We aimed to evaluate the association among race and ethnicity, statin prescription, and clinical outcomes among hospitalized patients with coronary heart disease (CHD), adjusted for confounders. Racial and ethnic disparities in CHD outcomes may be related to differential uptake of preventive medications, but data from real-world settings are limited. This was a 1-year prospective study of patients with preexisting CHD without a documented contraindication to statin (n = 3,067, 35% black or Hispanic, 65% white or Asian, 35% women) who participated in an National Heart, Lung and Blood Institute clinical outcome study of patients admitted to a cardiovascular service. Baseline clinical and medication data and 30-day and 1-year outcomes (death or rehospitalization) were documented by electronic medical record, National Death Index, and/or standardized mail survey. Logistic regression was used to evaluate associations among race and ethnicity, statin prescription, and outcomes adjusted for demographics and co-morbidities. Black and Hispanic patients were more likely to be dead or rehospitalized at 1 year (odds ratio [OR] 1.23, 95% confidence interval [CI] 1.06 to 1.43) and less likely to report statin use before admission (62% vs 72%, adjusted OR 0.64, 95% CI 0.54 to 0.76) than whites and Asians; statin prescription was similar at discharge among blacks and Hispanics (81%) versus whites and Asians (84%). Black and Hispanic patients were more likely to have hypertension, diabetes, or renal failure and less likely to have health insurance than whites and Asians (p <0.05). The increased 1-year odds of death or rehospitalization in minorities versus whites and Asians were explained by demographics and co-morbidities not by differential statin prescription (adjusted OR 1.10, 95% CI 0.93 to 1.30). In conclusion, in this study of hospitalized patients with preexisting CHD, differential statin prescription did not explain racial and ethnic disparities in 1-year outcomes. Efforts to reduce CHD rehospitalizations should consider the greater burden of co-morbidities among racial and ethnic minorities.
It is well documented that statins reduce risk for major vascular events and all-cause mortality in patients with existing coronary heart disease (CHD) ; recent meta-analysis of >50 cross-sectional studies, cohort studies, and randomized controlled clinical trials that quantified statin adherence showed that uptake of statins was less among racial and ethnic minorities compared with whites. Racial and ethnic differences in CHD clinical outcomes may be attributable to differences in the uptake of statin therapy, but this is not established. The purpose of this study was to evaluate the association between race and ethnic group, statin prescription, and the rates of death and hospital readmission in the short term (30 days) and long term (1 year) among patients with preexisting CHD admitted to the cardiology service at a major university hospital, adjusted for demographic factors and co-morbid medical conditions.
Methods
The study cohort consisted of 3,067 patients consecutively admitted to the cardiovascular disease (CVD) service at New-York Presbyterian Hospital/Columbia University Medical Center (NYPH/CUMC) who took part in the Family Cardiac Caregiver Investigation To Evaluate Outcomes (FIT-O) study sponsored by the National Heart, Lung and Blood Institute. The design and methods of FIT-O have been previously published. Briefly, FIT-O was a prospective observational study among 4,500 patients hospitalized for CVD designed to evaluate the association between having a caregiver and clinical outcomes 1 year after hospitalization. Consecutive patients were recruited from November 2009 to June 2010 and were excluded from participation if they were unable to read English or Spanish, lived in a full-time nursing facility, were unable to participate because of mental status, or refused to participate for any reason. The overall enrollment rate was 93%. Participants were included in this analysis if they had a documented medical history of CHD and/or a CHD-equivalent diagnosis of diabetes mellitus, peripheral vascular disease, abdominal aortic aneurysm, or other atherosclerotic diseases before admission (n = 3,260). Potential participants were excluded if they did not have a race and ethnic group documented in their medical record (n = 175) or if they had a documented contraindication to lipid-lowering agents in their medical record (n = 18). This study was approved by the CUMC Institutional Review Board.
Standardized electronic chart review was conducted by trained research staff who documented (1) demographic characteristics (race and ethnicity, age, gender, and health insurance status), (2) co-morbid medical conditions (hypertension, diabetes, renal failure or dialysis, peripheral vascular disease, chronic obstructive pulmonary disease, heart failure, stroke, history of myocardial infarction or coronary artery bypass surgery, and admission type [surgical (cardiac) vs not surgical]), and (3) smoking status. International Classification of Diseases, ninth revision billing codes and physician or nurse practitioner notes were used to classify medical conditions and were validated by an external physician reviewer. Standardized questionnaires were used to determine caregiver status, defined as a paid professional or a nonpaid person who assisted the cardiac patients with their medical and/or preventive care.
Self-reported statin prescriptions before admission and statins prescribed at discharge were documented by standardized electronic chart review of admission and discharge records. Statin prescription was defined as either a single or combination statin prescription.
