Self-Rated Health and Age-Related Differences in Ambulatory Blood Pressure: The Mediating Role of Behavioral and Affective Factors

Abstract

Objective:

Despites the well-established association between SRH and health, little is known about the potential psychobiological mechanisms responsible for such links, and if these associations differ by age. The main goals of this study were to investigate the links between SRH and ambulatory blood pressure (ABP), if age moderated the risk, and the health behavior/affective mechanisms responsible for such links.

Methods:

188 men and women (94 married couples, ages 18 to 63) completed a standard measure of self-rated health and a one-day ABP assessment. Multilevel models were run to examine if SRH was associated with daily ABP and if these links were moderated by age. The Monte Carlo method was used to construct confidence intervals for mediation analyses.

Results:

Results indicated that poor SRH was associated with higher ambulatory SBP (b=3.14, SE=.68, p<.001) and DBP (b=1.34, SE=.43, p=.002) levels. Age also moderated the links between SRH and ambulatory SBP (b=.19, SE=.08, p=.01) and DBP (b=.14, SE=.05, p=.004), with links being stronger in relatively older individuals. However, only daily life negative affect significantly mediated the age X SRH interaction for both ambulatory SBP and DBP.

Conclusions:

These results highlight the potential psychobiological mechanisms linking SRH to longer-term health outcomes. Such work can inform basic theory in the area as well as intervention approaches that target such pathways.

Despite its simplicity, single-item measures of self-rated health (SRH) are robust predictors of morbidity and mortality [1,2]. In one meta-analysis, poorer SRH showed a graded association to higher mortality rates [3]. Although SRH is a partial subjective summary of one’s physical health, statistical adjustments for functional status and co-morbid diseases attenuate, but do not eliminate, the association between SRH and mortality. In fact, there is still a 2-fold increase in mortality risk that is not explained by objective health indicators [3].

Although the links between SRH and mortality are well-documented, the biological mechanisms potentially responsible for such associations are less clear. SRH is linked to composite indices of cardiovascular risk [4–6] so one might expect it to be related to biological pathways implicated in cardiovascular disease. Blood pressure is one of the strongest predictor of stroke and cardiovascular risk (besides age) and hence an important potentially modifiable pathway in terms of risk which manifests itself in everyday life [7–9]. To date, several studies have linked poorer SRH to higher clinic blood pressure levels [5,10,11]. However, clinic blood pressure levels may not represent an individual’s daily blood pressure levels as they are susceptible to white coat and masked hypertension confounds [12]. As a result, ambulatory blood pressure (ABP) levels are more strongly linked to future cardiovascular risk compared to clinic blood pressure levels [7,13]. To date, there do not appear to be any studies that have examined if ABP is a potential biological pathway responsible for links between SRH and disease outcomes. Thus, one aim of this study was to examine if poorer SRH was related to higher levels of daily life ABP.

A second aim of this study was to examine the moderation of these associations via age. Lifespan models argue for the importance of conceptualizing age-related differences and how they might influence health outcomes (e.g., Uchino, 2009). For instance, if age moderates the link between SRH and ABP it would be consistent with a process by which SRH cumulatively impacts disease risk over time. Such a finding would be consistent with the enduring self-concept view in which SRH is conceptualized as a stable perception of one’s health goals and motivation which influences diseases risk over time [15,16]. This view argues that for some individuals having good health is an important goal and thus individuals may proactively address health issues consistent with this self-view. For instance, one study found that older adults who had high levels of SRH engaged in more active coping strategies in response to age-related challenges 4 years later [17]. Although the study focused on older adults, this model implies that some of the risks associated with SRH may occur much earlier in life than previously examined. It should be noted, that moderation via age might also suggest that relatively older individuals are more sensitive to health issues or have more variability in SRH which in turn influence associations with cardiovascular risk. The lack of moderation via age, however, would suggest that SRH has an impact on disease risk regardless of the stage of the lifespan. It is also important to note that an examination of age differences can also inform at what point in life SRH starts to impact cardiovascular disease risk.

