Elsevier

Annals of Epidemiology

Volume 17, Issue 8, August 2007, Pages 591-596
Annals of Epidemiology

Social Avoidance and Long-Term Risk for Cardiovascular Disease Death in Healthy Men: The Western Electric Study

https://doi.org/10.1016/j.annepidem.2007.03.010Get rights and content

Purpose

Although personality traits may contribute to risk for cardiovascular disease (CVD), inconsistent findings have prompted efforts to refine their measurement to include only the hostile and aggressive components. Data are sparse on the “social avoidance” (SA) subscale that measures more indirectly negative traits such as shyness. Thus, we sought to examine the association between SA and CVD, coronary heart disease (CHD), and non-CVD death.

Methods

A total of 2107 men (ages 40–55 years) free of baseline CVD were enrolled in 1957 in the Western Electric Study. SA was measured at study entry using the four-item subscale of the Cook-Medley hostility scale to divide the cohort into four groups according to the degree of social avoidance. CHD mortality, CVD mortality, and non-CVD mortality were determined by death certificate.

Results

After 30 years of follow-up, SA was associated with CVD mortality for the highest vs. the lowest SA group in age-adjusted models (hazard ratio 1.39; 95% confidence interval [95% CI] 1.04–1.84) and after adjustment for traditional CVD risk factors (hazard ratio 1.49; 95% CI 1.12–2.00). After further adjustment for measures of hostility, the findings were similar. Findings for CHD mortality were similar. However, there was no significant association between SA and non-CVD mortality.

Conclusions

Social avoidance is associated with CVD mortality but not with non-CVD mortality in middle-aged men. These findings suggest the hypothesis that social avoidance might promote CVD through physiologic, non-behavioral mechanisms.

Introduction

It has long been suspected that personality traits may contribute to risk for cardiovascular disease (CVD). However, inconsistent associations between the type A behavior pattern and CVD over several decades of research suggest that the search for broad behavioral or personality constructs can sometimes result in the erroneous combinations of unrelated, independent variables (1). The combination of time-urgency and hostility within the Type A behavior construct obscured the unique contributions of these component traits to the risk for CVD 2, 3. Efforts to refine the understanding of the role personality traits have continued with investigation of the subcomponents of the measure of the Cook–Medley hostility scale, the most widely studied measure of hostility in the literature 4, 5, 6.

The Cook–Medley hostility scale measures a broad variety of negative beliefs, attitudes, and behaviors. The heterogeneous content of this scale reflects an absence of a theoretical framework that is consistent with its origin, as it was empirically derived to identify teachers with difficult student relationships (7). In an effort to improve upon the measurement of hostility, Barefoot performed a focused analysis of the item content, leading to the creation of separate subscales within the Cook–Medley hostility scale (see Table 1) (5). Although four of these scales reflect hostile attitudes and behaviors, one subscale, termed “social avoidance,” was interpreted as conceptually distinct from hostility: in contrast to the overtly negative dimensions of the four hostility subscales, the social avoidance subscale described indirectly negative behaviors with a tendency to avoid social contact. Since the publication of these subscales, investigators have used some of the hostility subscales as a more specific measure of hostility in CVD research 4, 6, 8. However, to our knowledge, there has been no analysis of the association between social avoidance and CVD.

The tendency to avoid social contact reflects a broader temperament pattern characterized by a generalized fear of novelty and unfamiliarity (9) that can be observed consistently throughout childhood (10) and into adulthood 11, 12. Furthermore, a large body of research has shown this style of temperament to be associated with heightened sympathetic responses to environmental stimuli (9). Thus, we hypothesized that social avoidance might be related to CVD and coronary heart disease (CHD) death in healthy individuals. The objective of the present study was to examine the association of social avoidance with 30-year CVD, CHD, and non-CVD mortality in a large cohort of healthy men.

Section snippets

Study Sample and Covariates

The Western Electric Study is a longitudinal epidemiologic study designed to study risk factors for CVD (13). In 1957, 5397 men who had been working for the Western Electric Company's Hawthorne Works in Chicago for a minimum of 2 years and were between the age of 40 and 55 were randomly sampled, producing 3102 potential candidates for enrollment into the study. Of these, 2080 men agreed to participate. In addition, there were 27 men from the original pilot study who were included, yielding a

Baseline Characteristics

The social avoidance scores ranged from 0 to 4 (mean = 1.13; SD 1.02). Because of the limited range of responses, the number of participants in each group was not equivalent. In general, the baseline characteristics across all four groups were similar (Table 2). However, body mass index was higher and educational status lower in the groups scoring higher on social avoidance. In addition, higher levels of social avoidance were associated with higher levels of hostility and depression.

