Abstract
Background and aims:
Psychosocial factors are associated with increased risk of cardiovascular disease (CVD). However, associations with peripheral artery disease (PAD) remain uncharacterized. We aimed to compare associations of psychosocial factors with the risk of PAD and two other major atherosclerotic CVD: coronary heart disease (CHD) and ischemic stroke, in the Atherosclerosis Risk in Communities (ARIC) Study.
Methods:
In 11,104 participants (mean age 56.7 [SD 5.7] years) without a clinical history of PAD and CHD/stroke at baseline (1990–1992), we evaluated four psychosocial domains: depressive/fatigue symptoms by the Maastricht Questionnaire, social support by the Interpersonal Evaluation List, social networks by the Lubben Scale, and trait anger by the Spielberger Scale. PAD was defined as hospitalizations with diagnosis or related procedures. CHD included adjudicated coronary heart disease and stroke included ischemic stroke.
Results:
We observed 397 PAD and 1,940 CHD/stroke events during a median follow-up of 23.1 years. Higher depressive/fatigue symptoms and less social support were significantly associated with incident PAD (adjusted hazard ratios for top vs. bottom quartile 1.65 [95%CI, 1.25–2.19] and 1.40 [1.05–1.87], respectively). When these factors were simultaneously modeled, only depressive/fatigue symptoms remained significant. Incident CHD/stroke was not associated with either of depressive/fatigue symptoms or social support. Social networks and trait anger were not independently associated with PAD or CHD/stroke.
Conclusions:
Depressive/fatigue symptoms and social support (especially the former) were independently associated with the risk of hospitalizations with PAD but not CHD/stroke in the general population. Our results support the importance of depressive/fatigue symptoms in vascular health and suggest the need of including PAD when studying the impact of psychosocial factors on CVD.
Keywords: psychosocial factors, depressive symptoms, peripheral artery disease, atherosclerotic disease, cardiovascular disease
Graphical Abstract
1. Introduction
Psychosocial factors are associated with the risk of cardiovascular disease (CVD). For example, systematic reviews showed depression as a risk factor for the development of coronary heart disease (CHD) and stroke.1–3 Social isolation, a condition closely related to depression, has also been associated with the risk of CHD and stroke.4 Several other studies have also reported a link of trait anger to increased risk of CHD.5, 6
There are plausible mechanisms by which psychosocial factors can increase the risk of CVD. For example, autonomic nervous system dysfunction, hypothalamic-pituitary-adrenal axis dysregulation, inflammation, and increased platelet reactivity can increase CVD risk.7–9 Behavioral mechanisms such as smoking, high-fat diet, poor adherence to medical treatment, and lack of physical exercise are likely to play a role as well.7–9
Despite a large body of evidence linking psychosocial factors with CVD, data regarding the association of psychosocial factors with the risk of peripheral artery disease (PAD) are sparse. To our knowledge, only one prospective study has reported positive associations of depressive/fatigue score and trait anger score with incident PAD in the general population.10 Since risk factor profiles are not necessarily consistent across different atherosclerotic CVD subtypes (e.g., lipids strongly related to CHD, blood pressure to stroke, and smoking and diabetes to PAD),11, 12 it is important to specifically evaluate the associations of psychosocial factors with PAD.
Therefore, we sought to comprehensively investigate the association of several psychosocial factors with the subsequent risk of hospitalizations with PAD in a bi-racial community-based cohort, the Atherosclerosis Risk in Communities (ARIC) Study. To understand potentially different associations with PAD and CHD/stroke, we contrasted their associations with psychosocial factors in a single study population.
2. Materials and methods
2.1. Study population
The ARIC study enrolled 15,792 participants aged 45–64 years from four US communities (Washington County, Maryland; suburban Minneapolis, Minnesota; Jackson, Mississippi; and Forsyth County, North Carolina) in 1987–1989 (visit 1).13 For the present study, we used visit 2 (1990–1992) as baseline because psychosocial factors were systematically evaluated for the first time in ARIC. Of the 14,348 participants who attended visit 2, we excluded participants with race other than White or Black (n = 42), and those with prior history of PAD (n = 641), CHD (n = 743), or ischemic stroke (n = 178). We also excluded participants with missing values for psychosocial factors (n = 654) and other covariates of interest (n = 986), leaving 11,104 participants in the present study. Compared to participants who were excluded due to missing data, our study participants were more likely to be men and less likely to be black. They also tended to have a healthier risk factor profile (e.g., lower prevalence of current smoking and diabetes) (Supplementary Table 1). Nonetheless, we did not recognize a substantial difference between these two groups in terms of several factors such as age, blood pressure, and kidney function. When we compared our study population and those who were excluded due to missing covariates (n = 986), our study population had better scores for depressive symptoms and social support. The study was approved by the Institutional Review Boards of all participating institutions, and all participants gave informed consent.
2.2. Assessment of psychosocial factors:
At ARIC visit 2, the following psychosocial factors were assessed: depressive/fatigue symptoms, social support, social network, and trait anger using self-administered questionnaires. Participants were not informed of the results of these questionnaires.