A third and final aim was to examine potential statistical mediators. There is sufficient prior work suggesting that SRH may influence future disease risk via its association to health behaviors [18–21]. For instance, the Canadian Health Measures Survey found that poorer SRH was related to lower exercise and higher alcohol consumption [19]. Less studied, but potentially equally important are affective processes. For instance, poor SRH is associated with higher depression and negative affect, as well as lower positive affect [22,23]. In addition, several studies have linked poor SRH to lower levels of self-esteem [24,25]. However, most prior work in the area has relied on global questionnaire assessments. A complementary approach would be to examine these processes as they unfold in everyday life [26,27]. Ambulatory diary assessments have several distinct advantages including multiple real-time assessments that occur as individuals are engaged in everyday life. As a result, they represent more specific affective processes and not aggregated global evaluations based on recollections and possibly selective recall [26]. SRH is likely to permeate the daily life of participants given its past links to outcomes that shape affective functioning [22,28]. It should be noted that it is likely that links between SRH and these potential mediators are bidirectional as positive and negative affect have been shown to predict changes in SRH [23]. However, to date, no studies appear to have also examined the association between SRH and daily life affective processes.

The current study thus had several aims. The first was to provide some of the first evidence that SRH might be related to ABP levels. ABP examines blood pressure levels in daily life and is a superior cardiovascular risk assessment compared to clinic blood pressure levels [7,12]. Given that SRH is related to cardiovascular mortality, it was predicted that poorer SRH would be related to higher ambulatory systolic and diastolic blood pressure. A second, major aim was to test if the links between SRH and ABP were moderated by age. Based on the enduring self-concept view, it was hypothesized that age would moderate the links with associations being stronger in older individuals. A final aim was to examine if SRH was related to ABP via its influence on health behaviors and/or daily affective functioning. It was predicted that health behaviors, daily positive affect, negative affect, and self-esteem would mediate the link between SRH and ABP.

Method

Participants

One hundred ninety-four individuals (97 married heterosexual couples) were recruited through newspaper ads and notices posted on campus and were paid $75 for a larger study on relationships processes and health [29]. This paper represents a secondary analyses of data collected in this study [29,30]. Data collection occurred between 06/2008 and 07/2009. Exclusion criteria included those who were not generally healthy or who had medical conditions with a cardiovascular component (e.g., no hypertension) or psychological problems for which they are being medically treated [31]. Most were White (83%), college educated (62%), and had an income over $40,000 per year (66%). Six individuals (3 couples) were eliminated from the study due to failure to follow protocol instructions resulting in a final total of 188 individuals (average age=29.4 years, age range 18 to 63 years, SD=8.49).

Procedure

Study procedures were approved by the institutional IRB at the University of Utah. Eligible couples arrived at the laboratory on the morning of a typical work day. Height, weight and demographic information were collected. Following informed consent, participants completed background questionnaires (i.e., demographic, medication/health questionnaires) as well a standard measure of self-rated health (see below).

Participants were fitted with the ABP monitor and given a Palm Pilot device. Participants were instructed to initiate a Palm Pilot diary reading within 5 minutes of each cuff inflation. Monitors were set to randomly obtain readings every 30 minutes from time of fitting until bedtime (approximately 10:30 pm). This random sampling procedure prevented participants from anticipating a reading and thus altering their activities. Total readings spaced across the workday and home ranged between 20 and 35. An appointment was set for the following day for participants to return the equipment and to receive compensation.

Measures

Self-Rated Health.

Participants were asked to rate their current health on a 1 to 5-point scale (1=excellent, 2=good, 3=fair, 4=poor, 5=bad). This simple measure of self-rated health has been shown to predict mortality above and beyond existing health conditions and physical limitations [3].

Health Behavior Assessment.

