1. Introduction
Hypertension is a ubiquitous and pressing issue of major public health significance due to its role as a major driver of cardiovascular diseases (CVD, including coronary heart disease and stroke), which are the leading causes of death and disability in the United States (U.S.) and globally [
1]. While cardiovascular disease incidence has decreased among older adults in the past few decades, current evidence has underscored a contrasting trend of increased cardiovascular disease burden in working populations, especially among younger individuals [
2]. Furthermore, recent data indicate an overall decline in working-age mortality rates among most economically developed nations since 2010, whereas in the U.S., mortality rates for working populations have shown the opposite pattern, worsening over time due especially to hypertensive heart disease [
3].
Past attempts to ameliorate the epidemic of hypertension have predominantly emphasized traditional risk factors such as smoking, diet, and exercise [
4]. More recent evidence has identified a key role of non-traditional risk factors such as psychosocial stressors as critical elements of hypertension etiology [
5]. Among these psychosocial factors, work stress and social isolation have received special attention [
6,
7,
8,
9,
10,
11]. Job strain, a well-established operationalization of work stress, has reliably been associated with hypertension in systematic reviews and meta-analyses [
6,
7]. Social isolation, defined as a lack of social contacts and shortage of social relationships, has been evidenced as a severe psychosocial stressor in adulthood that demonstrates robust associations with CVD and consistently predicts increased hypertension risk [
8,
9,
10,
11].
In line with previous attempts to assess the interplay of psychosocial stressors across childhood and adulthood [
12,
13,
14], we include in our exposure models adverse childhood experiences (ACEs, e.g., parental abuse/neglect) as indicators of early life stress, and both job strain and social isolation as metrics of work and nonwork related stress in mid-life. To our knowledge, only few studies based solely on European cohorts have attempted to investigate potential interactions between ACEs, work-related stress, and adulthood health, reporting mixed and inconsistent findings, which, along with no data from the, U.S. present a gap in knowledge [
15,
16,
17,
18,
19,
20]. Importantly, the majority of previous studies assessing psychosocial stressors separated work and non-work-related factors when investigating associations with hypertension; notably, one Canadian study found that the combination of work stress and social stress improved risk estimates for high blood pressure over a 5-year follow-up period [
21]. Therefore, cumulative adulthood stress deserves further exploration, and hence, we combine job strain and social isolation to form the construct of adulthood psychosocial disadvantages (APDs) as an index of cumulative adulthood stress. Our analytic approach was designed to elicit the relative contributions of psychosocial exposures to cardiometabolic disease burden across the life course, examining ACEs in childhood and positing the construct of APDs in mid-life adulthood.
The overall objective of this study was to study associations of work and non-work-related psychosocial stressors with incident hypertension, using data from the national, population-based Mid-life in the United States (MIDUS) [
22,
23] study with prospective cohort design. Furthermore, while prior research has suggested that APDs act as mediators between associations of ACEs and adulthood health outcomes [
14], another line of evidence has argued that ACEs in fact act as effect modifiers, moderating associations of adulthood stress with disease [
15,
16]. Therefore, our aims were two-fold: first, to assess prospective associations of ACEs and APDs with incident hypertension; and second, to examine effect modification of associations between APDs and hypertension by ACEs exposure. We hypothesize that higher ACEs and APDs at baseline exposures will be associated with higher risk of incident hypertension, compared to lower exposure levels, and that incident hypertension risk will be higher when both ACEs and APDs are present.
4. Discussion
This was the first study to assess prospective associations of ACEs and APDs at baseline with incident hypertension in a national, population-based cohort of U.S. workers. Detailed information about the participants’ early life experiences was used to measure ACEs, and a measure of APDs was constructed by combining a classic measure of job strain based on Karasek’s demand–control model [
29] with a well-validated measure of social isolation known as the Berkman–Syme Social Network Index [
32]. While early life exposure to ACEs was not associated with incident hypertension, we found that exposure to APDs at baseline was associated with a significantly elevated risk of incident hypertension within 20 years of follow-up. These results suggest a pathological influence of psychosocial stressors in the etiology of hypertension. Our hypotheses were therefore partially supported by the findings.
These results are consistent with the literature on job strain and social isolation—the individual constituents of exposure to APDs—which has demonstrated robust and stable associations with cardiovascular diseases and hypertension [
5,
6,
7,
8,
9,
10,
11,
40]. Indeed, social isolation in adulthood has been linked to drastically increased CVD mortality risk, as well as hypertension [
8,
9,
11]. Similarly, job strain is well-established as a major contributor to hypertension, with an extensive body of evidence demonstrating consistent and robust associations [
6,
7].
