Abstract
Background
Veterans are at an increased cardiovascular risk compared to age- and sex-matched non-Veterans. Cardiac rehabilitation (CR) can improve outcomes in cardiovascular disease, but its use in men and women Veterans is not well understood.
Objectives
This study aimed to examine CR participation by sex and socioeconomic status among Veterans.
Methods
The authors conducted a retrospective cohort study from January 1, 2021, to December 31, 2023, using a national electronic health record database. The primary outcome was participation in at least 1 CR session among patients within 1 year of myocardial infarction, percutaneous coronary intervention, or coronary artery bypass surgery. Multivariable logistic regression models accounted for patient-level (demographics, medical/psychiatric comorbidities) and community-level factors. Area deprivation indices (ADIs) (analyzed as quartiles) assessed socioeconomic status.
Results
Among 82,496 CR-eligible Veterans (3.6% women), CR participation was low (10.4%) and similar by sex (women = 10.2%, men = 10.4%). Women Veterans did not differ significantly in CR participation compared to men Veterans after adjusting for patient-level and community-level characteristics, including age, race, cardiac and comorbidities, mental health risk factors, rural-urban status, and ADI (adjusted OR: 0.90; 95% CI: 0.79-1.03; P = 0.121). Veterans in the most deprived ADI quartile were less likely to participate vs the least deprived quartile (adjusted OR: 0.82; 95% CI: 0.75-0.89; P < 0.001).
Conclusions
CR participation among U.S. Veterans remains low, far below that of the Medicare population (10.4% vs 28%), with no significant differences in initiation by sex. However, low socioeconomic status is associated with decreased uptake. Further research is needed to explore innovative, Veteran-specific CR delivery models.
Key words: Area Deprivation Index, cardiac rehabilitation, sex differences, Veterans
Central Illustration
Background
Cardiac rehabilitation (CR) is a secondary preventive intervention that includes exercise training, risk factor modification, and counseling, and is shown to reduce cardiovascular mortality and improve functional outcomes among patients with coronary artery disease.1 Clinical guidelines recommend CR for patients who have had acute coronary syndromes, coronary artery bypass surgery (CABG), and percutaneous coronary intervention (PCI).2,3 Nationally, however, CR remains underutilized, with estimates of only 28% of eligible Medicare beneficiaries participating in at least 1 session in 2017.4 Among Medicare beneficiaries, there was a marked decline in in-person CR participation after the onset of the COVID-19 pandemic, with incomplete recovery over the following years.5,6
CR utilization is not well characterized, particularly for under-represented populations, including Veterans, women, and individuals of lower socioeconomic status.7,8 CR participation among U.S. Veterans has historically been low (13.2% in 2015),9 despite an increased risk for cardiovascular disease (CVD).8,10,11 Women Veterans, in turn, comprise a growing proportion of the Veteran population and experience a significant burden of CVD. In 2023, women Veterans comprised about 11% of the Veteran population, with a projected increase to 19% by 2053.12 Between 2000 and 2017, although civilian women experienced a decline in both crude and age-adjusted CVD mortality rates, women Veterans saw stable crude mortality rates and a 5.3% increase in age-adjusted mortality from cardiac disease.10 Women Veterans have a differing burden of risk factors for CVD compared to men Veterans, with an increased prevalence of several important risk factors such as obesity, diabetes, tobacco use, depression, and post-traumatic stress disorder (PTSD).13, 14, 15, 16 In addition, military sexual trauma (MST), which is estimated to affect 20 to 68% of women in the military, increases the risk of cardiovascular risk factors, including diabetes, hypertension, depression, and PTSD.13,17, 18, 19 For these reasons, understanding the delivery of cardiovascular care to this high-risk cohort is particularly timely. In addition, the interplay of socioeconomic status on sex is not well understood. Cost is often cited as a key barrier to CR participation among women,20, 21, 22 and studies suggest that women of lower socioeconomic status are the least likely to attend CR.23
Given the increased burden of CVD among Veterans, particularly women, we sought to critically evaluate CR utilization to address knowledge gaps and inform policy. We assessed current CR initiation among men and women Veterans, conducted sex-based analyses of utilization, and examined the impact of both patient-level and community-level factors, including socioeconomic status, on participation.
Methods
Study design and data sources
We performed a retrospective study of national Veterans Affairs Electronic Health Record data of the VA Corporate Data Warehouse,24 which includes both administrative and clinical data. We identified all Veterans who had an eligible qualifying event of myocardial infarction (MI), CABG, or PCI by Current Procedural Terminology (CPT) and International Classification of Diseases-10 (ICD-10) codes (Supplemental Table 1). To avoid temporal ambiguity, we defined a “virtual” recruitment period between January 1, 2021, and December 31, 2022, to include Veterans with at least 1 relevant ICD-10 or CPT code as the primary or secondary diagnosis code and25 followed up for CR participation up to 1 year (censored at first CR participation or December 31, 2023, whichever occurred first). For Veterans who had multiple qualifying diagnoses or events within the study period, the earliest qualifying event was used to determine CR eligibility. Veterans may have had multiple qualifying events within 21 days of the index event (e.g. MI and PCI). We excluded Veterans who died within 30 days of the initial qualifying event and those discharged to skilled nursing facilities, long-term acute care, or hospice after the index hospitalizations. The research was approved by the Institutional Review Board of the Veterans Affairs Connecticut Healthcare System.
