International Journal of Obesity (2005) 29, 244–250
& 2005 Nature Publishing Group All rights reserved 0307-0565/05 $30.00
www.nature.com/ijo
PAPER
Stress and weight gain in parents of cancer patients
AW Smith1*,w, A Baum1 and RR Wing2
1
University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA; and 2Brown Medical School/The Miriam Hospital,
Providence, RI, USA
OBJECTIVE: To investigate the effects of chronic stress on weight changes and related behavioral changes in parents with a child
who had just been diagnosed with cancer compared to parents with healthy children.
DESIGN: Longitudinal case–control study with assessments occurring over a three-month period following the child’s diagnosis
of cancer.
SUBJECTS: In total, 49 parents of healthy children and 49 parents of cancer patients aged 19–58 y.
MEASUREMENTS: Body weight, diet, physical activity, self-reported mood and stress.
RESULTS: Parents of cancer patients were more likely to gain weight, and experienced significantly greater weight gain over the
3 months than parents of healthy children. The magnitude of weight gain was related to the degree of psychological distress
that the parents experienced. Parents of cancer patients reported lower levels of physical activity and lower caloric intake than
parents of healthy children, with the most marked differences between groups occurring in the area of physical activity.
CONCLUSION: Findings suggest that a major stressor, such as a child’s diagnosis of cancer, is associated with weight gain.
Further research is needed to determine how long these weight gains persist and whether other types of stress also produce
weight gains. Such studies should focus not only on the effect of stress on eating behavior but also on physical activity.
International Journal of Obesity (2005) 29, 244–250. doi:10.1038/sj.ijo.0802835
Keywords: chronic stress; weight change; physical activity; cancer
Introduction
Stress is often cited as an explanation for weight gain,1 but
few studies have actually tested this relationship. Laboratory
paradigms have been used to examine the effect of acute
stressors such as viewing an unpleasant film or delivering a
speech, on subsequent food intake.1–3 These studies have
identified characteristics of the stressors (eg ego-threatening)
and individual characteristics (high restraint or high cortisol
reactivity) that lead to increased consumption after stress.2–5
Fewer studies have examined the impact of long-term, reallife stressors on eating, exercise, or body weight. High levels of
stress have been associated with poorer diets,6 but the
associations between stress and physical activity are less
consistent. In a cross-sectional study of over 12 000 adults,
perceived stress was associated with a higher fat diet and lower
levels of physical activity.7 Stress has also been associated with
increased abdominal obesity8 and weight gain. In a prospective
*Correspondence: AW Smith, National Cancer Institute, Executive Plaza
North, Suite 4001, 6130 Executive Boulevard, MSC 7344, Bethesda, MD
20892-7344, USA.
E-mail: [email protected]
w
Currently a Cancer Prevention Fellow in the Division of Cancer
Prevention, National Cancer Institute, National Institutes of Health.
Received 1 March 2004; revised 19 July 2004; accepted 26 August 2004
study of over 5000 twin pairs, higher stress levels predicted
weight gain over 6 y of follow-up.9 Caregivers have been used as
a model of the effect of ongoing stress.10 Caregivers have been
shown to have higher body mass index (BMI) than controls and
to report substantial weight gain since becoming caregivers.11
The present study was a prospective investigation of the
effects of chronic stress on health behaviors and resulting
weight changes. Parents of a child who had just been
diagnosed with cancer were selected as a group we expected
to be experiencing high levels of stress. Previous studies have
shown that diagnosis and treatment of cancer in a child is
associated with anxiety, stress, somatic complaints, marital
difficulties, and poor adjustment in parents and families.12–17
These parents were studied immediately after the child’s
diagnosis and again 3 months later, and compared to parents
with healthy children. We hypothesized that parents with
children with cancer would report more distress, more calories
consumed, and less physical activity and that they would
exhibit greater weight change over the 3-month follow-up.