2.2.1. Depressive/fatigue symptoms:
The Maastricht Questionnaire was designed to measure symptoms of depression and fatigue among employees of the city of Rotterdam14 and consists of 21 questions (Supplementary Table 2).15 The total score ranges from 0 to 42 points, with a higher score indicating greater depressive and fatigue symptoms. The Maastricht Questionnaire has shown a correlation of 0.62 with attitudes and symptoms of depression measured by the Beck Depression Inventory,16 a widely used test to discriminate individuals with depression.17 The Maastricht Questionnaire has a high internal consistency with Cronbach’s alpha of 0.89.14
2.2.2. Social support:
Perceived social functional support was evaluated by the Interpersonal Support Evaluation List Short Form with sixteen questions (Supplementary Table 3).18 The Form had four functions (1) appraisal support, (2) self-esteem, (3) belonging support, and (4) tangible support. The total scale ranges from 0 to 48 points, with the lower scale indicating less social support.
2.2.3. Social network:
The magnitude of social structural network was evaluated using the Lubben Social Network Scale (Supplementary Table 4).19 This scale consists of 10 questions about the size of the participant’s active social network of family, relatives, friends, and neighbors, with a total score ranging from 0 to 50 points. A lower score indicates smaller social network.
2.2.4. Trait anger:
Trait anger was evaluated using the 10-item Spielberger Trait Anger Scale (Supplementary Table 5).20 This questionnaire includes 10 questions and asks typical experience with anger on a 4-point scale. The total score ranges from 10 to 40 points. A higher score implies higher trait anger.
2.3. Outcomes:
Participants were followed until administrative censoring on September 30, 2015, date of outcomes of interest, or loss to follow-up, whichever came first. PAD events were identified from hospitalizations with diagnostic codes for PAD (the International Classification of Diseases Code, Ninth Revision [ICD-9] 440.2, 440.3, 440.4) or procedure codes for leg revascularization (38.18, 39.25, 39.29, 39.50).21 This definition has been used in previous studies22–24 but is likely to miss a number of PAD cases without symptoms or with mild intermittent claudication (Fontaine classification Stage I or II).25 To contrast the impact of psychosocial factors on the risk of PAD, we considered CHD or stroke as secondary outcomes. CHD was defined as definite or probable myocardial infarction and fatal coronary heart disease adjudicated by a physician panel.13 Ischemic stroke was defined as definite or probable cases adjudicated by a physician panel.13
2.4. Other variables of interest:
All covariates were assessed at visit 2, except for physical activity and family income evaluated at visit 1. Age, sex, race, marital status, health insurance, education level, family income, smoking status, alcohol drinking status, and physical activity were based on self-report. Marital status was categorized as married, divorced, never married, separated, and widowed. Education level was classified as basic (less than high school), intermediate (high school graduate or vocational school), or advanced (college, graduate school, or professional school). Family income was categorized as <$25,000, $25,000 −$49,999, or ≥$50,000 per year. Pack-years of smoking were calculated as the average number times the duration of smoking. Smoking status and alcohol drinking status were defined as current, former, and never. Physical activity was categorized as poor, intermediate, and ideal based on the guidelines of American Heart Association.26 Body mass index (BMI) was defined as weight (kg) divided by the square of height (m2). Hypertension was defined as the use of antihypertensive medication, systolic blood pressure ≥140 mmHg, or diastolic blood pressure ≥90 mmHg. Diabetes was defined as taking an antidiabetic drug, a fasting glucose ≥126 mg/dL, non-fasting glucose ≥200 mg/dL, or self-reported physician diagnosis of diabetes. Total cholesterol and high-density lipoprotein cholesterol were determined using enzymatic methods. Antidepressant medication included monoamine oxidase inhibitors, tricyclic antidepressants, and other antidepressants. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration creatinine equation.27
2.5. Statistical analyses:
Baseline characteristics were summarized as mean (standard deviation) for continuous variables (median [interquartile interval] if skewed distribution), and number (proportion) for categorical variables across quartiles of each psychosocial factor and compared using analysis of variance and chi-square test, as appropriate.
We used the Kaplan Meier method to estimate the cumulative incidence of PAD and CHD/stroke by quartiles of each psychosocial factor. Then, we performed Cox proportional hazards models to quantify the independent association of psychosocial factors with PAD and CHD/stroke. We implemented three models to evaluate the impact of potential confounders. Model 1 was adjusted for demographic variables (age, sex, race, and study center). Model 2 added marital status, socioeconomic status (health insurance, education levels, and family income), and behavioral factors (pack-years of smoking, ever smoking status, alcohol drinking, and physical activities). Model 3 additionally adjusted for cardiovascular risk factors (BMI, systolic blood pressure, antihypertensive medication use, diabetes, total cholesterol, high-density lipoprotein cholesterol, cholesterol-lowering medication use, and eGFR). p for trend was obtained by modeling the quartile as an ordinal variable.
Since depressive/fatigue symptoms demonstrated the most robust association with PAD in the entire study population, as described below, we added a post-hoc analysis to evaluate the influence of other psychosocial measures on the association of depressive/fatigue symptoms and incident PAD. Specifically, we further adjusted for antidepressant use (Model 4) and all of the other psychosocial factors as continuous variables (e.g., social support, social network, and trait anger) (Model 5). We also explored the association using questions specific to depressive symptoms in the Maastricht Questionnaire (Supplementary Table 2: Questions 2,3,7,10,16,18, and 19).15 We also conducted subgroup analysis by key demographic, socioeconomic status, and clinical factors (e.g., age, race, sex, smoking status, hypertension, diabetes, marital status, and family income.) for this psychosocial factor to confirm the robustness of its association. The interactions were tested by log-likelihood ratio tests of models with and without interaction terms. We also ran models with time-varying covariates whenever available at subsequent visits (visit 3 [1993–1995] and visit 4 [1996–1998]) or annual phone interview (semi-annual since 2012).
