Skip to main content

Table 2 Reasons provided to support conducting an IPD 1

From: Systematic review of methods for individual patient data meta- analysis with binary outcomes

Reason N (%)
To perform subgroup analyses 13 (50%)
To improve consistency across studies (in terms of inclusion criteria, outcome definition, etc.) 4 (15%)
To consider other outcomes 4 (15%)
To adjust for confounding variables 1 (4%)
To estimate diagnostic accuracy 5 (19%)
To identify predictors of an outcome 2 (8%)
Unclear 6 (23%)
  1. 1Percentages do not total to 100 because some studies reported more than one reason for conducting an IPD-MA.