Systematic Review vs. Literature Review vs. Meta-Analysis: What’s the Difference?

A literature review, a systematic review, and a meta-analysis all draw on existing research, but they are different kinds of work with different goals, methods, and standards. A literature review summarizes what is known about a topic, a systematic review answers a focused question using a structured and reproducible method, and a meta-analysis is a statistical technique that combines the numerical results of several studies. The terms are often used interchangeably, which leads researchers to choose the wrong approach or to describe their work inaccurately in a manuscript.

This article covers the following:

• What a literature review, a systematic review, and a meta-analysis each involve
• How a systematic review differs from a meta-analysis, and why the two are not the same thing
• Where scoping reviews and other review types fit among the categories of literature review
• A side-by-side comparison of the main review types
• How to choose the right type of review, and common mistakes to avoid

What Is a Literature Review?

A literature review is a summary and critical discussion of existing research on a topic. The term is broad. It covers the background section of a research paper or thesis, which places a study in the context of earlier work, and it also covers standalone review articles. A literature review that does not follow a formal, predefined search protocol is often called a narrative or traditional review.

The main purpose of a literature review is to give readers an overview of the topic, show how ideas have developed, and identify gaps that new research could address. Because authors usually choose which sources to include without a fixed protocol, a narrative review offers flexibility but carries a greater risk of selection bias. How to approach one is covered in How to Write a Literature Review.

What Is a Systematic Review?

A systematic review is a research method for answering a specific, focused question by identifying, appraising, and synthesizing all the relevant studies in a systematic and reproducible way. The difference between a narrative review vs. systematic review is one of rigor. A systematic review typically includes:

• A clearly defined research question
• A written protocol, which in health research is often registered in a registry such as PROSPERO before the review begins
• A comprehensive search of several databases, using documented search terms
• Predefined criteria for including and excluding studies
• An assessment of the quality or risk of bias of each included study
• A transparent account of how studies were selected, usually reported with a flow diagram

Systematic reviews are usually reported following the PRISMA guidelines, which set out what a review report should include. The process for the flow diagram is explained in How to Create an Effective PRISMA Flow Diagram, and the broader role of such checklists is covered in What are Reporting Guidelines for Research?. The findings of a systematic review may be summarized in words or, where the studies are similar enough, combined statistically in a meta-analysis.

What Is a Meta-Analysis?

A meta-analysis is a statistical technique that combines the quantitative results of multiple independent studies to estimate an overall effect. Pooling results gives a more precise estimate than any single study can, and it allows researchers to examine how much the results differ between studies, a property called heterogeneity. Meta-analyses are commonly presented with a forest plot, which shows the result of each study and the pooled estimate.

A meta-analysis is only as reliable as the studies and the selection process behind it. Because studies with significant results are more likely to be published, a meta-analysis can overestimate an effect if the search misses unpublished work. The article Assessing and Avoiding Publication Bias in Meta-analyses explains how this problem is detected and handled.

Systematic Review vs. Meta-Analysis: The Key Difference

The difference between a systematic review and a meta-analysis is one of category. A systematic review is a research design, a structured way of finding and evaluating evidence. A meta-analysis is a statistical method for combining numerical results. The two are related, since many meta-analyses are carried out as part of a systematic review, but neither depends on the other.

A systematic review does not always include a meta-analysis. When the included studies differ too much in design, populations, or outcomes to be combined meaningfully, the results are synthesized narratively instead. A meta-analysis, in turn, should be based on a systematic search. A meta-analysis of a hand-picked set of studies can produce a misleading estimate, because the selection of studies is itself a source of bias.

Scoping Reviews and Other Types of Literature Review

The categories of literature review extend beyond these three. A scoping review maps the extent and nature of research on a broad topic, often to clarify concepts or identify gaps. It uses a systematic search, but it usually does not assess the quality of the studies or pool their results. The choice between a scoping review vs. systematic review is covered in What is a Scoping Review?, which explains how the two compare.

Other types include integrative reviews, which combine different kinds of evidence, such as experimental and non-experimental studies, and rapid reviews, which streamline systematic methods to produce results in less time. The right type of review depends on the question being asked.

