Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Preventive Medicine Reports 27 (2022) 101764
Available online 17 March 2022
2211-3355/© 2022 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
COVID-19 stressors and health behaviors: A multilevel longitudinal study
across 86 countries
Shian-Ling Keng
a
,
bq
,
*
,
1
, Michael V. Stanton
b
,
1
, LeeAnn B. Haskins
c
, Carlos A. Almenara
d
,
Jeannette Ickovics
e
,
bq
, Antwan Jones
f
, Diana Grigsby-Toussaint
g
, Maximilian Agostini
h
,
Jocelyn J. B´
elanger
i
, Ben Gützkow
h
, Jannis Kreienkamp
h
, Edward P. Lemay Jr.
am
,
Michelle R. vanDellen
c
, Georgios Abakoumkin
j
, Jamilah Hanum Abdul Khaiyom
k
,
Vjollca Ahmedi
l
, Handan Akkas
m
, Mohsin Atta
n
, Sabahat Cigdem Bagci
o
, Sima Basel
i
,
Edona Berisha Kida
l
, Allan B.I. Bernardo
p
, Nicholas R. Buttrick
q
, Phatthanakit Chobthamkit
r
,
Hoon–Seok Choi
s
, Mioara Cristea
t
, S´
ara Csaba
u
, Kaja Damnjanovic
v
, Ivan Danyliuk
w
,
Arobindu Dash
x
, Daniela Di Santo
y
, Karen M. Douglas
z
, Violeta Enea
aa
, Daiane G. Faller
bn
,
Gavan Fitzsimons
ab
, Alexandra Gheorghiu
aa
, ´
Angel G´
omez
ac
, Ali Hamaidia
ad
, Qing Han
ae
,
Mai Helmy
af
,
bo
, Joevarian Hudiyana
ag
, Bertus F. Jeronimus
h
, Ding–Yu Jiang
ah
,
Veljko Jovanovi´
c
ai
, ˇ
Zeljka Kamenov
aj
, Anna Kende
u
, Tra Thi Thanh Kieu
ak
, Yasin Koc
h
,
Kamila Kovyazina
al
, Inna Kozytska
w
, Joshua Krause
h
, Arie W. Kruglanski
am
, Anton Kurapov
w
,
Maja Kutlaca
an
, N´
ora Anna Lantos
u
, Cokorda Bagus Jaya Lesmana
ao
, Winnifred R. Louis
ap
,
Adrian Lueders
aq
, Marta Maj
ar
, Najma Iqbal Malik
n
, Anton Martinez
as
, Kira O. McCabe
at
,
Jasmina Mehuli´
c
aj
, Mirra Noor Milla
ag
, Idris Mohammed
au
, Erica Molinario
bp
,
Manuel Moyano
av
, Hayat Muhammad
aw
, Silvana Mula
y
, Hamdi Muluk
ag
, Solomiia Myroniuk
h
,
Reza Naja
ax
, Claudia F. Nisa
i
, Bogl´
arka Nyúl
u
, Paul A. O’Keefe
bq
, Jose Javier Olivas Osuna
br
,
Evgeny N. Osin
ay
, Joonha Park
az
, Gennaro Pica
ba
, Antonio Pierro
y
, Jonas Rees
bb
,
Anne Margit Reitsema
h
, Elena Resta
y
, Marika Rullo
bc
, Michelle K. Ryan
h
,
bd
, Adil Samekin
be
,
Pekka Santtila
bf
, Edyta M. Sasin
i
, Birga M. Schumpe
bu
, Heyla A. Selim
bg
, Wolfgang Stroebe
h
,
Samiah Sultana
h
, Robbie M. Sutton
z
, Eleftheria Tseliou
j
, Akira Utsugi
bh
,
Jolien Anne van Breen
bi
, Caspar J. Van Lissa
bj
, Kees Van Veen
h
, Alexandra V´
azquez
ac
,
Robin Wollast
bs
, Victoria Wai–lan Yeung
bk
, Somayeh Zand
bt
, Iris Lav ˇ
Zeˇ
zelj
v
, Bang Zheng
bl
,
Andreas Zick
bb
, Claudia Zú˜
niga
bm
, N. Pontus Leander
h
,
bv
a
Monash University Malaysia, Malaysia
b
California State University, East Bay, USA
c
University of Georgia, USA
d
Universidad Peruana de Ciencias Aplicadas, Peru
e
Yale University, USA
f
The George Washington University, USA
g
Brown University, USA
h
University of Groningen, The Netherlands
i
New York University Abu Dhabi, United Arab Emirates
j
University of Thessaly, Volos, Greece
k
International Islamic University Malaysia, Gombak, Malaysia
l
University of Pristina, Pristina, Kosovo
m
Ankara Science University, Ankara, Turkey
n
University of Sargodha, Sargodha, Pakistan
* Corresponding author at: Monash University Malaysia, Jalan Lagoon Selatan, Bandar Sunway, 47500 Subang Jaya, Selangor, Malaysia.
E-mail address: keng.sl@monash.edu (S.-L. Keng).