Research methods and procedures
Participants
Participants consisted of 49 parents of pediatric cancer
patients (32 female, 17 male) and 49 parents of healthy
Stress and weight
AW Smith et al
245
children (28 female, 21 male). Parents of children with
cancer were recruited through Children’s Hospital of Pittsburgh within 2 weeks after diagnosis. They were identified by
a social worker who asked if they would be willing to hear
about a study of the impact of their child’s health. Parents of
healthy children (defined as those who had a child who had
not been diagnosed with a major illness in the past 3
months) were recruited through neighborhood flyers and
local newspaper advertisements. They were asked to participate in a study examining the effects of child health on
parents’ behaviors. Eligibility included caring for a child
between ages 1 and 18, and the caregivers could not
themselves have been diagnosed with a major illness in the
past 3 months. Female participants were excluded if they
reported pregnancy or were planning to become pregnant
during the 3-month follow-up period.
Procedure
Participants were assessed within 2 weeks of the child’s
cancer diagnosis (Time 1) and again 3 months later (Time 2).
For parents with ill children, sessions were scheduled around
inpatient or outpatient visits. Sessions were held at Children’s Hospital of Pittsburgh and the University of Pittsburgh
Cancer Institute; participants were reimbursed $15 at each
session.
Measures
Body weight was assessed using a portable digital scale that
was calibrated with a balance beam scale at the Obesity and
Nutrition Resource Center at the University of Pittsburgh.
Participants were weighed in street clothes without shoes.
Eating behavior was assessed with a 24-h dietary recall asking
about food eaten the day before the visit. Parents were
interviewed about the amount of food and drink they
consumed, the method of preparation, and the time and
place of consumption. These assessments were then analyzed
using the Minnesota Nutrition Data System (NDS) that
allowed for computation of calories and fat consumed.
Previous research has shown that this approach is more
reliable and accurate than food frequency questionnaires.18
Participants also completed the Three-Factor Eating Questionnaire,19 which assesses the level of dietary restraint (the
degree to which individuals cognitively control their calorie
consumption), disinhibition (susceptibility to loss of control
of eating) and hunger.
Physical activity was assessed using the Paffenbarger
Activity Questionnaire.20 This questionnaire asks about stairs
climbed, blocks walked, and light, medium, and heavy
recreational activities during the preceding week and
provides an estimate of energy expenditure per week in
kilocalories. This measure has been widely used in studies of
exercise and weight change21 and is correlated with weight
changes over time.22 Due to the skewed distribution of
physical activity, this measure was log transformed for
analysis. We also assessed time (hours) spent watching
television and time spent sitting during the week.
Current stress was measured using the short form of the
Perceived Stress Scale.23 This scale measures the extent to
which respondents appraise their lives as unpredictable,
uncontrollable, and overloaded, and has been used in
research examining caregiver stress.10 Background stress
was measured with the Recent Life Changes Questionnaire.24
This scale was designed to assess prospective life change.
Participants indicated whether they had experienced any of
the list of 55 major life events in the past 3 months. Scores
were then generated based on the reported number and
severity of recent life events. One item on this questionnaire
measures a major change in the health or behavior of a
family member. The severity score of this item was calculated
and used separately in analyses to indicate the specific effects
of the child’s diagnosis (Impact of Child’s Illness). Mood was
measured using the Profile of Mood States.25 The 65-item
scale contains six subscales measuring Fatigue, Tension–
Anxiety, Vigor, Anger–Hostility, Depression–Dejection and
Confusion–Bewilderment. This scale has been widely used in
healthy individuals, as well as in cancer patients and their
families.26
Data analysis
Demographic characteristics of parents with healthy children and those with a child with cancer were compared
using one-way ANOVAs and w2 statistics. In all, 70 of the
parents across both groups were married or partnered,
creating 35 parent-pairs. Since analyses indicated that there
were no significant correlations between two parents from
the same family in baseline weight (r ¼ 0.002, P ¼ 0.99) or
weight change (r ¼ 0.17, P ¼ 0.33), all cases were treated
independently.