Types of Literature Review at a Glance

<table style="width:100%; border-collapse:collapse;">
  <thead>
    <tr>
      <th style="border:1px solid #ddd; padding:10px; text-align:left;">Criteria</th>
      <th style="border:1px solid #ddd; padding:10px; text-align:left;">Literature (Narrative) Review</th>
      <th style="border:1px solid #ddd; padding:10px; text-align:left;">Systematic Review</th>
      <th style="border:1px solid #ddd; padding:10px; text-align:left;">Meta-analysis</th>
      <th style="border:1px solid #ddd; padding:10px; text-align:left;">Scoping Review</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Main purpose</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">Overview of a topic and its context</td>
      <td style="border:1px solid #ddd; padding:10px;">Answer a focused question with a transparent method</td>
      <td style="border:1px solid #ddd; padding:10px;">Combine numerical results into an overall estimate</td>
      <td style="border:1px solid #ddd; padding:10px;">Map the breadth of evidence on a broad topic</td>
    </tr>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Question</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">Broad</td>
      <td style="border:1px solid #ddd; padding:10px;">Focused</td>
      <td style="border:1px solid #ddd; padding:10px;">Focused, with comparable outcomes</td>
      <td style="border:1px solid #ddd; padding:10px;">Broad</td>
    </tr>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Search</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">Not necessarily systematic</td>
      <td style="border:1px solid #ddd; padding:10px;">Comprehensive and reproducible</td>
      <td style="border:1px solid #ddd; padding:10px;">Usually part of a systematic review</td>
      <td style="border:1px solid #ddd; padding:10px;">Systematic</td>
    </tr>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Quality assessment of studies</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">Not required</td>
      <td style="border:1px solid #ddd; padding:10px;">Yes</td>
      <td style="border:1px solid #ddd; padding:10px;">Yes, usually from the systematic review</td>
      <td style="border:1px solid #ddd; padding:10px;">Usually not performed</td>
    </tr>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Synthesis</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">Narrative</td>
      <td style="border:1px solid #ddd; padding:10px;">Narrative, and sometimes statistical</td>
      <td style="border:1px solid #ddd; padding:10px;">Statistical</td>
      <td style="border:1px solid #ddd; padding:10px;">Descriptive or narrative</td>
    </tr>
    <tr>
      <td style="border:1px solid #ddd; padding:10px;"><strong>Typical reporting guideline</strong></td>
      <td style="border:1px solid #ddd; padding:10px;">None required</td>
      <td style="border:1px solid #ddd; padding:10px;">PRISMA</td>
      <td style="border:1px solid #ddd; padding:10px;">PRISMA</td>
      <td style="border:1px solid #ddd; padding:10px;">PRISMA Extension for Scoping Reviews (PRISMA-ScR)</td>
    </tr>
  </tbody>
</table>

How to Choose the Right Type of Review

The choice depends on the research question and on the resources available.

• To introduce a study or a thesis, a literature review is the appropriate choice. It places the work in context and shows the gap the study addresses.
• To answer a specific question about effectiveness, risk, or association, a systematic review is the standard approach, and a meta-analysis can be added if the studies can be combined.
• To see what has been studied on a broad topic, or to decide whether a systematic review is feasible, a scoping review is a good fit.
• To estimate an overall effect across comparable studies, a meta-analysis within a systematic review is appropriate.

A systematic review is labor-intensive. Screening and appraisal are typically done by more than one reviewer, which takes time and planning. This is a practical reason to scope a question carefully before committing to a full systematic review.

Common Mistakes to Avoid

• Calling a review “systematic” without a protocol, a documented search, and predefined inclusion criteria
• Treating a meta-analysis and a systematic review as the same thing
• Combining studies that are too different to be pooled meaningfully, without addressing heterogeneity
• Searching only one database, which can miss relevant studies and bias the results
• Ignoring publication bias when interpreting a meta-analysis
• Skipping the relevant reporting guideline, such as PRISMA, when submitting a systematic review
• Choosing a systematic review for a broad topic that a scoping review would suit better

Final Thoughts

Literature reviews, systematic reviews, meta-analyses, and scoping reviews serve different purposes, and the labels carry specific expectations that journals and reviewers check. Choosing the right type at the start of a project, and describing it accurately, saves time and avoids methodological criticism later. Researchers preparing a review manuscript for submission can also benefit from a language and clarity check so that the method is communicated precisely, and AJE’s editing services help ensure a manuscript is clear to the editors and reviewers who assess it.