One-stage and two-stage meta-analysis of individual participant data led to consistent summarized evidence: lessons learned from combining multiple databases.

OBJECTIVE: Combining multiple health-care databases (DBs) allows comparing the effects of a wide variety of health-care services. There is a growing interest in methods for combining the results from multiple DBs. We attempted to learn lessons about the performance of one- and two-stage approaches from the reanalysis of data drawn from two studies of pharmacoepidemiology based on multiple DBs.

STUDY DESIGN AND SETTING: Two nested case-control studies were carried out for estimating the tricyclic antidepressants (TCAs)-arrhythmia and etoricoxib-heart failure associations, respectively, from the Italian Group for Appropriate Drug Prescription in the Elderly and the European Safety of Non-Steroidal Anti-Inflammatory programs. The associations of interest were modeled by conditional logistic regression for matched case-control sets, fitting fixed-effect and random-effect models with both one- and two-stage approaches.

RESULTS: One- and two-stage approaches gave very similar results, showing uncertainty of TCA-arrhythmia association (random-effect odds ratios [ORs], 95% confidence interval [CI], 1.26, 0.71-2.24, and 1.30, 0.66-2.55, respectively) and statistical evidence for etoricoxib-heart failure association (fixed-effect OR, 95% CI, 1.53, 1.41-1.66, and 1.54, 1.42-1.66, respectively).

CONCLUSION: Our study offers further evidence that two-stage approach generates estimates very similar as those from one-stage approach, even in the case of between-DB exposure heterogeneity and when several covariates must be concurrently considered. As current rules limit the free movement of electronic health data, our findings open the door of treating data within the country where they are generated and then to apply conventional techniques for summarizing estimates, which is the two-stage approach for meta-analysis using individual participant data.