidelines - Environment International [https://www.sciencedirect.com/journal/environment-international/about/policies-and-guidelines/guidance-notes-for-authors-of-narrative-reviews]
• 5. The methodological rigour of systematic reviews in ... [https://pubmed.ncbi.nlm.nih.gov/35793403/]
• 6. 8. Synthesize Your Results - Systematic Reviews [https://hslguides.osu.edu/systematic_reviews/synthesize]
• 7. Synthesise - Systematic reviews - All guides at RMIT University [https://rmit.libguides.com/systematicreviews/synthesise]
• 8. Quantitative synthesis in systematic reviews - PubMed - NIH [https://pubmed.ncbi.nlm.nih.gov/9382404/]
• 9. PRISMA 2020 flow diagram [https://www.prisma-statement.org/prisma-2020-flow-diagram]
• 10. 1.1 History of toxicology [https://fiveable.me/toxicology/unit-1/history-toxicology/study-guide/FKXmYZyTO53Warh1]
• 11. Poisons and the development of Toxicology in the 19th ... [https://www.rcseng.ac.uk/library-and-publications/library/blog/poisons-and-toxicology/]
• 12. A primer on systematic reviews in toxicology - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC5489636/]
• 13. Quantitative systems toxicology - PMC - PubMed Central - NIH [https://pmc.ncbi.nlm.nih.gov/articles/PMC5754001/]
• 14. Scientific integrity issues in Environmental Toxicology and Chemistry... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7313240/]
• 15. Working with Quantitative Data - LibGuides ... [https://libguides.law.ucla.edu/c.php?g=1382972&p=10262492]
• 16. Types of Quantitative Research Methods and Designs [https://www.gcu.edu/blog/doctoral-journey/types-quantitative-research-methods-designs]
• 17. Increasing the reliability and reproducibility of aquatic ... [https://www.sciencedirect.com/science/article/abs/pii/S0147651318305311]
• 18. Scope and Delimitations in Research - Enago Academy [https://www.enago.com/academy/scope-and-delimitations/]
• 19. Understanding the Scope and Limitations of Expert Opinions [https://www.expertinstitute.com/resources/insights/understanding-the-scope-and-limitations-of-expert-opinions/]
• 20. Considerations for Causal Inference Studies - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC12060746/]
• 21. Causal Inference - an overview [https://www.sciencedirect.com/topics/mathematics/causal-inference]
• 22. Causal inference [https://en.wikipedia.org/wiki/Causal_inference]
• 23. Research Advance of Causal Inference in Clinical Medicine [https://pmc.ncbi.nlm.nih.gov/articles/PMC12077415/]
• 24. A primer on systematic reviews in toxicology [https://link.springer.com/article/10.1007/s00204-017-1980-3]
• 25. Identifying the PECO: A framework for formulating good ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6908441/]
• 26. Doing a systematic review: Step 1: creating a protocol ... [https://libguides.reading.ac.uk/systematic-review/protocol]
• 27. Systematic Review Guide: Step 1: Creating a Protocol (PICO ... [https://wit-ie.libguides.com/c.php?g=711810&p=5139961]
• 28. A narrative review of recent tools and innovations toward ... [https://www.sciencedirect.com/science/article/pii/S186592172300140X]
• 29. Draft Protocol for Systematic Review in TSCA Risk ... [https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/draft-protocol-systematic-review-tsca-risk-evaluations]
• 30. Trifluoroacetic Acid: A Narrative Review on Physico- ... [https://www.mdpi.com/2076-3298/12/8/277]
• 31. Advanced Journal of Environmental Sciences (AJES) [https://zapjournals.com/Journals/index.php/ajes/article/view/2374]
• 32. A comprehensive guide to conduct a systematic review and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12366998/]
• 33. Mendeley - Reference Management Software [https://www.mendeley.com/]
• 34. Tools to Manage a Systematic Review - Research Guides [https://unmc.libguides.com/systematicreview/tools]
• 35. Cite RevMan in a reference list - RMW Knowledge Base [https://documentation.cochrane.org/revman-kb/cite-revman-in-a-reference-list-110242006.html]
• 36. 7 Best AI Tools for Literature Review in 2025 [https://paperguide.ai/blog/ai-tools-for-literature-review/]
• 37. Principles and framework for assessing the risk of bias for ... [https://environmentalevidencejournal.biomedcentral.com/articles/10.1186/s13750-022-00264-0]
• 38. Risk of Bias Tool - National Toxicology Program - NIH [https://ntp.niehs.nih.gov/research/assessments/noncancer/riskbias]
• 39. Study sensitivity: Evaluating the ability to detect effects in ... [https://pubmed.ncbi.nlm.nih.gov/27156196/]
• 40. Principles and framework for assessing the risk of bias for ... [https://assessments.epa.gov/risk/document/&deid%3D359635]
• 41. Risk of bias tools - ROBINS-E tool [https://www.riskofbias.info/welcome/robins-e-tool]
• 42. Species sensitivity distribution modeling for ecotoxicity ... [https://www.sciencedirect.com/science/article/pii/S004896972502515X]
