Content uploaded by Bryce Hruska
Author content
All content in this area was uploaded by Bryce Hruska on Aug 07, 2020
Content may be subject to copyright.
Running head: AMBULATORY HEART RATE & VACATIONING 1
This is an original preprint of an article published by Taylor & Francis in Psychology & Health
on November 6, 2019, available online: https://doi.org/10.1080/08870446.2019.1687699.
Do Vacations Alter the Connection Between Stress and Cardiovascular Activity?
The Effects of a Planned Vacation on the Relationship Between
Weekly Stress and Ambulatory Heart Rate
Bryce Hruska, Ph.D., Sarah D. Pressman, Ph.D.,
Kestutis Bendinskas, Ph.D., Brooks B. Gump, MPH Ph.D.
Author Note
Bryce Hruska, Department of Public Health, Food Studies and Nutrition, Syracuse
University; Sarah D. Pressman, Department of Psychological Science, University of California,
Irvine; Kestutis Bendinskas, Department of Chemistry, State University of New York at
Oswego; Brooks B. Gump, Department of Public Health, Food Studies and Nutrition, Syracuse
University.
Funding Details: This work was supported by Project: Time Off under funding
agreement 28570.
Corresponding author: Brooks B. Gump, PhD, MPH, 344H White Hall
Department of Public Health, Food Studies and Nutrition, Syracuse University, Syracuse, NY
13244, Ph: 315-443-2208, Fax: 315-342-2046, Email: bbgump@syr.edu
Words: 5,033
Tables: 2
Figures: 3
Running head: AMBULATORY HEART RATE & VACATIONING 2
Abstract
Objective: To examine how the association between psychological stress and ambulatory heart
rate varies in the weeks before and after a planned vacation. We hypothesized that the impact of
stress on heart rate would weaken in the weeks leading up to the vacation and return to normal
levels in the weeks following the vacation.
Method: Fifty-four workers eligible for paid vacation time were recruited; stress ratings
obtained via weekly surveys and ambulatory heart rate readings obtained via a wrist-worn
consumer device were collected before and after the vacation.
Results: A statistically significant interaction was observed between weekly stress and the time
period leading up to the vacation on ambulatory heart rate (b = -0.51, SE = 0.21, 95% CI = -0.91,
-0.10, p = 0.01). A plot of predicted values demonstrated that the relationship between weekly
stress and heart rate was stronger when the vacation was further away in the future and imparted
less of an effect as the vacation approached.
Conclusions: Vacations may have physical health benefits that extend beyond the vacation
experience by reducing the association between stress and ambulatory heart rate in the weeks
leading up to a planned vacation.
Keywords: heart rate; psychological stress; recreation
Running head: AMBULATORY HEART RATE & VACATIONING 3
Acronyms
CVD = cardiovascular disease
ICC = intraclass correlation coefficient
AIC = Akaike information criterion
BIC = Bayesian information criterion
TWH = Total Worker Health
Running head: AMBULATORY HEART RATE & VACATIONING 4
Do Vacations Alter the Connection Between Stress and Cardiovascular Activity?
An Examination of the Effects of a Planned Vacation on the Relationship
Between Weekly Stress and Ambulatory Heart Rate
The far reaching effects of chronic psychological stress on multiple facets of mental and
physical health are well documented (1). The association between psychological stress and
cardiovascular disease (CVD) – the leading cause of death for both men and women in the
United States – is particularly notable (2). Large scale meta-analyses of individual participant
data indicate that people exposed to job strain have 1.2-1.5 times the risk of developing
cardiovascular disease (3). When psychological stress is defined more broadly, large-scale case-
control studies suggest that nearly one-third of myocardial infarctions may be attributable to the
effects of stress, making it comparable to other conventional risk factors including smoking,
diabetes, hypertension, and abdominal obesity (4).
The association between psychological stress and CVD has been hypothesized to occur
due to the “wear and tear” incurred from repeated cardiovascular activation (i.e., increases in
heart rate and blood pressure) in response to stressor exposure (5). Laboratory-based research
demonstrates that psychological stress induced under controlled conditions is reliably associated
with increases in heart rate and blood pressure that are predictive of future cardiovascular health
status (6). For example, heart rate activity in response to perceived threat has been shown to
predict subsequent resting heart rate 10-15 years later (7). Furthermore, in the large-scale British
Cohort study, heart rate activity in response to psychological stress at age 10 was found to
predict a heightened risk for hypertension in adulthood (8).
Additional evidence comes from research utilizing ecological momentary assessment, in
which the circumstances surrounding a stressor and associated CVD are measured under real life
Running head: AMBULATORY HEART RATE & VACATIONING 5
conditions, in close proximity to their actual occurrence. For example, in a recent meta-analysis
work stress was shown to be associated with elevated ambulatory blood pressure both at work
and at home during non-work hours (9). Further, more comprehensive assessments, not limited
to the workplace environment, have shown that elevations in stress across the day are associated
with concomitant elevations in ambulatory heart rate and blood pressure, which are in turn
associated with subsequent increases in subclinical markers of CVD (e.g., carotid-intima media
thickness) (10,11).
Given the evidence connecting psychological stress with CVD incidence and associated
disease processes, a growing literature is considering mitigating factors that may offset the health
risks imposed by chronic psychological stress. According to Geurts and Sonnentag (2006),
recovery opportunities occurring throughout the day may provide an occasion for the
physiological responses associated with psychological stress to return to resting levels. Indeed,
workers reporting relaxing experiences when away from work also report lower levels of distress
and physical complaints (13,14). These effects extend beyond the immediate day on which they
are experienced: Relaxation that occurs during the weekend is associated with increases in well-
being and decreases in negative affective experiences reported at the end of the subsequent week,
suggesting that the beneficial effects of recovery opportunities extend beyond the termination of
the experience (15).
