Association of heart attack or stroke history with current mental health symptoms among adults in the USA: cross-sectional analyses of nationally representative samples

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This study used nationally representative samples of US adults from the National Health Interview Survey.

  • Validated screening instruments were used to assess symptoms of depression, anxiety and psychological distress.

  • Poisson regression models with robust SEs were employed to estimate adjusted prevalence ratios.

  • The cross-sectional study design limits causal inference regarding the relationship between cardiovascular history and mental health symptoms.

  • Unmeasured confounding and self-reported data may introduce bias.

Introduction

Cardiovascular diseases remain the leading cause of death and disability in the USA.1 Heart attack (myocardial infarctions) and stroke are among the most common acute cardiovascular events.1–3 According to the 2017–2020 National Health and Nutrition Examination Survey, approximately 3.2% of US adults aged 20 and older have experienced a heart attack and 3.3% have had a stroke in their lifetime.1 These conditions not only cause significant morbidity but also impose a substantial economic burden. In 2019–2020, the direct medical costs associated with heart disease and stroke in the USA totalled an estimated US$154.7 billion, representing about 7% of all healthcare expenditures.1

There is growing recognition of the bidirectional relationship between cardiovascular diseases and mental health. Mental health disorders are well-documented risk factors for the development of cardiovascular conditions.4–6 Existing research has highlighted biological, behavioural and social mechanisms through which poor mental health contributes to the onset of cardiovascular diseases.6–8 Conversely, experiencing an acute cardiovascular event, such as heart attack or stroke, can increase the risk of developing mental health conditions due to the profound physical and emotional stress involved.9–11 Comorbid mental illnesses in patients with cardiovascular diseases are associated with increased healthcare utilisation, decreased quality of life and elevated mortality.12–15 For example, one study in Germany found that patients hospitalised for cardiovascular conditions with psychiatric comorbidities incurred €2500 higher inpatient costs than patients without psychiatric comorbidities.13 A systematic review of studies involving 2477 patients with stroke found that depression was associated with a 52% increase in post-stroke mortality.15

Despite this known relationship, there is limited population-level evidence on the mental health burden faced by adults with a history of heart attack or stroke in the USA. A study using the 2016–2017 Medical Expenditure Panel Survey found that 18.3% of survivors of stroke experienced depression, and 11.3% experienced serious psychological distress.16 However, to our knowledge, no research has examined the national prevalence of mental health conditions among survivors of heart attack. Furthermore, prior studies often lacked a comparison group, limiting their ability to quantify the incremental mental health burden associated with heart attack or stroke.16 17

This study uses data from the 2019 and 2018 National Health Interview Survey (NHIS) to estimate the prevalence of current mental health symptoms among adults with a history of heart attack or stroke in the USA.18 19 Specifically, our outcomes included moderate-to-severe symptoms of depression, anxiety and serious psychological distress. This study makes several important contributions to the existing literature. First, it is the first population-based study to examine the prevalence of mental health symptoms among US adults with a history of heart attack. Second, it estimates the relative prevalence of mental health symptoms associated with heart attack and stroke after adjusting for socio-demographic and health characteristics known to influence both mental and cardiovascular outcomes.20 The results will shed light on the extent to which survivors of heart attack or stroke experience greater mental health burden than their counterparts. Third, the NHIS uses validated instruments to capture clinically meaningful mental health conditions, allowing for an accurate and comprehensive assessment of the mental health burden among the population of interest.

Methods

Data

We analysed data from the sample adult interviews of the 2019 and 2018 NHIS.18 19 The NHIS is an annual cross-sectional survey conducted by the National Center for Health Statistics to monitor population health.21 22 It uses an address-based complex design to generate a nationally representative sample of non-institutionalised civilians every year.21 22 Data are collected by field agents via computer-assisted interviewing either face-to-face or by telephone.21 22 We analysed the 2019 NHIS for anxiety and depression symptoms, and the 2018 NHIS for serious psychological distress, as the measures were available in the respective years.23

Sample

Our sample included adults aged 18 years and older, excluding those with missing mental health outcomes, heart attack or stroke indicators or covariates (figure 1). Less than 4% of adults were excluded due to missing variables. Our final analytical sample included 30 872 adults from the 2019 NHIS for anxiety and depression symptoms, and 24 593 adults from the 2018 NHIS for serious psychological distress. The research was reviewed by the USA Centers for Disease Control and Prevention and it was deemed as not involving human subjects. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for cross-sectional studies.24

Figure 1
Figure 1

Sample derivation flow charts. COPD, chronic obstructive pulmonary disease; GAD-7, seven-item Generalised Anxiety Disorder scale; NHIS, National Health Interview Survey; PHQ-8, eight-item Patient Health Questionnaire. Note: Both A and B report unweighted sample counts and percentages.