The main outcome of interest was change in body weight
over time. The first analysis examined the percentage of
parents of children with cancer vs parents of healthy children
who gained weight, using w2 analysis. Body weight was then
examined in a 2 (group: healthy, not healthy) 2 (time: Time
1, Time 2) repeated measures ANOVA. Behavioral variables
including total calories expended in physical activity, hours
spent watching television, calories consumed and percent of
calories from fat were also evaluated with repeated measures
ANOVA. Analyses then examined differences in physical
activity and diet between the 20% of the sample, who gained
the most weight compared to the 20% of the sample who lost
the most weight using 2 (group: weight gain, weight loss) 2
(time: Time 1, Time 2) repeated measures ANOVAs.
To confirm that parents of children with cancer were more
stressed, univariate ANOVAs were used to compare parents of
children with cancer to parents of healthy children on Time
1 scores of Impact of Child’s Illness. Analyses examining the
effects of psychological distress across various measures
(including stress, depression, anxiety, number and severity
of life events) were performed using multivariate tests
International Journal of Obesity
Stress and weight
AW Smith et al
246
(MANOVA). Correlations between these outcome measures
ranged from 0.28 to 0.87; therefore, all distress variables were
examined within one model. Univariate and follow-up tests
were reported when relevant. Analyses were then performed
examining differences in distress variables in the 20% of the
sample that gained the most weight compared to the 20%
that lost the most weight using similar MANOVA models.
Finally data were analyzed to examine explanations for
observed changes in weight. Correlations were computed
between the behavioral variables, the psychosocial variables,
and weight change (Time1 to Time 2) for each group
separately and across groups.
differences between the two groups on any other demographics. Therefore, education and smoking status were
included as covariates in all analyses.
A significantly greater percentage of parents of cancer
patients (63%) gained weight over the 3 months following
their child’s diagnosis than did parents with healthy children
(31%), w2(2) ¼ 12.02, Po0.01 (see Figure 1). Stressed parents
also had a broader range of weight change, from 3.6 to
8.9 kg (8 to 20 lbs), compared to 3.9 to 3.2 kg (9 to
7.0 lbs), in control parents (see Figure 2). Moreover, there was
a significant Group Time interaction F(1,94) ¼ 15.03,
Results
Table 1 compares the parents of healthy children and
children with cancer on demographic variables. Parents
ranged in age from 19 to 58 y (median ¼ 38), and had an
average BMI of 27.30 kg/m2. In all, 85% of the participating
parents were Caucasian and 14% were African American.
Parents of healthy children were more educated than parents
of cancer patients, w2(1) ¼ 4.80, Po0.05, and were also less
likely to be smokers, w2(1) ¼ 5.24, Po0.05, but there were no
Table 1
Means (s.d.) and counts for Time 1 demographic characteristics
Parents of cancer Parents of healthy
patients (n ¼ 49) children (n ¼ 49)
Age
Weight (kg)
Body mass index
Current smoker
(no/yes)
Gender (female/male)
Ethnicity (Caucasian/
African American)
Education
(rHS/ZCollege)
35.47 (8.98)
79.74 (18.78)
27.61 (6.59)
31/18
38.41 (8.22)
78.77 (20.30)
27.00 (6.10)
41/8
F(1,96) ¼ 2.86
F(1,96) ¼ 0.06
F(1,96) ¼ 0.22
w2(1) ¼ 5.24*
32/17
43/6
28/21
40/8
w2(1) ¼ 0.69
w2(1) ¼ 0.38
20/29
10/39
w2(1) ¼ 4.80*
*Po0.05.
Figure 2 Parental weight change from Time 1 to Month 3.
International Journal of Obesity
Test statistic
Figure 1 Percent of parents who gained weight.
Stress and weight
AW Smith et al
247
Table 2
Behavioral variables by parent group
Analysis of variancea
Parents of cancer patients
Variable
Weight (kg)
M
s.d.