Outcome
The outcome was participation in at least 1 CR session, identified using the CPT codes 93797, 93798, G0422, and G0423, within 1 year of qualifying event.25
Covariates
The electronic health record database includes age at time of initial cardiac event, sex, race, ethnicity, and geographic region by VA Veterans Integrated Service Network (categorized as Continental, Midwest, Northeast, Pacific, and Southeast).26
We also evaluated medical and psychiatric conditions using ICD codes. Comorbidities of interest, based on clinical relevance and previous work,5 included atrial fibrillation, cerebrovascular disease, heart failure, hypertension, peripheral vascular disease, tobacco use, anxiety disorder, dementia, depression, MST history, PTSD, cancer, chronic kidney disease, end-stage renal disease, chronic obstructive pulmonary disease, diabetes, and obesity. Community-level characteristics of rural-urban status and Area Deprivation Index (ADI) were obtained through data linkage. Rural-urban status was obtained through the geo-coded home address data in the patient’s demographic data. The ADI is a metric that ranks neighborhoods according to socioeconomic disadvantage by incorporating measures in the domains of housing quality, income, education, and employment.27 The ADI for the fiscal year quarter was linked to each Veteran’s home address census block closest to the index CR date (ADI years included 2015, 2019, 2020, 2021, and 2022).28 Rankings were then categorized into quartiles.
Data validation
Veteran CR participation can occur at the local primary VA or through centers in the community. To ensure that we captured both of these groups, we compared data in our identified cohort of CR participants with our internal CR database at the VA Connecticut Healthcare System. We identified 8 Veterans in the internal database for validation—4 who attended CR in the community and 4 who attended CR at the VA. We identified all 8 Veterans within our cohort for CR, regardless of location of CR.
Statistical analyses
The demographical, clinical, and geographical characteristics of the study cohort were summarized using frequency and percentage for discrete variables and mean and SD for quantitative variables. Differences in these factors were compared according to participation in CR for men and women Veterans, respectively, using chi-square test or Student’s t-test, as appropriate.
We used a multilevel logistic regression with a random intercept to account for the VA facilities where Veterans received their usual care. Starting with a model with sex as the sole explanatory variable (model 1), we first introduced a set of a priori selected veteran characteristics as covariates, including age, sex, race, ethnicity, qualifying event, comorbidities, and MST history (model 2). Adjusted sex estimates should be interpreted as direct associations conditional on MST history. Next, we included state-level ADI quartiles as our primary contextual explanatory variable, along with variables for rural-urban status, and geographic region (model 3), informed by published reports highlighting disparities in access related to these factors.7 Adjusted OR (aOR) and 95% CIs were estimated with the use of robust variance estimator. Potential interactions of clinical interest were tested in the fully adjusted model by adding product terms of sex with ADI quartiles, rural-urban status, and geographic region. Adjusted ORs (95% CI) were estimated for ADI quartiles among men and women Veterans, respectively.
All analyses were conducted in SAS (version 9.4; SAS Institute). A value of P < 0.05 was considered to be statistically significant.
Results
Baseline characteristics
We identified 82,496 Veterans eligible for CR, with 10.4% participating in CR (Table 1). Among these Veterans, 10.2% of women (303/2,965) and 10.4% of men participated in CR (8,244/79,531). Women Veterans within the cohort were younger than men Veterans, and participants were younger than nonparticipants in both groups (women: 62.3 ± 8.9 participants, 64.2 ± 12.6 nonparticipants; men: 68.5 ± 9.7 participants, 71.8 ± 11.2 non-participants). MI was the qualifying diagnosis for CR for most Veterans—87.1% of women and 79.6% of men. For Veterans who qualified for CR by PCI procedures at a VA facility, 23.8% of women Veterans and 16.7% of men Veterans participated in CR. Among the potential participants, due to VA-based CABG procedures, 31.4% of women Veterans and 39.1% of men Veterans participated in CR.
