Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Objectives: To estimate and compare annual medical expenditures and utilization attributed to hypertension and its complications in men and women. We also investigated predisposing, enabling and need predictors of gender differences in hypertension attributable medical expenditures and utilization.

Methods: Procedural equity was assessed by applying a theoretical framework that incorporates equity, efficiency and effectiveness as three major criterions to evaluate health policy (Aday et.al, 2004). Data were extracted from the 2001-2004 Medical Expenditure Panel Survey (MEPS), a panel study of the U.S. non-institutionalized population. Hypertensive individuals were identified using ICD-9 codes or patients who self-reported of being diagnosed of hypertension. A cross-sectional study design was used. The unit of analysis was person/year. The four equation regression model was used to predict total medical expenditures and the two equation regression model was used to estimate expenditures for inpatient stay, emergency room visits, outpatient visits, office based visits and prescription drug. Medical inflation rate was used to convert 2001-2003 dollars to 2004 dollars. Negative binomial regressions were used to predict utilization measures. Interaction between hypertension and gender was used to study gender differences. In addition, interactions between hypertension and other control variables (p<|t|?0.10) were used to differentiate different levels of health care utilization between hypertensive and non-hypertensive individuals. Incremental health care utilizations attributed to hypertension was estimated as the difference in utilization and expenditures between hypertensive and non-hypertensive individuals. Percentage changes in total and attributable utilization and expenditures of hypertensive men relative to hypertensive women were also estimated Results: Sample comprised of 81,307 individuals (weighted sample of 748 million individuals) ?18 years of age. Approximately 23.6% of the sample reported hypertension. The weighted mean of total medical expenditure for hypertensive women and hypertensive men was $5,517 and $5,276 respectively. However, total expenditure attributable to hypertension for hypertensive women and hypertensive men was $3,279 and $3,918 respectively. Therefore, hypertensive men had 19.5% more attributable expenditures relative to hypertensive women. After controlling for other factors, hypertensive women had significantly (p<0.05) more office visits, prescription drug use and outpatient visits. However, similar to total expenditure, hypertensive men had 27.3% more office visits, 2.91% more prescription drug, and 10.9 % more outpatient visits that is attributable to hypertension than hypertensive women. Conclusion: Gender remained a significant predictor of resource use even after controlling for predisposing, enabling and need factors. Hypertensive women had more utilization and expenditures than hypertensive men. However, hypertensive men had more utilization and expenditure attributed to hypertension than hypertensive women. This implies that hypertensive men and women with similar health status and co-morbid conditions did not have similar hospital utilization, office visits and other measures of realized access or health care utilization regardless of their gender, age, income or insurance. Thus, according to procedural equity, hypertensive men and women were not being treated equally.