Contents lists available at ScienceDirect
Preventive Medicine Reports
journal homepage: www.elsevier.com/locate/pmedr
https://doi.org/10.1016/j.pmedr.2022.101764
Received 7 July 2021; Received in revised form 6 March 2022; Accepted 13 March 2022
Preventive Medicine Reports 27 (2022) 101764
2
o
Sabanci University, Istanbul, Turkey
p
De La Salle University, Manila, Philippines
q
University of Virginia, Charlottesville, USA
r
Thammasat University, Bangkok, Thailand
s
Sungkyunkwan University, Seoul, South Korea
t
Heriot Watt University, United Kingdom
u
E¨
otv¨
os Lor´
and University (ELTE), Budapest, Hungary
v
University of Belgrade, Belgrade, Serbia
w
Taras Shevchenko National University of Kyiv, Kiev, Ukraine
x
Leuphana University Luneburg, Lüneburg, Germany
y
Sapienza University of Rome, Rome, Italy
z
University of Kent, Canterbury, UK
aa
Alexandru Ioan Cuza University of Iasi, Iasi, Romania
ab
Duke University, Durham, USA
ac
Universidad Nacional de Educaci´
on a Distancia (UNED), Madrid, Spain
ad
University Setif 2, S´
etif, Algeria
ae
University of Bristol, Bristol, UK
af
Menoua University, Al Minuyah, Egypt
ag
Universitas Indonesia, Depok, Indonesia
ah
National Chung-Cheng University, Chiayi, Taiwan
ai
University of Novi Sad, Novi Sad, Serbia
aj
University of Zagreb, Zagreb, Croatia
ak
HCMC University of Education, Ho Chi Minh City, Viet Nam
al
Independent Researcher, Nur-Sultan, Kazakhstan
am
University of Maryland, College Park, USA
an
Durham University, Durham, UK
ao
Udayana University, Denpasar, Indonesia
ap
University of Queensland, Brisbane, Australia
aq
University of Limerick, Ireland
ar
Jagiellonian University, Krak´
ow, Poland
as
University of Shefeld, Shefeld, UK
at
Carleton University, Canada
au
Usmanu Danfodiyo University Sokoto, Sokoto, Nigeria
av
University of Cordoba, C´
ordoba, Spain
aw
University of Peshawar, Peshawar, Pakistan
ax
University of Padova, Italy
ay
National Research University Higher School of Economics, Moscow, Russia
az
NUCB Business School, Nagoya, Japan
ba
University of Camerino, Camerino, Italy
bb
University of Bielefeld, Bielefeld, Germany
bc
University of Siena, Siena, Italy
bd
University of Exeter, Exeter, UK
be
School of Liberal Arts, M. Narikbayev KAZGUU University, Nur-Sultan, Kazakhstan
bf
New York University Shanghai, Shanghai, China
bg
King Saud University, Riyadh, Saudi Arabia
bh
Nagoya University, Nagoya, Japan
bi
Leiden University, Leiden, The Netherlands
bj
Utrecht University, Utrecht, The Netherlands
bk
Lingnan University, Tuen Mun, Hong Kong
bl
Imperial College London, London, UK
bm
Universidad de Chile, Santiago, Chile
bn
National University of Singapore, Singapore, Singapore
bo
Sultan Qaboos University, Egypt
bp
Florida Gulf Coast University, USA
bq
Yale-NUS College, Singapore, Singapore
br
National Distance Education University, Spain
bs
Universit´
e Clermont-Auvergne, France
bt
University of Milano-Bicocca, Italy
bu
University of Amsterdam, The Netherlands
bv
Wayne State University, USA
ARTICLE INFO
Keywords:
COVID-19
Health behaviors
Infection risk
Economic burden
ABSTRACT
Anxiety associated with the COVID-19 pandemic and home connement has been associated with adverse health
behaviors, such as unhealthy eating, smoking, and drinking. However, most studies have been limited by
regional sampling, which precludes the examination of behavioral consequences associated with the pandemic at
a global level. Further, few studies operationalized pandemic-related stressors to enable the investigation of the
impact of different types of stressors on health outcomes. This study examined the association between perceived
risk of COVID-19 infection and economic burden of COVID-19 with health-promoting and health-damaging
behaviors using data from the PsyCorona Study: an international, longitudinal online study of psychological
and behavioral correlates of COVID-19. Analyses utilized data from 7,402 participants from 86 countries across
three waves of assessment between May 16 and June 13, 2020. Participants completed self-report measures of
COVID-19 infection risk, COVID-19-related economic burden, physical exercise, diet quality, cigarette smoking,
1
Shian-Ling Keng and Michael Stanton are co-rst authors on this paper.
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
3
sleep quality, and binge drinking. Multilevel structural equation modeling analyses showed that across three time
points, perceived economic burden was associated with reduced diet quality and sleep quality, as well as
increased smoking. Diet quality and sleep quality were lowest among respondents who perceived high COVID-19
infection risk combined with high economic burden. Neither binge drinking nor exercise were associated with
perceived COVID-19 infection risk, economic burden, or their interaction. Findings point to the value of
developing interventions to address COVID-related stressors, which have an impact on health behaviors that, in
turn, may inuence vulnerability to COVID-19 and other health outcomes.
1. Introduction
The COVID-19 pandemic has caused profound adverse health, eco-
nomic, and psychological consequences. To contain the spread of the
pandemic, many countries have imposed lockdowns, limiting citizens’
participation in regular social and physical activities. Though essential
to slow the rate of infection, lockdowns have been found to be positively
associated with negative mental health consequences, such as depres-
sion and anxiety (Huang and Zhao, 2020; Nguyen et al., 2020).
Furthermore, such measures are likely to impact health-related behav-
iors: restricted mobility decreases physical activity, and heightened
psychological distress increases the propensity to engage in unhealthy
eating, smoking, and binge drinking (Grzywacz and Almeida, 2008;
Kassel et al., 2003). These unhealthy behaviors are risk factors for non-
communicable diseases, including obesity, diabetes, and cardiovascular
diseases (Thornton et al., 2016; Stang et al., 2000; Hu et al., 2000),
which in turn increase the risk of contracting COVID-19 and greater
disease severity and may eventually lead to increased mortality (Esai,
2020; Zheng et al., 2020).
To date, results are mixed across extant cross-sectional studies
looking at the relationship between stress related to COVID-19 and
unhealthy behaviors. In the United States, pandemic-related psycho-
logical distress was positively associated with alcohol use, with women
being signicantly more likely to consume greater amounts of alcohol on
a typical evening and during their recent heaviest drinking occasion
(Rodriguez et al., 2020). In Vietnam, fear of COVID-19 was associated
with greater alcohol consumption and smoking among college students
(Nguyen et al., 2020). In contrast, a study based in Spain reported less
alcohol consumption and better dietary behaviors during the COVID-19
lockdown (Rodríguez-P´
erez et al., 2020). In China, pandemic-related
home isolation was associated with improvements in dietary behaviors
and sleep quality, even though time spent being sedentary increased
during lockdown compared to pre-lockdown (Wang et al., 2020). These
varying associations could in part be attributed to regional variations in
lockdown policies, which affect ease of access to health-relevant re-
sources such as exercise facilities, and outdoor dining options.
Even though these studies provide some insight into the potential
impact of the pandemic on health behaviors, several caveats can be
identied. First, the majority of the studies are regionally focused and do
not explore global trends. One exception is a study involving over 1000
adults in Asia, Europe, and Africa, which documented a decrease in
physical activity and binge drinking and an increase in unhealthy food
consumption during COVID-19 home connement (Ammar et al., 2020).
The analyses however did not control for potential confounding vari-
ables, such as gender, age, and education that may have explained the
changes in these health behaviors. Though most individuals likely
experienced heightened anxiety about contracting COVID-19, the de-
gree of anxiety and perceived risk may also vary globally depending on
access to protective measures, as well as perceived effectiveness of the
government and/or the community in curbing the pandemic.