Table 1.
Baseline Characteristics of CR Participants and Nonparticipants According to Sex
| Women |
Men |
|||||
|---|---|---|---|---|---|---|
| CR Nonparticipants (n = 2,662) | CR Participants (n = 303) | P Value | CR Nonparticipants (n = 71,287) | CR Participants (n = 8,244) | P Value | |
| Patient characteristics | ||||||
| Age, y | 64.2 ± 12.6 | 62.3 ± 8.9 | 0.015 | 71.8 ± 11.2 | 68.5 ± 9.7 | <0.001 |
| Age groups, y | <0.001 | <0.001 | ||||
| 18-54 | 564 (21.2) | 12 (4.0) | 5,285 (7.4) | 830 (10.1) | ||
| 55-64 | 829 (31.1) | 47 (15.5) | 11,811 (16.6) | 1,815 (22.0) | ||
| 65-74 | 822 (30.9) | 127 (41.9) | 27,604 (38.7) | 3,580 (43.4) | ||
| 75-84 | 292 (11.0) | 91 (30.0) | 18,295 (25.7) | 1,789 (21.7) | ||
| ≥85 | 155 (5.8) | 26 (8.6) | 8,292 (11.6) | 230 (2.8) | ||
| Race | 0.748 | <0.001 | ||||
| White | 1,545 (58.0) | 185 (61.1) | 50,806 (71.3) | 6,368 (77.2) | ||
| Black | 854 (32.1) | 90 (29.7) | 13,863 (19.5) | 1,128 (13.7) | ||
| Asian, Pacific Islander, or American Indian | 63 (2.4) | 7 (2.3) | 1,619 (2.3) | 201 (2.4) | ||
| Other/unknown | 200 (7.5) | 19 (6.3) | 4,999 (7.0) | 547 (6.6) | ||
| Ethnicity, Hispanic | 102 (3.8) | 9 (3.0) | 0.454 | 4,748 (6.7) | 474 (5.8) | |
| Region | 0.028 | <0.001 | ||||
| Continental | 462 (17.4) | 49 (16.2) | 11,493 (16.1) | 1,183 (14.4) | ||
| Midwest | 520 (19.5) | 73 (24.1) | 15,248 (21.4) | 2,703 (32.8) | ||
| Northeast | 581 (21.8) | 82 (27.1) | 15,682 (22.0) | 1,747 (21.2) | ||
| Pacific | 442 (16.6) | 44 (14.5) | 12,491 (17.5) | 1,207 (14.6) | ||
| Southeast | 627 (23.6) | 54 (17.8) | 15,546 (21.8) | 1,311 (15.9) | ||
| Unknown | 30 (1.3) | 1 (0.3) | 827 (1.2) | 93 (1.1) | ||
| Qualifying event | ||||||
| MI | <0.001 | <0.001 | ||||
| MI only | 2387 (89.7) | 198 (65.4) | 58,732 (82.4) | 4,610 (55.9) | ||
| MI with PCI | 108 (4.1) | 6 (2.0) | 3836 (5.4) | 928 (11.3) | ||
| MI with CABG | 3 (0.11) | 45 (14.9) | 533 (0.8) | 352 (4.3) | ||
| MI, PCI, and CABG | 1 (0.04) | 0 | 30 (0.04) | 8 (0.10) | ||
| PCI only | 138 (5.2) | 43 (14.2) | 6,549 (9.2) | 1,317 (16.0) | ||
| CABG only | 24 (0.9) | 11 (3.6) | <0.001 | 1,590 (2.2) | 1,022 (12.4) | <0.001 |
| Comorbidities | ||||||
| Cardiovascular | ||||||
| Atrial fibrillation | 482 (18.1) | 27 (8.9) | <0.001 | 23,461 (32.9) | 2,300 (27.9) | <0.001 |
| Cerebrovascular disease | 550 (20.7) | 53 (17.5) | 0.194 | 15,762 (22.1) | 1,476 (17.9) | <0.001 |
| Heart failure | 905 (34.0) | 84 (27.7) | 0.028 | 33,797 (47.4) | 3,264 (39.6) | <0.001 |
| Hypertension | 2,174 (81.7) | 254 (83.8) | 0.355 | 63,751 (89.4) | 7,500 (91.0) | <0.001 |
| Peripheral vascular disease | 495 (18.6) | 60 (19.8) | 0.610 | 22,506 (31.6) | 2,324 (28.2) | <0.001 |
| Tobacco use | 904 (34.0) | 102 (33.7) | 0.973 | 22,043 (30.9) | 2,140 (26.0) | <0.001 |
| Psychiatric/neurologic | ||||||
| Anxiety disorder | 299 (11.2) | 28 (9.2) | 0.294 | 3,517 (4.9) | 471 (5.7) | 0.002 |
| Dementia | 207 (7.8) | 3 (1.0) | <0.001 | 8,801 (12.4) | 305 (3.7) | <0.001 |
| Depression | 1,169 (43.9) | 137 (45.2) | 0.666 | 17,803 (25.0) | 2,124 (25.8) | 0.117 |
| Military sexual trauma | 858 (32.2) | 103 (34.0) | 0.535 | 1890 (2.7) | 247 (3.0) | 0.067 |
| Post-traumatic stress disorder | 775 (29.1) | 103 (34.0) | 0.078 | 13,569 (19.0) | 1,827 (22.2) | <0.001 |
| Other | ||||||
| Cancer | 344 (12.9) | 21 (6.9) | 0.003 | 15,025 (21.1) | 1,372 (16.6) | <0.001 |
| Chronic kidney disease | 335 (12.6) | 18 (5.9) | <0.001 | 12,801 (18.0) | 903 (11.0) | <0.001 |
| End-stage renal disease | 535 (18.9) | 32 (10.6) | <0.001 | 20,756 (29.1) | 1,697 (20.6) | <0.001 |
| COPD | 747 (28.1) | 67 (22.1) | 0.028 | 23,723 (33.3) | 2,068 (25.1) | <0.001 |
| Diabetes | 1,127 (42.3) | 137 (45.2) | 0.337 | 37,666 (52.8) | 4,410 (53.5) | 0.258 |
| Obesity | 944 (35.5) | 138 (45.5) | <0.001 | 18,529 (26%) | 2,875 (35%) | <0.001 |
| Community characteristics | ||||||
| Rural-urban status | 0.0749 | <0.001 | ||||
| Rural | 682 (25.6) | 92 (30.4) | 23,407 (32.8) | 3,121 (37.9) | ||
| Urban | 1980 (74.4) | 211 (69.6) | 47,880 (67.2) | 5,123 (62.1) | ||
| Area Deprivation Index | ||||||
| First quartile | 583 (21.9) | 74 (24.4) | 0.433 | 15,787 (22.2) | 2,020 (24.5) | <0.001 |
| Second quartile | 829 (31.1) | 86 (28.4) | 22,246 (31.2) | 2,739 (33.2) | ||
| Third quartile | 594 (22.3) | 61 (20.1) | 15,970 (22.4) | 1,847 (22.4) | ||
| Fourth quartile | 613 (23.0) | 79 (26.0) | 16,121 (22.6) | 1,551 (18.8) | ||
Values are mean ± SD or n (%), unless otherwise indicated.