Physical activity
M
s.d.
Hours of TV
M
s.d.
Calories
M
s.d.
% cal/fatb
M
s.d.
Time 1
Time 2
Parents of healthy children
Time 1
Time 2
79.74
18.78
81.50
19.32
78.77
20.30
78.93
20.55
403.61
640.38
593.13
741.08
1503.19
1911.04
1433.08
1334.02
4.23
1.56
3.50
1.58
3.43
1.16
3.47
1.06
2022.41
1007.89
2287.75
1177.52
2584.51
1172.38
2311.89
1566.76
33.11
9.75
34.57
10.47
34.51
10.12
33.60
9.34
Group
Time
Group Time
F(1,94)
F(1,94)
F(1,94)
0.19
0.22
15.65**
43.38**
0.02
6.04*
1.22
0.83
5.84*
4.48*
2.99
2.46
0.04
0.01
1.13
a
Covariates include education and smoking. bPercent of calories from fat. *Po0.05, **Po0.0001.
Po0.0001. Parents of cancer patients gained 1.76 kg over the
3 months, t(48) ¼ 4.49, Po0.0001, whereas parents of
healthy children had no weight change, mean ¼ 0.16 kg;
t(48) ¼ 0.69, P40.05.
There was no evidence that weight changes in parents of ill
children were related to level of restraint, (r ¼ 0.10, P40.05),
disinhibition (r ¼ 0.17, P40.05), or hunger (r ¼ 0.05, P40.05).
In all, 50% of those participants above the median in restraint
gained weight and 45% of those below the median in restraint.
Examination of behavioral variables including physical
activity and inactivity, and eating behavior showed differences between parent groups and over time (see Table 2). For
physical activity, there was a highly significant main effect of
Group, F(1,94) ¼ 43.38, Po0.0001. Parents of cancer patients
reported only 400–500 kcal/week of physical activity compared to 1400–1500 kcal/week in parents of healthy children.
There was also a significant Group Time interaction
(F(1,94) ¼ 6.04, Po0.05); parents of cancer patients increased
their activity physical activity from Time 1 to Time 2,
t(48) ¼ 2.50, P ¼ 0.01, while parents of healthy children
reported no changes in activity level, t(48) ¼ 0.49, P40.05.
Analyses of sedentary behavior (hours spent watching
television) showed a significant Group x Time interaction,
F(1,94) ¼ 5.84, P ¼ 0.01. Follow-up tests suggested that
parents with ill children watched more hours of television
at Time 1 than parents of healthy children (t(48) ¼ 2.54,
P ¼ 0.01), but the groups were similar at Time 2. Differences
between groups were also found for diet. In contrast to our
original hypothesis, parents of cancer patients reported
consuming fewer calories than control parents, F(3.91) ¼
4.48, Po0.05. There were no significant differences between
groups on the percentage of calories consumed from fat.
We then examined differences in physical activity and diet
by the 20% who gained the most vs the 20% who lost the
most in the whole sample. There were no significant
differences between weight gainers and weight losers in
terms of physical activity, time spent watching television or
on either dietary variable (calories or percentage of calories
consumed from fat).
The next set of analyses examined effects of distress by
parent group. As expected, parents of cancer patients
reported more extensive impact of having an ill child at
Time 1 than did parents of healthy children, F(1,88) ¼ 97.71,
Po0.0001. Parents of cancer patients also reported more
stress, depression, anxiety, and a greater number and severity
of major life events over time than did control participants.
Multivariate analyses of all distress measures indicated main
effects of Group F(5,73) ¼ 6.780, Po0.0001; Time,
F(5,73) ¼ 2.72, Po0.05; and a Group Time interaction,
F(5,73) ¼ 9.50, Po0.0001. Univariate analyses showed Group
x Time interactions on the number and severity of recent life
changes, on depression and anxiety (Pso0.05; see Table 3).
