Background
Health-related quality of life (HRQOL) is becoming increasingly used for evaluating health services in clinical practice [
1]. The World Health Organization has defined the HRQOL as an individual’s view of himself or herself in terms of his/her hopes, objectives, values, and worries [
2]. The HRQOL reflects both an individual’s physical and psychological health. Aside from this, the HRQOL also reflects an individual’s social status and the quality of his/her surrounding environment. All in all, an individual’s physical health, psychological health, social status, and the quality of his/her surrounding environment, can all influence an individual’s well-being.
A growing body of research indicates that people who report lower HRQOL have higher all-cause mortality [
3‐
6]. Some researchers have shifted the focus of their HRQOL research to its relationship with specific disease incidence and disease outcomes to increase the understanding of why this association exists. For example, some research teams first explored whether lower HRQOL is a risk factor for cardiovascular disease (CVD). As a follow up inquisition, others also wondered that if the above mentioned hypothesis proved to be true, which realms of the HRQOL acts as a predictor of CVD. As they have anticipated, the results of their studies have shown that impaired HRQOL was related to high prevalence of overt CVD [
7] and subsequent CVD incidence [
8‐
10]. In addition, they all approved that impaired HRQOL in the physical domain (HRQOL
physical) may increase CVD risks independently of traditional CVD risk factors. However, the mechanisms associated with the link between this novel risk factor and CVD are yet to be established.
The score of the HRQOL
physical is rated by the participant intuitively. Thus, HRQOL
physical reflects the bodily physiological health determined by certain factors perceived by the individual. This process of perception relies on an integral body-mind connection. When mind–body dissonance or disconnection [
11] occurs, participants might report an unrealistically high HRQOL
physical to the investigators though they in fact had poor physical conditions [
12]. In humans, the autonomic nervous system (ANS) plays important roles in flexibly adjusting the response of the body to a range of environmental demands and internal stimuli [
13,
14]. A series of past studies have shown altered ANS activity among subjects with psychosomatic symptoms, which include anxiety [
15], depression [
16] and chronic fatigue syndrome [
17]. All these findings support the idea that the ANS activity reflects the interaction between mind and body. So far various techniques have been developed to detect the ANS activity. In the last two decades, short-term frequency-domain analysis of heart rate variability has been developed as a sophisticated and non-invasive tool for detecting ANS activity [
18]. Heart rate variability (HRV) refers to the complex beat-to-beat variation in heart rate measured by electrocardiogram. Variability in heart rate is produced by the sympathetic nerves, which accelerate heart rate, and the parasympathetic (vagus) nerves, which decelerate it. An increased level of HRV reflects a healthy ANS that is able to respond to environmental demands [
13]. Individuals with higher levels of resting HRV have been shown to have greater abilities of emotion regulation [
19]. In contrast, low HRV is an indicator of autonomic inflexibility [
20] and a predictor of poor health status [
21]. In the cardiovascular territory, a high degree of HRV aids healthy cardiac activity and provides a protective effect against myocardial infarction and heart failure [
22], whereas decreased parasympathetic tone [
23] is associated with an increased risk of CVD and mortality. All in all, HRV is a biologically plausible candidate for investigating the mechanisms underlying the contribution of HRQOL
physical to CVD risks. This in turn prompts us to raise the important question of whether there is an association between HRQOL
physical and HRV.
Most studies supporting an association existing between the HRQOL and HRV have been conducted in clinical samples with chronic illness [
24‐
26]. The physiologic consequences of chronic illness could influence the relationship observed between HRQOL and HRV. Few studies have focused on healthy individuals in order to avoid overestimation of the association between HRQOL and HRV, but these studies’ results were inconsistent [
11,
27,
28]. Furthermore, these studies have the limitation of lacking a procedure to control the physical and psychiatric conditions that can confound HRV profiles. Overall, these studies have only tried to prove whether HRV predicts HRQOL. To the extent of our knowledge, no study has examined whether HRQOL
physical and/or other domains independently affect the HRV profiles in healthy individuals.