A standardized health questionnaire provided information on the following potential health-related variables: weekly exercise habits, use of tobacco products (no, yes), and weekly alcohol consumption. The health behavior questionnaire has been used in a large longitudinal study on the chronic stress of caregiving for a relative with Alzheimer’s Disease and its effects on physiological function [32].

Ambulatory Blood Pressure (ABP).

The Oscar 2 (Suntech Medical Instruments, Raleigh, NC) was used to estimate ambulatory SBP and DBP. The Oscar was designed specifically for ambulatory assessments and is validated to international standards of reliability [33]. Outliers associated with artifactual readings were identified using the criteria by Marler, Jacobs, Lehoczky, and Shapiro [34]. These include: (a) SBP < 70 mmHg or > 250 mmHg, (b) DBP < 45 mmHg or > 150 mmHg, and (c) SBP / DBP < [1.065 + (.00125 × DBP)] or > 3.0. Less than 3% of the ABP readings were discarded due to these criteria.

Ambulatory Diary record (ADR).

The ADR includes information on standard control variables in ABP studies including posture, activity level, talking, temperature, prior consumption of caffeine/nicotine/alcohol/meal, prior exercise, and location. It was programmed into a Palm Pilot device that allowed for easy downloading for data reduction and analyses [35]. The ADR also assessed affective processes during the day and evening including state positive affect (4-items: excited, interested, active, alert), state negative affect (4-items: sad, frustrated, stressed, upset) , and state self-esteem (6-items: confident about abilities, worried about what others think of me, satisfied with my body, feel as smart as others, concerned about impression I am making, pleased about my appearance) mostly taken from the PANAS and state self-esteem scales [36,37]. The internal consistencies of these abbreviated scales in the present study were acceptable at .63 (positive affect), .82 (negative affect), and .70 (self-esteem). The time / date stamp allowed us to verify that the ADR was completed soon after each programmed ambulatory cardiovascular reading. Readings were examined to ensure compliance and were discarded if not instigated within 5 minutes of a blood pressure reading. 286 diary items (7.15%) that were not instigated within the 5-minute period were discarded. The average participant had less than one ADR dropped (M =.78, with a range from 0 to 7).

Statistical Analysis Plan

PROC MIXED (SAS institute) was used to examine if SRH was associated with ABP levels and if these associations were moderated by age. All factors were treated as fixed and grand mean centered prior to analyses [38]. The covariance structure for the two repeated measures factors of dyad (i.e., husband, wife) and measurement occasion (i.e., reading number) was modeled using the Kronecker product [39]. The Satterthwaite approximation was used to determine the appropriate degrees of freedom [40]. Consistent with prior work, a priori statistical adjustments included age, body mass, measurement occasion, posture, activity level, temperature, alcohol, recent meals, exercise, and talking [29,41] which were simultaneously entered into the model with the main study variables. The statistical interactions between SRH and age were tested by using the main effects models above but also entering the cross-product term based on the centered main effects. Subsequent mediational analyses were conducted using the Monte Carlo method [42]. This method uses parameter estimates and standard errors from the main models which account for the dependency within couples. These data points are then used in the MCMED macro [43] to construct 95% confidence intervals for each indirect effect using 5000 resamples.

Results

Descriptive Analyses

Descriptive data for the main study variables separated by sex are detailed in Table 1. The average self-rated health was 1.73 (SD=0.66) for women and 1.89 (SD=0.68) for men. The mean ambulatory SBP and DBP were 125.5 mmHg (SD=17.4) and 76.2 mmHg (SD=13.2), respectively for women and 137.7 mmHg (SD=18.2) and 79.5 mmHg (SD=12.5) for men. Mean levels of daily life positive affect were moderate (2.06 and 2.03) while state negative affect was relatively lower (1.30 and 1.32) based on the 1 to 4-point scale. Average state self-esteem was relatively high (over 3.5) based on the 1 to 5-point scale.

Table 1.