Finally, we categorized the psychosocial factors based on cutoff points used in other studies. Specifically, the Maastricht Questionnaire was dichotomized as ≤13 or >13 points.14 The Lubben Social Network Scale was divided into four categories; large (31 to 50 points), moderate large (26 to 30 points), moderate small (21 to 25 points), and small (0 to 20 points).19 The Spielberger Trait Anger Scale was divided into three categories; lower (10 to 13 points), moderate (14 to 21 points), and higher (22 to 40 points).28 We used tertiles for social support score since we could not find any commonly used cutoffs. All analyses were performed with Stata version 15.0, and a p-value <0.05 was considered statistically significant.
3. Results
3.1. Baseline characteristics:
The mean age of our study participants was 56.7 (SD 5.7) years, with 44.2% male and 22.4% Black. Baseline characteristics were compared by the quartiles of depressive/fatigue symptoms (Table 1), social support score (Supplementary Table 6), social network score (Supplementary Table 7), and trait anger score (Supplementary Table 8). There were dose-response relationships between all the adverse psychosocial factors and the proportion of current smokers, lower physical activity, and antidepressant medication use. Those with higher depressive/fatigue score were more likely to be older, female, Black, have lower education, lower family income, and have worse cardiovascular risk factor profiles (e.g., obesity and higher prevalence of hypertension and diabetes). They were also less likely to be married, alcohol drinkers and have health insurance (Table 1).
Table 1.
Baseline characteristics according to the quartiles of the depressive/fatigue score.
| Q1(0–4) | Q2(5–8) | Q3(9–14) | Q4(15–42) | ||
|---|---|---|---|---|---|
| Lowest depressive/fatigue | Highest depressive/fatigue | ||||
| Number of participants | 3699 | 2239 | 2423 | 2743 | |
| Age (years) | 56.4 (5.6) | 56.8 (5.7) | 56.8 (5.7) | 57.0 (5.7) | |
| Male | 2213 (59.8%) | 1018 (45.5%) | 894 (36.9%) | 779 (28.4%) | |
| Black | 610 (16.5%) | 474 (21.2%) | 612 (25.3%) | 792 (28.9%) | |
| Marital status | Married | 3165 (85.6%) | 1801 (80.4%) | 1913 (79.0%) | 1995 (72.7%) |
| Divorced | 271 (7.3%) | 188 (8.4%) | 202 (8.3%) | 260 (9.5%) | |
| Never married | 64 (1.7%) | 61 (2.7%) | 44 (1.8%) | 58 (2.1%) | |
| Separated | 61 (1.6%) | 54 (2.4%) | 76 (3.1%) | 110 (4.0%) | |
| Widowed | 138 (3.7%) | 135 (6.0%) | 188 (7.8%) | 320 (11.7%) | |
| No health insurance | 203 (5.5%) | 147 (6.6%) | 209 (8.6%) | 294 (10.7%) | |
| Education level | Basic | 461 (12.5%) | 353 (15.8%) | 516 (21.3%) | 805 (29.3%) |
| Intermediate | 1444 (39.0%) | 986 (44.0%) | 1021 (42.1%) | 1235 (45.0%) | |
| Advanced | 1794 (48.5%) | 900 (40.2%) | 886 (36.6%) | 703 (25.6%) | |
| Family income | <$25,000 | 824 (22.3%) | 687 (30.7%) | 911 (37.6%) | 1277 (46.6%) |
| $25,000–49,999 | 1511 (40.8%) | 936 (41.8%) | 909 (37.5%) | 998 (36.4%) | |
| ≥$50,000 | 1364 (36.9%) | 616 (27.5%) | 603 (24.9%) | 468 (17.1%) | |
| Alcohol drinking status | Current | 2430 (65.7%) | 1353 (60.4%) | 1368 (56.5%) | 1359 (49.5%) |
| Former | 636 (17.2%) | 394 (17.6%) | 474 (19.6%) | 646 (23.6%) | |
| Never | 633 (17.1%) | 492 (22.0%) | 581 (24.0%) | 738 (26.9%) | |
| Smoking status | Current | 646 (17.5%) | 436 (19.5%) | 573 (23.6%) | 713 (26.0%) |
| Former | 1550 (41.9%) | 866 (38.7%) | 871 (35.9%) | 884 (32.2%) | |
| Never | 1503 (40.6%) | 937 (41.8%) | 979 (40.4%) | 1146 (41.8%) | |
| Physical activity | Poor | 1019 (27.5%) | 745 (33.3%) | 915 (37.8%) | 1296 (47.2%) |
| Intermediate | 908 (24.5%) | 578 (25.8%) | 612 (25.3%) | 673 (24.5%) | |
| Ideal | 1772 (47.9%) | 916 (40.9%) | 896 (37.0%) | 774 (28.2%) | |
| Pack-years of smoking | 3.4 (0.0, 24.0) | 2.5 (0.0, 26.0) | 4.0 (0.0, 26.2) | 2.6 (0.0, 29.0) | |
| Body mass index (kg/m2) | 27.3 (4.6) | 27.7 (5.1) | 27.9 (5.4) | 28.8 (6.2) | |
| Systolic blood pressure (mmHg) | 119.9 (17.4) | 121.6 (18.0) | 121.1 (18.6) | 121.9 (19.8) | |
| Diastolic blood pressure (mmHg) | 72.4 (9.9) | 72.6 (10.0) | 71.8 (10.4) | 71.8 (10.5) | |
| Antihypertensive medication use | 709 (19.2%) | 525 (23.4%) | 620 (25.6%) | 862 (31.4%) | |
| Diabetes mellitus | 381 (10.3%) | 281 (12.6%) | 329 (13.6%) | 481 (17.5%) | |
| Total cholesterol (mg/dl) | 207.1 (37.1) | 209.0 (39.2) | 209.0 (39.5) | 211.9 (41.4) | |