Given that they are typically longer, formal vacations from work may represent a
recovery opportunity capable of stronger effects than the relatively short recovery periods
occurring during and after the work day (16). Indeed, secondary analyses of large-scale cohort
studies have demonstrated a dose-response relationship between the amount of vacationing a
person engages in and risk for myocardial infarction, cardiovascular-specific mortality, and all-
Running head: AMBULATORY HEART RATE & VACATIONING 6
cause mortality (17,18). Furthermore, stress and burnout among vacationers have been shown to
decrease following a period of time off both within vacationers as well as relative to matched
non-vacationers (19,20). This effect on stress is associated with corresponding enhancements in
subjective health: A past meta-analysis indicates that upon returning to work, vacationers report
improvements in their physical and emotional health that last up to one month following their
vacation and that include reduced exhaustion and health complaints as well as increases in life
satisfaction (21).
In addition to the effect that vacations have on health in the period following vacationing,
they may also have an effect on physiological activity in the period preceding the vacation. For
example, the anticipation of an approaching vacation may alter the interpretation of stressful
work conditions, leading to a reduction in associated cardiovascular activity as the vacation
nears. Indeed, research shows that reframing physiological arousal in response to an upcoming
public performance task as excitement rather than anxiety can improve performance on the task
as well as decrease associated cardiovascular activity (22,23). Additionally, in anticipation of
their occurrence, experiential purchases like vacations (relative to material purchases) have been
shown to increase positive affect and subjective well-being even before they occur (24).
Collectively, this research suggests that vacations may have implications for cardiovascular
health prior to their occurrence. However, existing vacation research has not adequately tested
the potential health effects that vacations may confer prior to vacationing due to either
intentionally excluding assessments during this period of time (25) or by failing to include
objective physiological measures that may better capture vacation health effects (20). Indeed,
noting these limitations, de Bloom et al. (2009) (21) recommended that to advance the field of
vacation research, both physiological measures of stress and repeated measurements occurring
Running head: AMBULATORY HEART RATE & VACATIONING 7
before and after a vacation are necessary to capture any within-person changes that spillover into
the time period surrounding the vacation episode. To date, no study in the field of vacation
research has incorporated both of these recommendations.
The objective of the current study is to extend the vacation literature by examining the
relationship between self-reported psychological stress – as indicated in a weekly-administered
survey – and cardiovascular activity – as indexed by ambulatory heart rate – in the weeks before
and after a planned vacation. While the literature suggests that both heart rate and blood pressure
are related to cardiovascular health, we elected to focus on heart rate given that this study is the
first of its kind and that ambulatory heart rate sensors are relatively unobtrusive compared to
ambulatory blood pressure monitors. This is an especially important consideration given the high
impact and water-based activities that might occur on a vacation as well as participants’ desire to
not have an obvious medical device on in vacation photographs. Thus, we anticipated that
compliance would be greater with a heart rate sensor.
We hypothesized that the strength of the relationship between psychological stress and
ambulatory heart rate would vary depending upon the proximity of the vacation. Specifically, we
predicted that:
(1) In the period preceding the vacation, the relationship between self-reported stress and
ambulatory heart rate would attenuate as the vacation approached.
(2) In the period following the vacation, the relationship between self-reported stress and
ambulatory heath rate would re-establish as the vacation became a more distant event
in the past.
Method
Participants
Running head: AMBULATORY HEART RATE & VACATIONING 8
Participants were recruited from the community between 2015-2016 using flyers, the
internet, and media advertisements. Inclusionary criteria included (a) being at least 18 years of
age, (b) being employed full-time, (c) being eligible for paid time off intended for vacationing,
(d) being able to read and understand English, (e) having access to the internet, (f) having a
working email address, and (g) having a vacation planned using at least three paid days that was
at least one month away. Individuals with vacations planned over Thanksgiving or Christmas
were excluded as these particular vacations represent unique experiences with likely mixed
benefits and costs (26). University faculty and school teachers were excluded from participation
given that their vacation behavior varies cyclically with the academic schedule of their students
and thus is not representative of the larger population of working adults. In addition, participants
had to be free of adrenal gland (e.g., Cushing’s disease, Addison’s disease), pituitary gland (e.g.,
tumors or cancers of the pituitary gland, acromegaly), and inflammatory/auto-inflammatory
disorders (e.g., rheumatoid arthritis, lupus). Finally, participants were excluded if they used any
medications having long-term effects on the hypothalamic pituitary adrenal axis (e.g., systemic
steroids, anticonvulsants, opioid agonists). In total, 211 individuals expressed interest in the
study. Of these, 67 met inclusionary criteria. The most common reason for ineligibility was
having a vacation planned that was not at least one month away. Of the 67 meeting all
inclusionary criteria, 63 (94.0%) were enrolled and participated in data collection. The four
individuals not included in the final sample were enrolled but declined participation prior to data
collection due to a lack of time to commit to the study.
For the current analyses, 4 participants were removed because it was discovered after
study completion that they took a non-qualifying vacation (e.g., < 3 paid days), 3 participants
were removed because they worked overnight shifts during the study, which is known to
Running head: AMBULATORY HEART RATE & VACATIONING 9
influence cardiovascular activity (27), and 2 were removed due to the presence of preexisting
cardiovascular disease. Thus, a final sample size of 54 was used in all analyses reported (see
Table 1 for participant characteristics).
