The definition of disability includes “impairments, activity limitations, and participation restrictions”1 and disability arises from the interaction between contextual factors and the health condition of the person.1 The prevalence of disability is considerable, being approximately 18.0% in lower-income countries and 11.8% in higher-income countries.1 Moreover, it has been suggested that several risk factors1 and secondary health conditions2 could be more prevalent in people with disability, and they might increase the risk of cardiovascular disease and its consequences in this population.3

A limited number of studies with long follow-ups have reported the association between disability and cardiovascular mortality,4–10 all of them finding relevant associations. However, some of these studies only focused on older adults4,5,7 and/or did not perform specific analyses for different sociodemographic groups.4,6,7,10 Furthermore, only 3 of them studied the association by disability severity categories8–10 and 1 of them found that cardiovascular mortality by disability severity categories increased only in males.9 Finally, only Forman-Hoffman et al.8 assessed the association by several specific causes of cardiovascular death.

Thus, the main objective of this study was to assess the associations between disability and cardiovascular mortality (overall and for several specific causes) among disability severity categories, and for the different sociodemographic groups, in a large representative cohort of the population aged 40 years or older in Spain.

In the community-dwelling Spanish population aged 40 years or older, the prevalence of disability (95%CI) was 16.2% (15.9-16.4), including 11.5% (11.3-11.8) with mild, 3.5% (3.4-3.7) with moderate, and 1.1% (1.0-1.2) with severe/complete disability. People with a higher degree of disability were more likely to be female, older, less educated, and to live alone in small municipalities, while people with any degree of disability were more likely to have lower household incomes. Among people with disability, the baseline prevalence of cardiovascular disease (particularly cerebrovascular disease) increased with the degree of disability (table 1).

During 977 082 person-years of follow-up, 5070 participants died from cardiovascular disease, 1467 from ischemic heart disease, and 1211 from cerebrovascular disease, corresponding to mortality rates of 5.1, 1.5, and 1.2 deaths per 1000 person-years. There were also 11 706 deaths from other causes, with a mortality rate of 12.0 deaths per 1000 person-years. After standardizing to the distribution of baseline sociodemographic characteristics in the overall community-dwelling Spanish population and accounting for competing deaths from other causes, cardiovascular disease mortality was consistently higher in people with higher degrees of disability over the entire follow-up (figure 2A). Compared with people without disability, the standardized 5-year risk ratios for cardiovascular disease mortality (95%CI) were 1.70 (1.49-1.93) for mild, 3.92 (3.23-4.76) for moderate, and 4.77 (3.42-6.66) for severe/complete disability, corresponding to 11.7, 49.0, and 63.1 excess deaths per 1000 persons, respectively. The increased risk of death from cardiovascular disease was maintained after 10 years of follow-up, with standardized 10-year risk ratios (95%CI) of 1.57 (1.41-1.74) for mild, 2.70 (2.27-3.22) for moderate, and 2.67 (1.92-3.72) for severe/complete disability compared with no disability (table 2). Regarding the potential residual confounding by unmeasured cardiovascular disease at baseline, bias analysis indicated that, once standardized for sociodemographic characteristics (mainly age and sex), the 10-year mortality risk ratio for cardiovascular disease mortality comparing people with and without disability would be biased away from the null (positive confounding) by 5.9% (results of the supplementary data and table 2 of the supplementary data).

Figure 2.

Central illustration. Standardized cumulative mortality (per 1000) from cardiovascular disease (A), ischemic heart disease (B), and cerebrovascular disease (C) by disability category in the community-dwelling Spanish population aged 40 years or older, 2007-2008 to 2017. Parametric cumulative mortality curves (smooth lines) were estimated from spline-based survival models, and nonparametric curves (step functions) were derived from Kaplan-Meier methods, both stratified by disability category, weighted by combined weights, and accounting for competing deaths from other causes. Combined weights were used to standardize cumulative mortality in each disability category to the distribution of baseline sociodemographic characteristics in the entire community-dwelling Spanish population aged 40 years or older, including sex, age, living with a partner, educational level, household income, municipality size, and geographical region.

(0.45MB).