• 43. Species Sensitivity to Toxic Substances: Evolution, Ecology ... [https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2020.588380/full]
• 44. Methodological Standards for Conducting High-Quality ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12383630/]
• 45. Systematic Reviews and Meta-Analyses: Get Started [https://guides.lib.vt.edu/SRMA/start]
• 46. How to Present Qualitative Data (Expert Tips + Tools) [https://heymarvin.com/resources/how-to-present-qualitative-data/]
• 47. Ecotoxicology and Environmental Safety | Vol 294, 1 April ... [https://www.sciencedirect.com/journal/ecotoxicology-and-environmental-safety/vol/294/suppl/C]
• 48. How to choose colors for data visualizations [https://www.atlassian.com/data/charts/how-to-choose-colors-data-visualization]
• 49. Ten simple rules to colorize biological data visualization - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC7561171/]
• 50. Mastering The Art of Data Visualization Color Palettes [https://www.datylon.com/blog/a-guide-to-data-visualization-color-palette]
• 51. A proposed framework for the systematic review and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4944316/]
• 52. Ten challenges for improved ecotoxicological testing in ... [https://www.sciencedirect.com/science/article/abs/pii/S0147651305002940]
• 53. Critical assessment and integration of separate lines of ... [https://link.springer.com/article/10.1007/s00204-019-02547-x]
• 54. Bayesian network-based framework for exposure-response ... [https://ehjournal.biomedcentral.com/articles/10.1186/s12940-019-0461-y]
• 55. A plausibility database summarizing the level of evidence ... [https://www.sciencedirect.com/science/article/pii/S143846392300202X]
• 56. Recommendations for the conduct of systematic reviews in ... [https://www.sciencedirect.com/science/article/pii/S016041202031881X]
• 57. A meta-review of transparency and reproducibility-related ... [https://link.springer.com/article/10.3758/s13428-021-01644-z]
• 58. Citation Errors in Scientific Research and Publications [https://pmc.ncbi.nlm.nih.gov/articles/PMC10307651/]
• 59. Manuscript Referencing Errors and Their Impact on Shaping ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7405306/]
• 60. Defining Bias #8: Lack of Transparency [https://honestreporting.com/news-literacy-defining-bias-lack-of-transparency/]
• 61. Appraising systematic reviews: a comprehensive guide to ... [https://www.frontiersin.org/journals/research-metrics-and-analytics/articles/10.3389/frma.2023.1268045/full]
• 62. Reporting of systematic reviews and meta-analysis of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8885144/]
• 63. Methodological Issues in Systematic Reviews and Meta ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12309503/]
• 64. Methods to systematically review and meta-analyse ... [https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-018-0495-9]
• 65. Use of Data Visualization To Present Findings of the Living ... [https://www.ncbi.nlm.nih.gov/books/NBK590592/]
• 66. a visual tool for overviews of systematic reviews [https://www.sciencedirect.com/science/article/pii/S0895435625002045]
• 67. Submission guidelines | Ecotoxicology [https://link.springer.com/journal/10646/submission-guidelines]
• 68. an updated guideline for reporting systematic reviews - PubMed [https://pubmed.ncbi.nlm.nih.gov/33782057/]
• 69. Recommendations for the conduct of systematic reviews in ... [https://21centurytoxicology.com/publications/recommendations-for-the-conduct-of-systematic-reviews-in-toxicology-and-environmental-health-research-coster/]
• 70. Recommendations for the conduct of systematic reviews in ... [https://ui.adsabs.harvard.edu/abs/2020EnInt.14305926W/abstract]
• 71. An updated guideline for reporting systematic reviews [https://www.equator-network.org/reporting-guidelines/prisma/]
• 72. LibGuides: Creating a PRISMA flow diagram: PRISMA 2020 [https://guides.lib.unc.edu/prisma]
• 73. PRISMA for Scoping Reviews [https://www.prisma-statement.org/scoping]
• 74. Guideline for reporting systematic reviews of outcome ... [https://pubmed.ncbi.nlm.nih.gov/38849061/]
• 75. Guideline for reporting systematic reviews of outcome ... [https://www.sciencedirect.com/science/article/pii/S089543562400177X]
• 76. PRISMA statement [https://www.prisma-statement.org/]
• 77. How to make a systematic review live up to its name [https://atm.amegroups.org/article/view/113532/html]
• 78. The role of reviewers in the era of systematic ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12499546/]
• 79. Guide for authors - Ecotoxicology and Environmental Safety [https://www.sciencedirect.com/journal/ecotoxicology-and-environmental-safety/publish/guide-for-authors]
