Study-type module for individual participant data (IPD) meta-analyses. Provides section-by-section guidance aligned with the PRISMA-IPD extension, covering IPD collection processes, data harmonisation, one-stage and two-stage analytical approaches, and individual-level subgroup analyses.
This module governs manuscript generation for individual participant data (IPD) meta-analyses. It supplements the core engine and section-specific skills with IPD-specific reporting requirements derived from the PRISMA-IPD extension statement. IPD meta-analyses are considered the gold standard of evidence synthesis because they allow consistent re-analysis of primary data, flexible subgroup exploration at the individual level, and avoidance of ecological bias in treatment-covariate interactions.
The primary reporting guideline for IPD meta-analyses is the PRISMA-IPD extension statement:
Stewart LA, Clarke M, Rovers M, et al. Preferred reporting items for a systematic review and meta-analysis of individual participant data: the PRISMA-IPD statement. JAMA. 2015;313(16):1657-1665.
This extension modifies and supplements the standard PRISMA 2020 items to address the unique features of IPD meta-analyses, including the process of obtaining IPD, data integrity checking, harmonisation procedures, and the analytical framework (one-stage versus two-stage approaches).
When citing the guideline in the Methods section, use the following phrasing:
"This systematic review and IPD meta-analysis was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Individual Participant Data (PRISMA-IPD) (Stewart et al., JAMA 2015;313:1657-1665)."
See references/prisma-ipd-checklist.md for the full IPD-specific checklist
items.
State the prospective registration and note whether the protocol specified the IPD collection approach. Template sentences:
"The protocol for this systematic review and IPD meta-analysis was registered prospectively with PROSPERO (registration number CRD42XXXXXXX) before the first data request was issued to study investigators. The protocol specified eligibility criteria, the IPD collection process, planned analyses including one-stage and two-stage approaches, and pre-specified treatment-covariate interactions."
"The study was coordinated by [collaborative group name] and followed established procedures for IPD meta-analyses as recommended by the Cochrane IPD Meta-Analysis Methods Group."
Define eligibility using the PICO framework with the additional requirement of IPD availability. Template sentences:
"Studies were eligible if they were randomised controlled trials comparing [intervention] with [comparator] in patients with [condition], reporting [outcome] with a minimum follow-up of [N] months. Both published and unpublished trials meeting these criteria were eligible regardless of whether IPD were ultimately obtainable, to permit assessment of availability bias."
"The eligibility criteria were applied identically to all studies identified by the systematic search. Studies for which IPD could not be obtained were retained in the review for descriptive comparison but excluded from the primary IPD meta-analysis."
Specify any additional IPD-specific criteria, such as minimum data fields required for harmonisation.
Describe both the systematic search and the direct contact with trialists. Template sentences:
"A systematic search was conducted across MEDLINE (via PubMed), Embase (via Elsevier), the Cochrane Central Register of Controlled Trials (CENTRAL), and [additional databases] from inception to [date]. Trial registries (ClinicalTrials.gov, WHO ICTRP, EU Clinical Trials Register) were searched to identify unpublished and ongoing trials. The full search strategy is provided in the Supplementary Materials (eTable 1)."
"In addition to the electronic search, principal investigators or data custodians of all eligible studies were contacted directly by email to request individual participant data. Up to three contact attempts were made over a period of [N] months. Collaborative agreements and data-sharing protocols were established with responding investigators."
Describe the two-phase selection: study identification and IPD acquisition. Template sentences:
"Study selection followed a two-phase process. In the first phase, two reviewers independently screened titles and abstracts, followed by full-text assessment against the eligibility criteria. In the second phase, authors of eligible studies were contacted to request IPD. Studies for which IPD were obtained were included in the primary IPD meta-analysis; studies with aggregate data only were included in supplementary aggregate-data analyses where possible."
This is a critical subsection for IPD meta-analyses. Describe the IPD request process, response rate, harmonisation, and quality checks. Template sentences:
"Individual participant data were requested from all eligible study investigators. Of [N] eligible studies, IPD were obtained from [N] ([X]%), representing [N] participants. Reasons for non-availability included: loss of contact with investigators ([N] studies), data no longer accessible ([N] studies), and declined participation ([N] studies)."
"Upon receipt, all IPD were subjected to a standardised quality-assurance protocol. Data integrity checks included: verification of randomisation codes against published results, reproduction of published summary statistics (sample sizes, event rates, treatment effects), identification of duplicates and out-of-range values, and cross-checking of baseline characteristics against published tables. Discrepancies were resolved by correspondence with the original investigators."
"Individual variables were harmonised across studies using a pre-specified coding manual. Continuous variables (e.g., age, baseline biomarker levels) were standardised to common units. Categorical variables (e.g., disease stage, comorbidity classification) were recoded to a common taxonomy. Where harmonisation was not possible for a specific variable, it was excluded from that analysis with a note in the data dictionary."
Describe both study-level and IPD-specific risk-of-bias assessments. Template