How Long Does a Systematic Review Take? Timeline, Stages and Milestones
Most rigorous systematic reviews take about a year, sometimes longer, to reach publication. One PROSPERO-based study of medical-intervention reviews found a mean of 67.3 weeks from registered start to publication. A 2025 study of animal-research reviews reported medians of 11.5 months to completion and 16.2 months to publication. These are sample averages, not a deadline you should force onto your own project.12
Medical-intervention reviews registered in PROSPERO, drawn from 192 reviews included in the timing analysis.1
Animal-study systematic reviews with registered protocols. The observed range ran from 0.13 to 44.9 months.2
The same animal-review study found roughly five extra months between completion and publication.2
These two studies drew on different populations and used different timing definitions, so don’t pool their numbers or treat either one as a universal benchmark.
There’s no single duration that fits every review
Cochrane points to several sources of variation: the topic, the number of included studies, the chosen methods, the effort spent chasing missing data, and how experienced the team already is. The work itself spans protocol development, searching, screening, data collection, risk-of-bias assessment, synthesis, interpretation, and reporting. A time chart with target dates can help a team schedule all of this, but it can’t predict it in advance.3
Search yield drives a big chunk of the workload on its own. Allen and Olkin suggested using the number of retrieved citations as a practical predictor of how long a meta-analysis will take, since every record adds work in screening and record management.4 Included studies add a second layer: each one can bring multiple reports, multiple outcomes, missing details, and disagreements that need resolving.
Methodological safeguards cost time too, and that’s by design. Cochrane requires two people to make final inclusion decisions independently, and it requires duplicate extraction of outcome data for intervention reviews.56 Those checks are exactly what a rushed schedule tends to quietly drop.
Look at the dependencies before you assign dates
The eight stages form a chain of decisions, not a checklist you can freely reorder. Click through a stage below to see its output, its gate, and the delay teams most often discover too late.
Systematic review stage explorer
This teaching tool walks through sequence and dependencies. The paired planner turns them into dates, owners, and progress checks.
Question, scope, and team
Turn the decision problem into a reviewable question. Nail down scope, users, eligible evidence, outcomes, and the expertise the project needs before anyone commits to a timetable.
Eight stages, and the milestone that closes each one
1. Frame the question and assemble the team
The question sets the workload from day one. A broad population, several interventions, many outcomes, or a handful of eligible study designs all expand the search and multiply later decisions. Cochrane expects a team with topic and methodological expertise, plus statistical and information-specialist input wherever the review actually needs it.7
Milestone: the team signs off on the review question, scope, roles, and decision process. A timetable built before this point rests on assumptions nobody has actually agreed to yet.
2. Develop and register the protocol
The protocol pins down eligibility criteria, information sources, selection methods, data collection, risk-of-bias methods, and the intended synthesis. PRISMA-P offers a reporting checklist for protocols.8 Cochrane’s whole point here is to lock in methods before any results can quietly nudge them.7
Milestone: the team freezes a review-ready protocol and records it in the appropriate registry or public repository. Registration doesn’t fix a weak protocol; it just creates a public, time-stamped record of the plan.
3. Design, peer review, and run the search
An information specialist can translate eligibility criteria into database syntax, test retrieval, adapt the strategy across sources, and document every search decision. PRESS gives reviewers a structured way to check electronic search strategies.9 PRISMA-S covers how to report literature searches.10
Milestone: the team approves a reproducible search package and imports every record into the review system. The number of retrieved records now replaces guesswork with an actual, visible screening workload.
4. Screen titles, abstracts, and full texts
Teams first drop records that fail the criteria at title-and-abstract level, then work through potentially eligible reports in full text. Cochrane treats the final inclusion call as a high-stakes judgement and requires two people working independently.5 Retrieval problems, fuzzy criteria, and poor calibration are what create queues here.
Milestone: the included-study set stops changing, exclusion reasons are on record, and the PRISMA flow counts add up.
5. Extract and reconcile the data
Pilot the extraction form before running it on everything. It has to follow the review question and the planned synthesis, nothing else. Multiple reports from one study need study-level linking, and missing or ambiguous results may mean contacting the original investigators. Cochrane requires two people to extract outcome data independently for intervention reviews.6
Milestone: the team signs off on a clean, traceable dataset that actually supports every planned table and analysis.
6. Assess risk of bias
Pick the appraisal tool based on the study design and the question, not on convenience. Train reviewers, test the decision rules, and resolve disagreements through whatever process the protocol laid out. A rushed appraisal here tends to resurface as inconsistent judgements during synthesis and peer review.
