Learning from local to global: An efficient distributed algorithm for modeling time-to-event data
- 1 July 2020
- journal article
- research article
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 27 (7) , 1028-1036
Abstract
Objective: We developed and evaluated a privacy-preserving One-shot Distributed Algorithm to fit a multicenter Cox proportional hazards model (ODAC) without sharing patient-level information across sites.Materials and Methods: Using patient-level data from a single site combined with only aggregated information from other sites, we constructed a surrogate likelihood function, approximating the Cox partial likelihood function obtained using patient-level data from all sites. By maximizing the surrogate likelihood function, each site obtained a local estimate of the model parameter, and the ODAC estimator was constructed as a weighted average of all the local estimates. We evaluated the performance of ODAC with (1) a simulation study and (2) a real-world use case study using 4 datasets from the Observational Health Data Sciences and Informatics network.Results: On the one hand, our simulation study showed that ODAC provided estimates nearly the same as the estimator obtained by analyzing, in a single dataset, the combined patient-level data from all sites (ie, the pooled estimator). The relative bias was <0.1% across all scenarios. The accuracy of ODAC remained high across different sample sizes and event rates. On the other hand, the meta-analysis estimator, which was obtained by the inverse variance weighted average of the site-specific estimates, had substantial bias when the event rate is <5%, with the relative bias reaching 20% when the event rate is 1%. In the Observational Health Data Sciences and Informatics network application, the ODAC estimates have a relative bias <5% for 15 out of 16 log hazard ratios, whereas the meta-analysis estimates had substantially higher bias than ODAC.Conclusions: ODAC is a privacy-preserving and noniterative method for implementing time-to-event analyses across multiple sites. It provides estimates on par with the pooled estimator and substantially outperforms the meta-analysis estimator when the event is uncommon, making it extremely suitable for studying rare events and diseases in a distributed manner.Keywords
All Related Versions
Funding Information
- National Institutes of Health (1R01LM012607, 1R01AI130460, 5R01LM010098, UL1TR001427, R21AG061431, R01CA246418, 1R01AI116794)
- Patient-Centered Outcomes Research Institute
- PEDSnet Clinical Research Infrastructure (RI-CRN-2020-007)
- Cancer Informatics and eHealth Core
- University of Florida Health Cancer Center
- NIH
This publication has 23 references indexed in Scilit:
- Survival analysis with electronic health record data: Experiments with chronic kidney diseaseStatistical Analysis and Data Mining: An ASA Data Science Journal, 2014
- PEDSnet: How A Prototype Pediatric Learning Health System Is Being Expanded Into A National NetworkHealth Affairs, 2014
- Launching PCORnet, a national patient-centered clinical research networkJournal of the American Medical Informatics Association, 2014
- PEDSnet: a National Pediatric Learning Health SystemJournal of the American Medical Informatics Association, 2014
- Grid Binary LOgistic REgression (GLORE): building shared models without sharing dataJournal of the American Medical Informatics Association, 2012
- Validation of a common data model for active safety surveillance researchJournal of the American Medical Informatics Association, 2012
- Achieving a Nationwide Learning Health SystemScience Translational Medicine, 2010
- Risk factors for myocardial infarction in women and men: insights from the INTERHEART studyEuropean Heart Journal, 2008
- Risk Factors for Acute Myocardial Infarction in Latin AmericaCirculation, 2007
- Coding of Stroke and Stroke Risk Factors Using International Classification of Diseases , Revisions 9 and 10Stroke, 2005