Further, few studies have operationalized stressors related to the
pandemic. Two critical stressors faced by many individuals during the
pandemic include infection risk and economic burden. During the
ongoing pandemic, many individuals experience varying degrees of
nancial impact, with millions facing unemployment and loss of income
and housing, which may adversely impact health-related behaviors and
outcomes. It remains to be examined whether perceived risk of infection
and economic burden may differentially impact health behaviors and
whether these stressors may interact to predict engagement in specic
health behaviors. Importantly, these effects should be assessed while
controlling for sociodemographic characteristics, which are known to
impact health behaviors, such as binge drinking, smoking, and healthy
eating (Wilsnack et al., 2018; Wardle et al., 2004; Bauer et al., 2007;
Cavelaars et al., 2000).
In this study, we utilized data from a multinational, longitudinal
online study on psychological and behavioral correlates of COVID-19 to
examine the association between perceived risk of infection and eco-
nomic burden with several health-promoting (exercise, diet quality,
sleep quality) and health-damaging (binge drinking, smoking) behav-
iors. We hypothesized that perceived risk of infection and economic
burden would be associated with reduced engagement in healthier be-
haviors. Specically, we predicted that higher levels of perceived
infection risk and economic burden would each independently be
associated with less exercise, poorer diet, and worse sleep quality, as
well as more binge drinking and smoking, independent of the effects of
demographic factors. Additionally, we expected the interaction between
perceived infection risk and economic burden would be a particularly
strong predictor of health-damaging behaviors. Recruitment of a large
international sample enabled us to observe the association between
pandemic-related stressors and health behaviors on a global scale.
2. Method
2.1. Participants and procedure
The sample consisted of adult participants (aged 18 and above) of an
online, longitudinal study as part of the PsyCorona project (htt
ps://psycorona.org/), a multinational research project examining
behavioral and psychological responses to the COVID-19 pandemic.
Research participants initially completed a baseline cross-sectional
survey, and a subset of participants signed up for a longitudinal study
involving follow-up surveys over the course of the pandemic (Jin et al.,
2021; Han et al., 2021; Romano et al., 2020). Our analysis focused on a
self-selected cohort of participants (N =7, 402) who completed Wave 7,
9, and 11 of assessments (administered in two-week intervals) between
May 16 and June 13 of 2020. Each assessment lasted approximately 10
min. The surveys were translated into 30 languages and distributed by
members of the research team (consisting of over 100 behavioral sci-
entists) in their respective countries using social media campaigns, press
releases, and social and academic networks.
This study complies with ethical regulations for research on human
subjects. All participants gave informed consent, as approved by the
Institutional Review Board at New York University Abu Dhabi (HRPP-
2020–42) and the Ethics Committee of Psychology at Groningen Uni-
versity (PSY-1920-S-0390).
2.2. Measures
2.2.1. Perceived Stressors: COVID-19 infection risk and economic burden
Perceived stress was measured by the item: “How likely is it that the
following will happen to you in the next few months?” (1) COVID-19
infection risk – “you will get infected with coronavirus”, and (2)
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
4
Economic burden – “your personal situation will get worse due to eco-
nomic consequences of coronavirus.” Responses were based on a Likert-
type scale of 1 (very unlikely) to 8 (already happened).
2.2.2. Health behaviors
Five health-related behaviors were measured with single-item
questions:
(1) Physical Exercise was measured with the question: “During the
past week, how many days did you do 20 min of vigorous
(sweating and pufng) or 30 min of moderate (increasing your
heart rate but not vigorous) physical activity?” (adapted from the
Brief Physical Activity Assessment Tool) (Marshall et al., 2005).
Participants responded using a range of 0 to 7 days.
(2) Diet quality was assessed with the question: “During the past
week, how healthy was your overall diet? Consider how many
sweets you have been eating as well as how many portions of fruit
and/or vegetables you ate each day” (adapted from National
Health and Nutrition Examination Survey Questionnaire) (Na-
tional Health and Nutrition Examination Survey Questionnaire,
2018). Participants were asked to provide a rating on a 1 (poor) to
5 (excellent) scale.
(3) Sleep quality was measured with the question: “During the past
week, how would you rate your sleep quality overall?” (adapted
from Pittsburgh Sleep Quality Index) (Buysse et al., 1989). Par-
ticipants were asked to provide a rating on a 1 (poor) to 5
(excellent) scale.
(4) Binge drinking was measured with the item: “During the past
week, how many days did you have>4 drinks in a day?” (adapted
from a screening test for unhealthy alcohol use recommended by
the National Institute on Alcohol Abuse and Alcoholism) (Smith
et al., 2009). Participants responded using a range of 0 to 7 days.
(5) Smoking was assessed with the item: “During the past week, how
many cigarettes did you smoke each day?”, with an open
response option (adapted from National Health and Nutrition
Examination Survey Questionnaire) (National Health and Nutri-
tion Examination Survey Questionnaire, 2018). This variable was
transformed into four categories: 0 cigarettes per day coded as
non-smoker, 1–10 cigarettes per day coded as light smoker,
11–19 cigarettes per day coded as moderate smoker, >=20 cig-
arettes per day coded as heavy smoker, following the criteria of
the Government of Canada (Government of Canada, 2008). After
a visual inspection of the dataset, plots, and measures of disper-
sion, we excluded outliers, particularly those who reported
smoking>75 cigarettes per day (n =37, n =24, and n =28, in
waves 7, 9, and 11, respectively).
2.2.3. Sociodemographic characteristics
Participants provided information about age, categorized on a scale
from 1 (18–24 years old) to 7 (75 +years old); education, categorized on
a scale from 1 (elementary) to 6 (doctorate); and gender, categorized as 1
(female), 2 (male), and 3 (other). For the purpose of our analyses, gender
was re-coded into a binary variable (0 =female, 1 =male, whereas
“other” was excluded from analyses).
2.3. Statistical analyses
Demographic information was assessed using SAS. Mplus 8.4 was
used to conduct multilevel structural equation modeling (MSEM)
bivariate correlations and regression. Data from Waves 7, 9, and 11
(time points; level 1) were nested within the participants (level 2). All
health behavior outcomes had sufcient variance across the two levels
(ICCs >0.68), so MSEM was employed to estimate the structural re-
lationships at both levels (i.e., within and between persons). Acknowl-
edging that participants were nested within geographical region (i.e.,
North America, Europe, Asia, Africa, Oceania, Caribbean, Central, and
South America) (United Nations. World Population Prospectus, 2019),
we evaluated the intraclass correlations (ICCs) of each of the health
behaviors by adding region as a level 3 variable (time points within
participants within region). We evaluated region as opposed to country
as a level 3 variable because of limited samples from some countries (e.
g., n <10), which precluded sufcient data for analyses of country as a
higher order variable. However, because all ICCs were at or below 0.05,
we did not include region as a level 3 variable in the nal MSEM ana-
lyses (LeBreton and Senter, 2008).