Rehospitalization was defined as admission to NYPH/ CUMC, or another hospital over the 1-year period after the index hospitalization associated with each participant’s baseline survey date. Rehospitalization at NYPH/CUMC was systematically collected by hospital electronic clinical information system, which was updated daily, at 30 days, and at 1 year; there was 100% ascertainment of readmission to NYPH/CUMC. Electronic data were supplemented by standardized survey completed by participants 1 year after the index hospitalization through mail or telephone to document hospitalization outside of NYPH/CUMC at 1 year (82% response rate). One-year results were not materially different when analysis was limited to NYPH/CUMC readmissions only; therefore, total readmissions based on clinical information system and survey data were reported. Death outcomes were obtained using the National Death Index, which was used to populate the hospital clinical information system, updated monthly.
Descriptive data are presented as frequencies and percentages. Race and ethnic group was dichotomized as black/Hispanic/other (racial and ethnic minority) versus white and Asian based on the greater risk of recurrent CVD events among black and Hispanic patients with CHD compared with white and/or Asian patients CHD, similar to national statistics. Chi-square tests were used to determine differences in baseline characteristics by race and ethnic group and evaluate the univariate associations between participant characteristics and clinical outcomes at 30 days and 1 year. Logistic regression was used to evaluate whether statin prescription at admission or discharge confounded the association between race and ethnic group and death or rehospitalization at 1 year and to adjust for other potential confounders. A stratified analysis and a test for interaction were conducted to determine whether the association between race and ethnic group and death or rehospitalization at 1 year was differential among those with versus without health insurance. All tests were 2-sided with significance set at p <0.05. Data were analyzed using SAS, version 9.3 (SAS Institute, Cary, North Carolina).
Results
The baseline characteristics of the study population, overall and by race and ethnic group, are listed in Table 1 . Among the 3,067 participants, 35% were women, 65% white or Asian, 21% Hispanic, 13% black, and 1% belonged to other races. Racial and ethnic minorities were more likely to smoke and have several co-morbid medical conditions including hypertension, diabetes, renal failure or dialysis, and stroke compared with whites and Asians. In contrast, racial and ethnic minority patients were less likely to have health insurance compared with white and Asian patients.
| Variable | All Participants, n (%) | Race and Ethnic Group | p | |
|---|---|---|---|---|
| Black/Hispanic/Other, n = 1,063 (%) | White/Asian, n = 2,004 (%) | |||
| Statin prescription | ||||
| At admission | 2,097 (68) | 656 (62) | 1,441 (72) | <0.0001 |
| At discharge | 2,549 (83) | 864 (81) | 1,685 (84) | 0.05 |
| Men | 1,994 (65) | 580 (55) | 1,414 (71) | <0.0001 |
| Age ≥65 yrs | 1,828 (60) | 565 (53) | 1,263 (63) | <0.0001 |
| No health insurance | 384 (13) | 239 (22) | 145 (7) | <0.0001 |
| Has a caregiver | 1,149 (37) | 450 (42) | 699 (35) | <0.0001 |
| Previous/current hypertension | 2,373 (77) | 864 (81) | 1,509 (75) | 0.0002 |
| Previous/current diabetes mellitus | 1,290 (42) | 547 (51) | 743 (37) | <0.0001 |
| Previous/current renal failure/dialysis | 173 (6) | 75 (7) | 98 (5) | 0.02 |
| Previous/current peripheral vascular disease | 476 (16) | 153 (14) | 323 (16) | 0.23 |
| Previous/current chronic obstructive pulmonary disease | 188 (6) | 46 (4) | 145 (7) | 0.001 |
| Previous/current heart failure | 566 (18) | 235 (22) | 389 (19) | 0.08 |
| Previous/current stroke | 313 (10) | 131 (12) | 196 (10) | 0.03 |
| Previous myocardial infarction | 987 (32) | 350 (33) | 637 (32) | 0.54 |
| Previous coronary bypass | 657 (21) | 159 (15) | 498 (25) | <0.0001 |
| Current smoker | 276 (9) | 124 (12) | 152 (8) | 0.0002 |
| Surgical admission | 365 (12) | 64 (6) | 301 (15) | <0.0001 |
Racial and ethnic minorities were less likely than white and Asian participants to report being prescribed a statin before admission (62% vs 72%, p <0.0001). This association remained significant after adjustment for demographic characteristics and co-morbid medical conditions (odds ratio [OR] 0.64, 95% confidence interval [CI] 0.54 to 0.76). Frequency of statin prescription at discharge did not significantly differ between racial and ethnic minority and whites and Asian patients (81% vs 84%, p = 0.05).