CABG = coronary artery bypass surgery; COPD = chronic obstructive pulmonary disease; CR = cardiac rehabilitation; MI = myocardial infarction; PCI = percutaneous coronary intervention.
Women Veterans demonstrated a higher prevalence of mental health risk factors of anxiety, depression, and PTSD compared to their male counterparts. The prevalence of these conditions did not substantially differ with CR participation. Among women, 9.2% of participants vs 11.2% of nonparticipants had an anxiety diagnosis, 45.2% of participants and 43.9% of nonparticipants had a diagnosis of depression, and PTSD was diagnosed in 34.0% participants and 29.1% nonparticipants. For men, anxiety was diagnosed in 5.7% of participants and 4.9% of nonparticipants, depression was diagnosed in 25.8% of participants and 25.0% of nonparticipants, and PTSD was diagnosed in 22.2% of participants and 19.0% of nonparticipants. MST was more prevalent in women Veterans and did not vary with CR participation among both sexes (women: 34.0% participants, 32.2% nonparticipants; men: 3.0% participants, 2.7% nonparticipants). In terms of community characteristics, most Veterans lived in urban areas (women: 69.6% participants, 74.4% nonparticipants; men: 62.1% participants, 67.2% nonparticipants). For ADI, similar proportions of men and women Veterans comprised the first, least deprived, quartile (women: 24.4% participants, 21.9% nonparticipants, men: 24.5% participants, 22.2% nonparticipants) and fourth, most deprived, quartile (women: 26% participants, 23% nonparticipants, men: 18.8% participants, 22.6% nonparticipants).
Multilevel logistic regression
Using multilevel logistic regression modeling, we adjusted for patient- and community-level characteristics (Table 2, Central Illustration). In terms of patient characteristics, the primary association of interest found CR participation to be similar between men and women Veterans (aOR: 0.90; 95% CI: 0.79-1.03; P = 0.121). Older Veterans were less likely to participate in CR (aOR: 0.86; 95% CI: 0.85-0.87; P < 0.001). Geographic location was associated with CR participation as Veterans in the Northeast region were 134% more likely to participate in CR than the Continental region (aOR: 2.34; 95% CI: 1.70-3.22; P < 0.001). Among those who had procedures at the VA, Veterans who qualified for CR due to CABG or PCI were significantly more likely to participate in CR than those with MI (CABG aOR: 7.79; 95% CI: 6.42-9.46; P < 0.001, PCI aOR: 2.58; 95% CI: 2.25-2.95; P < 0.001). When evaluating comorbidities of interest, MST was associated with a slightly increased likelihood of CR participation, although this relationship was borderline significant (aOR: 1.16; 95% CI: 1.00-1.35; P = 0.045). Rural-urban status did not significantly influence CR participation (aOR: 1.07; 95% CI: 1.00-1.16; P = 0.054). Veterans in the most deprived ADI quartile were found to be less likely to participate in CR than those within the least deprived quartile (aOR: 0.82; 95% CI: 0.75-0.89; P < 0.001).
Table 2.
Logistic Regression Models
| OR | 95% CI | P Value | |
|---|---|---|---|
| Unadjusted (sex-only) | |||
| Men (Ref.) | — | — | — |
| Women | 0.98 | 0.90-1.15 | 0.758 |
| Multilevel modeling | |||
| Patient characteristics | |||
| Age, indexeda | 0.86 | 0.85-0.87 | <0.001 |
| Sex | |||
| Men (Ref.) | — | — | — |
| Women | 0.90 | 0.79-1.03 | 0.121 |
| Race | |||
| White (Ref.) | — | — | — |
| Black | 0.83 | 0.77-0.90 | <0.001 |
| Asian, Pacific Islander, or American Indian | 1.01 | 0.87-1.17 | 0.904 |
| Other/unknown | 0.92 | 0.70-1.19 | 0.510 |
| Ethnicity, Hispanic | 1.05 | 0.94-1.17 | 0.407 |
| Region | |||
| Continental (Ref.) | 1.00 | — | — |
| Midwest | 1.24 | 0.84-1.85 | 0.285 |
| Northeast | 2.34 | 1.70-3.22 | <0.001 |
| Pacific | 1.21 | 0.87-1.67 | 0.260 |
| Southeast | 1.09 | 0.74-1.59 | 0.673 |
| Qualifying event | |||
| MI (Ref.) | — | — | — |
| CABG | 7.79 | 6.42-9.46 | <0.001 |
| PCI | 2.58 | 2.25-2.95 | <0.001 |
| Tobacco use | 0.70 | 0.66-0.75 | <0.001 |
| Composite index of comorbidities | 0.95 | 0.94-0.97 | <0.001 |
| Military sexual trauma | 1.16 | 1.00-1.35 | 0.045 |
| Community characteristics | |||
| Rural-urban status | |||
| Urban (Ref.) | — | — | — |
| Rural | 1.07 | 1.00–1.16 | 0.054 |
| Area Deprivation Indexb | |||
| First quartile (Ref.) | — | — | — |
| Second quartile | 0.94 | 0.88-1.00 | 0.044 |
| Third quartile | 0.92 | 0.86-0.99 | 0.019 |
| Fourth quartile | 0.82 | 0.75-0.89 | <0.001 |
Central Illustration.
Observational Study: Cardiac Rehabilitation Among Veterans—A Sex-Based Analysis
The figure summarizes the population, data source, outcome, analysis, and findings of the study. Notably, low participation levels of 10.4% without significant differences by sex after multilevel logistic regression analyses were found. aOR = adjusted OR.
