Findings from the study were previously presented at the 28th International Conference on Pharmacoepidemiology & Therapeutic Risk Management, Barcelona, Spain in August 2012, and a Brookings Institution Roundtable on Active Medical Product Surveillance webinar in October 2012.
Multivariable confounding adjustment in distributed data networks without sharing of patient-level data†
Article first published online: 22 JUL 2013
Copyright © 2013 John Wiley & Sons, Ltd.
Pharmacoepidemiology and Drug Safety
Volume 22, Issue 11, pages 1171–1177, November 2013
How to Cite
Toh, S., Reichman, M. E., Houstoun, M., Ding, X., Fireman, B. H., Gravel, E., Levenson, M., Li, L., Moyneur, E., Shoaibi, A., Zornberg, G. and Hennessy, S. (2013), Multivariable confounding adjustment in distributed data networks without sharing of patient-level data. Pharmacoepidem. Drug Safe., 22: 1171–1177. doi: 10.1002/pds.3483
- Issue published online: 21 OCT 2013
- Article first published online: 22 JUL 2013
- Manuscript Revised: 21 JUN 2013
- Manuscript Accepted: 21 JUN 2013
- Manuscript Received: 12 FEB 2013
- active surveillance;
- distributed data network;
- propensity scores;
- disease risk scores;
It is increasingly necessary to analyze data from multiple sources when conducting public health safety surveillance or comparative effectiveness research. However, security, privacy, proprietary, and legal concerns often reduce data holders' willingness to share highly granular information. We describe and compare two approaches that do not require sharing of patient-level information to adjust for confounding in multi-site studies.
We estimated the risks of angioedema associated with angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), and aliskiren in comparison with beta-blockers within Mini-Sentinel, which has created a distributed data system of 18 health plans. To obtain the adjusted hazard ratios (HRs) and 95% confidence intervals (CIs), we performed (i) a propensity score-stratified case-centered logistic regression analysis, a method identical to a stratified Cox regression analysis but needing only aggregated risk set data, and (ii) an inverse variance-weighted meta-analysis, which requires only the site-specific HR and variance. We also performed simulations to further compare the two methods.
Compared with beta-blockers, the adjusted HR was 3.04 (95% CI: 2.81, 3.27) for ACEIs, 1.16 (1.00, 1.34) for ARBs, and 2.85 (1.34, 6.04) for aliskiren in the case-centered analysis. The corresponding HRs were 2.98 (2.76, 3.21), 1.15 (1.00, 1.33), and 2.86 (1.35, 6.04) in the meta-analysis. Simulations suggested that the two methods may produce different results under certain analytic scenarios.
The case-centered analysis and the meta-analysis produced similar results without the need to share patient-level data across sites in our empirical study, but may provide different results in other study settings. Copyright © 2013 John Wiley & Sons, Ltd.