CEDAR: Communication Efficient Distributed Analysis for Regressions

Abstract

Electronic health records (EHRs) offer great promises for advancing precision medicine and, at the same time, present significant analytical challenges. Particularly, it is often the case that patient‐level data in EHRs cannot be shared across institutions (data sources) due to government regulations and/or institutional policies. As a result, there are growing interests about distributed learning over multiple EHRs databases without sharing patient‐level data. To tackle such challenges, we propose a novel communication efficient method that aggregates the optimal estimates of external sites, by turning the problem into a missing data problem. In addition, we propose incorporating posterior samples of remote sites, which can provide partial information on the missing quantities and improve efficiency of parameter estimates while having the differential privacy property and thus reducing the risk of information leaking. The proposed approach, without sharing the raw patient level data, allows for proper statistical inference. We provide theoretical investigation for the asymptotic properties of the proposed method for statistical inference as well as differential privacy, and evaluate its performance in simulations and real data analyses in comparison with several recently developed methods.

Distributed Statistical Analyses: A Scoping Review and Examples of Operational Frameworks Adapted to Healthcare

“…The majority of articles address a setting where a CC exists external to the nodes, as exemplified by articles such as [28], [51], and [58]. In contrast, as mentioned above, some articles designate one of the nodes to assume this central role, as demonstrated in [9].…”

“…tey et al[6]; Fan, Guo and Wang[17]; Guo, Sun and Jiang[19]; Chen and Xie[11]; Lin and Xi[29]; Rosenblatt and Nadler[41]; Zhang, Duchi and Wainwright[60]; Chang, Bu and Long[9]; Wu et al[56]; Hector and Song[20] Huang and Huo[21]; Jordan, Lee and Yang[24]; Mozafari-Majd and Koivunen[35][36]; Yue, Kontar and Gómez[58]; Duan, Ning and Chen[13]; Duan et al[14]; Tong et al[47]; Di, Wang and Lian[12]; Edmondson et al[16]; Luo and Li[33]; Shu, Young and Toh[44] Semi-parametric regression Zhao, Cheng and Liu[61]; Park et al[38] …”