We evaluated the interaction between Veterans’ sex and rural-urban status, geographic region, and ADI quartiles. Only the associations for ADI were found to be statistically significant for men (Table 3). When evaluated for men, as ADI worsened by quartile, CR participation was found to be less likely, and statistically significant when comparing the third and fourth quartiles with the first (third quartile vs first quartile, aOR: 0.93; 95% CI: 0.87-0.99; P = 0.0331, fourth quartile vs first quartile, aOR: 0.81; 95% CI: 0.74-0.87; P < 0.001).
Table 3.
Interaction for Area Deprivation Index by Sex
| OR | 95% CI | P Value | |
|---|---|---|---|
| Men | |||
| Area Deprivation Index | |||
| First quartile (Ref.) | — | — | — |
| Second quartile | 0.94 | 0.88-1.00 | 0.055 |
| Third quartile | 0.93 | 0.87-0.99 | 0.033 |
| Fourth quartile | 0.81 | 0.74-0.87 | <0.001 |
| Women | |||
| Area Deprivation Index | |||
| First quartile (Ref.) | — | — | — |
| Second quartile | 0.87 | 0.59-1.28 | 0.480 |
| Third quartile | 0.80 | 0.56-1.17 | 0.254 |
| Fourth quartile | 1.13 | 0.78-1.64 | 0.510 |
Logistic regression model adjusted for age, race, ethnicity, region, qualifying event, composite index of comorbidities, military sexual trauma, rural/urban status, and region. OR and 95% CIs estimated with robust variance estimator.
Abbreviation as in Table 2.
Conclusions
This is the first known study to evaluate CR participation among Veterans with CVD since 2015, and the first sex-based analysis of CR participation among Veterans. Our study had 3 novel findings. First, we found that CR initiation remains very low among men and women Veterans at 10.4% between 2021 and 2023, and well below national goals of 70% participation as recommended by the Center for Disease Control and Prevention’s Million Hearts CR Collaborative.7,29 Second, after adjusting for patient-level and community-level covariates among men and women Veterans, CR participation was not significantly different. Third, Veterans living in areas of severe socioeconomic deprivation participated less in CR than those living in the least deprived areas; in the model evaluating sex-based interactions, this association remained significant for men but not for women.
CR participation rates among both women and men Veterans were lower than rates reported in 2015 (13.2%).9 Overall, CR participation among Veterans (10.4%) was less than half the rate reported among Medicare Fee-for-Service beneficiaries (28%).4,30 Among Medicare beneficiaries, women were 9% less likely than men to participate in CR,30 and a meta-analysis found women 36% less likely to enroll in CR.31 We found no statistical difference between CR participation between men and women Veterans. The lack of difference in CR participation rates between men and women Veterans should not be interpreted as the absence of barriers—instead, it may reflect that both men and women Veterans encounter substantial challenges, potentially exceeding those faced by non-Veterans, given lower participation rates. A study of women Veterans reported cost, access, and lack of respect in health care settings as barriers to obtaining CVD prevention care.32 More research is needed to evaluate these specific barriers, including examining patient-level and system-level factors such as referrals, access to care, and logistical or personal obstacles that impact enrollment. Identification of these barriers will help inform targeted strategies to improve CR accessibility for Veterans.
A potential barrier investigated in this study was the impact of MST on CR participation. However, in our analysis, a slightly increased likelihood of participation was observed among those with a history of MST, warranting further investigation. MST is known to be under-reported among women Veterans due to factors such as stigma, fear of retaliation, and confidentiality concerns.18,19,33 However, for those who do disclose this history, our findings may reflect increased monitoring or access to care within the VA system and the impact of trauma-informed care and treatments,34 which may ultimately facilitate CR participation. However, additional research is needed to confirm this association and evaluate the underlying mechanisms.
When compared with findings from the non-Veteran population, our study found similar decreased likelihood of participation among older adults.35, 36, 37 We also observed regional differences in participation rates based on geographic location—patterns consistent with previous research on Medicare and Veteran populations.38 Veterans who underwent PCI and CABG at VA facilities were far more likely to participate in CR than those diagnosed with MI, although overall participation fell well below Million Heart’s goals at 70%.39 Given that these Veterans receive their care for their primary cardiac conditions through the VA, this finding represents an area where critical change is needed. Including CR referrals in the postprocedure discharge orders may be an effective strategy to increase participation, as this approach has been shown to triple the likelihood of CR enrollment in non-VA settings.29,40
The impact of rural-urban status and ADI on CR has been previously reported in the non-Veteran published reports. Among rural communities, significant disparities have been observed in CR access.5,35,41 In this study, no significant difference in CR participation was observed between Veterans living in rural and urban communities, which underscores the low participation among Veterans in both of these communities. However, it may also indicate efforts to improve CR access among rural Veterans—especially through home-based CR, now offered at 40 VA facilities and supported by the Office of Rural Health.8,42,43 Home-based CR delivers CR remotely to Veterans and is an alternate mechanism by which Veterans can access CR. It has shown promise in delivering care to Veterans who have transportation difficulties, varied schedules, or who live in areas without in-person CR centers.9
Residents of the most socially deprived areas of the United States are significantly less likely to participate in CR than those in the least socially deprived areas.5,44 Our findings indicate this association holds true among Veterans; indeed, Veterans living in lower socioeconomic status neighborhoods have been found to have poorer health status and a greater risk of all-cause mortality.45,46 Area deprivation may compound some of the other barriers to CR access, including transportation difficulties, work responsibilities, and out-of-pocket costs.22,39,47 Veterans attending CR may have copays based on financial means or private insurance coverage.48 However, as financial means is taken into account, the burden of out-of-pocket costs is less than that of the non-Veteran population.49 In this study, socioeconomic barriers, as reflected by ADI, disproportionately impacted men Veterans’ CR participation. Identifying and addressing these barriers requires targeted strategies to improve access and engagement, particularly for men living in the most socioeconomically disadvantaged areas.