Because the current research interest was to evaluate the effects of
COVID-19 stressors on health behaviors across individuals, all results
reported are at the between-person level and over three time periods. As
part of preliminary analyses, we conducted MSEM bivariate correla-
tional analyses to examine the association between demographic factors
and COVID-19 related stressors, as well as each of the health behaviors.
Next, we conducted MSEM regression with random intercepts and xed
slopes to examine the role of perceived infection risk, economic burden,
and their interaction as predictors of each of the health behaviors. All
MSEM regression analyses included age, gender, and education as
between-person covariates. Analyses were conducted using full-
information maximum likelihood estimation, which provides standard
errors that are robust to data non-normality and non-independence
(Heck and Thomas, 2015).
3. Results
3.1. Sample characteristics and preliminary analyses
The sample consisted of 7,402 participants from 86 countries.
Table 1 provides a detailed breakdown of demographic information in
this sample. Sixty-seven percent (n =4959) of the participants were
female. Regionally, more than one-half of the sample was based in
Europe (60.9%), followed by North America (14.8%) and Asia (6.7%).
There was a relatively even distribution of individuals across age groups:
63.1% were between 18 and 54 years. More than half (56.2%) had at
Table 1
Sample Characteristics (N =7402).
Variable n (percentage)
Gender
Female 4959 (67%)
Male 2443 (33%)
Age
18 to 24 years old 794 (10.73%)
25 to 34 years old 1235 (16.68%)
35 to 44 years old 1260 (17.02%)
45 to 54 years old 1386 (18.72%)
55 to 64 years old 1400 (18.91%)
65 to 74 years old 1143 (15.44%)
75 and older 184 (2.48%)
Region
Europe 4510 (61.01%)
North America 1387 (18.74%)
Asia 633 (8.56%)
Caribbean, Central and South America 486 (6.57%)
Oceania 197 (2.67%)
Africa 179 (2.42%)
Country Not Indicated 10 (0.14%)
Education
Elementary and Secondary Education 907 (12.25%)
Vocational Education 831 (11.23%)
Higher Education (Without a Bachelor’s Degree) 1504 (20.32%)
Bachelor’s Degree 2018 (27.26%)
Master’s degree 1590 (21.48%)
Doctorate Degree 552 (7.46%)
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
5
least a college degree. A list of all countries included in this study is
provided in S1, under Supplementary Materials. Table 2 presents the
descriptive statistics of COVID-19 stressors and health behavior out-
comes across the whole sample.
We next examined demographic factors (age, gender, and education)
as potential correlates of the two COVID-19 stressors and each of the
health behaviors (see Table 3). Older age predicted signicantly lower
perceived COVID infection risk and economic burden, better diet and
sleep quality, and more cigarettes smoked in the past week, all ps <
0.001. Being male was associated with lower perceived infection risk,
better perceived diet and sleep quality, and more smoking and binge
drinking, all ps <0.01. Higher education levels were associated with
signicantly greater perceived COVID infection risk, better diet quality,
more days spent engaging in moderate to vigorous exercise, and fewer
cigarettes smoked, all ps <0.01.
3.2. Perceived infection Risk, economic Burden, and their interaction as
predictors of each health behavior and outcome
Between-person results of the multilevel structural equation
modeling analyses are presented in Table 4. Post hoc power analyses
were conducted to determine achieved power for each parameter coef-
cient in the ve models. Power analysis was conducted using Monte
Carlo simulation with 500 replications using Robust Maximum Likeli-
hood (MLR) estimation in Mplus. The analyses indicated adequate power
(>80%) to detect the majority of effects, with the exception of physical
exercise, binge drinking, and select parameter estimates for smoking.
Within-person results are reported in S2, under Supplementary
Materials.
COVID-related infection risk and economic burden were both nega-
tively associated with perceived diet quality during the previous week.
These main effects were qualied by a signicant interaction between
perceived infection risk and perceived economic burden, b =0.01, SE =
0.01, p <.05. As shown in Fig. 1, those who reported high economic
burden (top 10%) reported lower diet quality regardless of levels of
perceived infection risk, b =0.008, SE =0.02, p =.693, whereas those
perceiving low economic burden (bottom 10%) reported better diet
quality if their perceived infection risk was also low, b =-0.057, SE =
0.02, p =.002.
COVID-related infection risk and perceived economic burden were
both negatively associated with sleep quality during the previous week.
These main effects were qualied by a signicant interaction, b =0.67,
SE =0.01, p <.001. As shown in Fig. 2, those who reported high eco-
nomic burden (top 10%) reported decreased sleep quality regardless of
levels of perceived infection risk, b =-0.02, SE =0.02, p =.325, whereas
people perceiving low economic burden (bottom 10%) reported better
sleep quality if their perceived infection risk was also low, b =-0.111,
SE =0.02, p <.001.
Perceived economic burden was positively associated with the
number of cigarettes smoked. COVID-related infection risk was not
associated with the number of cigarettes smoked in the previous week.
There was no signicant interaction between infection risk and eco-
nomic burden in predicting the number of cigarettes smoked.
No relationship was observed between perceived COVID-related
infection risk, economic burden or their interaction and the number of
days spent binge drinking or the number of days spent exercising
moderately or vigorously. Across these analyses, none of the associa-
tions at the within-person level were signicant, indicating stability in
participants’ responses over time.
4. Discussion
This longitudinal study of health behaviors during the COVID-19
pandemic found that two pandemic-related stressors – perceived infec-
tion risk and perceived economic burden – were associated with a range
of health-related behaviors and outcomes. In particular, perceived eco-
nomic burden related to the pandemic was found to have the most
consistent negative impact across several health behavior outcomes,
including diet quality, sleep quality, and cigarette smoking. Economic
burden may lead to individuals engaging in unhealthy behaviors as a
Table 2
Descriptive Statistics for COVID-19 Stressors and Health Behaviors.
