- University of Mary United States
- Washington State University United States
- Universität Hamburg Germany
- National Institutes of Health United States
- University Medical Ctr Hamburg-Eppendorf
- University of Chicago Finland
- Kaiser Permanente United States
- University of Chicago United States
- CASE WESTERN RESERVE UNIVERSIT
- Louisiana State University Health Sciences Center United States
- Louisiana State University Health Sciences Center New Orleans United States
- National Human Genome Research Institute United States
- University Medical Center Hamburg-Eppendorf
- CASE WESTERN RESERVE UNIVERSITY
- The University of Chicago
- University Medical Center Hamburg-Eppendorf Germany
- Case Western Reserve University United States
- University Medical Center Hamburg-Eppendorf Germany
- University of British Columbia Canada
- University of Washington United States
- The University of Texas MD Anderson Cancer Center United States
- Oklahoma Medical Research Foundation United States
The underlying ethos of dbGaP is that access to these data by secondary data analysts facilitates advancement of science. NIH has required that genome-wide association study data be deposited in the Database of Genotypes and Phenotypes (dbGaP) since 2003. In 2013, a proposed updated policy extended this requirement to next-generation sequencing data. However, recent literature and anecdotal reports suggest lingering logistical and ethical concerns about subject identifiability, informed consent, publication embargo enforcement, and difficulty in accessing dbGaP data. We surveyed the International Genetic Epidemiology Society (IGES) membership about their experiences. One hundred and seventy five (175) individuals completed the survey, a response rate of 27%. Of respondents who received data from dbGaP (43%), only 32% perceived the application process as easy but most (75%) received data within five months. Remaining challenges include difficulty in identifying an institutional signing official and an overlong application process. Only 24% of respondents had contributed data to dbGaP. Of these, 31% reported local IRB restrictions on data release; an additional 15% had to reconsent study participants before depositing data. The majority of respondents (56%) disagreed that the publication embargo period was sufficient. In response, we recommend longer embargo periods and use of varied data-sharing models rather than a one-size-fits-all approach.