Final sample characteristics (n=188)

Variable	Women	Men
Mean (SD)
Age (years)	28.5 (8.33)	30.5 (8.78)
BMI (kg/m2)	24.8 (5.38)	26.5 (4.85)
Self-rated health	1.73 (0.66)	1.89 (0.68)
Ambulatory SBP (mm/Hg)	125.5 (17.4)	137.7 (18.2)
Ambulatory DBP (mm/Hg)	76.2 (13.2)	79.5 (12.5)
State positive affect	2.03 (0.57)	2.06 (0.56)
State negative affect	1.30 (0.52)	1.32 (0.48)
State self-esteem	3.57 (0.55)	3.72 (0.55)
Hours of weekly exercise	3.75 (3.15)	4.83 (4.37)
Weekly alcohol consumption	1.44 (2.84)	2.39 (3.87)
Frequency
Ethnicity (% White)	81.5%	83.9%
Income Over $40,000	65.2%	65.6%
Education (some College or higher)	69.2%	56.4%
Nonsmoker	94.7%	93.6%

Main Analyses

The first aim of this paper was to examine if SRH was related to ABP levels. Consistent with predictions, poorer SRH was associated with higher levels of ambulatory SBP (b=3.14, SE=.68, p<.001) and DBP (b=1.34, SE=.43, p=.002), independent of covariates. The second aim of this study was to test if age moderated the link between SRH and ABP. As shown in Table 2, analyses revealed a statistical interaction between age and SRH for both ambulatory SBP (b=0.19, SE=.08, p=.011) and DBP (b=0.14, SE=.05, p=.005). As shown in Figure 1, plotting predicted values one SD above and below the mean for age and SRH showed that the association between SRH and ABP was stronger for the relatively older compared to the relatively younger group. Simple slope analyses of these patterns revealed that for relatively younger participants, poorer self-rated health was associated with marginally higher ambulatory SBP (p=.069) but not DBP (p=.67). However, the link between poor SRH and ABP in relatively older individuals was significant for both SBP and DBP (p’s<.001).

Table 2.

Main results for covariates and self-rated health on ambulatory systolic and diastolic blood pressure.^*

Variable	SBP			DBP
	b	SE	p	b	SE	p
Measurement number	.02	.05	.62	−.11	.03	<.001
Sex (male-female)	−9.92	.84	<.001	−1.87	.53	<.001
Posture (sitting-standing)	4.65	.49	<.001	5.40	.39	<.001
Posture (lying-standing)	5.43	.75	<.001	6.61	.59	<.001
Temp. change	2.09	.42	<.001	1.61	.35	<.001
Alcohol consumption	5.25	1.81	.004	4.77	1.42	<.001
Meal	1.57	.58	.006	1.86	.47	<.001
Activity	2.35	.37	<.001	1.03	.29	<.001
Talking	1.38	.41	<.001	1.41	.33	<.001
Age	.19	.05	<.001	.20	.03	<.001
BMI	1.00	.09	<.001	.50	.05	<.001
Self-rated health	3.25	.68	<.001	1.45	.43	<.001
Age X Self-rated health	.19	.08	.011	.14	.05	.004

^*Model analyzed using proc mixed with simultaneous variable entry.

Figure 1.

Predicted SBP (top panel) and DBP (bottom panel) one SD above and below the mean as a function of self-rated health and age.

The association between poor SRH and ABP was marginally significant or nonsignificant in the relatively younger group. Exploratory analyses examined at what age range SRH was associated with higher ambulatory SBP and DBP by starting at the lowest age group and extending the range by 5 years at a time. SRH did not predict either SBP (p=.46) or DBP (p=.43) in the 18 to 23 age range. However, for SBP, the link with poor SRH became evident for individuals between the ages of 18 and 28 years (p=.055). SRH did not predict DBP in the 18 to 28 (p=.74), 18 to 33 (p=.23), and 18–38 (p=.18) age ranges. Poor SRH started to predict higher DBP in the ages between 18 to 43 years (p=.022). These data suggest that the link between SRH and ABP becomes evident earlier for SBP compared to DBP.