Procedure
All study procedures were approved by the Human Subjects Review Board of Syracuse
University. The procedures described here were part of a larger observational study examining
the effects of past vacationing behavior on current health outcomes, as well as the effects of an
upcoming vacation on acute health outcomes before and after the vacation (28). The analyses
reported in this manuscript are based upon the data examining the effects of an upcoming
vacation.
Participants visited the study’s laboratory to provide written informed consent, to
complete a battery of assessments, and to receive training on the maintenance and operation of
the ambulatory heart rate devices. The duration of time during which ambulatory heart rate data
were collected was the same for each participant; however, the initiation and termination of data
collection in relation to the vacation varied according to assignment to one of four groups: for
group 1 collection began 4 weeks pre-vacation and ended 1 week post-vacation (n = 11); for
group 2 collection began 3 weeks pre-vacation and ended 2 weeks post-vacation (n = 12); for
group 3 collection began 2 weeks pre-vacation and ended 3 weeks post-vacation (n = 16); for
group 4 collection began 1 week pre-vacation and ended 4 weeks post-vacation (n = 15). This
design allowed us to collect data on an extended period of time both before and after the
vacation, while decreasing participant burden by limiting the active data collection window for
each participant.
Running head: AMBULATORY HEART RATE & VACATIONING 10
A 4 week collection period after the vacation was selected on the basis of past research
demonstrating that self-reported vacation health benefits remain for up to one month following
the return to work (21). Given the absence of prior research examining health effects prior to
vacationing, we also adopted a 4 week assessment period before the vacation episode; in effect,
standardizing the sampling period around the vacation episode.
Group status was determined prior to the study’s initiation using pseudo-random
assignment via a computer algorithm (i.e., group status was assigned to each participant
number). Upon initiation of study recruitment, deviation from this pre-determined assignment
occurred in 6 instances (4 instances being due to the four individuals who initially agreed to
participate in the study, but subsequently declined prior to data collection [see Participants
section for more details] and 2 instances due to participant schedule conflicts).
Each week leading up to the vacation, immediately before the vacation, and each week
after the vacation, participants received an email containing a link to a survey administered
through the Qualtrics survey platform (29). This weekly survey was used to assess psychological
stress experienced in the preceding week. Participants also completed a survey (at one time only)
that inquired about demographic information and their current health status.
Measures
Ambulatory heart rate. Heart rate was recorded using a wrist-worn consumer device
equipped with an optical heart rate sensor capable of displaying current heart rate readings in
real time (a Fitbit Charge HR) (30). Consistent with guidelines, participants were instructed to
wear their devices at least one finger width above their wrist bone. Research suggests that
wearable consumer devices may be valid and useful instruments for collecting and monitoring
health information in a research context (31). Participants’ second-by-second heart rate data
Running head: AMBULATORY HEART RATE & VACATIONING 11
were collected via the device and synced with a research account created for the purposes of the
study. Data were extracted from the research account using Fitabase (32). Heart rate data were
restricted to the period of time during the day when participants were awake as indicated by the
movement detection algorithms of the device and were aggregated over the previous 7 days to
ensure correspondence with the period of time being assessed by the weekly surveys.
Weekly stress. Consistent with research performed by Miller and colleagues (33),
psychological stress was assessed using two items that were adapted from the Perceived Stress
Scale (34). One item asked participants to rate how “stressed” they had felt in the preceding
week and one item asked how “overwhelmed” they had felt during the preceding week. Both
items were rated on a scale ranging from 0 (“Not at all”) to 4 (“Extremely”). These items were
averaged together to yield a single stress rating for each weekly survey. Cronbach’s alpha ranged
from 0.75 to 0.91 across the weekly administrations.
Covariates. Demographic characteristics (i.e., age, sex), preexisting disease status (e.g.,
presence of hypertension, heart disease, or diabetes), and blood pressure medication status were
assessed by survey while physical activity was assessed using total weekly step counts as
collected by the wrist-worn consumer device. Overall, 18.5% of the sample reported taking
blood pressure medication. Hypertension (20.4%) and diabetes (3.6%) status were combined to
create a composite item reflecting the presence of these diseases (0 = hypertension or diabetes
not present; 1 = hypertension or diabetes present). Participants’ daily total step count was
collected via the device and synced with the research account created for the study. Similar to the
ambulatory heart rate data, step data were extracted from the research account using Fitabase
(Small Steps Labs, LLC) and aggregated over the previous 7 days to ensure correspondence with
the period of time being assessed by the weekly surveys. We selected these covariates
Running head: AMBULATORY HEART RATE & VACATIONING 12
(demographics, preexisting disease status, blood pressure medication status, and physical
activity) based upon their known associations with cardiovascular activity (35–41).
Given that participants’ might alter their behavior in response to being monitored (42), in
addition to these covariates, supplemental analyses included two variables assessing the extent to
which reactivity to wearing the heart rate sensor might have impacted the results. These variables
consisted of two survey questions presented at the end of the study that asked participants to rate
how often they monitored their health using the device during the course of the study (0 = “Not
at all”, 4 = “Every day during the week”) and to what degree they perceived that their health
behaviors had been influenced by wearing the device (0 = “Not at all”, 4 = “A lot”).