At 30 days, 10% of patients had been readmitted (n = 301) and 2% had died (n = 49). The overall rate of death or readmission at 30 days did not vary between racial and ethnic minority (11%) and white and Asian patients (11%, p = 0.62). There was no association between taking a statin before admission and death or readmission at 30 days (OR 1.02, 95% CI 0.80 to 1.31). Patients who were prescribed a statin at discharge were significantly less likely to be readmitted or dead at 30 days compared with those not prescribed a statin at discharge, independent of demographic characteristics and co-morbid medical conditions (OR 0.59, 95% CI 0.45 to 0.77).
At 1 year, 1,497 participants (49%) had been rehospitalized and 238 had died (8%), including 91 participants who were rehospitalized and had died within the 1-year follow-up. Racial and ethnic minority participants were 23% more likely than white and Asian participants to have died or have been rehospitalized at 1 year ( Table 2 ). Statin prescription before admission or at discharge was not associated with death or rehospitalization at 1 year. Significant demographic predictors of death and rehospitalization at 1 year included age ≥65 years, having a caregiver, and lack of health insurance.
| Variable | Rehospitalized or Dead at 1 yr, OR (95% CI) | p |
|---|---|---|
| Black/Hispanic/other ∗ | 1.23 (1.06–1.43) | 0.008 |
| Statin prescription | ||
| At admission | 1.11 (0.95–1.29) | 0.20 |
| At discharge | 0.94 (0.78–1.14) | 0.56 |
| Men | 0.87 (0.75–1.01) | 0.07 |
| Age ≥65 yrs | 1.30 (1.12–1.50) | 0.0004 |
| No health insurance | 1.41 (1.13–1.75) | 0.002 |
| Has a caregiver | 1.56 (1.35–1.81) | <0.0001 |
| Previous/current hypertension | 1.12 (0.94–1.33) | 0.19 |
| Previous/current diabetes mellitus | 1.58 (1.37–1.83) | <0.0001 |
| Previous/current renal failure/dialysis | 2.09 (1.50–2.92) | <0.0001 |
| Previous/current peripheral vascular disease | 1.50 (1.23–1.83) | <0.0001 |
| Previous/current chronic obstructive pulmonary disease | 1.33 (0.99–1.80) | 0.06 |
| Previous/current heart failure | 2.31 (1.91–2.78) | <0.0001 |
| Previous/current stroke | 1.19 (0.95–1.50) | 0.14 |
| Previous myocardial infarction | 1.19 (1.02–1.39) | 0.02 |
| Previous coronary bypass | 1.41 (1.19–1.68) | 0.0001 |
| Current smoker | 0.91 (0.71–1.16) | 0.49 |
| Surgical admission | 0.67 (0.53–0.83) | 0.0003 |
Race and ethnic group was no longer a statistically significant predictor of death and rehospitalization at 1 year after adjustment for co-morbid medical conditions ( Table 3 ). Age ≥65 years, having a caregiver, and lack of health insurance were significant multivariate predictors of death and rehospitalization at 1 year.
| Variable | Model 1: Adjusted for Statin, OR (95% CI) | Model 2: Demographic Adjusted, OR (95% CI) | Model 3: Demographic and Co-morbidity Adjusted, OR (95% CI) |
|---|---|---|---|
| Black/Hispanic/other ∗ | 1.25 (1.07–1.45) | 1.17 (1.00–1.38) | 1.10 (0.93–1.30) |
| Statin prescription | |||
| At admission | 1.20 (1.01–1.42) | 1.16 (0.97–1.37) | 1.08 (0.91–1.30) |
| At discharge | 0.86 (0.70–1.07) | 0.89 (0.71–1.10) | 0.99 (0.79–1.23) |
| Men | 0.93 (0.80–1.09) | 0.92 (0.78–1.07) | |
| Age ≥65 yrs | 1.27 (1.09–1.47) | 1.20 (1.03–1.40) | |
| No health insurance | 1.33 (1.06–1.66) | 1.27 (1.07–1.60) | |
| Has a caregiver | 1.48 (1.28–1.72) | 1.26 (1.08–1.47) | |
| Previous/current hypertension | 0.95 (0.80–1.14) | ||
| Previous/current diabetes mellitus | 1.37 (1.17–1.60) | ||
| Previous/current renal Failure/dialysis | 1.50 (1.06–2.13) | ||
| Previous/current peripheral vascular disease | 1.33 (1.07–1.64) | ||
| Previous/current chronic obstructive pulmonary disease | 1.07 (0.78–1.46) | ||
| Previous/current heart failure | 1.93 (1.58–2.36) | ||
| Previous/current stroke | 0.98 (0.77–1.25) | ||
| Previous myocardial infarction | 1.09 (0.93–1.28) | ||
| Previous coronary bypass | 1.28 (1.06–1.55) | ||
| Current smoker | 1.02 (0.79–1.33) | ||
| Surgical admission | 0.64 (0.51–0.81) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
