SIMILARITIES AND DIFFERENCES BETWEEN REAL WORLD PATIENT DATA SOURCES: A GLOBAL CASE STUDY

• Published in: Value in health 2017-05, Vol.20 (5), p.A318
• Main Authors: Jaffe, DH, Haskell, T, Feldman, B, Bramlett, J, Bachrach, A, Hyunjung, A, Kamauu, AW, Friedel, H, Pignot, M
• Format: Article
• Language: English
• Subjects: Allopurinol; Claims; Computerized medical records; Data processing; Data sources; Demography; Diagnosis; Disease management; Drugs; Electronic medical records; Geography; Gout; Health care industry; Health insurance; Health records; Health services; Health services utilization; Health status; Help seeking behavior; Medical diagnosis; Polls & surveys; Resource utilization; Rheumatism
• ISSN: 1098-3015; 1524-4733
• DOI: 10.1016/j.jval.2017.05.005

OBJECTIVES: To assess and highlight variations in patients' demographic and clinical characteristics, disease management and healthcare utilization across geographies and data sources using patients with gout as an example. METHODS: A cross-sectional study of gout patients ages 18+ years was conducted using electronic health records (EHRs) from US (a healthcare system from the Anolinx eResearch Network, Anolinx, and Kantar Health Ambulatory EHR, KH) and Israel (Clalit Health Services, CHS); healthcare claims from Germany (BKK) and South Korea (Health Insurance Review & Assessment, HIRA); and patient-reported survey data from the National Health and Wellness Survey (NHWS) from US, Japan, and from five European Union (EU5) countries (France, Germany, Italy, Spain, and UK). Gout patients were identified as having at least one gout diagnosis or healthcare encounter for a single year for each data source, during 2014-2016. Patients from EHR data were required to have an additional diagnosis or gout medication purchase to insure an 'active' (non-historic) gout patient. Demographics, clinical characteristics, disease management and healthcare resource utilization data were examined. RESULTS: 109,975 adult gout patients were identified using EHR (Anolinx n=2148; KH n=51,722; CHS n=10,234), claims (BKK n=31,162; HIRA n=7696), and patient-reported survey data (NHWS-US n=3457; NHWS-Japan n=1172; NHWS-EU5 n=2384). Patient characteristics and healthcare use varied by data source and geography. For example, the proportion of patients 65+ years was similar between EHR datasets (Anolinx=53.9%; KH=51.7%; CHS=51.9%), but varied by data source and geography for claims (BKK=65.1%; HIRA=24.6%) and patient-reported survey data (NHWS-US=42.9%; NHWS-Japan=61.2%; NHWS-EU5=54.2%). Allopurinol use (≥1 prescription/purchase during the study year) was higher in EHR (KH and CHS~66%) compared to patient-reported survey (NHWS-Japan and -EU5 -22%) cohorts. CONCLUSIONS: Collecting and analyzing data from diverse patient populations across geographies, healthcare systems, data sources, and cultures is essential for providing informative real-world evidence of disease management and progression and health outcomes.