Data Analysis
Data analyses were conducted using Stata IC 13 (43). Given the multilevel nature of the
data, with weekly measurements at the level 1 unit of analysis and nested within participants at
the level 2 unit of analysis, a multilevel piecewise regression model using restricted maximum
likelihood estimation was used to examine the relationship between weekly stress and heart rate
as the vacation approached and after it was over. Piecewise regression is an analytic technique
that allows for the representation of different periods of time – and thus different relationships
between predictors and outcomes at different periods of time – within the same statistical model
(44). We initially tested a random intercept-only model to determine the between- and within-
person variability on the outcome as represented by the intraclass correlation coefficient (ICC),
which reflects the proportion of variance in the outcome attributable to between-person
differences. Next, we examined a full model consisting of the time variables, weekly stress, time
by stress interactions, and covariates. Finally, in addition to this main model, we also conducted
Running head: AMBULATORY HEART RATE & VACATIONING 13
a supplemental analysis that included two variables controlling for participant reactivity to
wearing the heart rate sensor.
In the current study, time was anchored to the date that participants’ completed each
weekly survey and was represented by two continuous variables: pre-vacation – representing the
number of weeks before the vacation took place – and post-vacation – representing the number
of weeks after the vacation took place. Thus, the model intercept represents the predicted weekly
heart rate occurring during the vacation, conditional on all other predictors and covariates present
in the model.
Continuous level 1 predictors (i.e., weekly stress and physical activity) were person-mean
centered and thus reflect the within-person relationship between each predictor and ambulatory
heart rate. Given evidence that a goal of 10,000 steps per day provides a favorable level of
physical activity for promoting health (45), each participant’s total weekly steps were divided by
10,000 in order to facilitate interpretation of the parameter estimate associated with this
predictor. Grand mean centering was used for continuous level 2 predictors (i.e., age), and
unweighted effects coding was used for dichotomous level 2 predictors (i.e., gender, preexisting
disease status, and blood pressure medication status). Given the nested structure of the data, with
repeated measurements taken at different points in time taken from the same participants, we
tested compound symmetry, autoregressive, and Toeplitz error structures to determine which
best represented the data.
Results
Preliminary Analyses
Overall, missing data on both the weekly surveys and the ambulatory measures were
minimal. Participants completed 100% (n = 263) of the weekly surveys delivered during the
Running head: AMBULATORY HEART RATE & VACATIONING 14
course of the study. Of the 263 weekly surveys, 248 (94.3%) surveys had complete ambulatory
data for all 7 days comprising the previous week. Of the remaining surveys in the analysis, 9
(3.4%) had data for 6/7 days; 2 (0.7%) had data for 5/7 days; 1 (0.4%) had data for 4/7 days; 1
(0.4%) had data for 3/7 days; 1 (0.4%) had data for 2/7 days; and 1 (0.4%) was missing
ambulatory data for the entire week and was not included in subsequent analyses. Thus, 262
weekly surveys with complete or partial ambulatory heart rate records were included in all
analyses reported. Participants with partial ambulatory heart rate data did not differ from
participants with complete ambulatory data in terms of sex (% female: 75.0 vs. 68.2, p = 0.75,
Fisher’s exact test), race (% Caucasian: 100.0 vs. 94.5, p = 1.00, Fisher’s exact test), age (t [26.7]
= 1.31, p = 0.20, M = 41.26 vs. M = 45.30), family income (% ≥ $75,000/year: 62.5 vs. 76.3, p =
0.33, Fisher’s exact test), or industry (% healthcare/higher education: 43.7 vs. 50.0, p = 0.77,
Fisher’s exact test).
On average, participants used 5.88 days (SD = 2.46) of paid leave for their vacation. The
average stress rating across all the weekly surveys was 1.48 (SD = 0.85; possible range = 0-4;
reported range = 0.1-3.5). Collapsed across the weeks preceding the vacation, the average stress
rating was 1.68 (SD = 0.98), while collapsed across the weeks following the vacation, the
average stress rating was 1.27 (SD = 0.87), a difference that was statistically significant (t [48] =
3.79, SE = 0.11, p < 0.001) (see Figure 1). Across the entire sampling period, the average
ambulatory heart rate was 80.91 (SD = 8.31), while it was 81.46 (SD = 8.40) in the weeks
preceding the vacation and 80.57 (SD = 8.61) on average in the weeks following the vacation (t
[48] = 1.90, SE = 0.47, p = 0.06). Finally, across the sampling period, the average total weekly
step count was 54,590.29 (SD = 18,461.20); it was 48,735.11 (SD = 21,693.08) in the weeks
before the vacation and 56,843.81 (SD = 21,744.19) in the weeks following the vacation, a
Running head: AMBULATORY HEART RATE & VACATIONING 15
difference that was statistically significant (t [48] = -3.03, SE = 2680.10, p = 0.004) (See Figure
2).
[Figures 1 and 2 near here]
Multilevel Piecewise Regression
The initial random-intercept only model indicated that the ICC was 0.90; thus, 90% of the
variance in weekly ambulatory heart rate was attributable to between-person differences and
10% was attributable to within-person differences. A first-order autoregressive error structure
was found to most parsimoniously represent the model residuals as indicated by both the Akaike
information criterion (AIC) and Bayesian information criterion (BIC) (compound symmetry:
AIC = 1479.95, BIC = 1494.23; Toeplitz: AIC = 1474.67, BIC = 1492.51; autoregressive: AIC =
1473.10, BIC = 1487.38).
[Table 1 near here]
Results from the full model are presented in Table 1. After controlling for demographics,
preexisting disease, blood pressure medications, and weekly physical activity, a statistically
significant interaction between weekly stress and the time leading up to the vacation on
ambulatory heart rate was observed (p = 0.018). In other words, the within-person relationship
between weekly stress and ambulatory heart rate varied depending upon proximity to the
upcoming vacation. To better understand the nature of this interaction, a plot of predicted values
was constructed in which ambulatory heart rate was plotted against low, medium, and high
values of weekly stress with time as the moderator. The figure revealed a positive relationship
between weekly stress and ambulatory heart rate that grew weaker as the vacation approached
(see Figure 3). In contrast, a statistically significant interaction was not detected between weekly
stress and the time after the vacation ended (p = 0.75).