Main effects of Group were seen on all distress variable (all
Pso0.01). Analyses also indicated significant effects of Time
on the number of recent life changes, as well as on anxiety
(Pso0.05). Follow-up tests revealed that at Time 1, parents of
cancer patients reported more life changes t(48) ¼ 3.96,
Po0.0001; higher life change scores, t(48) ¼ 4.32,
Po0.0001, more depression, , t(45) ¼ 5.87, Po0.0001, and
more anxiety, t(45) ¼ 8.98, Po0.0001 than at Time 2. Parents
International Journal of Obesity
Stress and weight
AW Smith et al
248
Table 3
Distress by parent group
Analysis of variance a
Parents of cancer patients
Variable
Parents of healthy children
Time 1
Time 2
Time 1
Time 2
Stress
M
s.d.
8.15
3.46
6.46
3.00
5.49
2.55
4.47
2.89
# RLCb
M
s.d.
7.41
3.93
5.06
3.93
4.16
2.99
3.94
3.61
RLCb score
M
s.d.
448.57
279.12
266.46
271.95
188.41
180.10
161.65
186.01
Depression
M
s.d.
42.50
25.99
24.02
22.06
15.73
16.13
13.61
14.79
Anxiety
M
s.d.
58.64
22.66
34.66
18.59
29.05
17.14
24.65
14.01
Child Ill c
M
s.d.
80.89
31.30
Group
Time
Group Time
F(5,73)
F(5,73)
F(5,73)
15.12***
0.14
0.95
12.46**
9.07**
3.63*
16.88***
0.78
5.47*
20.73***
2.36
17.00***
27.08***
5.70*
35.40***
30.21***
15.85
30.95
a
Multivariate F-scores presented; covariates include education and smoking. bRecent life changes. cChild Ill ¼ impact of child’s illness, df (3,88). *Po0.05, **Po0.01,
***Po0.0001.
of healthy children did not report differences in these
distress measures (all Ps40.05).
Multivariate analyses comparing differences in distress
between weight gainers vs weight losers revealed significant
differences by Group F(5,29) ¼ 3.25, Po0.05 and Time
F(5,29) ¼ 3.66, P ¼ 0.01, but no significant interaction. Univariate tests showed significant differences in anxiety, with
weight gainers reporting more anxiety (M ¼ 48.00,
s.e. ¼ 3.81) than weight losers (M ¼ 34.51, s.e. ¼ 4.61),
Po0.5. No significant group differences were found for other
distress variables.
Correlations were then computed between weight change
and the behavioral and distress variables for each of the
parent groups. Among parents of healthy children, hours
spent watching television was significantly associated with
weight gain (r ¼ 0.35, Po0.05). Among the parents of cancer
patients, weight change was not significantly associated with
any of the behavioral variables. Self-reported impact of having
an ill child (r ¼ 0.30, Po0.05) and severity of recent life events
at Time 1 (r ¼ 0.32, Po0.05) were the strongest predictors of
weight gain among the parents of cancer patients.
Discussion
In the current study, parents of children recently diagnosed
with cancer reported marked psychological distress, anxiety
International Journal of Obesity
and depression. Following these parents prospectively can
thus provide important information about stress-associated
changes in eating and exercise behavior and weight changes.
We found that parents with children diagnosed with cancer
reported less physical activity, more sedentary behavior, and
lower caloric intake than parents with healthy children.
Most importantly, we found that parents of children with
cancer were more likely to gain weight and gained more
weight over 3 months than parents of healthy children.
Investigating a stressor of this nature and magnitude is
challenging and imposes a number of limitations on study
design and implementation. Most importantly, we were not
permitted to meet with these parents prior to their diagnosis.
Thus, we do not have a prediagnosis baseline assessment;
and therefore we cannot rule out prediagnosis differences
between parent groups. Although our original goal was to
recruit friends of the parents of cancer patients as a
comparison group, to control for socioeconomic factors,
this proved to be logistically impossible. Moreover, in order
to maximize participation, we needed to meet with parents
of cancer patients in the hospital or outpatient clinic. Thus,
study personnel were not blinded to parent group status.