Milestone: every included study has a documented judgement with supporting text, and the team has worked through its conflicts.
7. Synthesize the evidence and assess certainty
The synthesis follows the protocol and the data, in that order. Meta-analysis only makes sense when the studies answer a genuinely compatible question with suitable data. Other reviews need structured synthesis without meta-analysis instead. The team has to explain any deviations, look at heterogeneity where it matters, and rate the certainty of the evidence with an appropriate framework.
Milestone: the results, sensitivity checks, evidence tables, and certainty judgements come together into a coherent answer to the review question.
8. Report, update, submit, and disseminate
PRISMA 2020 guides complete reporting once the review is finished.11 Update the search if too much time has passed since the original one, since new eligible studies can reopen several earlier stages. Journal review then adds its own separate delay that the research team has no real control over.
Milestone: the manuscript, checklist, flow diagram, supplementary files, data, and code are all ready for submission and public access.
Use milestone gates instead of one distant deadline
A single final submission date hides the real state of the work. Milestone gates instead show plainly whether the next stage can start without dragging unresolved decisions along with it.
Scope gate
The question, users, review type, team, and roles all have written approval.
Protocol gate
The methods are complete enough to run, audit, and register.
Search gate
The strategies pass peer review and produce a documented record set.
Selection gate
Included studies and exclusion reasons reconcile with the flow diagram.
Dataset gate
Extraction, study links, and critical outcome data pass duplicate checks.
Appraisal gate
Reviewers resolve disagreements and back up each judgement with evidence.
Synthesis gate
The analyses answer the protocol question and account for uncertainty.
Submission gate
The search is current and the reporting package is complete.
Some work can overlap; some decisions genuinely can’t
Parallel work saves real calendar time when each task has a clear owner and clear inputs. It creates rework the moment a team starts a dependent task before the upstream decision has actually closed.
| Work pair | Overlap? | Condition |
|---|---|---|
| Protocol drafting and search development | Yes | Eligibility criteria and concepts need to stay under shared change control. |
| Reference-system setup and protocol drafting | Yes | Agree on file naming, identifiers, permissions, and backups before importing records. |
| Extraction-form design and full-text screening | Partly | Pilot the form on known examples, then freeze it once the included evidence is clear. |
| Background writing and screening | Yes | Keep the introduction separate from any conclusions about the review results. |
| Final synthesis and incomplete extraction | No | The analysis needs a verified dataset and resolved study links first. |
| Final manuscript and an outdated search | No | Run the planned update and process new records before locking the results. |
The delays that deserve real space in the plan
Most delays actually sit between tasks rather than inside them. A reviewer finishes a batch, then waits on conflict resolution. The search is ready, then waits on peer review. Authors get asked for missing data, then everyone waits on a reply.
| Delay | Early signal | Planning response |
|---|---|---|
| Scope drift | Eligibility rules change mid-screening | Close the scope gate and route amendments through one decision owner. |
| Search redesign | Known eligible studies don’t show up | Test retrieval before launch and arrange a PRESS review. |
| Large search yield | Record counts blow past the planning assumption | Re-estimate after deduplication; add trained screeners rather than loosen criteria. |
| Full-text access | Reports still unresolved near the screening deadline | Start retrieval during title-and-abstract screening and track every request. |
| Multiple reports per study | Several papers describe the same participants | Link reports to study IDs before extraction begins. |
| Unusable outcome reporting | Measures, time points, or denominators stay unclear | Predefine contact procedures and document your assumptions. |
| Reviewer disagreement | Calibration rounds keep producing the same conflicts | Clarify the decision rules and repeat training before scaling up. |
| Analysis mismatch | Extracted data doesn’t fit the planned synthesis | Bring the methodologist and statistician in during protocol and pilot stages. |
| Stale search | Submission falls well beyond the final search date | Reserve time for an update search and its downstream loop. |
| Journal process | Formatting and submission requirements surface late | Track completion, submission, acceptance, and publication as separate dates. |
A shorter deadline may call for a different method entirely
A rapid review answers a genuine need for timely evidence by restricting selected methods and documenting those restrictions openly. Updated Cochrane guidance sets a six-month ceiling for Cochrane rapid reviews and explicitly rejects one fixed recipe for every project.12
A living systematic review, by contrast, has no final completion date while it stays active: the team runs a baseline review, watches for new evidence, and updates the synthesis on a published schedule. A scoping review might skip some tasks an intervention-effect review needs, yet its broad search and charting work can still eat plenty of time. Review type should follow the decision need. A tight deadline doesn’t quietly turn a standard review into a rapid one just by wishing it were faster.