Running head: AMBULATORY HEART RATE & VACATIONING 16
[Figure 3 near here]
Supplemental Analysis
When controlling for the participant’s response to wearing the heart rate sensor, the
interaction between time leading up to the vacation and weekly stress on ambulatory heart rate
remained statistically significant (b = -0.60, SE = 0.26, 95% CI = -1.11, -0.08, p = 0.023), while
the interaction between time after the vacation and weekly stress remained statistically non-
significant (b = 0.10, SE = 0.48, 95% CI = -0.84, 1.05, p = 0.83).
Discussion
Psychological stress is consistently associated with negative physical health outcomes.
As a potential source of restoration, vacations may have health effects that extend beyond the
vacation episode by protecting against the impact that psychological stress has on cardiovascular
activity. The current study examined how the relationship between self-reported weekly stress
and ambulatory heart rate differed before and after a period of planned vacation. Results revealed
an interaction between stress and the time period leading up to the vacation. More specifically, as
the vacation approached, the relationship between psychological stress and heart rate weakened.
On the other hand, an interaction was not detected between stress and the time period after the
vacation.
The observed interaction between psychological stress and time leading up to the
vacation on ambulatory heart rate may reflect an anticipatory effect of the upcoming vacation on
stressor appraisals. In other words, as the vacation approaches, stressors that are encountered
may be interpreted as less threatening or demanding due to the anticipation of an extended leave
from work. Indeed, cognitive appraisals of stressful events are important determinants of
cardiovascular responses (46,47). Furthermore, altering the appraisals ascribed to stressors has
Running head: AMBULATORY HEART RATE & VACATIONING 17
been shown to improve cardiovascular stress reactivity, demonstrating the central role played by
the appraising process in shaping physiological stress responses (48). The finding from the
current study is also consistent with stress buffering hypotheses that suggest that positive factors
such as social support or positive emotions can reduce the negative downstream effects that
stress can impart on physical health (49,50). Applied to the current study, a stress buffering
account would suggest that vacation anticipation may buffer the relationship between self-
reported stress and cardiovascular activity by altering the appraisal of events that are typically
perceived as challenging or demanding.
While it is possible that workers encountered less stressors in the weeks leading up to the
vacation, this is unlikely given that concerns about increased workload are often cited as
vacation deterrents (51). In addition, stress ratings reported before the vacation were statistically
significantly greater than the ratings reported after the vacation, further arguing against this
possibility. The absence of a statistically significant interaction between stress and time after the
vacation may have been due to this vacation effect on stress levels upon returning. While this
finding is consistent with prior research suggesting that vacationing can reduce psychological
stress in the weeks following a vacation – as well as suggest that the stress buffering effects of a
vacation may last longer than the assessment period of the study – it also likely constrained our
ability to detect an interaction. Thus, assessment periods that extend beyond four weeks or a
finer grained stress measure inquiring about stressful experiences at a lower level of analysis –
such as the day level – may be necessary to detect any effects present in the period following
vacationing.
The findings from the present study have implications for workplace interventions
targeting worker health. In 2011 the National Institute for Occupational Safety and Health
Running head: AMBULATORY HEART RATE & VACATIONING 18
implemented its Total Worker Health (TWH) initiative. TWH represents a holistic approach to
reducing work-related health hazards and promoting the well-being of workers. While still
relatively small in number, interventions utilizing this approach are demonstrating the
effectiveness of TWH in promoting worker health outcomes (52). Results from the current study
suggest that encouraging vacationing among workers may reduce cardiovascular risks and serve
as an important component of these interventions. Since paid vacation time is already available
to over three-quarters of American workers, encouraging workers to utilize this benefit may be a
relatively easy strategy to implement to reduce cardiovascular health risks (53), especially given
other studies showing ties between vacation behavior and heart disease mortality (18).
Furthermore, given the important role played by supervisors in work culture formation, managers
may be especially important figures who can ensure that their employees use all paid vacation
time during the calendar year (54,55).
Although requiring further examination in future research, the finding that participants
engaged in more physical activity following the vacation (as indexed by their step count) suggest
that vacations may have an effect on health behaviors related to cardiovascular risk. If replicated,
this finding suggests that in addition to impacting stress levels, vacations may confer beneficial
health effects by improving workers’ physical activity levels.
A number of limitations should be noted for the findings observed in the present study.
First, the study utilized a weekly stress measure that was reported retrospectively and aggregated
over the previous week. A measure using a finer grained level of analysis (e.g., day level) would
have allowed us to reduce the amount of time between when stress was experienced and
reported, while also providing greater power to detect smaller within-person effects that may
have been operating. This is an especially important consideration given that only 10% of the
Running head: AMBULATORY HEART RATE & VACATIONING 19
variance in ambulatory heart rate was attributable to within-person differences in the present
study. Thus, this level of within-person variance provides support for the idea that examination at
a lower level of analysis may provide the opportunity to detect the effect that vacations have on
the relationship between stress and cardiovascular activity in the period after the vacation ends.
On the other hand, this relatively small amount of within-person variance makes our findings
more notable because it suggests that the effect of vacationing still had a detectable and
statistically significant impact in the weeks preceding the vacation even though there was not a
large amount of within-person variability present.
Third, the addition of a comparison group of non-vacationers with survey and ambulatory
measures taken during a parallel period of time would strengthen the observational design
employed here by allowing for the examination of between-person differences on the
relationships reported in the current study. In addition, the sample size used was relatively small
and predominantly Caucasian, which may constrain the study’s generalizations given the
elevated cardiovascular disease risk among African Americans (56). Thus, future research
employing more complex research designs in more diverse samples is needed in order to
strengthen the findings from present study.