In addition to adulthood stressors, emerging evidence in recent years has revealed that exposure to early life adversity, or ACEs, can have pronounced deleterious impacts on multiple adult cardiometabolic health outcomes, including hypertension [
12,
13]. Life-course exposure models have indicated that early childhood is a critical period that greatly influences responses to environmental stressors later in life [
41,
42]. The present findings regarding ACEs contrast with the prevailing literature documenting the adverse impacts of ACEs on a variety of health conditions. These inconsistencies may be in part explained by the restriction of our sample to the working population, given that most studies on ACEs and adulthood cardiometabolic diseases are in the general population, especially ageing people [
12,
13]. In addition, due to the lack of direct associations of ACEs and risk of hypertension in our study, according to the traditional assumptions of mediation analysis, it is unlikely that a mediating effect by other variables (such as APDs) was present [
43]. Nevertheless, the stratified analyses demonstrate effect modification of ACEs on associations of APDs with incident hypertension. In other words, the hypertension risk associated with APDs exposure was significantly higher among participants with higher exposure to ACEs, compared to those with lower exposure to ACEs. Such examples of moderating effects of ACEs have been previously substantiated [
15,
16]. For instance, a prospective cohort study of Finnish employees followed from childhood to adulthood reported that while job demands—a fundamental component of Karasek’s job strain model—predicted depressive symptoms across 6 years, this association was moderated by ACEs [
15]. The study found that participants with three or more ACEs were more susceptible to developing depressive symptoms elicited by high job demands, compared to those with fewer ACEs. Similarly, a Swedish cohort study showed effect modification by early life adversity, wherein associations of job strain with increased allostatic load were detected only among participants who had experienced adversity in adolescence [
16]. However, another Finnish study and an analysis based on the Whitehall II study found that pre-employment factors such as early life adversity did not moderate associations of job strain with CVD [
19,
20].
The findings of the stratified analyses are consistent with the vulnerability hypothesis, which emphasizes differential susceptibility to adversity between individuals based on genetic factors such as vulnerability and risk alleles, and environmental influences [
44]. Potential explanations underlying observed increases in stress vulnerability with ACEs exposure include heightened systemic responses to stressors [
45] and the increased appraisal of hostile intent in social interactions with others, which is likely to contribute to social isolation [
46]. Evidence also indicates that ACEs may lead to more adverse working conditions and augment perceptions of stressful work environments in adulthood [
47]. Furthermore, childhood adversity has been explicitly and mechanistically linked to adulthood hypertension risk, with proinflammatory mediators and vasoactive factors identified as principal biological drivers of hypertensive pathogenesis [
13]. The preponderance of evidence indicates that early life adversity shapes and characterizes both autonomic physiological and psychological stress responses, exacerbating the impacts of adulthood stressors [
48].
The overall findings regarding exposure to ACEs, APDs, and increased hypertension risk are biologically plausible and mechanistically sound, as the pathways underlying associations of psychosocial stressors with cardiometabolic health conditions have been clearly delineated. Encompassing the entirety of the autonomic nervous system, mechanisms involved in chronic stress response include heightened sympathetic arousal such as increased heart rate and blood pressure, neuroendocrine changes such as increased secretion of cortisol and noradrenaline, and the diffuse perturbation of the hypothalamic-pituitary-adrenal axis [
49,
50]. ACEs specifically have also been found to impair stress reactivity and regulation, with one major consequence being dampened cardiometabolic responses to stress, ultimately leading to increased chronic disease susceptibility [
48]. Cumulatively, these pathways constitute allostatic load, a measure of wear and tear of the body due to stress demands [
51]. With chronic exposures over many years, high levels of psychosocial stress from combined ACEs and APDs are likely to increase allostatic load and hence result in persistent cardiovascular burden.
4.1. Strengths
The major strengths of this study come from the sample population and well-validated measures used. The MIDUS study sample is large and highly diverse, including participants across a range of demographics and occupations, and has a long follow-up length of 20 years. The exposure measure of job strain was based on the well-established Karasek’s demand–control model [
28], and the similarly well-validated Berkman–Syme Social Network Index was used to assess social isolation [
32]. We also accounted for several important confounders and risk factors for hypertension in our multivariable analyses, including smoking, alcohol consumption, physical activity, and MDE [
4,
37]. Furthermore, the sensitivity analyses using an expanded definition of baseline hypertension demonstrated not only the same pattern of associations but an increase in effect size, increasing confidence in the robustness and stability of the results.
4.2. Limitations
There are several limitations in this study. While the reliability of adult retrospective reports of ACEs has been questioned in the past due to potential recall bias, evidence suggests that retrospective reports are generally valid, with robust test–retest reliability ranging from 0.45 to 0.90, and adequate kappa coefficients ranging from 0.52 to 0.72 [
52,
53]. In a similar vein, all exposure information was collected at baseline, and hence our results may be impacted by exposure misclassification bias due to potential changes in APDs during follow-up. Another limitation is the use of self-reported hypertension as the outcome, as opposed to clinically observed hypertension; however, self-reported hypertension has been shown to have good validity, particularly in large-scale epidemiological studies [
54]. Additionally, our study participants were middle aged, predominantly Whites, and had higher levels of education than average levels in the U.S. Therefore, our findings cannot be generalized to those who are non-Whites, younger or older adults, and have lower levels of education. We also did not include other factors which may impact adulthood psychosocial stress furthermore risk of hypertension. For example, female workers with family responsibilities may be more vulnerable to psychosocial stress and negative physical health outcomes than counterparts without such responsibilities. Furthermore, our results may be affected by selection bias, as a substantial number of participants were lost to follow-up from MIDUS I to MIDUS III, and those impacted by attrition may have been systematically different from those who were followed up. The 965 participants lost to follow-up were more likely to be socially isolated, experience greater APDs, a racial or ethnic minority, less educated, lower income, less physically active, smokers, and heavy drinkers. However, there were no significant differences in job strain, prevalence of MDE, or hypertension prevalence at baseline (details available upon request). Finally, while these results offer promising evidence on ACEs and APDs for hypertension incidence, they raise more questions regarding the role of these stress exposures in the broader context of cardiometabolic health. Hence, the contributions of ACEs and APDs to manifested diseases and relevant biomarkers deserve further investigation.