Significant innovation and policy change are required to increase CR participation among Veterans, particularly as Veteran CR participation is lagging significantly behind the national Medicare population. These findings are particularly alarming considering the cardiovascular burden experienced by men and women Veterans. Waiting to intervene risks widening these gaps in care experienced by Veterans. Virtual models of delivering CR, including home-based CR, have proven effective in reaching eligible individuals who may face barriers such as transportation issues or a lack of nearby CR centers.7,39 Our findings suggest a need to expand home-based CR and explore other care delivery models to improve CR access among Veterans. In addition, this study focused on CR initiation. Understanding CR engagement and factors that hinder and promote adherence are also essential to improving CR utilization.
Study Limitations
First, we are unable to ascertain CR referrals in this database; therefore, whether low referral rates may explain the low participation rates is unknown. Further study is needed to evaluate the impact of CR referrals on participation among Veterans. Second, women Veterans represented a small proportion of the overall cohort, which may have limited the ability to detect certain differences in our analyses. As the number of women Veterans receiving care at the VA grows over time and the current cohort ages, we expect future studies will provide clearer insights into these differences. Third, our study found that most qualifying events in this Veteran cohort were due to MI. This finding reflects the unique patterns of care that Veterans receive. For example, Veterans may receive emergent care and procedures in non-VA settings closer to home but then follow up within the VA system. Therefore, these findings should not be interpreted to reflect whether coronary revascularization occurred after an MI diagnosis, as data are only from VA facilities. Fourth, we are unable to differentiate between the type of CR (whether a referral to the community or VA-based) that a Veteran attended. Further evaluation may clarify the barriers to CR utilization on the facility level. Fifth, although ADI and rural-urban status provide context on community-level factors, individual-level factors such as social support, transportation access, and distance to CR facility were not available in this data set and may influence participation.
Conclusions
Current participation in CR among U.S. Veterans remains low, at just 10.4%, which is significantly lower than the Medicare population. Characteristics of CR-eligible Veterans differed by sex in our analyses, although participation did not differ significantly between women or men. Our findings underscore the need for further research to identify Veteran-specific barriers to CR enrollment, particularly among the most socioeconomically disadvantaged, and expand novel CR delivery models, including home-based CR.
Funding support and author disclosures
Dr Varghese is supported by the Department of Veterans Affairs, Veterans Health Administration, VISN 1 Career Development Award. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States Government.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
Appendix
For an expanded Methods section as well as supplemental tables, please see the online version of this paper.
Supplementary material
References
- 1.Dibben G., Faulkner J., Oldridge N., et al. Exercise-based cardiac rehabilitation for coronary heart disease. Cochrane Heart Group. Cochrane Database Syst Rev. 2021;11(11) doi: 10.1002/14651858.CD001800.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lawton J.S., Tamis-Holland J.E., Bangalore S., et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization. J Am Coll Cardiol. 2022;79(2):e21–e129. doi: 10.1016/j.jacc.2021.09.006. [DOI] [PubMed] [Google Scholar]
- 3.Rao S.V., O’Donoghue M.L., Ruel M., et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI guideline for the management of patients with acute coronary syndromes. J Am Coll Cardiol. 2025;85(22):2135–2237. doi: 10.1016/j.jacc.2024.11.009. [DOI] [PubMed] [Google Scholar]
- 4.Keteyian S.J., Jackson S.L., Chang A., et al. Tracking cardiac rehabilitation utilization in medicare beneficiaries: 2017 update. J Cardiopulm Rehabil Prev. 2022;42(4):235–245. doi: 10.1097/HCR.0000000000000675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Varghese M.S., Beatty A., Song Y., et al. Cardiac rehabilitation and the COVID-19 pandemic: persistent declines in CR participation and access among US medicare beneficiaries. Circ Cardiovasc Qual Outcomes. 2022;15(12) doi: 10.1161/CIRCOUTCOMES.122.009618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Varghese M.S., Song Y., Xu J., et al. Availability and use of In-Person and virtual cardiac rehabilitation among US medicare beneficiaries: a post-pandemic update. J Cardiopulm Rehabil Prev. 2023;43(4):301–303. doi: 10.1097/HCR.0000000000000803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Beatty A.L., Brown T.M., Corbett M., et al. Million hearts cardiac rehabilitation think tank: accelerating new care models. Circ Cardiovasc Qual Outcomes. 