Aims of the study
The present study aimed to test the following hypotheses: (a) there is a meaningful correlation between the HRQOLphysical and HRV; (b) in healthy subjects, the HRQOLphysical can independently contribute to HRV; and (c) the subjects with low levels of HRQOLphysical show lower HRV as compared to those with high levels of HRQOLphysical.
Discussion
Numerous studies have stressed that ANS activity has significant impact on HRQOL among patient population, mainly because altered ANS activity is a frequent physiological consequence of the diseases, thereby relating ANS activity to patient perceptions of HRQOL in a unidirectional manner [
12,
24‐
26,
45]. To date, this is the first study to examine whether HRQOL
physical can independently contribute to HRV in a well-defined healthy population. Our study is also pioneering in raising the possibility that the relationship between ANS activity and HRQOL could be bidirectional. The main results of our study are summed up as follows.
Our study showed a significant contribution of HRQOL
physical to variances in HRV after excluding potential confounding factors. Consistent with this, when our participants were divided into two groups by their scores of HRQOL
physical, those with low HRQOL
physical exhibited significantly lower HRV as compared with those with high HRQOL
physical. The significant differences were not due to the levels of self-reported depression and/or anxiety because adjusting for psychological states did not alter the significance. Moreover, it should be stressed that the difference in the mean BAI scores between high and low HRQOL
physical groups may be statistically significant due to a large sample size, but not be clinically significant because mean BAI scores of both groups are only within the borderline of minimal to mild level of anxiety (score 7–8). The same applied to the mean BDI scores. In the present study, the HRQOL
physical includes seven items that comprise the extent of physical pain, discomfort, energy, fatigue, and quality of sleep. The association of low HRQOL
physical scores on HRV may be driven by the overall contributions of the impact of each item on HRV since data in the literature indicated that compared to healthy controls individuals with low energy expenditure [
46,
47], sleep disturbance [
48], chronic fatigue [
17], or chronic pain [
49] were associated with low HRV. A working conceptual model conceptualizes HRV as a process involving regulatory mechanisms of ANS, which may structurally, as well as functionally, link psychological and/or cognitive processes with health-related physiology [
50]. Our study results lend further support to this model and suggest that in a healthy population HRQOL
physical may be a sensitive reflection of one’s current autonomic balance or a proxy for a subclinical state or disease that are not clinically manifested yet–– it can be a measure of one’s awareness of the symptoms or disease risk factors that may impact upon his or her future health outcomes especially when these factors sustained over a long period of time.
It is well known that eighty percent of CVD risk could be explained by traditional CVD risk factors including non-modifiable (i.e., age and sex) and modifiable (i.e., hypertension, smoking, diabetes, obesity and high cholesterol) ones [
51,
52]. Little is known about the major determinants for the remaining twenty percent of risk in CVD. Researchers have identified reduced HRV [
21,
53] and other nontraditional cardiovascular risk factors [
54,
55] as pointers in helping to improve risk assessment. Recent studies reported that HRQOL
physical was strongly associated with cardiovascular events and CVD-specific death, and that the association was independent of traditional risk factors [
56,
57]. One possibility is that the items of HRQOL
physical mainly reflect the bodily physiological health determined by other factors unmeasured in the aforementioned traditional risk calculations but perceived by individuals [
8]. Another possibility is that the sum expression of the diverse items of HRQOL
physical is conceptually close to self-rated health (SRH), a simple measure of subjective health status. Much evidence already shows that SRH is an independent predictor of CVD incidence and cardiovascular mortality [
58,
59], and that individuals who assess their health as poor have a higher mortality risk or cardiovascular events than those whose assessment is excellent [
60]. However, studies in this area did not depict a clear underlying mechanism for HRQOL
physical to confer increased risk for CVD. Our study results may provide a new insight on why HRQOL
physical strongly predicts CVD health and death.