Measures

We examined three dichotomous measures of current mental health symptoms among adults. Using data from the 2019 NHIS, we constructed binary indicators (yes/no) for moderate-to-severe symptoms of depression based on the eight-item Patient Health Questionnaire (PHQ-8), and for moderate-to-severe symptoms of anxiety based on the seven-item Generalised Anxiety Disorder scale (GAD-7). In both instruments, respondents reported how often they experienced specific symptoms in the past 2 weeks, with options ranging from ‘not at all’ (0) to ‘nearly every day’ (3) and higher scores indicating severer symptoms.25 Following established guidelines, we defined scores of 10 or higher as indicative of moderate-to-severe symptoms.26 27 The PHQ-8 and GAD-7 demonstrated strong internal consistency, with Cronbach’s alpha values of 0.88 and 0.92, respectively.28 29 Prior studies have shown that, at this cut-off, the PHQ-8 has a sensitivity of 58% and a specificity of 83% for identifying major depressive disorders, while the GAD-7 has a sensitivity of 89% and a specificity of 82% for detecting generalised anxiety disorders.28 29

We additionally used data from the 2018 NHIS to create a binary measure (yes/no) of serious psychological distress, based on the six-item Kessler Psychological Distress Scale (Kessler-6).25 This measure captures severe mental health symptoms that impair daily functioning and require treatment.30 Respondents were asked how often they experienced symptoms such as feeling hopeless or unable to carry out routine activities over the past 30 days, using a scale from 0 (‘none of the time’) to 4 (‘all of the time’) and yielding a total score from 0 to 24.25 Consistent with existing research, we classified scores of 13 or higher as indicative of serious psychological distress.31 The Kessler-6 has a Cronbach’s alpha of 0.89.31 At the 13-point cut-off, it correctly identified serious mental illness in 36% of cases and its absence in 96% of cases.31 We included the PHQ-8, GAD-7 and Kessler-6 questionnaires in online supplemental tables S1-S3.

Supplemental material

We created two dichotomous key independent variables indicating adults’ lifetime history of heart attack (yes/no) or stroke (yes/no). The NHIS asked adults if they had ‘ever been told by a doctor or health professional that (they) had a heart attack (yes/no)’ and if they had ‘ever been told by a doctor or health professional that (they) had a stroke (yes/no)’ in both 2018 and 2019.25

We adjusted for demographic and socioeconomic variables that have been documented to be associated with both cardiovascular history and mental health outcomes.1 20 Demographic covariates included sex (male [reference] vs female), race (non-Hispanic white [reference], Hispanic, non-Hispanic black or non-Hispanic other), age (≥65 years [reference], 45–59 years, or 18–44 years), marital or partner status (not married and not living with a partner [reference] vs married or living with a partner) and census region (Northeast [reference], Midwest, South or West). Socioeconomic covariates included working status in the previous week (did not work [reference] vs worked), family income relative to the federal poverty level (FPL) (at or greater than 400% FPL [reference], 200%–399% FPL or 0%–199% FPL), and current health insurance coverage (private insurance only excluding Medicare [reference], any public insurance excluding Medicare, any Medicare, uninsured or unknown). Education level was operationalised as the highest education among family members (less than high school [reference], high school graduate, some college or college degree or more).

We further adjusted for adult health behaviours and comorbidities that may influence both cardiovascular and mental health outcomes.1 32–34 Health behaviour was measured using current smoking frequency (not at all [reference], some days or every day). Comorbidities included self-reported history of chronic obstructive pulmonary disease (yes/no), cancer (yes/no), diabetes (yes/no) and connective tissue disease such as arthritis (yes/no). We also measured body mass index (BMI), categorised as underweight (<18.5), healthy weight (18.5–25.0) [reference], overweight (25.0–30.0) or obese (at or >30.0).

