A systematic review abstract is a structured summary that reports the objectives, methods, results, and conclusions of a systematic review in approximately 250 to 350 words. The PRISMA 2020 Abstract Checklist, published by Page et al. (2021), specifies 12 reporting items that every systematic review abstract must address, including the review question, eligibility criteria, information sources, risk of bias assessment, number of included studies, synthesis results, limitations, registration number, and funding. Writing a strong abstract matters because it is the single most-read section of any published paper, it determines whether journal editors send your manuscript for peer review, and it is the only content indexed in full by PubMed, Cochrane Library, and other databases. Researchers who follow the PRISMA abstract checklist produce abstracts that are more complete, more transparent, and more likely to survive peer review without requests for revision.
The 12 Items on the PRISMA 2020 Abstract Checklist
The PRISMA 2020 statement, authored by Page et al. (2021) and endorsed by hundreds of journals worldwide, includes a dedicated abstract checklist with 12 reporting items organized across six domains. Each item corresponds to a specific piece of information that readers, reviewers, and database indexers need to evaluate your review at a glance.
Item 1: Title. Identify the report as a systematic review, a meta-analysis, or both. The word "systematic review" must appear in the title. If you conducted quantitative pooling, add "and meta-analysis" as well.
Item 2: Objectives. State the research question using a structured format such as PICO (Population, Intervention, Comparison, Outcome). Be specific enough that readers understand exactly what the review aimed to answer.
Item 3: Eligibility criteria. Describe the inclusion criteria for studies: types of participants, interventions or exposures, comparators, outcomes, and study designs. Mention any restrictions such as language or date limits.
Item 4: Information sources. Name the databases searched and the date of the most recent search. At minimum, report the specific databases (for example, PubMed, Embase, Cochrane CENTRAL) and the search date range.
Item 5: Risk of bias assessment. Specify the tool or tools used to assess risk of bias in included studies, such as the Cochrane Risk of Bias 2 (RoB 2) tool for randomized trials or the Newcastle-Ottawa Scale for observational studies.
Item 6: Included studies. Report the number of studies and participants included in the review. If the review includes a meta-analysis, state the number of studies contributing to each pooled estimate.
Item 7: Synthesis of results. Present the main results of the review, including summary effect estimates with confidence intervals and measures of statistical heterogeneity (such as the I-squared statistic). For narrative syntheses, summarize the direction and consistency of findings.
Item 8: Limitations. Describe the main limitations of the evidence, such as risk of bias across studies, imprecision, inconsistency, indirectness, or publication bias. Also note any limitations of the review process itself.
Item 9: Interpretation. Provide a general interpretation of the results in the context of other evidence and the implications for practice, policy, or future research.
Item 10: Funding. Specify the primary funding source for the review, or state that no specific funding was received.
Item 11: Registration. Provide the registration number (typically from PROSPERO) and, if available, the protocol DOI or URL. Learn how to register your review in our PROSPERO registration guide.
Item 12: Systematic review methods. Briefly describe the methods used for data synthesis, such as random-effects meta-analysis, narrative synthesis, or a specific synthesis framework.
These 12 items apply to every systematic review abstract regardless of journal, specialty, or whether the review includes a meta-analysis. Beller et al. (2013) first demonstrated that systematic review abstracts frequently omit critical information, and the PRISMA 2020 abstract checklist was designed to close those gaps.
Structured Versus Unstructured Abstract Formats
Journals differ in how they want you to organize your abstract, and understanding the format before you begin writing prevents wasted revision time.
Structured abstracts use labeled sections, typically Background, Objectives, Methods, Results, and Conclusions. Most medical and health sciences journals require structured abstracts because they force authors to address each domain explicitly. The Cochrane Database of Systematic Reviews uses a structured format with standardized headings, and journals such as The BMJ, JAMA, and The Lancet follow similar conventions.
Unstructured abstracts present the same information in a single flowing paragraph without section headings. Some social science, education, and interdisciplinary journals prefer this format. Even without headings, you must still cover all 12 PRISMA abstract checklist items. The difference is organizational, not substantive.
How to choose the right format. Always check your target journal's author guidelines before drafting. Look for three pieces of information: (1) whether the abstract must be structured or unstructured, (2) the maximum word count, and (3) whether the journal specifies its own heading labels. Never assume one journal's format applies to another.
Mapping PRISMA items to structured headings. If your journal requires a structured abstract, the mapping is straightforward. The Objectives heading covers PRISMA items 1 and 2. The Methods heading covers items 3, 4, 5, and 12. The Results heading covers items 6 and 7. The Conclusions heading covers items 8 and 9. Registration and funding (items 10 and 11) are usually placed at the end of the abstract or in a separate line below it, depending on journal requirements.
Word Count Limits by Journal and How to Stay Within Them
One of the most common challenges when writing a systematic review abstract is fitting all 12 PRISMA items into a tight word limit. Here are the typical limits for major journals and practical strategies for staying within them.
Journal word count reference table. The BMJ allows up to 250 words for structured abstracts. JAMA permits 350 words. The Lancet allows 300 words. Cochrane reviews allow up to 400 words. PLoS Medicine allows 300 words. The Annals of Internal Medicine allows 275 words. Most specialty journals fall between 200 and 350 words. Always verify the current limit in the author guidelines because journals update requirements periodically.
Strategy 1: Eliminate filler phrases. Remove introductory padding such as "In this systematic review, we aimed to" and replace it with a direct statement: "This systematic review assessed." Cut phrases like "it is well known that" or "there is a growing body of evidence suggesting." Every word in an abstract must carry information.
Strategy 2: Combine eligibility criteria with the research question. Instead of devoting separate sentences to objectives and eligibility criteria, merge them. For example: "We searched for randomized controlled trials comparing cognitive behavioral therapy with pharmacotherapy for generalized anxiety disorder in adults." This single sentence addresses objectives, study design, intervention, comparison, condition, and population.
Strategy 3: Report only the primary outcome in the abstract. If your review assessed five outcomes, report the primary outcome with full statistical detail (effect estimate, confidence interval, I-squared) and mention secondary outcomes only as supporting text. Reserve detailed secondary outcome reporting for the full-text results section.
Strategy 4: Use numerals for all numbers. Most journals allow numerals in abstracts even when style guides require words for numbers below ten in the main text. Writing "7 studies" instead of "seven studies" saves characters that accumulate across the abstract.
Strategy 5: Front-load each sentence with the most important information. Place the key finding or method at the beginning of the sentence, not at the end. This writing pattern makes the abstract scannable and allows you to trim trailing qualifications if you need to cut words.