Finally, the current study used a consumer device (a Fitbit Charge HR) to assess heart
rate. While research examining such devices suggests that they may be useful, economical
choices for heart rate monitoring (30), findings from the current study would be strengthened if
replicated in future studies using devices with well established performance specifications that
assess both heart rate and blood pressure.
Despite these limitations, the current study is the first to document a relationship between
stress and cardiovascular activity that varies as a function of a planned vacation through the use
Running head: AMBULATORY HEART RATE & VACATIONING 20
of repeated assessments before and after the vacation and the incorporation of objective,
physiological measures. Furthermore, it demonstrates that vacations may have physiological
effects that have health consequences that extend beyond the vacation experience, making them a
potentially important workplace factor capable of mitigating the negative effects of workplace
stress on cardiovascular health.
Running head: AMBULATORY HEART RATE & VACATIONING 21
Acknowledgements: The authors thank Alexandrah Gichingiri, Ivan Castro, Rachel Zajdel,
Aylonna Whitney, Samantha Henderson, Ian Thompson, Tatiana Gregory, Tessila Abbott,
Barbara Samson, and Jessica Fleming for their assistance in ensuring the successful completion
of the project.
Running head: AMBULATORY HEART RATE & VACATIONING 22
Disclosure Statement: As part of the funding agreement, Project: Time Off retained no legal
authority regarding data analysis, manuscript preparation, or publication decisions.
Running head: AMBULATORY HEART RATE & VACATIONING 23
References
1. Schneiderman N, Ironson G, Siegel SD. Stress and Health: Psychological, Behavioral, and
Biological Determinants. Annu Rev Clin Psychol. 2005;1(1):607–28.
2. Centers for Disease Control and Prevention. Heart Disease Facts [Internet]. 2017 [cited
2018 Nov 5]. Available from: https://www.cdc.gov/heartdisease/facts.htm
3. Kivimaki M, Batty GD, Ferrie JE, Kawachi I. Cumulative meta-analysis of job strain and
CHD. Epidemiology. 2014;25(3):464–5.
4. Yusuf PS, Hawken S, Ôunpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially
modifiable risk factors associated with myocardial infarction in 52 countries (the
INTERHEART study): Case-control study. Lancet. 2004;364(9438):937–52.
5. Seeman TE. Price of Adaptation—Allostatic Load and Its Health Consequences. Arch
Intern Med. 1997;157(19):2259.
6. Chida Y, Steptoe A. Greater cardiovascular responses to laboratory mental stress are
associated with poor subsequent cardiovascular risk status: A meta-analysis of prospective
evidence. Hypertension. 2010;55(4):1026–32.
7. Light KC, Dolan CA, Davis MR, Sherwood A. Cardiovascular responses to an active
coping challenge as predictors of blood pressure patterns 10 to 15 years later. Psychosom
Med. 1992;54:217–30.
8. Hamer M, Kivimäki M, Batty GD. Pulse rate reactivity in childhood as a risk factor for
adult hypertension. J Hypertens. 2016 Sep;34(9):1804–7.
9. Landsbergis PA, Dobson M, Koutsouras G, Schnall P. Job strain and ambulatory blood
pressure: A meta-analysis and systematic review. Am J Public Health. 2013;103(3):61–71.
10. Kamarck TW, Shiffman SM, Smithline L, Goodie JL, Paty JA, Gnys M, et al. Effects of
Running head: AMBULATORY HEART RATE & VACATIONING 24
task strain, social conflict, and emotional activation on ambulatory cardiovascular activity:
Daily life consequences of recurring stress in a multiethnic adult sample. Heal Psychol.
1998;17(1):17–29.
11. Kamarck TW, Muldoon MF, Shiffman SS, Sutton-Tyrrell K. Experiences of demand and
control during daily life are predictors of carotid atherosclerotic progression among
healthy men. Heal Psychol. 2007;26(3):324–32.
12. Geurts SAE, Sonnentag S. Recovery as an explanatory mechanism in the relation between
acute stress reactions and chronic health impairment. Scand J Work Environ Heal.
2006;32(6):482–92.
13. De Bloom J, Kinnunen U, Korpela K. Recovery Processes during and after Work:
Associations with Health, Work Engagement, and Job Performance. J Occup Environ
Med. 2015;57(7):732–42.
14. Shimazu A, Sonnentag S, Kubota K, Kawakami N. Validation of the Japanese version of
the recovery experience questionnaire. J Occup Health. 2012;54(3):196–205.
15. Fritz C, Sonnentag S, Spector PE, McInroe JA. The weekend matters: Relationships
between stress recovery and affective experiences. J Organ Behav. 2010;31(8):1137–62.
16. De Bloom J, Geurts S, Kompier M. Vacation from work as prototypical recovery
opportunity. Gedrag Organ. 2010;23:333–49.
17. Eaker ED, Pinsky J, Castelli WP. Myocardial infarction and coronary death among
women: Psychosocial predictors from a 20-year follow-up of women in the Framingham
Study. Am J Epidemology. 1992;135:854–64.
18. Gump BB, Matthews KA. Are Vacations Good for Your Health ? The 9-Year Mortality
Experience After the Multiple Risk Factor Intervention Trial. Psychosom Med.
Running head: AMBULATORY HEART RATE & VACATIONING 25
2000;612(62):608–12.
19. Etzion D. Annual vacation: Duration of relief from job stressors and burnout. Anxiety
Stress Coping. 2003;16(2):213–26.