Variable N Scale Mean SD
Perceived Infection
Risk
7402 1 (very unlikely) −8 (already
happened)
3.56 1.33
Perceived Economic
Burden
7402 1 (very unlikely) −8 (already
happened)
3.93 1.76
Exercise 7401 Days in the past week 2.54 2.19
Diet Quality 7401 1 (poor) −5 (excellent) 3.00 0.96
Sleep Quality 7400 1 (poor) −5 (excellent) 2.73 1.04
Binge Drinking 7401 Days in the past week (0–7) 0.65 1.49
Variable Scale Frequency
(Percentage)
Smoking 4664 0 =Non-smoker 3654
(78.34%)
1 =Light Smoker 495 (10.61%)
2 =Moderate Smoker 213 (4.57%)
3 =Heavy Smoker 282 (6.05%)
Table 3
Bivariate Relationships among Demographic Variables, COVID-19 Stressors, and Health Behaviors.
Age Gender Education Perceived
Infection Risk
Perceived Economic
Burden
Physical
Exercise
Diet
Quality
Sleep
Quality
Binge
Drinking
Smoking
Age –
Gender 0.18*** –
Education -0.28*** -0.04*** –
Perceived Infection
Risk
-0.27*** -0.04*** 0.18*** –
Perceived Economic
Burden
-0.31*** -0.02 -0.04 0.67*** –
Exercise 0.06 0.03 0.41*** -0.02 -0.25*** –
Diet Quality 0.20*** 0.02** 0.11*** -0.10*** -0.27*** 0.65*** –
Sleep Quality 0.15*** 0.04*** 0.04 -0.19*** -0.39*** 0.35*** 0.39*** –
Binge Drinking 0.05 0.09*** -0.03 -0.02 0.07 0.05 -0.02 0.00 –
Smoking 0.09*** 0.02*** -0.15*** -0.04** 0.13*** -0.15*** -0.04*** -0.01 0.14*** –
Notes. Gender is coded as 0 (female) and 1 (male); Education is coded on a scale from 1 (elementary) to 6 (doctorate); **p <.01; ***p <.001.
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
6
coping mechanism, consistent with theoretical and empirical work
demonstrating an association between stress and health-damaging be-
haviors (Park and Iacocca, 2014). A recent report suggests that cash-
based assistance in the form of stimulus check in the United States
was linked to a robust 20% reduction in symptoms of depression and
Table 4
Test Statistics for Multilevel Regression with Each Health Behavior Predicted by
Infection Risk, Economic Burden, and Their Interaction.
Physical Exercise
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk
0.06 0.06 0.34 −0.06 0.18 0.25
Economic
Burden
−0.06 0.05 0.25 −0.16 0.04 0.35
Infection
Risk*
Economic
Burden
−0.01 0.01 0.43 −0.04 0.02 0.20
Diet Quality
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk
−0.08 0.03 0.004 −0.13 −0.03 0.98
Economic
Burden
−0.14 0.02 <
0.001
−0.19 −0.09 >0.99
Infection
Risk*
Economic
Burden
0.01 0.01 0.028 0.00 0.03 0.90
Sleep Quality
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk
−0.15 0.03 <
0.001
−0.20 −0.09 >0.99
Economic
Burden
−0.20 0.03 <
0.001
−0.25 −0.15 >0.99
Infection
Risk*
Economic
Burden
0.02 0.01 0.002 0.01 0.03 0.99
Binge Drinking
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk
−0.06 0.04 0.14 −0.14 0.02 0.50
Economic
Burden
−0.00 0.04 0.93 −0.07 0.07 0.06
Infection
Risk*
Economic
Burden
0.01 0.01 0.27 −0.01 0.03 0.33
Smoking
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk
−0.04 0.02 0.075 −0.08 0.00 0.55
Economic
Burden
0.06 0.02 0.002 0.02 0.10 0.97
0.00 0.01 0.96 −0.01 0.01 0.05
Table 4 (continued )
Physical Exercise
b SE p 95% CI
(Lower)
95% CI
(Upper)
Achieved
Power to
Detect
Parameter
Estimate
Infection
Risk*
Economic
Burden
Note. The above analyses included age, gender, and education as covariates.
1.5
2
2.5
3
3.5
12345678
Diet Quality
Infection Risk
Low Economic Burden
High Economic Burden
Fig. 1. Interaction between Infection Risk and Economic Burden in Predicting
Diet Quality. Note: Low economic burden is represented as the 10th percentile,
equal to 1.67 on the economic burden scale of 1 to 8; High economic burden is
represented as the 90th percentile, equal to 6.33 on the economic burden scale
of 1 to 8. Thin dotted lines represent 95% condence intervals.
1.5
2
2.5
3
3.5
12345678
Sleep Quality
Infection Risk
Low Economic Burden
High Economic Burden
Fig. 2. Interaction between Infection Risk and Economic Burden in Predicting
Sleep Quality. Note: Low economic burden is represented as the 10th percentile,
equal to 1.67 on the economic burden scale of 1 to 8; High economic burden is
represented as the 90th percentile, equal to 6.33 on the economic burden scale
of 1 to 8. Thin dotted lines represent 95% condence intervals.
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
7
anxiety during the pandemic (Fottrell, 0000). Therefore, economic
burden might be related to unhealthy behaviors through symptoms of
depression or anxiety, and when economic burden is alleviated, this may
reduce unhealthy behaviors as well.
The nding that economic burden was associated with greater
cigarette use is in line with previous research demonstrating a positive
association between nancial stress and tobacco use across households
of varying incomes (Siahpush et al., 2003). Notably, the association
between perceived economic burden and negative health outcomes may
be bi-directional: heightened economic stress may increase smoking
behaviors, and greater expenditure on acquiring tobacco products may
pose further economic strain.
Consistent with past research, the present study documented a
negative association between COVID-19 economic burden and sleep
quality (Hall et al., 2009; Onder et al., 2020). This association may be
accounted for by an increased tendency to engage in nancial rumina-
tion and worry (de Bruijn and Antonides, 2020) which have been found
to predict worsened sleep quality and mental health outcomes (Thor-
steinsson et al., 2019). Financial stress may also be linked to unem-
ployment, which affords greater unstructured time and likely more time
for smoking and drinking, and fewer resources available for healthy food
consumption (French and McKillop, 2017). In the context of the COVID-
19 pandemic, stress and isolation resulting from government-imposed
lockdowns and home quarantine may leave individuals more prone to
engaging in unhealthy coping behaviors.
Importantly, the study found that perceived economic burden
interacted with COVID-19 infection risk to predict worsened diet and
sleep quality. This suggests that the main effects of perceived COVID-19-
related stressors can only be meaningfully examined in the context of an
interaction between the stressors. This nding highlights the need to
develop interventions that address these stressors simultaneously to
mitigate the negative impact of the COVID-19 pandemic on health
outcomes. Specically, economically disadvantaged populations are
likely to be disproportionately impacted by the pandemic. There is
therefore an urgent need to develop measures to lower their infection
risk and economic burden, in order to mitigate the pandemic’s long-term
negative health consequences.