Principal Findings

In a cohort of middle-aged men, we observed that social avoidance was associated with CHD and CVD death. After multivariable adjustment for baseline covariates and established CVD risk factors, the association persisted and appeared to strengthen, if anything. The association with CHD and CVD death persisted even after further adjustment for baseline hostility measures. Of note, there was no association between social avoidance and non-CVD mortality. Thus, these findings suggest the possibility

References (23)

  • J. Barefoot et al.

    The Cook-Medley hostility scale: item content and ability to predict survival

    Psychosom Med

    (1989)
  • Cited by (19)

    • A shy heart may benefit from everyday life social interactions with close others: An ecological momentary assessment trial using Bayesian multilevel modeling

      2020, Biological Psychology
      Citation Excerpt :

      Surprisingly, to the authors’ knowledge, studies relating shyness to HRV are lacking. However, a prospective cohort study across 30 years could observe that socially avoidant (i.e., shy) men had a significantly higher risk to die from cardiovascular diseases (Berry, Lloyd-Jones, Garside, Wang, & Greenland, 2007), thus substantiating the view that shyness and behavioral inhibition could harm cardiovascular health, presumably via impaired vagal control, among others (e.g., Alvares et al., 2013; Chalmers et al., 2014; Pittig et al., 2013; Tully, Cosh, & Baune, 2013). Importantly, there is also increasing evidence that social interactions and support from close others could buffer the adverse cardiac effects of negative affective states and traits on vagally-mediated HRV.

    • Association of recent major psychological stress with cardiac arrest: A case-control study

      2018, American Journal of Emergency Medicine
      Citation Excerpt :

      Because that the reliability of history for depression from representatives was not guaranteed because of the very reluctant culture for psychiatric service and strong stigma on people with mental illness in South Korea, we did not include information for clinical depression in-depth surveillance. The measure of social avoidance or social network was a potential proxy for those vulnerable; a few studies have reported on the association between the incidence of cardiovascular disease and these measurements [28,29], but adaption to cardiac arrest was very difficult to ascertain because there was no evidence for the reliability of those measurements from representatives. Finally, because the cohorts were defined by time periods, a priori sample size analysis was not performed.

    • Explaining heterogeneity in the predictive value of Type D personality for cardiac events and mortality

      2016, International Journal of Cardiology
      Citation Excerpt :

      While Type D studies reporting null findings focused on all-cause mortality [10–13], positive studies also used cardiac endpoints [7], suggesting that Type D may be more related to fatal and non-fatal cardiac events [14]. Others showed that social avoidance (a trait that is closely related to Type D) predicted cardiac death but not non-cardiac death [18]. The age composition of the sample is another potential effect modifier, and thus a potential explanation for heterogeneity in findings.

    • Social isolation, social activity and loneliness as survival indicators in old age; A nationwide survey with a 7-year follow-up

      2012, European Geriatric Medicine
      Citation Excerpt :

      Religious participation predicted better survival among oldest old Chinese [16] but there are also contrary findings [17]. Since then, the studies on social networks and prognosis have extended to explore its effects on health and mortality in various diseases [5,17–22], on cognitive decline [7] and emerging disability [23]. According to the newest meta-analysis including results of 87 studies, the cancer patients having high levels of perceived social support, larger social network, and being married were associated with decreases in relative risk for mortality of 25, 20, and 12%, respectively [22].

    View all citing articles on Scopus

    The investigators acknowledge support by the American Heart Association, Dallas, Texas, and its Chicago and Illinois affiliates; the National Heart, Lung, and Blood Institute (NHLBI), Bethesda, MD, grants R01-HL 15174, R01-HL 21010, and R01-HL 03387; and the Chicago Health Research Foundation, Chicago, IL. A list of colleagues who contributed to earlier aspects of this work has been published (Cardiology. 1993;82:191–222). Dr. Berry received support from a NRSA fellowship in the Department of Preventive Medicine, Northwestern University Feinberg School of Medicine (T32HL069771).

    View full text