20. Westman M, Eden D. Effects of a respite from work on burnout: Vacation relief and fade-
out. J Appl Psychol. 1997;82(4):516–27.
21. de Bloom J, Kompier M, Geurts S, de Weerth C, Taris T, Sonnentag S. Do we recover
from vacation? Meta-analysis of vacation effects on health and well-being. J Occup
Health. 2009;51(1):13–25.
22. Brooks AW. Get excited: Reappraising pre-performance anxiety as excitement. J Exp
Psychol Gen. 2014;143(3):1144–58.
23. Jamieson JP, Mendes WB, Nock MK. Improving Acute Stress Responses. Curr Dir
Psychol Sci. 2013;22(1):51–6.
24. Gilovich T, Kumar A. We’ll Always Have Paris: The Hedonic Payoff from Experiential
and Material Investments. 1st ed. Vol. 51, Advances in Experimental Social Psychology.
Elsevier Inc.; 2015. 147–187 p.
25. Strauss-Blasche G, Ekmekcioglu C, Marktl W. Moderating effects of vacation on
reactions to work and domestic stress. Leis Sci. 2002;24:237–49.
26. Greenberg A, Berktold J. Holiday stress [Internet]. 2006. Available from:
https://www.apa.org/news/press/releases/2006/12/holiday-stress.pdf
27. Steptoe A. Night shift work and the cardiovascular health of medical staff. European
Heart Journal. 2009;30(21):2560–1.
28. Hruska B, Pressman SD, Bendinskas K, Gump BB. Vacation frequency is associated with
metabolic syndrome and symptoms. Psychol Heal. 2019;1–15.
Running head: AMBULATORY HEART RATE & VACATIONING 26
29. Qualtrics [Internet]. 2017. Available from: http://www.qualtrics.com
30. Fitbit Charge HR Product Manual. San Francisco, CA;
31. Dooley EE, Golaszewski NM, Bartholomew JB. Estimating Accuracy at Exercise
Intensities: A Comparative Study of Self-Monitoring Heart Rate and Physical Activity
Wearable Devices. JMIR mHealth uHealth. 2017 Mar 16;5(3):e34.
32. Small Steps Lab LLC. Fitabase [Internet]. 2018 [cited 2018 Dec 21]. Available from:
https://www.fitabase.com/
33. Miller GE, Cohen S, Pressman S, Barkin A, Rabin BS, Treanor JJ. Psychological Stress
and Antibody Response to Influenza Vaccination: When Is the Critical Period for Stress,
and How Does It Get Inside the Body? Psychosom Med. 2004;66(2):215–23.
34. Cohen S, Kamarck T, Mermelstein R. A Global Measure of Perceived Stress. J Health Soc
Behav. 1983 Dec;24(4):385.
35. Carnethon MR, Yan L, Greenland P, Garside DB, Dyer AR, Metzger B, et al. Resting
heart rate in middle age and diabetes development in older age. Diabetes Care.
2008;31(2):335–9.
36. Fox K, Borer JS, Camm AJ, Danchin N, Ferrari R, Lopez Sendon JL, et al. Resting Heart
Rate in Cardiovascular Disease. J Am Coll Cardiol. 2007;50(9):823–30.
37. Komai R, Obara T, Ohkubo T, Kato T, Kikuya M, Metoki H, et al. Factors affecting heart
rate as measured at home among treated hypertensive patients: the Japan home versus
office blood pressure measurement evaluation (J-HOME) study. Hypertens Res.
2007;30(11):1051–7.
38. Kostis JB, Moreyra AE, Amendo MT, Di Pietro J, Cosgrove N, Kuo PT. The effect of age
on heart rate in subjects free of heart disease. Circulation. 1982;65(1):141–5.
Running head: AMBULATORY HEART RATE & VACATIONING 27
39. Ramaekers D, Ector H, Aubert AE, Rubens A, Van De Werf F. Heart rate variability and
heart rate in healthy volunteers: Is the female autonomic nervous system cardioprotective?
Eur Heart J. 1998;19(9):1334–41.
40. Reule S, Drawz PE. Heart rate and blood pressure: Any possible implications for
management of hypertension. Curr Hypertens Rep. 2012;14(6):478–84.
41. Strath SJ, Swartz AM, Bassett DR, O’Brien WL, King GA, Ainswoth BE. Evaluation of
heart rate as a method for assessing moderate intensity physical activity. Med Sci Sport
Exerc. 2000;32(Supplement):S465–70.
42. Barta WD, Tennen H, Litt MD. Measurement reactivity in diary research. In: Mehl MR,
Conner TS, editors. Handbook of Research Methods for Studying daily Life. New York,
NY: The Guilford Press; 2012. p. 108–23.
43. StataCorp. Stata Statistical Software: Release 13. 2013. 2013.
44. Hoffman L. Longitudinal Analysis: Modeling Within-Person Fluctuation and Change.
New York, NY: Routledge; 2015.
45. Kang M, Marshall SJ, Barreira T V., Lee J-O. Effect of pedometer-based physical activity
interventions: A meta-analysis. Res Q Exerc Sport. 2009;80(3):648–55.
46. Maier KJ, Waldstein SR, Synowski SJ. Relation of cognitive appraisal to cardiovascular
reactivity, affect, and task engagement. Ann Behav Med. 2003;26(1):32–41.
47. Tomaka J, Blascovich J, Kelsey RM, Leitten CL. Subjective, Physiological, and
Behavioral Effects of Threat and Challenge Appraisal. J Pers Soc Psychol.
1993;65(2):248–60.
