Learning from electronic health records across multiple sites: A communication-efficient and privacy-preserving distributed algorithm

Abstract

Objectives We propose a one-shot, privacy-preserving distributed algorithm to perform logistic regression (ODAL) across multiple clinical sites. Materials and Methods ODAL effectively utilizes the information from the local site (where the patient-level data are accessible) and incorporates the first-order (ODAL1) and second-order (ODAL2) gradients of the likelihood function from other sites to construct an estimator without requiring iterative communication across sites or transferring patient-level data. We evaluated ODAL via extensive simulation studies and an application to a dataset from the University of Pennsylvania Health System. The estimation accuracy was evaluated by comparing it with the estimator based on the combined individual participant data or pooled data (ie, gold standard). Results Our simulation studies revealed that the relative estimation bias of ODAL1 compared with the pooled estimates was <3%, and the ratio of standard errors was <1.25 for all scenarios. ODAL2 achieved higher accuracy (with relative bias <0.1% and ratio of standard errors <1.05). In real data analysis, we investigated the associations of 100 medications with fetal loss during pregnancy. We found that ODAL1 provided estimates with relative bias <10% for 85% of medications, and ODAL2 has relative bias <10% for 99% of medications. For communication cost, ODAL1 requires transferring p numbers from each site to the local site and ODAL2 requires transferring (p×p+p) numbers from each site to the local site, where p is the number of parameters in the regression model. Conclusions This study demonstrates that ODAL is privacy-preserving and communication-efficient with small bias and high statistical efficiency.

Exploring the Feasibility of Using Real-World Data from a Large Clinical Data Research Network to Simulate Clinical Trials of Alzheimer’s Disease

“…The consistent findings from our analysis suggested the ODAL method can be used to identify clinically relevant risk factors for OUD using distributed real-world data. Moreover, due to the low event rates of OUD in the clinical sites, the estimates provided by the ODAL algorithm outperform those of meta-analysis, which is consistent with the findings in Duan et al 20 . ODAL may be especially valuable for studying rare outcomes or exposures in a multicenter analysis.…”

“…In multicenter studies, sharing data is a major challenge due to privacy concerns 16 . To circumvent the issue of sharing individual patient-level data, many distributed algorithms have been developed to jointly study multiple datasets by communicating only summary-level statistics [17][18][19][20] . Among them, Duan et al [19][20] proposed a privacy-preserving Oneshot Distributed Algorithm for Logistic regression (ODAL), which can be used to identify risk factors of a binary healthcare outcome of interest.…”

“…To circumvent the issue of sharing individual patient-level data, many distributed algorithms have been developed to jointly study multiple datasets by communicating only summary-level statistics [17][18][19][20] . Among them, Duan et al [19][20] proposed a privacy-preserving Oneshot Distributed Algorithm for Logistic regression (ODAL), which can be used to identify risk factors of a binary healthcare outcome of interest. Compared to existing methods, ODAL requires only one round of communication across sites and can achieve high accuracy as a pooled analysis in which a logistic regression is fitted on the combined dataset 20 .…”

“…Among them, Duan et al [19][20] proposed a privacy-preserving Oneshot Distributed Algorithm for Logistic regression (ODAL), which can be used to identify risk factors of a binary healthcare outcome of interest. Compared to existing methods, ODAL requires only one round of communication across sites and can achieve high accuracy as a pooled analysis in which a logistic regression is fitted on the combined dataset 20 . However, Duan et al 20 evaluated the performance of ODAL using simulated data and random partition of a real-world dataset from a single health system.…”

Review of Clinical Research Informatics

“…A radically different concept in privacy preserving analysis is distributed computation. A University of Pennsylvania-led team describes two performant algorithms [99,100], differentiated by resource requirements and performance, which analyze data behind their home firewalls and aggregate statistical results, mainly regression models. Their particular success lies in controlling for data source heterogeneity and maintaining high faithfulness to gold standard (i.e., analysis of aggregated data).…”