Contrary to our hypotheses, the study found no signicant associa-
tion between perceived infection risk and binge drinking, and only a
trending, positive association between infection risk and smoking. It is
plausible that attempts to drink or smoke may be driven more by general
distress associated with the pandemic, as suggested by a study by
Rodriguez and colleagues (Rodriguez et al., 2020), as opposed to the
perception of infection risk, per se. The nding does not rule out the
possibility that perceived infection risk is linked with more drinking that
does not reach the threshold of a binge. The absence of a signicant
association between perceived infection risk and these behaviors may
also reect individual variations in response to infection risk: while
some may be motivated to reduce engagement in health-damaging be-
haviors following awareness of high infection risk, others may engage in
more of such behaviors as a coping mechanism (Park and Iacocca,
2014). Likewise, the lack of an association between the stressors and
physical exercise may be attributable to signicant individual variations
in exercise habits during the pandemic, along with varying access to
exercise facilities due to lockdowns.
The present study also identied a few demographic correlates of
COVID-19 stressors and associated health behaviors. In particular, older
individuals reported lower levels of perceived infection risk and eco-
nomic burden, as well as better sleep and diet quality. The perception of
lower infection risk could be due to several factors, such as the fact that
older adults are less socially mobile. Compared to younger adults, they
are also more likely to engage in prosocial COVID-19 protective be-
haviors like social distancing and mask-wearing (Jin et al., 2021). The
nding that older individuals have better sleep quality suggests they
may be less psychologically impacted by the pandemic, consistent with
other studies’ ndings that older adults experience lower levels of
psychological symptoms and stress reactivity compared to younger
adults, likely due to a higher degree of resilience (Nwachukwu et al.,
2020; Nelson et al., 20212021). Relative to females, males tend to
perceive lower infection risk, in line with other research nding similar
gender differences in the perception of seriousness of the COVID-19
pandemic (Galasso et al., 2020). Compared to females, males also
smoke a greater number of cigarettes and spend more days binge
drinking. Lastly, higher levels of education are identied consistently as
a correlate of greater engagement in health-promoting behaviors and
lower engagement in health-damaging behaviors. These ndings point
to the value of tailoring public campaigns to certain demographics such
as young males, in order to reduce infection risk and likelihood of
engaging in health-damaging behaviors.
This study is characterized by several strengths, such as recruitment
of a large, multinational sample, a longitudinal design, and use of a
multilevel analytical approach that takes into consideration potential
variances accounted for by region and within-person variances across
time. Limitations of the study include lack of representativeness and use
of self-report measures, subject to recall and social desirability biases.
Although several of the outcome measures were single-item, several of
them were derived from established and validated scales. Due to limi-
tations in survey length, some measures such as income and general
mental health were not available. We did not examine patterns of
behavior change over time because each of the 86 participating coun-
tries were in a different stage of dealing with the pandemic at the time of
the surveys.
Future research could examine health behaviors using multimodal
and/or objective measures (e.g., food diaries to assess diet, poly-
somnography to assess sleep quality). Future work should control for the
effects of generalized anxiety or mental health symptoms to examine the
unique effects of perceived infection risk and economic burden on health
behaviors. Beyond infection risk and economic burden, social isolation
is an additional stressor that should be examined as a potential
contributor to health outcomes. Future research could also examine
coping styles that may moderate the effects of pandemic-related
stressors on health behaviors. Efforts should be made to examine spe-
cic communities (e.g., lower income groups) who may be at higher risk
for contracting COVID-19 due to jobs that may not support social
distancing. It would be of value to examine mechanisms underlying the
associations between COVID-19 related stressors and health behaviors,
including decisions about vaccinations, which were not yet available at
the time of the surveys.
The COVID-19 pandemic persists, with>410 million conrmed cases
and 5.8 million deaths globally as of February 14, 2022 (World Health
Organization, 2021). Vaccination roll-out is moving quickly in a few
countries, with marked delays in many more. Moreover, coronavirus
variants are of grave concern. As such, it is critical that each country
develops effective interventions tailored to the context of the local
community, particularly to those who are economically disadvantaged
and/or at higher infection risk, to mitigate the negative impact of the
pandemic on health behaviors (Han et al., 2021; Nisa et al., 2021).
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
The authors would like to acknowledge Maleyka Mammadova for
her assistance with literature review and data coding. This research
received support from the New York University Abu Dhabi (VCDSF/75-
71015), the University of Groningen (Sustainable Society & Ubbo
Emmius Fund), and the Instituto de Salud Carlos III (COV20/00086).
The COVID-19 risk perception item measured at baseline was previously
S.-L. Keng et al.
Preventive Medicine Reports 27 (2022) 101764
8
reported in unrelated test of effects on subjective well-being and mental
health.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.pmedr.2022.101764.
References
Ammar, A., Brach, M., Trabelsi, K., et al., 2020. Effects of COVID-19 home connement
on physical activity and eating behaviour Preliminary results of the ECLB-COVID19
international online-survey. Nutrients.
Bauer, T., G¨
ohlmann, S., Sinning, M., 2007. Gender differences in smoking behavior.
Health Econ. 16 (9), 895–909.
Buysse, D.J., Reynolds, C.F., Charles, F., Monk, T.H., Berman, S.R., Kupfer, D.J., 1989.
The Pittsburgh sleep quality index: a new instrument for psychiatric practice and
research. Psychiatry Res. 28 (2), 193–213.
Cavelaars, A.E.J.M., Kunst, A.E., Geurts, J.J.M., et al., 2000. Educational differences in
smoking: international comparison. BMJ 320 (7242), 1102–1107.
de Bruijn, E.-J., Antonides, G., 2020. Determinants of nancial worry and rumination.
J. Econ. Psychol. 76.
Esai, Selvan M., 2020. Risk factors for death from COVID-19. Nat. Rev. Immunol. 20 (7),
407.
Fottrell Q. Stimulus checks played a major role in nationwide decline in anxiety and
depression, new analysis says. MSN2021.
French, D., McKillop, D., 2017. The impact of debt and nancial stress on health in
Northern Irish households. Journal of European Social Policy. 27 (5), 458–473.