Compared to the strong link observed with HRQOL in the physical domain and HRV, the associations between psychological, social, and environmental domain and HRV were weaker and statistically not significant. This is consistent with a recent study reporting that these domains did not predict the incidence CV events [
8]. This may be due to the fact that these domains are indirect indicators that reflect physical health rather than an independent predictor of autonomic balance. In our study, the psychological domain includes the extent of positive and negative feelings, thinking, self-esteem, body image, and spirit. Some constructs (e.g. pessimism and hopelessness) in this domain have been associated with increased CVD incidence and CVD-specific mortality [
60,
61]. However, most studies have similarly reported a null association between the mental domain of HRQOL and CVD outcomes [
8,
56]. A possible explanation for the lack of an association is that WHOQOL-BREF is better at assessing mental health well-being, and is not designed to sensitively detect the psychological factors of cardiophysiologic significance, e.g. depression and anxiety.
Our study clearly demonstrated significant associations between psychological factors (depress and anxiety) and HRV in adjusted models (Table
3). Depression and anxiety have been related to low HRV among cardiac [
62] and psychiatric patients [
15,
16], but this relationships do not consistently exist among healthy or otherwise unselected samples [
63‐
66]. It is conceivable that any associations observed in healthy samples may be driven by a subset of individuals who have not been evaluated formally but nonetheless meet the criteria for mood or anxiety disorder. All of our participants were evaluated with a structured diagnostic interview. This type of screening ruled out current or past psychiatric disorders. Furthermore, all participants underwent relevant laboratory investigations in addition to self-reported data of physical health. Our recent studies have emphasized this objective procedure to exclude subjects with physical co-morbidities, since subjects might underestimate their biological risk factors (e.g., elevated glucose and atherogenic lipid profile) for cardiac autonomic dysregulation when these factors were self-reported [
67]. Overall, our sample was well suited for studying the relationship between HRV and the factors investigated, as the effects of potential confounding factors were minimized. We believe that the above-mentioned strengths reinforce the reliability of our results.
The implications of this study’s findings are potentially important to clinicians and researchers. HRQOL is the sum expression of diverse influencing factors and is not easy to determine. Our study provides a clinically helpful option of identifying a specific HRQOL domain of cardiophysiologic significance and clinical usefulness in cardiovascular prevention. If replicated, HRQOLphysical may serve as a valuable screening tool in the healthy adult population to identify those who are potentially at-risk for CVD. It may also help identify and/or elaborate those interventions to lower the incidence of CVD mortality in some people. Specifically, a 5-min HRV analysis can be done for individuals with a low HRQOLphysical scores to provide a rapid screening of systemic autonomic disturbance without much burden on them; then as a next step, those who are found to have low HRV may benefit from cardiovascular risk reduction strategies.
The main finding of the independent contribution of HRQOL
physical to explaining variance in HRV was mostly attributable to the influence that HRQOL
physical has on the VLF and LF component of spectral HRV (Table
3). Because the definite physiological meaning of VLF is under debate, we were unable to accurately interpret the finding of VLF component; nevertheless, our finding that individuals with low HRQOL
physical were associated with lower LF-HRV is of great importance. Some researchers argue that LF power in supine subjects principally reflects baroreflex sensitivity, which is a measure of the gain of the baroreflex [
43,
68]. The arterial baroreflex is the main mediator of HRV [
69]. Therefore, it is possible that low HRQOL
physical contributes to lower baroreflex functioning, which in turn leads to reduced HRV.
The present study has two limitations. First, causality cannot be inferred from our cross-sectional data. Second, the presented regression analyses in our study revealed several statistically significant associations but the level of correlation (or regression coefficient) was low and insofar may be not physiologically relevant. The exploratory information about possible relations provided by our results requires further confirmation. Second, we classified the healthy individuals into high or low HRQOLphysical groups according to only the quartiles of the sample HRQOLphysical scores. The optimum HRQOLphysical scores to dichotomize healthy adults and to produce the largest contribution of a prognostic predictor to the physiologically relevant reduction in HRV should be explored in future research.