A final aim was to examine potential mediators of the SRH link. In preliminary analyses, SRH was not significantly related to weekly exercise (p=.15), weekly alcohol consumption (p=.96), or tobacco use (p=.37). However, poorer SRH was associated with higher state negative affect (b=.07, SE=.02, p<.001) and lower state positive affect (b=−.08, SE=.02, p<.001). Daily life state self-esteem was also lower for individuals with poor SRH (b=−.18, SE=.03, p<.001). Mediational analyses focused on the age X SRH interaction given it was the highest order term which was significant. These analyses showed that neither state positive affect or state self-esteem mediated the age X SRH interaction on ambulatory blood pressure. More specifically, the indirect (mediated) association was not significant for SBP (positive affect: aXb=−0.003, 95% CI [−0.010, 0.001], self-esteem: aXb=−0.0001, 95% CI [−0.009, 0.009] or DBP (positive affect: aXb=−0.002, 95% CI [−0.006, 0.001], self-esteem: aXb=−0.000, 95% CI [−0.008, 0.009]). However, as shown in Figure 2, negative affect did statistically mediate the age X SRH association with ambulatory blood pressure as the indirect association was significant for both SBP (aXb=0.011, 95% CI [0.003, 0.023]) and DBP (aXb=0.006, 95% CI [0.001, 0.013]).

Figure 2.

Mediational model with SRH X Age on SBP (top panel) and DBP (bottom panel) via negative affect.

Discussion

The mains goals of the current study were to examine (a) the link between SRH and ABP, (b) if age moderated the risk, and (c) potential mediators. No studies to date appeared to have tested these main aims and consistent with predictions poorer SRH was related to higher ambulatory SBP and DBP. Furthermore, these associations were stronger in older individuals, but were still evident in relatively younger individuals as well. The age X SRH association was also statistically mediated by negative affect during daily life. These results lend insight into the potential biological pathways linking SRH to greater cardiovascular risk and mortality, as well as possible mechanisms responsible for such links [3,28,44].

Very little work has examined the biological correlates that could link SRH to significant health outcomes. Given the association between SRH and cardiovascular risk, it was predicted that SRH would be related to ABP. Prior studies examining blood pressure have focused on clinic blood pressure readings [10,11]. Daily ABP measures, however, provide a superior assessment of blood pressure that is less influenced by white coat and masked hypertension issues [7,12]. Importantly, ABP appears to be a more sensitive and independent predictor of future cardiovascular risk compared to clinic blood pressure readings [13]. In fact, elevated blood pressure is one of the strongest modifiable predictors of future stroke and cardiovascular risk [7,8]. The emergence of SRH as a predictor of SBP at younger ages in this study is also important as ambulatory SBP tends to have good prognostic value for cardiovascular risk [45,46].

Results also indicated that the link between SRH and ABP became stronger at relatively older ages. Exploratory analyses also showed that SRH was associated with ABP from early adulthood (SBP) to later adulthood (DBP). Although limited by the cross-sectional design and exploratory nature of these analyses, these findings are consistent with the enduring self-concept view that seeks to examine why SRH predicts morbidity and mortality [15]. This perspective views SRH as a stable construct and implies that the link between SRH and health might be evident earlier then has been previously studied. For instance, individuals with better perceptions of SRH may proactively manage their psychosocial and behavioral risk to facilitate healthy outcomes (e.g., directly coping with health threats, Menec et al., 1999). Thus, over time these patterns accrue which leaves individuals protected or vulnerable to age-associated health declines. Longitudinal studies will thus be needed to test this model. Although this perspective suggests a developmental pathway for such links, the cross-sectional design limits inferences as it is also possible that older individuals are more sensitive to cues and/or risks associated with poor SRH [47]. Inconsistent with the first possibility, age is typically associated with decreases in interoceptive ability and accuracy [48]. It is also possible that age is simply associated with greater variability in SRH which enhances the ability to detect associations. Ancillary analyses, however, showed that SRH and age were not correlated in the current study (r=−.01 in men and r=−.04 in women, p’s>.74) and splitting the sample at the median for age suggest slightly greater variability for SRH in the relatively younger group (<28 years old, SD=0.73) compared to the relatively older group (SD=0.61).