| High-density lipoprotein (mg/d) | 48.3 (16.1) | 50.2 (17.2) | 50.9 (17.4) | 50.9 (16.8) | |
| Cholesterol lowering medication use | 184 (5.0%) | 113 (5.0%) | 129 (5.3%) | 166 (6.1%) | |
| Antidepressant medication use | 43 (1.2%) | 42 (1.9%) | 75 (3.1%) | 140 (5.1%) | |
| Estimated glomerular filtration rate (mL/min/1.73m2) | 95.9 (13.8) | 96.4 (14.6) | 97.1 (15.9) | 97.5 (16.8) | |
Values for categorical variables are number (percentage). Values for continuous variables are mean (standard deviation) or median (interquartile interval).
Similarly, those with lower social support and smaller social network were more likely to have lower education, lower family income, and worse cardiovascular risk factor profiles (higher prevalence of hypertension and diabetes), and less likely to be married. But they were more likely to be male (Supplementary Table s 6 and 7). Higher trait anger demonstrated slightly different patterns compared to the other psychosocial factors, e.g., more likely to be younger, male, married, and drinkers (Supplementary Table 8).
3.2. Association of psychosocial factors with incident PAD and CHD/stroke:
During a median follow-up of 23.1 years, there were 397 cases of incident hospitalizations with PAD and 1,940 cases of incident CHD/stroke. Higher depressive/fatigue score was associate with higher cumulative incidence of PAD and CHD/stroke in a graded fashion (Figure 1). A largely similar pattern was observed for social support score and PAD, but the separation across its quartiles was not that evident for social support and CHD/stroke, especially after ~20 years of follow up (Supplementary Fig. 1). In contrast, social network score showed a significant relationship with CHD/stroke but not necessarily with PAD (Supplementary Fig. 2). Trait anger did not show clear relations regardless of outcomes (Supplementary Fig. 3).
Figure 1.
Cumulative incidence of peripheral artery disease (PAD) (A) and CHD/stroke (B) by the quartiles of the depressive/fatigue score.
The significant associations of depressive/fatigue score with incident PAD and CHD/stroke remained consistent even after adjustment for demographic factors (e.g., hazard ratios [HRs] in top vs. bottom quartile 2.43 [95%CI, 1.86–3.18] and 1.45 [95%CI, 1.28–1.64], respectively) (Model 1 in Table 2). The associations were attenuated but remained significant for both outcomes after adjustment for behavioral factors (Model 2 in Table 2). Once we accounted for clinical factors, the significant association was seen for PAD (HR, 1.65 [95%CI, 1.25–2.19]) but not necessarily for CHD/stroke (1.14 [95%CI, 0.99–1.29]) (Model 3 in Table 2). We observed generally consistent results across the subgroups tested, without any significant statistical interaction (Figure 2). The associations of depressive/fatigue score with incident PAD was consistent even when we further adjusted for antidepressant medication and all other psychosocial factors (Supplementary Table 9) or when we relied on the questions specific to depression in the Maastricht Questionnaire (Supplementary Table 10). The association was slightly attenuated but still significant when we adjusted for time-varying covariates using the same covariates as Model 3 (Supplementary Table 11). Overall, the results were consistently more evident for PAD than for CHD/stroke.
Table 2.
Hazard ratios (95%CIs) of incident peripheral artery disease (PAD) and CHD/stroke by quartiles of the depressive/fatigue symptoms.
| Depressive/fatigue | Q1 (0–4) | Q2 (5–8) | Q3 (9–14) | Q4 (15–42) | p for trend |
|---|---|---|---|---|---|
| Number of participants | 3699 | 2239 | 2423 | 2743 | |
| PAD | |||||
| Cases | 98 | 65 | 91 | 143 | |
| Model 1 | Ref. | 1.17 (0.85–1.60) | 1.60 (1.19–2.13)* | 2.43 (1.86–3.18)** | <0.001 |
| Model 2 | Ref. | 1.04 (0.76–1.44) | 1.35 (1.00–1.81)* | 1.84 (1.40–2.43)** | <0.001 |
| Model 3 | Ref. | 1.01 (0.73–1.38) | 1.29 (0.96–1.73) | 1.65 (1.25–2.19)** | <0.001 |
| CHD/stroke | |||||
| Cases | 599 | 379 | 436 | 526 | |
| Model 1 | Ref. | 1.12 (0.99–1.28) | 1.27 (1.12–1.44)** | 1.45 (1.28–1.64)** | <0.001 |
| Model 2 | Ref. | 1.06 (0.93–1.20) | 1.15 (1.01–1.31)* | 1.22 (1.07–1.38)* | 0.001 |
| Model 3 | Ref. | 1.02 (0.90–1.16) | 1.11 (0.98–1.26) | 1.14 (0.99–1.29) | 0.030 |
Model 1: age, sex, race, and study center.