The risk of death from ischemic heart disease increased similarly among people with moderate and severe/complete disability (figure 2B), with standardized risk ratios (95%CI) at 5 and 10 years of follow-up of 3.68 (2.70-5.01) and 2.71 (2.05-3.57) for moderate disability and 4.52 (2.45-8.35) and 2.73 (1.39-5.39) for severe/complete disability compared with no disability (table 2). However, mortality risk from cerebrovascular disease increased progressively with the degree of disability severity (figure 2C). Compared with people without disability, the standardized 5-year risk ratio for cerebrovascular disease mortality (95%CI) rose from 4.05 (2.66-6.15) for moderate to 7.55 (4.70-12.1) for severe/complete disability, and the 10-year risk ratio rose from 3.16 (1.99-5.00) to 4.69 (3.06-7.18), respectively (table 2). Bias analysis showed that, after controlling for sex and age, failure to adjust for unmeasured heart disease at baseline would result in a minor upward bias of 3.6% in the 10-year risk ratio for ischemic heart disease mortality associated with disability, whereas failing to adjust for baseline cerebrovascular disease would introduce a moderate upward bias of 12.5% in the association between disability and cerebrovascular mortality (results of the supplementary data and table 2 of the supplementary data).

In subgroup analyses, the risk of death from cardiovascular disease increased progressively with the degree of disability in all sociodemographic subgroups, albeit risk ratios associated with disability were larger among younger individuals (P for homogeneity <.001), those living with a partner (P <.001), and more educated individuals (P=.004), corresponding to population subgroups with lower baseline risks. The standardized 5-year risk ratios for cardiovascular disease mortality comparing mild, moderate, and severe/complete disability with no disability reached 3.02, 7.16, and 13.10 among individuals younger than 65 years, 2.24, 5.06, and 6.17 among individuals living with a partner, and 2.66, 6.18, and 6.11 among individuals with at least secondary education, respectively (figure 3). The standardized 5-year risk ratios for cardiovascular disease mortality among disability categories were roughly similar for women and men (P for homogeneity=.18).

Figure 3.

Standardized 5-year risk ratios for cardiovascular disease mortality among disability categories by subgroup in the community-dwelling Spanish population aged 40 years or older, 2007-2008 to 2017. Subgroup-specific risk ratios (squares with area inversely proportional to the variance) and 95% confidence intervals (95%CI, horizontal lines) were obtained from spline-based survival models stratified by disability category and sociodemographic subgroup, weighted by combined weights, and accounting for competing deaths from other causes. Combined weights were used to standardize cardiovascular disease mortality in each disability category to the distribution of baseline sociodemographic characteristics in the entire community-dwelling Spanish population aged 40 years or older, including sex, age, living with a partner, educational level, household income, municipality size, and geographical region.

(0.44MB).

Subgroup analyses for ischemic heart disease mortality closely resembled those for cardiovascular disease, with higher risk ratios comparing disability categories among younger individuals (P for homogeneity=.03) and those living with a partner (P=.02) (figure 2 of the supplementary data). Mortality risk from cerebrovascular disease rose progressively among disability categories in all subgroups except for individuals with secondary or higher education (P for homogeneity=.02), whose standardized 5-year risk ratio for cerebrovascular disease mortality comparing severe/complete with no disability dropped to 1.33 (95%CI, 0.33-5.27) (figure 3 of the supplementary data).

In this large adult population-based cohort study, the cumulative cardiovascular mortality (overall and for each specific diagnosis) was higher among individuals with any level of disability compared with individuals without a disability at any follow-up time. Additionally, the risk of death from cardiovascular disease increased progressively with the degree of disability in all the sociodemographic subgroups studied.

Our results are consistent with the findings of 2 representative studies of the general population8,9 and additional cohort studies4,5,7,10 that have assessed this association in older populations. However, it is important to highlight that the measurement of disability is not homogeneous among the studies, hampering comparison of disability research.3 This is coherent with the statement that a common definition of disability would strengthen research on this subject.3,23 For example, there is a marked difference in cardiovascular mortality rates among people with disability between our study (5.1 per 1000 person-years) and that of Son et al.9 (1.3 per 1000 person-years). While both cohorts were fairly similar in age and sex distribution and length of follow-up, their corresponding baseline prevalence of people with disability differed notably: 16.2% vs 0.39%.9 It seems that Son et al.9 used a stricter definition of disability than our own. Forman-Hoffman et al.8 reported a prevalence similar to that found in the present study, above 16% for a younger population, as they also included individuals aged 18 to 40 years. Other studies reported higher prevalences but studied older populations,4,10 and it is known that older individuals have a higher prevalence of disability.1

According to our results, the risk of death from cardiovascular disease increased progressively with the degree of disability in all sociodemographic subgroups studied. This finding was previously reported by Plichart et al.10 and Forman-Hoffman et al.,8 but not by Son et al.9

In line with our study, Plichart et al.10 used a global variable for disability (dichotomous or categorized by severity gradient). However, Son et al.9 and Forman-Hoffman et al.8, in addition to a global measure of disability, further analyzed the association by type of disability, although they made aggregations for different types of disability to increase sample sizes. Moreover, they reached different conclusions. According to Son et al. people with disability had higher cardiovascular mortality than people without disability, irrespective of the type of disability.9 In contrast, Forman-Hoffman et al.8 only found this association for those with mental/substance use impairment or movement disability.

