Patterns and costs for hypertension treatment in the United States - Clinical, lifestyle and socioeconomic predictors from the 1987 National Medical Expenditures Survey

Objective: To estimate the impact of clinical and non-clinical predictors o f patterns of medication use and expenditures for the treatment of hyperten sion in the USA. Data Sources: The 1987 National Medical Expenditures Survey was used to ide ntify 6398 individuals with hypertension over the age of 18 years. Pharmaco logical treatment was identified through patient self-reports of antihypert ensive medications. Study Design: This retrospective, cross-sectional study used a multivariate two-stage decision model to estimate the demand for antihypertensive medic ations conditional on receipt of at least one antihypertensive prescription drug. Results: Women and the elderly were more likely to obtain medications and h ad greater expenditures on antihypertensive medications. Privately insured patients were 59% (if non-elderly) or 163% (if elderly with Medicare) more likely to receive drug therapy than uninsured patients. Patients with only Medicaid coverage were 126% more likely to receive drug therapy than uninsu red patients. Compared with patients characterised as lower risk-takers, ve ry high and high risk-takers were 38% and 24% less likely to be on drug the rapy, respectively. Black, non-Hispanics were 30% more likely to be on drug therapy than White, non-Hispanics, but had lower annual expenditures on an tihypertensive drugs. Severely overweight individuals [bodymass index (BMI) >30] were 62% more likely than patients with a BMI <27 to be on drug thera py and also had higher drug expenditures. Conclusions: Insurance had a more striking effect on access to antihyperten sive drug therapy than on patterns of drug use or expenditures. Race/ethnic ity and patient attitudes towards, risk were important determinants of acce ss to antihypertensive drug therapies, as well as patterns of drug use and expenditures.