Model 2: Model 1 + marital status, education level, health insurance, family income, and behavioral factors (pack-years of smoking, ever smoking status, alcohol drinking status, and physical activity).
Model 3: Model 2 + cardiovascular risk factors (body mass index, systolic blood pressure, antihypertensive medication use, total cholesterol level, high-density lipoprotein level, cholesterol lowering medication use, and estimated glomerular filtration rate).
p-value <0.05.
p-value <0.001.
Figure 2.
Adjusted hazard ratios (95%CI) of incident peripheral artery disease according to the depressive/fatigue score (Q1–3 [0–14] vs Q4 [15–42]) by demographic and clinical subgroups.
The hazard ratios were adjusted for age, sex, race, study center, marital status, education level, health insurance, family income, pack-years of smoking, ever smoking status, alcohol drinking status, physical activity, body mass index, systolic blood pressure, antihypertensive medication use, diabetes, total cholesterol level, high-density lipoprotein level, cholesterol lowering medication use, estimated glomerular filtration rate, as appropriate.
Less social support demonstrated a significant positive association only with PAD in all Models (Table 3). Neither social network score nor trait anger showed independent associations with PAD or CHD/stroke in Models 2 or 3 (Supplementary Tables 12 and 13). After the additional adjustment for antidepressant medication and depressive/fatigue score, less social support was no longer independently associated with incident PAD, while a significant relation was observed when the second bottom quartile was used as a reference (Supplementary Table 14).
Table 3.
Hazard ratios (95%CIs) of incident peripheral artery disease (PAD) and CHD/stroke by quartiles of the social support score.
| Social support | Q1(43–48) | Q2(39–42) | Q3(34–38) | Q4(2–33) | p for trend |
|---|---|---|---|---|---|
| Number of participants | 2467 | 2752 | 2890 | 2995 | |
| PAD | |||||
| Cases | 74 | 75 | 96 | 152 | |
| Model 1 | Ref. | 0.91(0.66–1.25) | 1.08(0.79–1.46) | 1.72(1.30–2.27)** | <0.001 |
| Model 2 | Ref. | 0.92(0.67–1.28) | 1.05(0.77–1.44) | 1.52(1.14–2.02)* | 0.001 |
| Model 3 | Ref. | 0.86(0.62–1.19) | 0.98(0.71–1.33) | 1.40(1.05–1.87)* | 0.003 |
| CHD/stroke | |||||
| Cases | 400 | 466 | 525 | 549 | |
| Model 1 | Ref. | 1.02(0.90–1.17) | 1.08(0.95–1.23) | 1.14(1.00–1.29) | 0.033 |
| Model 2 | Ref. | 1.00(0.87–1.14) | 1.00(0.88–1.14) | 1.00(0.88–1.14) | 0.97 |
| Model 3 | Ref. | 0.96(0.84–1.10) | 0.96(0.84–1.09) | 0.96(0.84–1.09) | 0.59 |
Model 1: age, sex, race, and study center.
Model 2: Model 1 + marital status, education level, health insurance, family income, and behavioral factors (pack-years of smoking, ever smoking status, alcohol drinking status, and physical activity).
Model 3: Model 2 + cardiovascular risk factors (body mass index, systolic blood pressure, antihypertensive medication use, total cholesterol level, high-density lipoprotein level, cholesterol lowering medication use, and estimated glomerular filtration rate).
p-value <0.05.
p-value <0.001.
The associations were consistent when we categorized psychosocial factors based on cutoff points used in other studies (Supplementary Tables 15–18).
4. Discussion
In this community-based cohort study, among the four domains of psychosocial factors tested, higher depressive/fatigue symptoms demonstrated the most robust association with incident hospitalizations with PAD, followed by the lack of social support. The association of depressive/fatigue symptoms with incident PAD remained consistent after additionally adjusting for antidepressant medication use or the other three psychosocial factors (social support, social network, and trait anger), and stratifying by demographic and clinical conditions. Neither small social network nor higher trait anger showed a strong association with incident PAD. CHD/stroke consistently demonstrated less evident associations with all psychosocial factors than PAD. Specifically, depressive/fatigue symptoms showed a significant association with incident CHD/stroke only in models adjusted for demographic and behavioral factors. None of the other three psychosocial factors was independently related to incident CHD/stroke.
The robust association of depressive/fatigue symptoms with the risk of PAD in our study is in line with an earlier ARIC report with a mean follow-up of ~10 years.10 However, there are several unique aspects of the current study. First, our study extended this observation over 25 years of follow-up. Second, we also uniquely compared the association of depressive/fatigue symptoms as well as the other three psychosocial factors with PAD and CHD/stroke in a single study population. Third, we confirmed largely consistent results in demographic and clinical subgroups.