48. Jamieson JP, Nock MK, Mendes WB. Mind over matter: reappraising arousal improves
cardiovascular and cognitive responses to stress. J Exp Psychol Gen. 2012
Running head: AMBULATORY HEART RATE & VACATIONING 28
Aug;141(3):417–22.
49. Cohen S, Pressman S. Stress Buffering Hypothesis. In: Encyclopedia of Health &
Behavior. 2nd ed. Sage Publications; 2004. p. 780–1.
50. Pressman SD, Cohen S. Does positive affect influence health? Psychol Bull.
2005;131(6):925–71.
51. U.S. Travel Association. The State of American Vacation 2018: How Vacation Became a
Casualty of Our Work Culture [Internet]. 2018. Available from:
https://projecttimeoff.com/reports/state-of-american-vacation-2018/
52. Kent Anger W, Elliot DL, Bodner T, Olson R, Rohlman DS, Truxillo DM, et al.
Effectiveness of total worker health interventions. J Occup Health Psychol.
2015;20(2):226–47.
53. Bureau of Labor Statistics. Leave Benefits: Access [Internet]. 2017 [cited 2018 Feb 8].
Available from:
https://www.bls.gov/ncs/ebs/benefits/2017/ownership/private/table32a.htm
54. Hämmig O. Health and well-being at work: The key role of supervisor support. SSM -
Popul Heal. 2017 Dec 1;3:393–402.
55. Payne J, Cluff L, Lang J, Matson-Koffman D, Morgan-Lopez A. Elements of a Workplace
Culture of Health, Perceived Organizational Support for Health, and Lifestyle Risk. Am J
Heal Promot. 2018;32(7):1555–67.
56. Benjamins MR, Hirschtick JL, Hunt BR, Hughes MM, Hunter B. Racial Disparities in
Heart Disease Mortality in the 50 Largest U.S. Cities. J Racial Ethn Heal Disparities
[Internet]. 2017;4(5):967–75. Available from: http://dx.doi.org/10.1007/s40615-016-0300-
6
Running head: AMBULATORY HEART RATE & VACATIONING 29
Table Captions
Table 1. Participant Characteristics (n = 54)
Table 2. Multilevel Piecewise Regression Examining the Interaction Between Weekly Stress and
Time on Ambulatory Heart Rate
Running head: AMBULATORY HEART RATE & VACATIONING 30
Table 1.
Age
M (SD)
44.1 (10.2)
Sex
% female
70.4
Race
% Caucasian
96.3
Income, %
$25,000-$34,999
$35,000-$49,999
$50,000-$74,999
$75,000-$99,999
$100,000 or more
3.7
7.4
16.7
35.2
37.0
Industry sector, %
Healthcare
Higher education
Management
Legal services
Banking/finance
Transportation
Other1
25.9
22.2
11.1
9.3
9.3
3.7
18.5
1Note. Various industries were represented by the remaining participants and could not be clearly
grouped together.
Running head: AMBULATORY HEART RATE & VACATIONING 31
Table 2.
b (SE)
95% CI
p
Fixed Effects
Intercept
80.38 (1.58)
77.29, 83.48
< 0.001
Pre-vacation
-0.44 (0.21)
-0.84, -0.03
0.035
Post-vacation
-0.10 (0.29)
-0.68, 0.47
0.73
Weekly stress
-0.37 (0.49)
-1.33, 0.59
0.45
Pre-vacation x weekly stress
-0.63 (0.27)
-1.16, -0.10
0.018
Post-vacation x weekly stress
0.16 (0.50)
-0.81, 1.13
0.75
Age
-0.25 (0.12)
-0.50, -0.01
0.042
Sex
1.79 (1.29)
-0.74, 4.32
0.17
Preexisting disease
-0.93 (4.30)
-9.35, 7.50
0.83
Blood pressure medications
2.66 (4.63)
-6.41, 11.73
0.57
Weekly steps
0.52 (0.12)
0.19, 0.65
< 0.001
Random Effects
Random intercept variance
67.71 (14.20)
45.89, 102.12
-
Residual variance
7.70 (1.16)
5.72, 10.35
-
Autocorrelation parameter
0.24 (0.12)
-0.01, 0.47
-
Note. SE = standard error; CI = confidence interval
Running head: AMBULATORY HEART RATE & VACATIONING 32
Figure Captions.
Figure 1. Box-and-Whiskers Plots Depicting Stress Rating in the Weeks Pre-Vacation and in the
Weeks Post-Vacation
Figure 2. Box-and Whiskers Plot Depicting Steps Taken in the Weeks Pre-Vacation and in the
Weeks Post-Vacation
Figure 3. Plot of Predicted Means Illustrating the Interaction Between Weekly Stress and Time
Leading Up to the Vacation on Ambulatory Heart Rate
Running head: AMBULATORY HEART RATE & VACATIONING 33
Figure 1.
Note. The difference between pre-vacation and post-vacation stress ratings was statistically significant (p
< 0.001).
0.00
1.00
2.00
3.00
4.00
Pre-Vacation Post-Vacation
Stress Rating
Running head: AMBULATORY HEART RATE & VACATIONING 34
Figure 2.
Note. The difference between pre-vacation and post-vacation weekly total steps was statistically
significant (p = 0.004).
0
20000
40000
60000
80000
100000
Pre-Vacation Post-Vacation
Steps
Running head: AMBULATORY HEART RATE & VACATIONING 35
Figure 3.
Note. Low = 1st percentile; Med = 50th percentile; High = 99th percentile.
77
78
79
80
81
82
83
84
85
86
87
Low Med High
Ambulatory Heart Rate
Weekly Stress
-4
-3
-2
-1
Weeks
Before
Vacation