Analytical strategy

We conducted both descriptive and multivariable regression analyses. Sample characteristics were summarised as unweighted counts and weighted proportions. Differences between groups were assessed using the Rao-Scott adjusted χ2 tests. All estimates incorporated the appropriate sample weights and accounted for the complex survey design of the NHIS.23

We used Poisson log-linear regression models with robust SEs to assess the relationship between a history of heart attack or stroke and mental health conditions. We chose this approach because, unlike logistic regression, Poisson regression with robust SEs yields accurate and interpretable estimates of prevalence ratios (PRs) particularly when the probability of the outcome is high.35 36 We adjusted for individual characteristics that could confound the relationship between cardiovascular history and mental health, and reported the results as adjusted PRs (aPRs). To assess the robustness of our findings, we repeated the analyses using logistic regression. Although logistic regression is frequently used for binary outcomes in cross-sectional studies, it can overestimate the strength of the associations when the outcome is common and it does not directly provide results as PRs.36

Patient and public involvement

Since we conducted secondary analyses of existing survey data, we did not involve patients or the public in the design or conduct of this research. There are currently no plans to disseminate this research to specific patient populations.

Results

Sample characteristics

Based on the 2019 NHIS, the self-reported lifetime prevalence of both heart attack and stroke was 3.1%. Among adults, 51.7% were female; 63.5% identified as non-Hispanic white, 16.5% as Hispanic and 11.6% as non-Hispanic black; and 21.1% were 65 or older (table 1). As shown in table 1, adults with a history of heart attack or stroke differed significantly in socio-demographic characteristics and health conditions from those without such a history. The 2018 sample showed similar characteristics as the 2019 sample (see online supplemental table S5). The unweighted sample counts of both years are presented in online supplemental tables S4 and S5.

Table 1

Characteristics of adults aged 18 years and older, the 2019 National Health Interview Survey (NHIS)

Prevalence of mental health symptoms among adults with a history of heart attack or stroke

Overall, 6.9% (95% CI: 6.5% to 7.3%) of US adults reported experiencing moderate-to-severe symptoms of depression, 6.0% (95% CI: 5.7% to 6.4%) reported moderate-to-severe anxiety symptoms and 3.9% (95% CI: 3.6% to 4.2%) reported serious psychological distress (table 2). Among adults with a history of heart attack, 15.7% (95% CI: 13.4% to 18.3%) reported moderate-to-severe depression symptoms, 11.8% (95% CI: 9.6% to 14.4%) reported anxiety symptoms and 8.4% (95% CI: 6.6% to 10.8%) experienced serious psychological distress. Among those with a history of stroke, 20.2% (95% CI: 17.0% to 23.8%) reported depression symptoms, 14.1% (95% CI: 11.6% to 17.0%) experienced anxiety symptoms and 11.3% (95% CI: 8.9% to 14.1%) had serious psychological distress (table 2). The unweighted counts of adults with the mental health condition of interest are reported in online supplemental table S6.

Table 2

Prevalence of mental health symptoms among adults by history of heart attack or stroke, the 2019 and 2018 National Health Interview Survey (NHIS)

Adjusted prevalence ratios of mental health symptoms associated with a heart attack or stroke

Tables 3 and 4 present aPRs of mental health symptoms by cardiovascular history. Controlling for covariates, a history of heart attack was associated with a 33% higher point prevalence of moderate-to-severe depression (aPR=1.33; 95% CI: 1.12 to 1.57) and a 40% higher point prevalence of moderate-to-severe anxiety (aPR=1.40; 95% CI: 1.15 to 1.71). No statistically significant association was found between heart attack history and serious psychological distress (table 3). In contrast, a history of stroke was associated with significantly elevated point prevalence for all three mental health outcomes. The prevalence rates increased by 59% for depression (aPR=1.59; 95% CI: 1.36 to 1.86), by 52% for anxiety (aPR=1.52; 95% CI: 1.24 to 1.86) and by 76% for serious psychological distress (aPR=1.76; 95% CI: 1.34 to 2.30) (table 4).