2021;14(10) doi: 10.1161/CIRCOUTCOMES.121.008215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Varghese M.S., Wu W.C., Drwal K.R., et al. Cardiac rehabilitation among Veterans: a narrative review. J Cardiopulm Rehabil Prev. 2025;45(2):78–84. doi: 10.1097/HCR.0000000000000932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schopfer D.W., Krishnamurthi N., Shen H., Duvernoy C.S., Forman D.E., Whooley M.A. Association of Veterans health administration home-based programs with access to and participation in cardiac rehabilitation. JAMA Intern Med. 2018;178(5):715–717. doi: 10.1001/jamainternmed.2017.8039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ebrahimi R., Yano E.M., Alvarez C.A., et al. Trends in cardiovascular disease mortality in US women Veterans vs Civilians. JAMA Netw Open. 2023;6(10) doi: 10.1001/jamanetworkopen.2023.40242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Assari S. Veterans and risk of heart disease in the United States: a cohort with 20 years of follow up. Int J Prev Med. 2014;5(6):703–709. [PMC free article] [PubMed] [Google Scholar]
- 12.National Center for Veterans Analytics and Statistics Population tables: age/gender. 2023. https://www.va.gov/vetdata/veteran_population.asp
- 13.Han J.K., Yano E.M., Watson K.E., Ebrahimi R. Cardiovascular care in women Veterans. Circulation. 2019;139(8):1102–1109. doi: 10.1161/CIRCULATIONAHA.118.037748. [DOI] [PubMed] [Google Scholar]
- 14.Breland J.Y., Phibbs C.S., Hoggatt K.J., et al. The obesity epidemic in the Veterans health administration: prevalence among key populations of women and men veterans. J Gen Intern Med. 2017;32(Suppl 1):11–17. doi: 10.1007/s11606-016-3962-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kilbourne A.M., Schumacher K., Frayne S.M., et al. Physical health conditions among a population-based cohort of Vietnam-Era women veterans: agreement between self-report and medical records. J Womens Health 2002. 2017;26(11):1244–1251. doi: 10.1089/jwh.2016.6069. [DOI] [PubMed] [Google Scholar]
- 16.Odani S., Agaku I.T., Graffunder C.M., Tynan M.A., Armour B.S. Tobacco product use among military veterans—United States, 2010–2015. Morb Mortal Wkly Rep. 2018;67(1):7–12. doi: 10.15585/mmwr.mm6701a2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Coyle B.S., Wolan D.L., Van Horn A.S. The prevalence of physical and sexual abuse in women veterans seeking care at a veterans affairs medical center. Mil Med. 1996;161(10):588–593. [PubMed] [Google Scholar]
- 18.Lutwak N., Dill C. Military sexual trauma increases risk of post-traumatic stress disorder and depression thereby amplifying the possibility of suicidal ideation and cardiovascular disease. Mil Med. 2013;178(4):359–361. doi: 10.7205/MILMED-D-12-00427. [DOI] [PubMed] [Google Scholar]
- 19.Suris A., Lind L. Military sexual trauma: a review of prevalence and associated health consequences in veterans. Trauma Violence Abuse. 2008;9(4):250–269. doi: 10.1177/1524838008324419. [DOI] [PubMed] [Google Scholar]
- 20.Firoozabadi M.G., Mirzaei M., Grace S.L., et al. Sex differences in cardiac rehabilitation barriers among non-enrollees in the context of lower gender equality: a cross-sectional study. BMC Cardiovasc Disord. 2023;23(1):329. doi: 10.1186/s12872-023-03331-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.de Ghisi G.L.M., Kim W.S., Cha S., et al. Women’s cardiac rehabilitation barriers: results of the international council of cardiovascular prevention and rehabilitation’s first global assessment. Can J Cardiol. 2023;39(11S):S375–S383. doi: 10.1016/j.cjca.2023.07.016. [DOI] [PubMed] [Google Scholar]
- 22.Grace S.L., Gravely-Witte S., Kayaniyil S., Brual J., Suskin N., Stewart D.E. A multisite examination of sex differences in cardiac rehabilitation barriers by participation status. J Womens Health 2002. 2009;18(2):209–216. doi: 10.1089/jwh.2007.0753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Khadanga S., Gaalema D.E., Savage P., Ades P.A. Underutilization of cardiac rehabilitation in women: barriers and solutions. J Cardiopulm Rehabil Prev. 2021;41(4):207. doi: 10.1097/HCR.0000000000000629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Veterans Affairs Corporate Data Warehouse (CDW) https://www.hsrd.research.va.gov/for_researchers/cdw.cfm
- 25.Million Hearts Cardiac Rehabilitation Collaborative Surveillance Workgroup Outpatient cardiac rehabilitation use surveillance methodology. https://millionhearts.hhs.gov/files/Cardiac-Rehab-Use-Surveillance-Guidance.pdf
- 26.Bernstein E., DeRycke E.C., Han L., et al. Racial, ethnic, and rural disparities in US veteran COVID-19 vaccine rates. AJPM Focus. 2023;2(3) doi: 10.1016/j.focus.2023.100094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kind A.J.H., Buckingham W.R. Making neighborhood-disadvantage metrics accessible — the neighborhood atlas. N Engl J Med. 2018;378(26):2456–2458. doi: 10.1056/NEJMp1802313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.University of Wisconsin School of Medicine and Public Health. Neighborhood Atlas. https://www.neighborhoodatlas.medicine.wisc.edu/login
