Living Systematic Reviews: Keeping Evidence Current in a Fast-Moving World

A traditional systematic review is a snapshot. It captures the evidence available at a single point in time, and the moment it is published, it begins to age. For slowly evolving topics, this is acceptable — the evidence base may not change meaningfully for years. But for rapidly evolving areas like infectious disease treatment, emerging technologies, or fast-moving clinical interventions, a static review can become outdated within months.

Living systematic reviews (LSRs) address this problem by building continuous updating into the review methodology itself.

What Is a Living Systematic Review?

A living systematic review is a systematic review that is continually updated, incorporating relevant new evidence as it becomes available. The core principle is simple: instead of conducting a review once and publishing a fixed manuscript, the review remains "alive" — with periodic search updates, screening, and re-analysis.

The concept was formalized by Elliott et al. in 2017 and has been adopted by Cochrane, the WHO, and several national guideline bodies. The COVID-19 pandemic accelerated adoption dramatically, as the pace of clinical trial publication made static reviews obsolete within weeks.

When Is a Living Review Warranted?

Not every systematic review should be living. The Cochrane Living Systematic Review Network identifies three conditions that should all be met:

The topic has high priority for decision-making (clinical guidelines, policy, or practice depend on up-to-date evidence)
There is ongoing uncertainty — the current evidence is insufficient to answer the question definitively
New evidence is likely — active research is being conducted and published regularly

If all three conditions are met, a living approach is justified. If the evidence base is stable and unlikely to change, a traditional review is more appropriate.

Examples of Appropriate Topics

Treatments for emerging infections (COVID-19, mpox, H5N1)
Rapidly evolving drug classes (GLP-1 agonists, immune checkpoint inhibitors)
Diagnostic technologies with ongoing validation studies
Interventions where early trials are being followed by larger confirmatory studies
Guidelines that need to respond to new evidence within months, not years

How Living Reviews Work

The Update Cycle

A living review follows a defined update cycle:

Baseline review: Conduct a complete systematic review using standard methods (search, screening, extraction, quality assessment, synthesis)
Ongoing search: Run the search strategy at regular intervals (typically monthly or quarterly)
Screening: Screen new records against the original eligibility criteria
Incorporation: Add newly eligible studies to the existing dataset
Re-analysis: Update the meta-analysis (if applicable) with the new data
Re-assessment: Update risk of bias and GRADE assessments as needed
Publication update: Publish an updated version with a clear date stamp and a description of what changed

Decision Points

At each update cycle, the team must decide:

Continue as living: New evidence is still expected, uncertainty remains, and the topic is still a priority
Pause: No new relevant evidence has been identified for several cycles; resume monitoring
Retire: The question has been answered definitively, no further updates are needed, or the topic is no longer a priority

Versioning and Transparency

Each update creates a new version of the review. Transparency requires:

Clear version numbering and dates
A summary of changes from the previous version
An archived copy of each previous version
Documentation of any changes to the protocol (eligibility criteria, outcomes, etc.)

Challenges of Living Reviews

Team and Resource Commitment

Living reviews require a sustained team commitment. Unlike traditional reviews, where the team disbands after publication, living reviews need ongoing personnel for searching, screening, and analysis. Funding models for living reviews are still developing.

Methodological Stability

Changing the protocol mid-review (adding outcomes, modifying eligibility criteria) creates methodological complexity. Protocol amendments must be documented and justified, and their impact on previous versions must be assessed.

Statistical Considerations

Repeated meta-analyses raise statistical concerns:

Multiplicity: Running the analysis repeatedly increases the risk of false positives. Sequential analysis methods (trial sequential analysis, cumulative meta-analysis with alpha-spending) help control this.
Changing heterogeneity: As new studies are added, heterogeneity may increase or decrease, potentially changing the interpretation.
Conditional power: At some point, enough evidence has accumulated that further updates are unlikely to change the conclusion. Monitoring boundaries help identify this point.

Publication and Indexing

Living reviews challenge traditional publication models. Some journals publish updated versions as new articles; others use online-only versioned formats. Cochrane maintains living reviews as continuously updated entries in the Cochrane Library. Citation tracking becomes complex when multiple versions exist.

The Technology Gap

Most systematic review tools were designed for the traditional, one-shot workflow. Living reviews require infrastructure that supports:

Round-based data management: Each update cycle is a "round" that preserves the previous state while adding new data
Incremental screening: New records should be screened without re-screening the entire database
Versioned analysis: The meta-analysis should be reproducible at any point in its history
Audit trails: Every decision (inclusion, exclusion, risk of bias judgment) must be traceable to a specific round
Automated search monitoring: Ideally, the tool should run saved searches periodically and flag new records

Without this infrastructure, teams resort to spreadsheets and manual version control — which is error-prone and unsustainable.

The Future of Evidence Synthesis

Living systematic reviews represent a broader shift in how the research community thinks about evidence synthesis. The traditional model — a large, labor-intensive project conducted once — is increasingly inadequate for fields where evidence evolves rapidly.

Several trends are converging:

AI-assisted screening: Machine learning models trained on a review's inclusion/exclusion decisions can prioritize new records, reducing the screening burden in update cycles
Automated search updates: Tools that monitor databases and preprint servers for new relevant publications
Real-time data integration: Connecting systematic reviews to clinical trial registries and results databases
Dynamic evidence maps: Visual representations of the evidence base that update automatically as new studies are incorporated

These capabilities are not speculative — they are being built now, and they will fundamentally change how evidence synthesis is conducted over the next decade.

Building for Living Reviews

mapped's data model includes infrastructure for living systematic reviews. The round-based architecture means that each update cycle can be tracked as a discrete unit while maintaining continuity with previous rounds. When living review support is fully released:

Each update cycle will be a new round within the same project
Previous rounds will be preserved and auditable
Screening will be incremental — only new records enter the pipeline
Meta-analysis will be versioned — you can view the pooled estimate at any point in the review's history
The PRISMA flow diagram will reflect cumulative totals across rounds

This is not a bolt-on feature — it is a design decision built into the platform's data model from the beginning.