Galasso, V., Pons, V., Profeta, P., Becher, M., Brouard, S., Foucault, M., 2020. Gender
differences in COVID-19 attitudes and behavior: Panel evidence from eight
countries. PNAS 117 (44), 27285–27291.
Government of Canada. Tobacco Use Statistics: Terminology. https://www.canada.ca/e
n/health-canada/services/health-concerns/tobacco/research/tobacco-use-statistics/
terminology.html. Published 2008. Updated 2008-11-28. Accessed.
Grzywacz, J.G., Almeida, D.M., 2008. Stress and binge drinking: A daily process
examination of stressor pile-up and socioeconomic status in affect regulation. Int. J.
Stress Manage. 15 (4), 364–380.
Hall, M.H., Matthews, K.A., Kravitz, H.M., et al., 2009. Race and nancial strain are
independent correlates of sleep in midlife women: The SWAN sleep study. Sleep 32
(1), 73–82.
Han, Q., Zheng, B., Agostini, M., B´
elanger, J.J., Gützkow, B., Kreienkamp, J.,
Reitsema, A.M., van Breen, J.A., Collaboration, PsyCorona, Leander, N.P., 2021.
Associations of risk perception of COVID-19 with emotion and mental health during
the pandemic. J. Affect. Disord. 284, 247–255.
Heck, R.H., Thomas, S.L., 2015. An introduction to multilevel modeling techniques: MLM
and SEM approaches using Mplus, 3rd ed. Routledge.
Hu, F.B., Stampfer, M.J., Manson, J.E., Grodstein, F., Colditz, G.A., Speizer, F.E.,
Willett, W.C., 2000. Trends in the incidence of coronary heart disease and changes in
diet and lifestyle in women. N. Engl. J. Med. 343 (8), 530–537.
Huang, Y., Zhao, N., 2020. Generalized anxiety disorder, depressive symptoms and sleep
quality during COVID-19 outbreak in China: a web-based cross-sectional survey.
Psychiatry Res. 288.
Jin, S., Balliet, D., Romano, A., Spadaro, G., van Lissa, C.J., Agostini, M., B´
elanger, J.J.,
Gützkow, B., Kreienkamp, J., Leander, N.P., Abakoumkin, G., Khaiyom, J.H.A.,
Ahmedi, V., Akkas, H., Almenara, C.A., Kurapov, A., Atta, M., Bagci, S.C., Basel, S.,
Kida, E.B., Buttrick, N.R., Chobthamkit, P., Choi, H.-S., Cristea, M., Csaba, S.,
Damnjanovic, K., Danyliuk, I., Dash, A., Di Santo, D., Douglas, K.M., Enea, V.,
Faller, D.G., Fitzsimons, G., Gheorghiu, A., G´
omez, ´
A., Han, Q., Helmy, M.,
Hudiyana, J., Jeronimus, B.F., Jiang, D.-Y., Jovanovi´
c, V., Kamenov, ˇ
Z., Kende, A.,
Keng, S.-L., Kieu, T.T.T., Koc, Y., Kovyazina, K., Kozytska, I., Krause, J.,
Kruglanski, A.W., Kutlaca, M., Lantos, N.A., Lemay, E.P., Lesmana, C.B.J., Louis, W.
R., Lueders, A., Malik, N., Martinez, A., McCabe, K.O., Mehuli´
c, J., Milla, M.N.,
Mohammed, I., Molinario, E., Moyano, M., Muhammad, H., Mula, S., Muluk, H.,
Myroniuk, S., Naja, R., Nisa, C.F., Nyúl, B., O’Keefe, P.A., Osuna, J.J.O., Osin, E.N.,
Park, J., Pica, G., Pierro, A., Rees, J., Reitsema, A.M., Resta, E., Rullo, M., Ryan, M.
K., Samekin, A., Santtila, P., Sasin, E., Schumpe, B.M., Selim, H.A., Stanton, M.V.,
Stroebe, W., Sultana, S., Sutton, R.M., Tseliou, E., Utsugi, A., van Breen, J.A., Van
Veen, K., van Dellen, M.R., V´
azquez, A., Wollast, R., Yeung, V.-L., Zand, S., ˇ
Zeˇ
zelj, I.
L., Zheng, B., Zick, A., Zú˜
niga, C., 2021. Intergenerational conicts of interest and
prosocial behavior during the COVID-19 pandemic. Personality Individ. Differ. 171.
Kassel, J.D., Stroud, L.R., Paronis, C.A., 2003. Smoking, stress, and negative affect:
Correlation, causation, and context across stages of smoking. Psychol. Bull. 129 (2),
270–304.
LeBreton, J.M., Senter, J.L., 2008. Answers to 20 questions about interrater reliability
and interrater agreement. Org. Res. Meth. 11 (4), 815–852.
Marshall, A.L., Smith, B.J., Bauman, A.E., Kaur, S., 2005. Reliability and validity of a
brief physical activity assessment for use by family doctors. Br. J. Sports Med. 39 (5),
294.
National Health and Nutrition Examination Survey Questionnaire. U.S. Department of
Health and Human Services, Centers for Disease Control and Prevention; 2017-2018.
https://wwwn.cdc.gov/Nchs/Nhanes/2017-2018/SMQRTU_J.htm.
Nelson, N.A., Bergeman, C.S., 2020. Daily stress processes in a pandemic: The effects of
worry, age, and affect. The Gerontologist.
Nguyen, H.T., Do, B.N., Pham, K.M., Kim, G.B., Dam, H.T.B., Nguyen, T.T., Nguyen, T.T.
P., Nguyen, Y.H., Sørensen, K., Pleasant, A., Duong, T.V., 2020. Fear of COVID-19
Scale: Associations of its scores with health literacy and health-related behaviors
among medical students. Int. J. Environ. Res. Public Health 17 (11), 4164.
Nguyen, H.C., Nguyen, M.H., Do, B.N., Tran, C.Q., Nguyen, T.T.P., Pham, K.M., Pham, L.
V., Tran, K.V., Duong, T.T., Tran, T.V., Duong, T.H., Nguyen, T.T., Nguyen, Q.H.,
Hoang, T.M., Nguyen, K.T., Pham, T.T.M., Yang, S.-H., Chao, J.-J., Duong, T.V.,
2020. People with suspected COVID-19 symptoms were more likely depressed and
had lower health-related quality of life: The potential benet of health literacy.