Table 3

Prevalence ratios of mental health symptoms associated with lifetime heart attack, the 2019 and 2018 National Health Interview Survey (NHIS)

Table 4

Prevalence ratios of mental health symptoms associated with lifetime stroke, the 2019 and 2018 National Health Interview Survey (NHIS)

As shown in tables 3 and 4, most covariates were also significantly associated with mental health outcomes, including sex, race/ethnicity, age, marital status, income, insurance status and comorbid conditions. Sensitivity analyses using logistic regression produced results in the same direction as those from the main Poisson regression analyses (online supplemental tables S5 and S6).

Discussion

Using nationally representative samples of adults, we found that the prevalence of moderate-to-severe symptoms of depression, anxiety and serious psychological distress was more than two times as high among individuals with a history of heart attack or stroke compared with those without such a history. Adults with prior cardiovascular events differed significantly in socio-demographic and health characteristics from the other adults. After adjusting for these differences, we observed that depression and anxiety symptoms were 33% and 40% more prevalent, respectively, among heart attack survivors, and 59% and 52% more prevalent among stroke survivors, compared with adults without these conditions. Additionally, stroke survivors had a 76% higher prevalence of serious psychological distress than those without a stroke history.

Our study contributes to the existing literature by estimating the population-level prevalence of mental health symptoms among US adults who have survived a heart attack. A 2018 systematic review of hospital-based studies reported a pooled depression prevalence of 26.0% (95% CI: 18.0% to 34.9%) among North American patients with heart attack, higher than our estimate of 15.7% (95% CI: 13.3% to 18.4%).37 This discrepancy may be explained by differences in study populations. Prior studies focused on individuals recently hospitalised for a heart attack, who may be at a higher risk for depression than our broader community-based sample.10

Our findings support existing evidence that a history of acute cardiovascular event is independently associated with poor mental health and that symptoms can persist long after discharge.9–11 According to a study analysing the Swedish patient registry, individuals were 174% more likely to be diagnosed with a mental health disorder in the first year after a cardiovascular event compared with their full siblings without a cardiovascular condition.10 This increased risk declined to 45% after the first year but remained elevated.10 Similarly, a study in Europe found 30% of patients in rehabilitation after a stroke experienced depression or anxiety 5 years later.11

Effective screening, prevention and treatment of mental health conditions may facilitate the recovery of individuals who experience a heart attack or stroke. Research emphasises the importance of using brief, validated and culturally sensitive screening tools within weeks of a cardiovascular event to guide treatment decisions.38 39 Collaborative care programmes, where mental health and medical professionals work together, have the potential to improve both mental and cardiovascular outcomes in a cost-effective manner.40 Coordination of inpatient treatment and outpatient psychosocial services may help enhance quality of life post-discharge.41 Additionally, cardiac rehabilitation programmes, which incorporate stress management and physical activity, may help prevent stress-related reactions following acute cardiovascular procedures.42 Integrating mental health services into rehabilitation programmes may further strengthen recovery and reduce the burden of mental illness among cardiovascular patients.42

This study has several limitations. First, its cross-sectional design limits our ability to establish causal relationships or determine the temporal order between cardiovascular events and mental health symptoms. Second, the study included only individuals who survived prior heart attacks or strokes, potentially limiting generalisability to those with a higher risk for mortality. Third, cardiovascular history was self-reported, which may have led to an under-representation of adults with severe mental illnesses or cognitive impairment who were unable to accurately account for their medical history. Fourth, unmeasured variables such as physical activity and diet may confound the observed associations. Finally, the use of pre-pandemic data from 2018 to 2019 may limit the applicability of our findings to the COVID-19 era. Future research should explore how the pandemic may have influenced mental health outcomes among individuals with a history of acute cardiovascular events.

Conclusion

Using data from the 2019 and 2018 NHIS, we found that US adults with a history of heart attack or stroke had significantly higher rates of moderate-to-severe symptoms of depression, anxiety and serious psychological distress than adults without these conditions. By leveraging nationally representative data, our study highlights the extent of these problems among US adults and illuminates the independent association between cardiovascular events and mental health outcomes. These findings underscore the importance of developing and implementing effective strategies to address mental health conditions among adults recovering from heart attack or stroke.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.

This post was originally published on https://bmjopen.bmj.com