In terms of potential mechanisms, we should recognize that MQ captures both fatigue and depressive symptoms. Therefore, it is possible our observation reflects somatic symptoms of chronic ischemia to the lower limb muscles. However, we confirmed a robust association when we restricted to questions specific to depression. Several mechanisms linking depression to atherosclerotic diseases have been proposed. For example, the hypothalamic-pituitary-adrenal axis can be activated in individuals with depression, which can cause insulin resistance29 and perivascular inflammation.30 Indeed, proinflammatory markers are reported to be activated by psychosocial factors.31, 32 Additionally, people with depression tended to be noncompliant to preventive therapies since they lack positive expectations in the efficacy of treatment.33 Finally, lower resilience to stress in depression34 may play a role since psychological stress is a risk factor for developing CVD.35
Among the remaining three psychosocial factors, only the lack of social support was associated with incident PAD independently of traditional risk factors, although the association was considerably attenuated after further adjustment for depressive/fatigue symptoms. We are not sure why these three other factors contributed to PAD less than depressive/fatigue symptoms, but similarly, CHD has been more robustly associated with depressive symptoms.36, 37 Potentially, depressive/fatigue symptoms can be located downstream of the adverse psychosocial related cascade. In this regard, depressive/fatigue symptoms may reflect an individual’s susceptibility to lack social support or network and thus may be more strongly associated with CVD.38 Additionally, a larger size of social relationship sometimes reflects the need for social support.
Potential reasons behind the stronger association of depressive/fatigue symptoms with PAD than CHD/stroke in our study would deserve some discussion. This may reflect different risk factor profiles between PAD and CHD/stroke. For example, diabetes and smoking are known as potent risk factors of PAD,12 while lipids and hypertension are tightly related to CHD and stroke, respectively. Of note, among traditional atherosclerotic risk factors, depression has a strong relation to diabetes and smoking.39, 40 Nonetheless, it is noteworthy that the association of depressive/fatigue symptoms with PAD remained robust even after accounting for diabetes and smoking, suggesting the presence of other mechanisms. This result should be confirmed in other settings and if so, future studies are needed to explore unique contributions of depressive/fatigue symptoms to PAD.
Our study has several clinical and research implications. First, among various psychosocial factors, our results further support the importance of evaluating depressive/fatigue symptoms as a potential risk factor for PAD. Second, given the stronger association of psychosocial factors with PAD than CHD/stroke in our study, studies of psychosocial factors and CVD should consider PAD as a relevant outcome. This is important since PAD has been overlooked41 and many community-based cohort studies have systems to ascertain CHD/stroke but not PAD. Third, although the management of traditional CVD risk factors is central for CVD prevention, our results suggest the importance of paying attention to psychosocial factors, especially depressive/fatigue symptoms, for PAD risk. Nonetheless, whether interventions to these psychosocial factors can reduce the risk of PAD needs future investigations.
Several limitations of this study should be acknowledged. First, PAD events were not adjudicated by a physician panel like CHD/stroke. Also, since they were based on hospitalization codes, our definition likely missed asymptomatic cases or patients with mild intermittent claudication (i.e., Fontaine classification Stage I or II). Second, we used the best available data on psychosocial factors in ARIC for our study question, and thus cannot deny the possibility that other instruments may demonstrate stronger associations with incident PAD than those in the present study. The data was self-reported without clinical interview by a psychiatrist or psychologist. Third, our results may not be generalized to other countries or other races since psychosocial conditions may be influenced by environmental and cultural factors. Finally, psychosocial factors may change over time, but our study was based on data at a single time point.
In conclusion, among the four psychosocial factors tested, the most robust association with hospitalizations with PAD was seen for depressive/fatigue symptoms. Overall, psychosocial factors were more strongly associated with PAD than CHD/stroke. These results support the etiological contribution of depressive/fatigue symptoms to the development of PAD and highlight the importance of including PAD when studying the relationship between psychosocial factors and CVD.
Supplementary Material
Highlights.
Depressive/fatigue symptoms showed a robust association with incident PAD.
Weaker than depressive symptoms, but social support was also associated with PAD.
Psychosocial factors were more strongly associated with PAD than ASCVD.
The inclusion of PAD is important when studying psychosocial factors on CVD.
Acknowledgments
The authors thank the staff and participants of The Atherosclerosis Risk in Communities Study for their important contributions.
Financial support
The Atherosclerosis Risk in Communities Study is carried out as a collaborative study supported by National Heart, Lung, and Blood Institute contracts (HHSN268201100005C, HHSN268201100006C, HHSN268201100007C, HHSN268201100008C, HHSN268201100009C, HHSN268201100010C, HHSN268201100011C, and HHSN268201100012C).