J. Clin. Med. 9 (4), 965.
Nisa, C.F., B´
elanger, J.J., Faller, D.G., Buttrick, N.R., Mierau, J.O., Austin, M.M.K.,
Schumpe, B.M., Sasin, E.M., Agostini, M., Gützkow, B., Kreienkamp, J.,
Abakoumkin, G., Abdul Khaiyom, J.H., Ahmedi, V., Akkas, H., Almenara, C.A.,
Atta, M., Bagci, S.C., Basel, S., Kida, E.B., Bernardo, A.B.I., Chobthamkit, P., Choi, H.-
S., Cristea, M., Csaba, S., Damnjanovi´
c, K., Danyliuk, I., Dash, A., Di Santo, D.,
Douglas, K.M., Enea, V., Fitzsimons, G., Gheorghiu, A., G´
omez, ´
A., Grzymala-
Moszczynska, J., Hamaidia, A., Han, Q., Helmy, M., Hudiyana, J., Jeronimus, B.F.,
Jiang, D.-Y., Jovanovi´
c, V., Kamenov, ˇ
Z., Kende, A., Keng, S.-L., Kieu, T.T.T., Koc, Y.,
Kovyazina, K., Kozytska, I., Krause, J., Kruglanski, A.W., Kurapov, A., Kutlaca, M.,
Lantos, N.A., Lemay, E.P., Lesmana, C.B.J., Louis, W.R., Lueders, A., Malik, N.I.,
Martinez, A., McCabe, K.O., Mehuli´
c, J., Milla, M.N., Mohammed, I., Molinario, E.,
Moyano, M., Muhammad, H., Mula, S., Muluk, H., Myroniuk, S., Naja, R., Nyúl, B.,
O’Keefe, P.A., Osuna, J.J.O., Osin, E.N., Park, J., Pica, G., Pierro, A., Rees, J.,
Reitsema, A.M., Resta, E., Rullo, M., Ryan, M.K., Samekin, A., Santtila, P., Selim, H.
A., Stanton, M.V., Sultana, S., Sutton, R.M., Tseliou, E., Utsugi, A., van Breen, J.A.,
Van Lissa, C.J., Van Veen, K., vanDellen, M.R., V´
azquez, A., Wollast, R., Yeung, V.-L.,
Zand, S., ˇ
Zeˇ
zelj, I.L., Zheng, B., Zick, A., Zú˜
niga, C., Leander, N.P., 2021. Lives versus
Livelihoods? Perceived economic risk has a stronger association with support for
COVID-19 preventive measures than perceived health risk. Sci. Rep. 11 (1).
Nwachukwu, I., Nkire, N., Shalaby, R., Hrabok, M., Vuong, W., Gusnowski, A.,
Surood, S., Urichuk, L., Greenshaw, A.J., Agyapong, V.I.O., 2020. COVID-19
pandemic: Age-related differences in measures of stress, anxiety and depression in
Canada. Int. J. Environ. Res. Public Health 17 (17), 6366.
Onder, G., Rezza, G., Brusaferro, S., 2020. Case-fatality rate and characteristics of
patients dying in relation to COVID-19 in Italy. JAMA 323 (18), 1775–1776.
Park, C.L., Iacocca, M.O., 2014. A stress and coping perspective on health behaviors:
theoretical and methodological considerations. Anxiety, Stress, & Coping. 27 (2),
123–137.
Rodriguez, L.M., Litt, D.M., Stewart, S.H., 2020. Drinking to cope with the pandemic: The
unique associations of COVID-19-related perceived threat and psychological distress
to drinking behaviors in American men and women. Addict. Behav. 110.
Rodríguez-P´
erez, C., Molina-Montes, E., Verardo, V., Artacho, R., García-Villanova, B.,
Guerra-Hern´
andez, E.J., Ruíz-L´
opez, M.D., 2020. Changes in dietary behaviours
during the COVID-19 outbreak connement in the Spanish COVIDiet study.
Nutrients. 12 (6), 1730.
Romano, A., Spadaro, G., Balliet, D., Joireman, J., Van Lissa, C.J., Jin, S., Leander, N.P.,
2020. Cooperation and trust across societies during the COVID-19 pandemic.
J. Cross-Cult. Psychol. in press.
Siahpush, M., Borland, R., Scollo, M., 2003. Smoking and nancial stress. Tobacco
Control. 12 (1), 60.
Smith, P.C., Schmidt, S.M., Allensworth-Davies, D., Saitz, R., 2009. Primary Care
Validation of a Single-Question Alcohol Screening Test. J. Gen. Intern. Med. 24 (7),
783–788.
Stang, P., Lydick, E., Silberman, C., Kempel, A., Keating, E.T., 2000. The Prevalence of
COPD. Chest 117 (5), 354S–359S.
Thornton, J.S., Fr´
emont, P., Khan, K., et al., 2016. Physical activity prescription: a
critical opportunity to address a modiable risk factor for the prevention and
management of chronic disease: a position statement by the Canadian Academy of
Sport and Exercise Medicine. Br. J. Sports Med. 50 (18), 1109.
Thorsteinsson, E.B., Brown, R.F., Owens, M.T., 2019. Modeling the Effects of Stress,
Anxiety, and Depression on Rumination, Sleep, and Fatigue in a Nonclinical Sample.
J. Nerv. Ment. Dis. 207 (5), 355–359.
United Nations. World Population Prospectus. https://population.un.org/wpp/Den
itionOfRegions/ Published 2019. Accessed May 2, 2021, 2021.
Wang, X., Lei, S.M., Le, S., Yang, Y., Zhang, B., Yao, W.u., Gao, Z., Cheng, S., 2020.
Bidirectional inuence of the COVID-19 pandemic lockdowns on health behaviors
and quality of life among Chinese adults. Int. J. Environ. Res. Public Health 17 (15),
5575.
Wardle, J., Haase, A.M., Steptoe, A., Nillapun, M., Jonwutiwes, K., Bellisie, F., 2004.
Gender differences in food choice: The contribution of health beliefs and dieting.
Ann. Behav. Med. 27 (2), 107–116.
Wilsnack, R.W., Wilsnack, S.C., Gmel, G., Kantor, L.W., 2018. Gender differences in binge
drinking: Prevalence, predictors, and consequences. Alcohol Research: Current
Reviews. 39 (1), 57–76.
World Health Organization. Coronavirus disease (COVID-19) pandemic. https://www.
who.int/emergencies/diseases/novel-coronavirus-2019. Published 2021. Accessed
March 3, 2021, 2021.
Zheng, Z., Peng, F., Xu, B., et al., 2020. Risk factors of critical & mortal COVID-19 cases:
A systematic literature review and meta-analysis. J. Infect. 81 (2), e16–e25.
S.-L. Keng et al.
