Online Program Home
My Program

Abstract Details

Activity Number: 197 - SPEED: Government and Health Policy
Type: Contributed
Date/Time: Monday, July 30, 2018 : 10:30 AM to 11:15 AM
Sponsor: Section for Statistical Programmers and Analysts
Abstract #332736
Title: Open Data Sharing and Its Statistical Limitations
Author(s): Pooja Iyer* and Barbara Do
Companies: RTI International and RTI International
Keywords: Open data; Public Use; NIH; Disclosure techniques; clinical trials

Since October 2003, researchers applying for NIH funding of $500,000 or more are expected to include a data sharing plan. Researchers releasing public or open data need ensure that the data have enough information to reproduce the main research findings while protecting private information from the study participants. Removal of basic identifiers is not enough to ensure study participants' confidentiality. As a result, more rigorous procedures are often needed to ensure privacy protection for all study participants. In this paper, we present different standard privacy preserving techniques (also known as disclosure techniques) for releasing data for public access that have been developed elsewhere. We discuss with detail the methods that are relevant to clinical trials, and illustrate each method with simulated clinical trial data. This paper evaluates each technique, and to what extent the technique limits utility and protects confidentiality. Finally, open data are recorded factual material accepted in the scientific community as necessary to document and validate research findings. We also discuss the components required for an NIH data release.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program