Declaration of interests
☒ The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
KM received research funding and personal fees from Fukuda Denshi outside of the work.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. Int J Geriatr Psychiatry. July 2007;22(7):613–26. doi: 10.1002/gps.1723 [DOI] [PubMed] [Google Scholar]
- 2.Pan A, Sun Q, Okereke OI, Rexrode KM, Hu FB. Depression and risk of stroke morbidity and mortality: a meta-analysis and systematic review. Jama. September 21 2011;306(11):1241–9. doi: 10.1001/jama.2011.1282 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gan Y, Gong Y, Tong X, et al. Depression and the risk of coronary heart disease: a meta-analysis of prospective cohort studies. BMC Psychiatry. December 24 2014;14:371. doi: 10.1186/s12888-014-0371-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Valtorta NK, Kanaan M, Gilbody S, Ronzi S, Hanratty B. Loneliness and social isolation as risk factors for coronary heart disease and stroke: systematic review and meta-analysis of longitudinal observational studies. Heart. July 1 2016;102(13):1009–16. doi: 10.1136/heartjnl-2015-308790 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Williams JE, Paton CC, Siegler IC, Eigenbrodt ML, Nieto FJ, Tyroler HA. Anger proneness predicts coronary heart disease risk: prospective analysis from the atherosclerosis risk in communities (ARIC) study. Circulation. May 2 2000;101(17):2034–9. [DOI] [PubMed] [Google Scholar]
- 6.Kawachi I, Sparrow D, Spiro A 3rd, Vokonas P, Weiss ST. A prospective study of anger and coronary heart disease. The Normative Aging Study. Circulation. November 1 1996;94(9):2090–5. [DOI] [PubMed] [Google Scholar]
- 7.Strike PC, Steptoe A. Psychosocial factors in the development of coronary artery disease. Progress in Cardiovascular Diseases. 2004;46(4):337–347. doi: 10.1016/j.pcad.2003.09.001 [DOI] [PubMed] [Google Scholar]
- 8.Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation. April 27 1999;99(16):2192–217. [DOI] [PubMed] [Google Scholar]
- 9.Hare DL, Toukhsati SR, Johansson P, Jaarsma T. Depression and cardiovascular disease: a clinical review. Eur Heart J. June 1 2014;35(21):1365–72. doi: 10.1093/eurheartj/eht462 [DOI] [PubMed] [Google Scholar]
- 10.Wattanakit K, Williams JE, Schreiner PJ, Hirsch AT, Folsom AR. Association of anger proneness, depression and low social support with peripheral arterial disease: the Atherosclerosis Risk in Communities Study. Vascular medicine (London, England). August 2005;10(3):199–206. doi: [DOI] [PubMed] [Google Scholar]
- 11.Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics−-2012 update: a report from the American Heart Association. Circulation. January 3 2012;125(1):e2–e220. doi: 10.1161/CIR.0b013e31823ac046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Price JF, Mowbray PI, Lee AJ, Rumley A, Lowe GD, Fowkes FG. Relationship between smoking and cardiovascular risk factors in the development of peripheral arterial disease and coronary artery disease: Edinburgh Artery Study. Eur Heart J. March 1999;20(5):344–53. doi: 10.1053/euhj.1998.1194 [DOI] [PubMed] [Google Scholar]
- 13.The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. The ARIC investigators. Am J Epidemiol. April 1989;129(4):687–702. [PubMed] [Google Scholar]
- 14.Appels A, Hoppener P, Mulder P. A questionnaire to assess premonitory symptoms of myocardial infarction. Int J Cardiol. October 1987;17(1):15–24. [DOI] [PubMed] [Google Scholar]
- 15.Williams JE, Mosley TH Jr., Kop WJ, Couper DJ, Welch VL, Rosamond WD. Vital exhaustion as a risk factor for adverse cardiac events (from the Atherosclerosis Risk In Communities [ARIC] study). Am J Cardiol. June 15 2010;105(12):1661–5. doi: 10.1016/j.amjcard.2010.01.340 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kopp MS, Falger PR, Appels A, Szedmak S. Depressive symptomatology and vital exhaustion are differentially related to behavioral risk factors for coronary artery disease. Psychosom Med. Nov-Dec 1998;60(6):752–8. [DOI] [PubMed] [Google Scholar]
- 17.Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Archives of general psychiatry. June 1961;4:561–71. doi: 10.1001/archpsyc.1961.01710120031004 [DOI] [PubMed] [Google Scholar]
- 18.Cohen SMR, Kamarck T, Hoberman HM. Measuring the Functional Components of Social Support. Theory, Research and Applications 1985;Netherlands: Springer.:73–94. [Google Scholar]
- 19.Lubben JE. Assessing social networks among elderly populations. Family & Community Health. 1988;11(3):42–52. [Google Scholar]
- 20.Spielberger CD JG, Russell S, Crane RS. Assessment of anger: The state-trait anger scale. In: Butcher JN, Spielberger CD, eds Advances in Personality Assessment. 1983; vol 2 [Google Scholar]
- 21.Garg PK, O’Neal WT, Mok Y, Heiss G, Coresh J, Matsushita K. Life’s Simple 7 and Peripheral Artery Disease Risk: The Atherosclerosis Risk in Communities Study. Am J Prev Med. November 2018;55(5):642–649. doi: 10.1016/j.amepre.2018.06.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.O’Hare AM, Newman AB, Katz R, et al. Cystatin C and incident peripheral arterial disease events in the elderly: results from the Cardiovascular Health Study. Archives of internal medicine. December 12-26 2005;165(22):2666–70. doi: 10.1001/archinte.165.22.2666 [DOI] [PubMed] [Google Scholar]