Although there is evidence linking poor SRH to negative health behaviors [19,49], weekly exercise, alcohol consumption, and tobacco use were not related to SRH after adjusting for the covariates and did not mediate the SRH X age interaction on ambulatory blood pressure. It is important to note that trends were in the expected direction (e.g., poorer SRH associated with lower weekly exercise). Thus, it is possible that the sample size in the current study was too small to detect the associated effect size given that prior research has tended to use large, national databases [19,49]. Although a reasonable estimate, the health behavior questionnaire used in the current study is limited (i.e., single-items self-report) although the accuracy of such reports depends on the timeframe and type of health behavior assessed [50,51]. Future research using larger samples and/or different health behavior assessments (e.g., independent reports, objective indices) will be needed to test these possibilities [52].

Daily negative affect, but not daily positive affect or self-esteem, emerged as a significant mediator of the age X SRH association for both SBP and DBP. This finding is consistent with the considerable literature linking negative affective processes to ambulatory biological risk assessments [41,53–55]. In a recent study, it was found that SRH was related to increased inflammation but depression as measured by the CES-D was not a mediator of such links [28]. It is possible that the increased sensitivity of the current study might be due to the ambulatory protocol which allowed for multiple assessments as they occur in daily life. Nevertheless, future work will be required to examine the specificity of such findings and if daily life positive affect and self-esteem might be mediators. However, the lack of mediation via positive affect is at least consistent with studies showing that negative affect/interactions are more closely related to changes in physical symptoms and cardiovascular function [56,57]. More generally, work will also be needed on the specific psychological and behavioral mechanisms that can explain these associations given that most of the work in this area has focused on antecedent processes such as actual physical health conditions and/or cognitive processes that give rise to self-rated health above and beyond objective indicators of health (e.g., symptom perception, [2,3]). Longitudinal studies will also be important to model the likely bidirectional association between SRH and affect [23].

These findings also suggest multiple entry points for intervention. Blood pressure is one of the strongest modifiable risk factors for both stroke and cardiovascular disease [8]. If blood pressure is a pathway ultimately linking SRH to health then there are lifestyle interventions that positively influence its management [58]. In addition, laboratory studies show that manipulations that impact negative moods can influence both somatic symptoms and cardiovascular reactions [59,60]. These studies suggest that cognitive behavioral stress management may be helpful as an intervention given it appears related to better cardiovascular outcomes [61]. Finally, these findings suggest the importance of primary prevention efforts aimed at relatively healthy individuals [62]. According to the enduring self-concept view individuals with better perceptions of health may proactive engage in strategies that improve their health [17]. Results from the present study suggest that health-relevant differences in SRH might be seen in early to mid-adulthood. Thus, early interventions aimed at improving blood pressure regulation via multiple entry points as noted above might help delay the onset of health problems.

There are also important limitations of the current work. First, the results of this study were consistent with mediation. However, it is limited by the cross-sectional design so longitudinal studies will be needed for strong inference and to model more complex links [63,64]. This study also included relatively healthy individuals so generalizations to clinical populations are limited. The sample size is also not large and precluded more complicated analyses such as segmented regression analyses which might pinpoint the more precise age in which SRH predicted ABP and how it might be impacted by different moderators. These limitations notwithstanding, this is one of the first papers demonstrating SRH is related to ABP and such an association emerges relatively early in relatively young, healthy adults. It also is consistent with the view that negative affect is an important mediator of such links. This research thus highlights the importance of future theoretical work focusing on potential biological pathways and how psychosocial factors might influence such risks.

Abbreviations:

SRH

self-rated health

BMI

body mass index

ABP

ambulatory blood pressure

SBP

systolic blood pressure

DBP

diastolic blood pressure

ADR

ambulatory diary record