Start the schedule with gates, not invented percentages
Picture a team reviewing an educational, clinical, environmental, or policy intervention. They know the question and expect a standard systematic review, but they have no idea yet what the search yield or included-study count will look like.
The first plan just assigns dates to the scope, protocol, and search gates. Once the team runs and deduplicates the search, it swaps the assumed screening workload for the real record count. After the extraction pilot, it revises the rest of the plan based on the number of included studies, reports per study, outcomes, appraisal needs, and how complex the synthesis is turning out to be.
That rolling estimate only claims to know what the team actually knows at each point. A twelve-month promise made before the search even runs might sound precise, but it carries less real information than a plan that states its assumptions plainly and names the next date it’ll be checked.
Turn the stages into a working project schedule
The Systematic Review Timeline and Milestone Planner helps you assign owners, dependencies, target dates, review gates, and progress checks without pretending there’s one universal timeline for every review.
Open the free timeline planner →Related planning and resource pages
Frequently asked questions
How long does a systematic review usually take?
Plan for about a year or longer if publication is your endpoint. A PROSPERO-based study of medical-intervention reviews found a mean of 67.3 weeks from registered project start to publication. A 2025 animal-review study found medians of 11.5 months to completion and 16.2 months to publication. Scope, record volume, methods, and team experience can move a project well outside those figures.
Can one researcher complete a systematic review alone?
Not a rigorous intervention review. Cochrane requires independent duplicate final inclusion decisions and duplicate extraction of outcome data, and a lone reviewer simply can’t meet those safeguards without bringing in additional reviewers.
Can a systematic review be completed in three months?
Some tightly focused projects can move fast, but a three-month promise is risky when the search yield and included-study count are still unknown. An urgent evidence need can justify a rapid review with declared methodological restrictions. It doesn’t justify silent shortcuts inside a standard systematic review.
Which systematic review stage takes the longest?
No single stage dominates every review. Screening grows with the number of retrieved records. Extraction and appraisal grow with the number and complexity of included studies. Synthesis gets harder when outcomes, designs, or reporting differ across studies.
Does PRISMA determine the systematic review timeline?
No. PRISMA 2020 is a reporting guideline for completed systematic reviews, and PRISMA-P supports protocol reporting. Neither one assigns project durations. The schedule should come from the review’s methods, dependencies, workload, and available expertise instead.
Can AI shorten a systematic review?
Automation can help with defined tasks like record management, screening assistance, and checks. The team still has to validate each use and keep human responsibility for inclusion, extracted data, appraisal, synthesis, and reporting. Tool use belongs in the audit trail either way.
Methodological guidance and empirical evidence
- Borah R, Brown AW, Capers PL, Kaiser KA. Analysis of the time and workers needed to conduct systematic reviews of medical interventions using data from the PROSPERO registry. BMJ Open. 2017;7:e012545. doi:10.1136/bmjopen-2016-012545
- Bugajska JV, Hild BF, Brüschweiler D, et al. How long does it take to complete and publish a systematic review of animal studies? BMC Med Res Methodol. 2025;25:226. doi:10.1186/s12874-025-02672-5
- Higgins JPT, Thomas J, Chandler J, et al, editors. Chapter II: Planning a Cochrane Review. In: Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2024. Current online chapter
- Allen IE, Olkin I. Estimating time to conduct a meta-analysis from number of citations retrieved. JAMA. 1999;282(7):634-635. doi:10.1001/jama.282.7.634
- Lefebvre C, Glanville J, Briscoe S, et al. Chapter 4: Searching for and selecting studies. In: Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2024. Current online chapter
- Li T, Higgins JPT, Deeks JJ, editors. Chapter 5: Collecting data. In: Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2024. Current online chapter
- Lasserson TJ, Thomas J, Higgins JPT. Chapter 1: Starting a review. In: Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2024. Current online chapter
- Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1. doi:10.1186/2046-4053-4-1
- McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75:40-46. doi:10.1016/j.jclinepi.2016.01.021
- Rethlefsen ML, Kirtley S, Waffenschmidt S, et al. PRISMA-S: an extension to the PRISMA Statement for reporting literature searches in systematic reviews. Syst Rev. 2021;10:39. doi:10.1186/s13643-020-01542-z
- Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71
- Garritty C, Hamel C, Trivella M, et al. Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. BMJ. 2024;384:e076335. doi:10.1136/bmj-2023-076335
