Meta-Analysis and Systematic Reviews

Lesson 1 • Role of Meta-Analysis in Research

Explains how meta-analysis quantitatively pools study results to increase precision. Connects statistical synthesis to the broader goal of reducing uncertainty in a field.

Lesson 2 • What Is a Systematic Review

Defines systematic reviews and contrasts them with narrative and scoping reviews. Anchors the chapter by establishing why reproducibility and transparency matter in evidence synthesis.

Lesson 3 • Ethical and Reporting Standards

Introduces reporting guidelines and registration requirements for systematic reviews. Establishes professional norms students must follow throughout the course.

Lesson 1 • Scoping and Feasibility Assessment

Guides a preliminary literature scan to estimate available evidence before full protocol commitment. Prevents resource waste on questions with insufficient or overly abundant literature.

Lesson 2 • Structured Question Frameworks

Introduces PICO, PICOS, and SPIDER frameworks for decomposing research questions. Correct framing prevents scope creep and ensures eligibility criteria are internally consistent.

Lesson 3 • Writing and Registering a Protocol

Covers the structure of a formal review protocol and submission to public registries. Registration locks in methods before data collection, reducing outcome-reporting bias.

Lesson 1 • Identifying Relevant Databases

Maps major bibliographic databases by discipline and coverage. Selecting the right combination of sources is the first step toward comprehensive retrieval.

Lesson 2 • Documenting and Reporting Searches

Establishes standards for recording search dates, strings, and result counts. Transparent documentation enables replication and satisfies PRISMA reporting requirements.

Lesson 3 • Searching Beyond Databases

Extends retrieval to grey literature, reference lists, and expert contacts. Supplementary searching reduces publication bias from database-only strategies.

Lesson 1 • Designing Data Extraction Forms

Teaches construction of standardized forms capturing study characteristics and outcomes. Well-designed forms reduce extraction errors and ensure consistent data across reviewers.

Lesson 2 • Conducting and Verifying Extraction

Implements dual extraction and verification workflows to minimize transcription errors. Accurate data are the foundation of valid meta-analytic calculations in later chapters.

Lesson 3 • Screening Titles and Abstracts

Covers two-reviewer independent screening and conflict-resolution procedures. Systematic screening prevents both over-inclusion and inadvertent exclusion of eligible studies.

Lesson 1 • Tools for Observational Studies

Covers ROBINS-I and Newcastle-Ottawa Scale for non-randomized designs. Observational studies require different bias domains than trials, reflecting confounding risks.

Lesson 2 • Integrating Quality Into Synthesis

Translates bias judgments into sensitivity analyses and GRADE certainty ratings. Quality assessment shapes the strength of conclusions drawn from pooled estimates.

Lesson 3 • Tools for Randomized Trials

Applies the Cochrane Risk of Bias 2 tool to randomized controlled trials. Structured domain-based assessment replaces subjective quality scoring for RCTs.

Lesson 1 • Fixed-Effect and Random-Effects Models

Contrasts the assumptions and formulas of fixed-effect and random-effects pooling. Model choice reflects assumptions about between-study heterogeneity and affects interval width.

Lesson 2 • Effect Size Measures and Calculation

Covers odds ratios, risk ratios, mean differences, and standardized mean differences. Selecting the correct effect measure is the first analytical decision in any meta-analysis.

Lesson 3 • Assessing and Quantifying Heterogeneity

Introduces Q statistic, I² index, and tau² to measure between-study variability. Quantifying heterogeneity determines whether pooling is appropriate and guides subgroup planning.

Lesson 1 • Sensitivity Analyses

Tests the robustness of pooled estimates by systematically varying analytical decisions. Sensitivity analyses reveal whether conclusions depend on specific studies or assumptions.

Lesson 2 • Interpreting and Reporting Findings

Synthesizes heterogeneity results into coherent narrative and tabular summaries. Accurate interpretation prevents overstatement of moderator effects in the final review.

Lesson 3 • Subgroup Analysis Principles

Defines a priori vs. post hoc subgroups and the risks of data dredging. Pre-specified subgroups are the only defensible basis for moderator claims in a review.

Lesson 1 • Detecting Publication Bias

Covers funnel plots, Egger's test, and Begg's test to identify asymmetry from selective reporting. Detecting bias is essential before interpreting pooled estimates as unbiased summaries.

Lesson 2 • Individual Participant Data Meta-Analysis

Explains the one-stage and two-stage approaches using raw participant-level data. IPD meta-analysis enables subgroup analyses impossible with aggregate data and reduces ecological bias.

Lesson 3 • Writing and Submitting the Review

Guides drafting, peer review, and journal submission of a completed systematic review. Adherence to PRISMA and journal-specific requirements maximizes acceptance and impact.

Lesson 4 • Adjusting for Publication Bias

Applies trim-and-fill and selection models to adjust pooled estimates for missing studies. Adjustment methods provide corrected estimates but carry their own assumptions and limitations.