- 23.Wattanakit K, Folsom AR, Selvin E, Coresh J, Hirsch AT, Weatherley BD. Kidney function and risk of peripheral arterial disease: results from the Atherosclerosis Risk in Communities (ARIC) Study. Journal of the American Society of Nephrology : JASN. February 2007;18(2):629–36. doi: 10.1681/asn.2005111204 [DOI] [PubMed] [Google Scholar]
- 24.Matsushita K, Kwak L, Yang C, et al. High-sensitivity cardiac troponin and natriuretic peptide with risk of lower-extremity peripheral artery disease: the Atherosclerosis Risk in Communities (ARIC) Study. Eur Heart J. July 1 2018;39(25):2412–2419. doi: 10.1093/eurheartj/ehy106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fontaine R, Kim M, Kieny R. [Surgical treatment of peripheral circulation disorders]. Helvetica chirurgica acta. December 1954;21(5–6):499–533. Die chirurgische Behandlung der peripheren Durchblutungsstörungen. [PubMed] [Google Scholar]
- 26.Piercy KL, Troiano RP, Ballard RM, et al. The Physical Activity Guidelines for Americans. Jama. November 20 2018;320(19):2020–2028. doi: 10.1001/jama.2018.14854 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Annals of internal medicine. May 5 2009;150(9):604–12. doi: 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lopez FG, Thurman CW. High-Trait and Low-Trait Angry College Students: A Comparison of Family Environments. Journal of Counseling & Development. 1993;71(5):524–527. doi: 10.1002/j.1556-6676.1993.tb02235.x [DOI] [Google Scholar]
- 29.Rosmond R, Björntorp P. The hypothalamic-pituitary-adrenal axis activity as a predictor of cardiovascular disease, type 2 diabetes and stroke. Journal of internal medicine. February 2000;247(2):188–97. doi: 10.1046/j.1365-2796.2000.00603.x [DOI] [PubMed] [Google Scholar]
- 30.Walker BR. Glucocorticoids and cardiovascular disease. European journal of endocrinology. November 2007;157(5):545–59. doi: 10.1530/eje-07-0455 [DOI] [PubMed] [Google Scholar]
- 31.Yang YC, McClintock MK, Kozloski M, Li T. Social isolation and adult mortality: the role of chronic inflammation and sex differences. Journal of health and social behavior. June 2013;54(2):183–203. doi: 10.1177/0022146513485244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Girard D, Tardif JC, Boisclair Demarble J, D’Antono B. Trait Hostility and Acute Inflammatory Responses to Stress in the Laboratory. PLoS One. 2016;11(6):e0156329. doi: 10.1371/journal.pone.0156329 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.DiMatteo MR, Lepper HS, Croghan TW. Depression Is a Risk Factor for Noncompliance With Medical Treatment: Meta-analysis of the Effects of Anxiety and Depression on Patient Adherence. Archives of Internal Medicine. 2000;160(14):2101–2107. doi: 10.1001/archinte.160.14.2101 [DOI] [PubMed] [Google Scholar]
- 34.Wild J, Smith KV, Thompson E, Béar F, Lommen MJ, Ehlers A. A prospective study of pre-trauma risk factors for post-traumatic stress disorder and depression. Psychological medicine. September 2016;46(12):2571–82. doi: 10.1017/s0033291716000532 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dimsdale JE. Psychological stress and cardiovascular disease. Journal of the American College of Cardiology. 2008;51(13):1237–1246. doi: 10.1016/j.jacc.2007.12.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Frasure-Smith N, Lespérance F. Depression and Other Psychological Risks Following Myocardial Infarction. Archives of general psychiatry. 2003;60(6):627–636. doi: 10.1001/archpsyc.60.6.627 [DOI] [PubMed] [Google Scholar]
- 37.Whittaker KS, Krantz DS, Rutledge T, et al. Combining psychosocial data to improve prediction of cardiovascular disease risk factors and events: The National Heart, Lung, and Blood Institute--sponsored Women’s Ischemia Syndrome Evaluation study. Psychosom Med. April 2012;74(3):263–70. doi: 10.1097/PSY.0b013e31824a58ff [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Utz RL, Swenson KL, Caserta M, Lund D, deVries B. Feeling lonely versus being alone: loneliness and social support among recently bereaved persons. J Gerontol B Psychol Sci Soc Sci. 2014;69(1):85–94. doi: 10.1093/geronb/gbt075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Anderson RJ, Freedland KE, Clouse RE, Lustman PJ. The prevalence of comorbid depression in adults with diabetes: a meta-analysis. Diabetes Care. June 2001;24(6):1069–78. doi: 10.2337/diacare.24.6.1069 [DOI] [PubMed] [Google Scholar]
- 40.Bandiera FC, Arguelles W, Gellman M, et al. Cigarette Smoking and Depressive Symptoms Among Hispanic/Latino Adults: Results From the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). Nicotine & Tobacco Research. 2014;17(6):727–734. doi: 10.1093/ntr/ntu209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Selvin E, Hirsch AT. Contemporary risk factor control and walking dysfunction in individuals with peripheral arterial disease: NHANES 1999–2004. Atherosclerosis. December 2008;201(2):425–33. doi: 10.1016/j.atherosclerosis.2008.02.002 [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.