Residents living in urban areas in the United States (US) have been shown to have higher mortality from myocardial infarction (MI) (118.2 deaths/1 000 000 people) than those residing in urban areas (106.2 deaths/1 000 000 people).1 In accordance with American Heart Association guidelines, cardiac rehabilitation (CR) is recommended for MI patients after discharge, as it has been shown to improve recovery and post-MI survival.2 Despite evidence of the benefit of CR existing literature shows national CR attendance in the US remains below 70% with lack of transportation, inability to miss work, and cost cited as potential barriers to attendance.3-5
Current initiatives to increase CR attendance (such as the Million Hearts program, a joint collaboration between the Centers for Disease Control and Prevention [CDC] and the Centers for Medicare & Medicaid Services to reduce cardiovascular events within a 5-yr period by 1 million) have sought to reduce the impact of CR barriers through interventions that include increasing NIH funding for programs testing the effectiveness of in-home CR and advocating for personnel that provide patients with additional support during the CR referral and admittance process.3-5 However, efforts to improve low CR attendance may be complicated by limited contemporary information about CR attendance across rural and urban areas in the US or within patient subgroups living in geographic areas.6,7 Our aim was to compare CR attendance among MI survivors by rural/urban residence and sex and age subgroups.
METHODS
We identified US adult MI survivors in the 2011, 2013, 2015, and 2017 Behavioral Risk Factor Surveillance System surveys (BRFSS), the most recent survey years that collected CR data.8,9 The BRFSS is an annual mobile and telephone survey conducted by the CDC that collects information on participant sociodemographic background, existing health conditions, and health behaviors.8,9 Due to the weighting scheme and oversampling of certain underrepresented groups, estimates from the BRFSS are representative of the rural/urban distribution found in the general population of the state.8,9
Individual MI status was determined using the survey question, "(Ever told) you had a heart attack, also called a myocardial infarction? (Yes/No)."8,9 The outcome CR attendance was identified by the survey question, "Following your heart attack, did you go to any kind of outpatient rehabilitation? This is sometimes called "rehab. (Yes/No)."8,9 Rural/urban residence was based on BRFSS metropolitan status codes.8 Self-reported information on sociodemographic factors (age, sex, race, education, income, health insurance status, and having a personal doctor) was obtained from the BRFSS.8,9
Logistic regression was used to examine the association between residence and CR attendance, with adjustment for sociodemographic factors. Subgroup analyses were conducted by sex and age (18-44, 45-65, and >65 yr). All analyses used survey weights and were conducted in SAS 9.4 (SAS Institute).
RESULTS
This study included 10 731 MI survivors (43.2% rural residents) from 20 participating states that included the CR survey questions. On average, MI survivors in rural areas were more likely to be White, have a lower income, and complete fewer years of education than their urban counterparts. Attendance in CR was comparable among rural (35.3%: 95% CI, 32.8-37.9%) and urban (37.5%: 95% CI, 35.2-39.7%) residents (Figure). The association between rural/urban residence and CR was nonsignificant in analyses that adjusted for age, sex, race, education, income, health insurance status, and having a personal doctor (OR = 0.95: 95% CI, 0.82-1.10). Women living in rural areas and younger patients 18-44 yr of age had lower CR attendance than other subgroups.
DISCUSSION
Overall, we found that approximately one-third of patients reported attending CR after their MI, with no difference in attendance by rural/urban residence. We also found low CR attendance for women living in rural areas compared with men, and CR attendance was lowest among patients 18-44 yr of age, with only one in seven of these younger MI patients attending CR. Our results are consistent with a prior CDC report using the 2005 BRFSS, which found that 30.3% (95% CI, 27.4-33.3%) of rural MI survivors attended CR compared with 38.1% of urban MI survivors; however, this study did not report attendance by age or sex subgroups within residence categories. The overall lack of an improvement in CR attendance over the past decade of BRFSS assessments is noteworthy.7 A study of Medicare beneficiaries hospitalized in 1997 for either MI or coronary bypass surgery found that urban residents were 36% less likely than rural residents to attend CR.6 Differences between our studies may reflect the advanced age of MI patients in the Medicare population and the assessment of attendance for both MI and coronary bypass surgery patients.6
Several limitations should be noted. The BRFSS surveys are based on self-report and do not include information on potential barriers to CR attendance identified in other studies including unavailability of local CR programs, differing CR referral patterns, lack of transportation, limited time, work demands, caregiving responsibilities, and cost associated with care among MI survivors.6,8 However, the BRFSS estimates have been shown to have high agreement with electronic health records and in-person measurements when validated by the CDC and non-CDC investigators.8
CONCLUSION
Our study highlighted gaps in CR attendance by age, sex, and urban/rural residence status. Additional research is needed to identify factors that contribute to low rates for subgroups of MI patients, particularly for women and younger patients.
Phoebe M. Tran, MS
Cenjing Zhu, BS
Department of Chronic Disease Epidemiology,
Yale University, New Haven, Connecticut
Rachel Dreyer, PhD
Department of Emergency Medicine, Yale University,
New Haven, Connecticut
Judith H. Lichtman, PhD, MPH
Department of Chronic Disease Epidemiology,
Yale University, New Haven, Connecticut
REFERENCES