%0 Journal Article %A Hieab HH Adams %A Hadie Adams %A Lenore J Launer %A Sudha Seshadri %A Reinhold Schmidt %A Joshua C Bis %A Stephanie Debette %A Paul A Nyquist %A Jeroen Van der Grond %A Thomas H Mosley, Jr %A Jingyun Yang %A Alexander Teumer %A Saima Hilal %A Gennady V Roshchupkin %A Joanna M Wardlaw %A Claudia L Satizabal %A Edith Hofer %A Ganesh Chauhan %A Albert Smith %A Lisa R Yanek %A Sven J Van der Lee %A Stella Trompet %A Vincent Chouraki %A Konstantinos A Arfanakis %A James T Becker %A Wiro J Niessen %A Anton JM de Craen %A Fabrice F Crivello %A Li An Lin %A Debra A Fleischman %A Tien Yin Wong %A Oscar H Franco %A Katharina Wittfeld %A J Wouter Jukema %A Philip L De Jager %A Albert Hofman %A Charles DeCarli %A Dimitris Rizopoulos %A WT Longstreth, Jr %A Bernard M Mazoyer %A Vilmundar Gudnason %A David A Bennett %A Ian J Deary %A M Kamran Ikram %A Hans J Grabe %A Myriam Fornage %A Cornelia M Van Duijn %A Meike W Vernooij %A M Arfan Ikram %A on behalf of the HD-READY Consortium %T Partial derivatives meta-analysis: pooled analyses when individual participant data cannot be shared %D 2016 %R 10.1101/038893 %J bioRxiv %P 038893 %X Joint analysis of data from multiple studies in collaborative efforts strengthens scientific evidence, with the gold standard approach being the pooling of individual participant data (IPD). However, sharing IPD often has legal, ethical, and logistic constraints for sensitive or high-dimensional data, such as in clinical trials, observational studies, and large-scale omics studies. Therefore, meta-analysis of study-level effect estimates is routinely done, but this compromises on statistical power, accuracy, and flexibility. Here we propose a novel meta-analytical approach, named partial derivatives meta-analysis, that is mathematically equivalent to using IPD, yet only requires the sharing of aggregate data. It not only yields identical results as pooled IPD analyses, but also allows post-hoc adjustments for covariates and stratification without the need for site-specific re-analysis. Thus, in case that IPD cannot be shared, partial derivatives meta-analysis still produces gold standard results, which can be used to better inform guidelines and policies on clinical practice. %U https://www.biorxiv.org/content/biorxiv/early/2016/02/07/038893.full.pdf