Our estimates for the risk of death were also higher for specific causes of cardiovascular mortality in people with disability. This finding is consistent with the results of Forman-Hoffman et al.8 According to our results, the risk for ischemic heart disease mortality was remarkably higher for individuals with moderate and/or severe/complete disability; while the risk for cerebrovascular mortality also increased markedly but followed a clear gradient with disability severity levels. To our knowledge, no other study has addressed the association between the severity of disability and the risk of death by specific causes of cardiovascular mortality.

Our findings could be supported by several biological and psychosocial mechanisms. Persons with disability have a higher prevalence of some unhealthy behavioral risk factors (substance use/abuse, smoking,24 obesity,25,26 reduced physical activity4,7,27,28)1; they are more likely to have depression,1,29 anxiety,2 sleep disturbances,2 mental distress,30 and type 2 diabetes31 and they also have a higher burden of health care disparities (including preventive screening disparities),27,32 isolation and loneliness,2,33 discrimination,1 and worse socioeconomic status.1 These factors may be related to a higher risk of cardiovascular diseases34–36 and mortality.37–40 Of note, people with disability who meet physical activity recommendations had lower cardiovascular mortality than those without disability who do not adhere to physical activity recommendations.4,7 Thus, the implementation of physical activity recommendations, among other measures, could reduce the mortality risk among people with disability. Finally, people with disability may have less access to health care or preventive services.41

We have also provided interesting subgroup analyses. The subpopulations with higher cardiovascular mortality risk ratios were the younger population, those living with a partner, and individuals with secondary or higher education. However, these results can be partly explained by the fact that those population subgroups had a lower baseline risk. Finally, we found roughly similar risks for both sexes, as opposed to Son et al.,9 who found that cardiovascular mortality increased only in men, which could be partially explained by the wide 95%CI in women.

The risk of cardiovascular mortality (overall and for every specific diagnosis) is higher among adults with any level of disability compared to adults without a disability at any follow-up time. Moreover, the risk of death from cardiovascular disease increased progressively with the degree of disability for all sociodemographic subgroups. Conditions that increase cardiovascular risk in people with a disability should be investigated. Preventive measures and health promotion initiatives are needed to reduce the risk of mortality in this population.

This work was supported by the Institute of Health Carlos III (grant number PI20CIII00045). The views expressed are those of the authors and not necessarily those of the Institute of Health Carlos III. The degree of accuracy or reliability of the quantitative information derived from this work is the sole responsibility of the authors and not of the institution that provides the data, the Spanish National Institute of Statistics (Instituto Nacional de Estadística).

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee Institute of Health Carlos III Research Ethics Committee (number CEI PI 17_2020) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent from the participants or their legal guardians/next of kin was not required for this study since it included participants in the Spanish Survey on Disability, Personal Autonomy, and Dependency 2008 (Encuesta de Discapacidad, Autonomía Personal y Situaciones de Dependencia 2008, EDAD-08) of the Spanish National Institute of Statistics (Instituto Nacional de Estadística) after asking for their collaboration (Instituto Nacional de Estadística. Survey on Disability, Personal Autonomy and Dependency Situations 200811). The authors had no role in the acquisition of information for the disability survey. The SAGER guidelines have been followed with respect to possible sex/gender bias.

Data are unavailable due to legal restrictions.

No artificial intelligence tools were used in the preparation of this manuscript.

J. Damián, R. Pastor-Barriuso, and F.J. García López designed the study and directed its implementation. R. Pastor-Barriuso, A. Padrón-Monedero, and F.J. García López wrote the manuscript. R. Pastor-Barriuso developed and wrote the methods. J. Damián, J. Almazán-Isla, and J. de Pedro-Cuesta provided relevant ideas during the implementation of the article and collaborated in the development of the manuscript. All authors have given final approval to the manuscript.

None declared.