Burden of Cardiovascular Disease in Asian Americans

S (ACE) AEP Vol. 20, No. 9 September 2010: 691–724 704 adolescents to reduce the burden associated with these mental disorders. P37 BURDEN OF CARDIOVASCULAR DISEASE IN ASIAN AMERICANS EC Wong, AT Holland, DS Lauderdale, LP Palaniappan, Palo Alto Medical Foundation Research Institute, Palo Alto, CA, University of Chicago, Chicago, IL PURPOSE: To compare the prevalence of stroke, coronary heart disease (CHD) and peripheral vascular disease (PVD) across Asian American subgroups (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese) and Non-Hispanic Whites (NHW) in an multispecialty, healthcare organization in California. METHODS: Electronic health records of 98,451 Asian and NHW patients were examined. Billing and problem list codes were used to define prevalent stroke (ICD-9: 430–438), CHD (410–414), and PVD (415.1, 440.2, 440.3, 443.9, 451, 453) from 2007–2010. Age-adjusted rates were calculated using logistic regression, and compared to NHWs as the referent group. RESULTS: The range of age-adjusted rates for men and women were as follows: stroke (0.8–1.2%; 0.5–1.2%), CHD (1.5–4.6%; 0.2–1.3%), PVD (0.2–1.5%; 0.4–1.6%). Compared to NHWs, ischemic stroke rates were higher in Filipina women (OR: 1.77, 95% CI: 1.16–2.69). CHD rates were higher in Asian Indian men (1.83, 1.48–2.26) and in Filipino men (1.45, 1.04–2.02) and women (1.64, 1.12– 2.40). CHD rates were lower in Chinese men (0.78, 0.65– 0.92) and women (0.74, 0.58–0.96). Lower rates of PVD were noted for men and women in nearly all Asian racial/ ethnic subgroups when compared to NHWs. CONCLUSION: There is considerable heterogeneity in Asian subgroups with regard to stroke, CHD and PVD. Filipinos andAsian Indians generally diverge from other Asian subgroups. While high rates of CHD among Asian Indians are increasingly recognized, Filipinos are less well studied. Future studies should strive to disaggregate Asian subgroups and disparate cardiovascular outcomes to better understand variation in these disease patterns. P38 CHANGE IN QUALITY OF LIFE AMONG CONGESTIVE HEART FAILURE PATIENTS MM Donneyong, CA Hornung, Department of Epidemiology, University of Louisville, Louisville, KY PURPOSE: To determine the combined effect of severity of congestive heart failure (CHF), type of patient management, and compliance to the AmericanHeart Association (AHA)/ American College of Cardiologists (ACC) treatment guidelines on change in quality of life among CHF patients. METHODS:Data was analyzed from a three-arm treatment, multi-site randomized control study. A total of 134 participants were enrolled in the study for 9 months with follow upat 12months.TheKansasCityCardiomyopathyQuestionnaire was administered at baseline and follow up to determine change in quality of life (QoL) among participants. Three categories of change in QoL (decline, no change, and improvement)were defined based onCohen’s small effect size parameter estimate (0.2 standard deviation). Multinomial logistic regression models were fitted between change in QoL (outcome) and arm of treatment, NHYA class, and compliance to AHA/ACC guidelines (as predictor variables) controlling for potential confounders. RESULTS: While 46% of the population had improved QoL, 18% and 36% had no change and decline in QoL respectively. NYHA class III&IV were associated with increased odds of decline as well as improvement in QoL compared to class I&II. Compliance to AHA/ACC guidelines was associated with increased odds of decline but reduced odds of improvement. CONCLUSION: NHYA class III&IV status suggested stability in QoL while compliance to AHA/ACC guidelines may lead to decline in QoL. P39 DIABETES AND KIDNEY CANCER: A DIRECT OR INDIRECT ASSOCIATION? RP Ojha, EL Evans, TN Offutt-Powell, BS Nuwayhid, ZD Mulla, Department of Epidemiology, UNT Health Science Center, Fort Worth, TX PURPOSE: Several studies have reported increased relative risks of kidney cancer among diabetics, but these estimates may not indicate the direct association because certain factors mediate this exposure-outcome relation. Therefore, we analyzed case-control data to estimate the total (i.e. indirect and direct) and direct associations between diabetes and kidney cancer. METHODS: Discharge data (with International Classification of Diseases – 9 codes) from 2001 for hospitals throughout Florida were used to construct a case-control population of inpatients aged 45 years. Cases (nZ1,909) were inpatients with malignant kidney cancer and controls (nZ6,451) were inpatients with motor vehicle injuries. Diabetes status was ascertained for cases and controls. Covariates that required adjustment to estimate the total (age, gender, ethnicity, obesity, and smoking) and direct (age, gender, ethnicity, obesity, smoking, hypertension, and kidney disease) associationswere identified in adirected acyclic graph.Unconditional logistic regression was used to estimate the adjusted total and direct odds ratios (ORs) and corresponding 95% confidence intervals (CIs) of kidney cancer for diabetics. RESULTS:The odds of kidney cancer were higher for inpatients with diabetes than inpatients without diabetes when