• 80. Guide for Authors | Environmental Chemistry and ... [https://www.keaipublishing.com/en/journals/environmental-chemistry-and-ecotoxicology/guide-for-authors/]
• 81. Improving the Quality of Toxicology and Environmental ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9472299/]
• 82. Narrative vs. Systematic Literature Review: Understanding ... [https://www.sourcely.net/post/narrative-vs-systematic-literature-review]
• 83. Application of new approach methodologies for nonclinical ... [https://www.nature.com/articles/s41573-025-01182-9]
• 84. New Approach Methods (NAMs) [https://www.fda.gov/food/toxicology-research/new-approach-methods-nams]
• 85. Systematic Review Process: Types of Reviews - AAH Library [https://library.aah.org/guides/systematicreview/types]
• 86. Common Types of Reviews [https://laneguides.stanford.edu/types-of-reviews]
• 87. HPLC 2025 Preview: The Present and Future of ... [https://www.chromatographyonline.com/view/hplc-2025-preview-the-present-and-future-of-automation-in-analytical-laboratories]
• 88. 8 Key Industrial Automation Trends in 2025 [https://www.rockwellautomation.com/en-us/company/news/the-journal/8-key-industrial-automation-trends-in-2025.html]
• 89. Understanding different types of review articles: A primer for ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12118787/]
• 90. Ecotoxicological implications of environmental contaminants ... [https://academic.oup.com/etc/advance-article/doi/10.1093/etojnl/vgaf209/8234492]
• 91. Mapping the ethical and sustainable transition in toxicology [https://link.springer.com/article/10.1007/s00204-025-04209-7]
• 92. Comparative Study of (Q)SAR Models for Predicting the ... [https://www.sciencedirect.com/science/article/pii/S3050620425000491]
• 93. Types of reviews [https://www.phdontrack.net/search-review-and-write/types-of-reviews/]
• 94. What is an evidence map? A systematic review of published ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4750281/]
• 95. Systematic evidence maps as a novel tool to support ... [https://www.sciencedirect.com/science/article/pii/S0160412019310323]
• 96. What are the different types of literature review? [https://www.litmaps.com/learn/what-are-the-different-types-of-literature-review]
• 97. Guidance to undertaking systematic evidence maps [https://www.sciencedirect.com/science/article/pii/S0160412025005781]
• 98. Mapping Review - Which review is that? A guide to review types [https://unimelb.libguides.com/whichreview/mappingreview]
• 99. Time to strengthen the governance of new contaminants in ... [https://www.nature.com/articles/s41467-025-63217-4]
• 100. Types of Literature Reviews: Methods, Strengths & Limitations [https://insights.pluto.im/types-of-literature-reviews-methods-strengths-and-limitations/]
• 101. Challenges and recommendations on the conduct of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7666644/]
• 102. Reviews in environmental health: How systematic are they? [https://www.sciencedirect.com/science/article/pii/S0160412021000982]
• 103. Use of Systematic Review and Meta-Analysis in ... [https://link.springer.com/article/10.1007/s40572-015-0062-z]
• 104. A Review of the Applications, Benefits, and Challenges ... [https://www.sciencedirect.com/science/article/pii/S2666027X25000180]
• 105. How we can make ecotoxicology more valuable to ... [https://www.sciencedirect.com/science/article/abs/pii/S004896971631614X]
• 106. A survey of prevalence of narrative and systematic reviews in ... [https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-017-0453-y]
• 107. Guidance on assessing the methodological and reporting ... [https://www.sciencedirect.com/science/article/pii/S0160412016300903]
• 108. A research synthesis of humans, animals, and ... [https://www.sciencedirect.com/science/article/pii/S016041202400446X]
• 109. From application in the fields to dispersion in the atmosphere [https://www.sciencedirect.com/science/article/abs/pii/S1309104225004222]
• 110. Pesticides have negative effects on non-target organisms [https://www.nature.com/articles/s41467-025-56732-x]
• 111. A systematic review of methods for the analysis of total per [https://www.sciencedirect.com/science/article/pii/S0048969725002785]
• 112. Environmental and Human Health Impacts of Agricultural ... [https://www.mdpi.com/1660-4601/22/11/1683]
• 113. Potential for biological effects of per- and polyfluoroalkyl ... [https://academic.oup.com/etc/article/44/6/1706/7943031]
• 114. Assessing the ecotoxicological effects of pesticides on non ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12380957/]
• 115. Literature Review Finds Heightened Risks for Thyroid ... [https://beyondpesticides.org/dailynewsblog/2025/11/literature-review-finds-heightened-risks-for-thyroid-cancer-in-women-with-pesticide-exposure/]