- 29.Ades P.A., Keteyian S.J., Wright J.S., et al. Increasing cardiac rehabilitation participation from 20% to 70%: a road map from the million hearts cardiac rehabilitation collaborative. Mayo Clin Proc. 2017;92(2):234–242. doi: 10.1016/j.mayocp.2016.10.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ritchey M.D., Maresh S., McNeely J., et al. Tracking cardiac rehabilitation participation and completion among medicare beneficiaries to inform the efforts of a national initiative. Circ Cardiovasc Qual Outcomes. 2020;13(1) doi: 10.1161/CIRCOUTCOMES.119.005902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Samayoa L., Grace S.L., Gravely S., Scott L.B., Marzolini S., Colella T.J.F. Sex differences in cardiac rehabilitation enrollment: a meta-analysis. Can J Cardiol. 2014;30(7):793–800. doi: 10.1016/j.cjca.2013.11.007. [DOI] [PubMed] [Google Scholar]
- 32.Orshak J., Ondoma C., Johnson H.M., Lauver D. Women veterans’ barriers to care-seeking for cardiovascular disease prevention. Nurs Res. 2024;73(1):26–36. doi: 10.1097/NNR.0000000000000684. [DOI] [PubMed] [Google Scholar]
- 33.Gaffey A.E., Burg M.M., Skanderson M., et al. Electronic health record concordance with survey-reported military sexual trauma among younger veterans: associations with health care utilization and mental health diagnoses. J Womens Health 2002. 2025;34(1):125–132. doi: 10.1089/jwh.2023.0993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Galovski T.E., Street A.E., Creech S., Lehavot K., Kelly U.A., Yano E.M. State of the knowledge of VA military sexual trauma research. J Gen Intern Med. 2022;37(Suppl 3):825–832. doi: 10.1007/s11606-022-07580-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Castellanos L.R., Viramontes O., Bains N.K., Zepeda I.A. Disparities in cardiac rehabilitation among individuals from racial and ethnic groups and rural communities—a systematic review. J Racial Ethn Health Disparities. 2019;6(1):1–11. doi: 10.1007/s40615-018-0478-x. [DOI] [PubMed] [Google Scholar]
- 36.Doll J.A., Hellkamp A., Thomas L., et al. Effectiveness of cardiac rehabilitation among older patients after acute myocardial infarction. Am Heart J. 2015;170(5):855–864. doi: 10.1016/j.ahj.2015.08.001. [DOI] [PubMed] [Google Scholar]
- 37.Doll J.A., Hellkamp A., Ho P.M., et al. Participation in cardiac rehabilitation programs among older patients after acute myocardial infarction. JAMA Intern Med. 2015;175(10):1700–1702. doi: 10.1001/jamainternmed.2015.3819. [DOI] [PubMed] [Google Scholar]
- 38.Beatty A.L., Truong M., Schopfer D.W., Shen H., Bachmann J.M., Whooley M.A. Geographic variation in cardiac rehabilitation participation in medicare and veterans affairs populations. Circulation. 2018;137(18):1899–1908. doi: 10.1161/CIRCULATIONAHA.117.029471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wall H.K., Stolp H., Wright J.S., et al. The million hearts initiative: catalyzing utilization of cardiac rehabilitation and accelerating implementation of new care models. J Cardiopulm Rehabil Prev. 2020;40(5):290–293. doi: 10.1097/HCR.0000000000000547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Grace S.L., Russell K.L., Reid R.D., et al. Cardiac Rehabilitation Care Continuity Through Automatic Referral Evaluation (CRCARE) investigators. Effect of cardiac rehabilitation referral strategies on utilization rates: a prospective, controlled study. Arch Intern Med. 2011;171(3):235–241. doi: 10.1001/archinternmed.2010.501. [DOI] [PubMed] [Google Scholar]
- 41.Valencia H.E., Savage P.D., Ades P.A. Cardiac rehabilitation participation in underserved populations. Minorities, low socioeconomic, and rural residents. J Cardiopulm Rehabil Prev. 2011;31(4):203–210. doi: 10.1097/HCR.0b013e318220a7da. [DOI] [PubMed] [Google Scholar]
- 42.VA|Diffusion Marketplace Home based cardiac rehabilitation. https://marketplace.va.gov/innovations/home-based-cardiac-rehabilitation
- 43.Drwal K.R., Wakefield B.J., Forman D.E., Wu W.C., Haraldsson B., El Accaoui R.N. Home-based cardiac rehabilitation: experience from the veterans affairs. J Cardiopulm Rehabil Prev. 2021;41(2):93. doi: 10.1097/HCR.0000000000000594. [DOI] [PubMed] [Google Scholar]
- 44.Bachmann J.M., Huang S., Gupta D.K., et al. Association of neighborhood socioeconomic context with participation in cardiac rehabilitation. J Am Heart Assoc. 2017;6(10) doi: 10.1161/JAHA.117.006260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Nelson K., Taylor L., Lurie N., Escarce J., McFarland L., Fihn S.D. Neighborhood environment and health status and mortality among veterans. J Gen Intern Med. 2011;26(8):862–867. doi: 10.1007/s11606-011-1710-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Nelson K., Schwartz G., Hernandez S., Simonetti J., Curtis I., Fihn S.D. The association between neighborhood environment and mortality: results from a national study of veterans. J Gen Intern Med. 2017;32(4):416–422. doi: 10.1007/s11606-016-3905-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Dunlay S.M., Witt B.J., Allison T.G., et al. Barriers to participation in cardiac rehabilitation. Am Heart J. 2009;158(5):852–859. doi: 10.1016/j.ahj.2009.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Veterans Affairs Fact sheet: veteran community care billing and payments. https://www.va.gov/COMMUNITYCARE/docs/pubfiles/factsheets/FactSheet_20-26.pdf#
- 49.Govier D.J., Than C.T., Chawla N., et al. Veterans health administration healthcare coverage and medical financial hardship in low-income veterans. Am J Prev Med. 2023;65(3):406–416. doi: 10.1016/j.amepre.2023.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.