Dataset Information

Comparison of privacy-protecting analytic and data-sharing methods: A simulation study.

ABSTRACT: PURPOSE:Privacy-protecting analytic and data-sharing methods that minimize the disclosure risk of sensitive information are increasingly important due to the growing interest in utilizing data across multiple sources. We conducted a simulation study to examine how avoiding sharing individual-level data in a distributed data network can affect analytic results. METHODS:The base scenario had four sites of varying sizes with 5% outcome incidence, 50% treatment prevalence, and seven confounders. We varied treatment prevalence, outcome incidence, treatment effect, site size, number of sites, and covariate distribution. Confounding adjustment was conducted using propensity score or disease risk score. We compared analyses of three types of aggregate-level data requested from sites: risk-set, summary-table, or effect-estimate data (meta-analysis) with benchmark results of analysis of pooled individual-level data. We assessed bias and precision of hazard ratio estimates as well as the accuracy of standard error estimates. RESULTS:All the aggregate-level data-sharing approaches, regardless of confounding adjustment methods, successfully approximated pooled individual-level data analysis in most simulation scenarios. Meta-analysis showed minor bias when using inverse probability of treatment weights (IPTW) in infrequent exposure (5%), rare outcome (0.01%), and small site (5,000 patients) settings. SE estimates became less accurate for IPTW risk-set approach with less frequent exposure and for propensity score-matching meta-analysis approach with rare outcomes. CONCLUSIONS:Overall, we found that we can avoid sharing individual-level data and obtain valid results in many settings, although care must be taken with meta-analysis approach in infrequent exposure and rare outcome scenarios, particularly when confounding adjustment is performed with IPTW.

SUBMITTER: Yoshida K

PROVIDER: S-EPMC6135666 | biostudies-literature | 2018 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Comparison of privacy-protecting analytic and data-sharing methods: A simulation study.

Yoshida Kazuki K Gruber Susan S Fireman Bruce H BH Toh Sengwee S

Pharmacoepidemiology and drug safety 20180718 9

<h4>Purpose</h4>Privacy-protecting analytic and data-sharing methods that minimize the disclosure risk of sensitive information are increasingly important due to the growing interest in utilizing data across multiple sources. We conducted a simulation study to examine how avoiding sharing individual-level data in a distributed data network can affect analytic results.<h4>Methods</h4>The base scenario had four sites of varying sizes with 5% outcome incidence, 50% treatment prevalence, and seven c ...[more]

PMID: 30022561

Similar Datasets

Project description:Data sharing between technology companies and academic health researchers has multiple health care, scientific, social, and business benefits. Many companies remain wary about such sharing because of unaddressed concerns about ethics, data security, logistics, and public relations. Without guidance on these issues, few companies are willing to take on the potential work and risks involved in noncommercial data sharing, and the scientific and societal potential of their data goes unrealized. In this paper, we describe the 18-month long pilot of a data-sharing program led by Crisis Text Line (CTL), a not-for-profit technology company that provides a free 24/7 text line for people in crisis. The primary goal of the data-sharing pilot was to design, develop, and implement a rigorous framework of principles and protocols for the safe and ethical sharing of user data. CTL used a stakeholder-based policy process to develop a feasible and ethical data-sharing program. The process comprised forming a data ethics committee; identifying policy challenges and solutions; announcing the program and generating interest; and revising the policy and launching the program. Once the pilot was complete, CTL examined how well the program ran and compared it with other potential program models before putting in place the program that was most suitable for its organizational needs. By drawing on CTL's experiences, we have created a 3-step set of guidelines for other organizations that wish to develop their own data-sharing program with academic researchers. The guidelines explain how to (1) determine the value and suitability of the data and organization for creating a data-sharing program; (2) decide on an appropriate data sharing and collaboration model; and (3) develop protocols and technical solutions for safe and ethical data sharing and the best organizational structure for implementing the program. An internal evaluation determined that the pilot satisfied CTL's goals of sharing scientific data and protecting client confidentiality. The policy development process also yielded key principles and protocols regarding the ethical challenges involved in data sharing that can be applied by other organizations. Finally, CTL's internal review of the pilot program developed a number of alternative models for sharing data that will suit a range of organizations with different priorities and capabilities. In implementing and studying this pilot program, CTL aimed both to optimize its own future data-sharing programs and to inform similar decisions made by others. Open data programs are both important and feasible to establish. With careful planning and appropriate resources, data sharing between big data companies and academic researchers can advance their shared mission to benefit society and improve lives.

Project description:BACKGROUND:Sharing research data uses resources effectively; enables large, diverse data sets; and supports rigor and reproducibility. However, sharing such data increases privacy risks for participants who may be re-identified by linking study data to outside data sets. These risks have been investigated for genetic and medical records but rarely for environmental data. OBJECTIVES:We evaluated how data in environmental health (EH) studies may be vulnerable to linkage and we investigated, in a case study, whether environmental measurements could contribute to inferring latent categories (e.g., geographic location), which increases privacy risks. METHODS:We identified 12 prominent EH studies, reviewed the data types collected, and evaluated the availability of outside data sets that overlap with study data. With data from the Household Exposure Study in California and Massachusetts and the Green Housing Study in Boston, Massachusetts, and Cincinnati, Ohio, we used k-means clustering and principal component analysis to investigate whether participants' region of residence could be inferred from measurements of chemicals in household air and dust. RESULTS:All 12 studies included at least two of five data types that overlap with outside data sets: geographic location (9 studies), medical data (9 studies), occupation (10 studies), housing characteristics (10 studies), and genetic data (7 studies). In our cluster analysis, participants' region of residence could be inferred with 80%-98% accuracy using environmental measurements with original laboratory reporting limits. DISCUSSION:EH studies frequently include data that are vulnerable to linkage with voter lists, tax and real estate data, professional licensing lists, and ancestry websites, and exposure measurements may be used to identify subgroup membership, increasing likelihood of linkage. Thus, unsupervised sharing of EH research data potentially raises substantial privacy risks. Empirical research can help characterize risks and evaluate technical solutions. Our findings reinforce the need for legal and policy protections to shield participants from potential harms of re-identification from data sharing. https://doi.org/10.1289/EHP4817.

Dataset Information

Comparison of privacy-protecting analytic and data-sharing methods: A simulation study.

Publications

Comparison of privacy-protecting analytic and data-sharing methods: A simulation study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets