The WHI data dictionary can be found here.
Type of Data
Speciality/Women’s Health Database
Years Available
The original WHI study began enrolling women and collecting data in 1993, with the original study ending enrollment in 2005. Extension studies have continued to provide follow-up data on enrolled women since 2005, with plans to continue through 2027. Data are updated yearly.
Short Description
The Women’s Health Initiative (WHI) is a long-term national health study funded by the National Heart, Lung, and Blood Institute (NHLBI). Original goals of the study were to research osteoporotic fractures, cardiovascular disease, and cancer outcomes in post-menopausal women. Additional research aims pertaining to outcomes such as frailty and aging have been studied as a part of extensions to the main WHI study as well as ancillary studies, such as the Life and Longevity After Cancer Study (WHI-LILAC) and the WHI Memory Study (WHIMS). Additional information on the WHI are available on WHI.org and the NHLBI website.
Strengths
- A large sample size and many variables and outcomes are available.
- Follow-up data encompass many years, and continue to be collected via extensions studies.
- The WHI has multiple ancillary studies that have collected additional data from subsets of women enrolled to WHI
Limitations
- Manuscript proposals must be submitted to the WHI Publications and Presentations committee and be approved before access to the data is granted.
- Data are available as SAS datasets, so experience is required to convert for use in other software.
- Women who are white and more educated are likely over-represented in the WHI population.
- Time points for follow-up visits differ between women enrolled in the observational study arm and the clinical trial arms. Frequencies of data collection for various study arms can be viewed here.
- Exact dates are not available in the standard WHI data. All time-to-event data are calculated as days from randomization/enrollment. WHI does offer a virtual data enclave to access sensitive data (such as Medicare data, geo-coded data, and exact dates), which requires additional approval.
SDC Members with Expertise
- Eric McLaughlin (2 publications)