Venous thromboembolism incidence, recurrence, and mortality based on Women's Health Initiative data and Medicare claims.

INTRODUCTION: Our objective was to compare Medicare claims to physician review and adjudication of medical records for identifying venous thromboembolism (VTE), and to assess VTE incidence, recurrence, and mortality in a large national cohort of post-menopausal women followed up to 19years. MATERIALS AND METHODS: We used detailed clinical data from the Women's Health Initiative (WHI) linked to Medicare claims. Agreement between data sources was evaluated among 16,003 women during 1993-2010. A claims-based definition was selected to analyze VTE occurrence and impact among 71,267 women during 1993-2012. RESULTS: Our VTE definition had 83% sensitivity. Positive predictive value was 69% when all records were included, and 94% after limiting Medicare records to those with a WHI hospitalization adjudicated. Annualized VTE incidence was 4.06/1000person-years (PY), recurrence was 5.30/100PY, and both rates varied by race/ethnicity. Post-VTE mortality within 1year was 22.49% from all causes, including 1.01% from pulmonary embolism, 10.40% from cancer, and 11.08% from other causes. Cancer-related VTE compared to non-cancer VTE had significantly (p<0.001) higher recurrence (9.86/100PY vs. 4.43/100PY) and mortality from all causes (45.89% vs. 12.28%), but not from pulmonary embolism (0.40% vs. 1.27%). CONCLUSIONS: Medicare claims compared reasonably well to physician adjudication. The combined data sources provided new insights about VTE burden and prognosis in older women.