COMBINING ELECTRONIC HEALTH RECORDS AND PATIENT-REPORTED INFORMATION IN THE UK- PRELIMINARY RESULTS FROM WASPS (WALES SAIL + PRO STUDY)
Author(s)
Cascade E1, Ford D2, Nixon M3, Brooks C2, Heaven M21MediGuard.org, Rockville, MD, USA, 2Swansea University, Swansea, United Kingdom, 3Quintiles Global Consulting, Bracknell, Berkshire, United Kingdom
Presentation Documents
OBJECTIVES: As demand for real-world data increases, the need to adopt more time and cost-efficient research methods will grow. Direct-to-patient studies (direct patient recruitment without physician sites) are a novel, efficient approach gaining proof in the United States. This study’s objective was to deploy a direct-to-patient observational study in the UK including both patient-reported and electronic health data. METHODS: In October/November 2011, UK MediGuard.org members with a cholesterol marker in their profile were invited to participate via email; enrollment was supplemented with digital outreach. Interested individuals clicked a link to access study information and screen based on self-reported diagnosis and willingness to have data linked to the SAIL warehouse. Study subjects completed an on-line questionnaire and provided identifiers to be used only for data linkage. The SAIL system’s NHS Trusted Third Party converted study identifiers into pseudo-identifiers allowing the linkage of patient-reported data with electronic health data. RESULTS: A total of 240 enrolled within 6 weeks: 98 from MediGuard and 142 through other digital channels. NHS matched 226 of 240 (94%) to a pseudo-identifier (69 exactly, 157 with probability>0.9). A total of 91 of 226 matched (41%) had at least one primary care record in SAIL and nearly all (89 of 91, 98%) had data suggesting a cholesterol problem. Matching cholesterol test dates (±6months) were located for 66 of 89 (74%, Spearman correlation=0.786, p<0.0001) and there was a high positive correlation of total cholesterol values amongst the 1/3 of patients able to self-report data (n=28, Pearson correlation=0.785, P<0.0001). Additional analyses exploring markers that predict patient's ability to self-report are underway. CONCLUSIONS: This study demonstrates the feasibility of linking patient reported data with electronic health information. UK individuals are interested in participating, provide truthful information, and their patient-reported data can be linked to electronic information in the SAIL warehouse in Wales, thus offering new research possibilities.
Conference/Value in Health Info
2012-06, ISPOR 2012, Washington, D.C., USA
Value in Health, Vol. 15, No. 4 (June 2012)
Code
PCV124
Topic
Study Approaches
Disease
Cardiovascular Disorders