Future investigators may be able to learn from our experiences to improve study methodology in research on weight
change following major life stressors.
Keeping these limitations in mind, findings in this study
are consistent with the hypothesis that stress is associated
Stress and weight
AW Smith et al
249
with weight gain and offer prospective, naturalistic confirmation of cross-sectional studies. On average, parents of
children with cancer gained more than 1.5 kg over 3 months,
or 0.5 kg (1 lb) per month for 3 months. Whether the weight
gain would have continued over long periods of time
remains unknown. In contrast to laboratory studies, the
present investigation found no evidence that dietary restraint, disinhibition or hunger was related to this weight
change. Perhaps, these variables are more strongly associated
with increased consumption in a single meal setting than
over the 3-month time period studied here.
Parents of ill children reported lower physical activity,
lower dietary intake, and lower amounts of television
watching than parents of healthy children. Interpretation
of these data is difficult for several reasons. First, the 24-h
recall and the Paffenbarger Activity Questionnaire both
provide data on current behavior, but there is no information about the eating and exercise behaviors of the two
groups before the diagnosis of cancer. Thus, change in these
behaviors from their prestress levels cannot be determined.
Moreover, self-report measures are known to be subject to
bias, with overestimation of physical activity and underestimation of diet.27 Inaccurate reporting may particularly
occur in stressed parents as their high level of stress may
interfere with their ability to recall their own behaviors;
likewise, inconsistency in their behaviors from day to day
may not be accurately captured in the diet and exercise
assessments.
Interestingly, the most dramatic difference in weight
related behavior between parents of children diagnosed with
cancer vs parents of healthy children occurred is the area of
physical activity (rather than diet). Parents with an ill child
reported expending 400–500 kcal/week in activity, compared
to 1400–1500 kcal/week in parents of healthy children. This
finding probably reflects the major change in daily routine
that has occurred after the cancer diagnosis (eg spending a
great deal of time in the hospital). Few, if any, previous
studies of stress have examined the role of decreases in
physical activity; rather the emphasis has been on changes in
diet, which may not reflect the main area of behavior change
related to body weight. Interventions targeting families of
cancer patients or others undergoing major life stressors may
therefore want to focus more on total energy balance and to
emphasize exercise programs.
In the present study, parents of cancer patients reported
eating less than parents of healthy children. Previous studies
of stress and eating have typically reported increased, rather
than decreased consumption1 and hence we had expected
such increases. The intensity of the stress studied here may
well have been related to the decreased, rather than the
increased consumption. The present study did not show that
the diet or exercise behaviors were correlated with the
changes in weight, perhaps due to the methodological issues
noted above. However, the energy balance data actually fit
quite well with the average weight gain observed. The
difference in energy balance between parents of healthy
and ill children averaged 700 kcal/week (1000 kcal/week less
exercise and 300 kcal/week less intake in parents of ill
children) over 12 weeks or an 8400 kcal difference overall.
This would correspond to a 1.1 kg difference in weight
change, whereas the observed difference was 1.76 kg. There
were no significant differences found in physical activity or
diet variables in those who gained the most weight vs weight
losers. However, this may have been due to the small sample
size, as these analyses were limited to 40% of the total
sample.
This study found that the 20% of participants who gained
the most weight reported more anxiety than weight losers.
Further, self-reported number and adjustment to severe life
events, depression and anxiety were all correlated with
weight gain, not weight loss. While we expected that this
relationship would be mediated through behavioral variables, behavioral factors were not associated with weight
change in our sample. This is likely due to methodological
limitations in measuring physical activity and diet, as well as
our small sample size. Our results were comparable to other
research showing stronger associations between depressive
symptoms and weight change than between behavioral
factors and weight change.28
Diagnostic criteria for clinical depression includes both
weight gain and weight loss.29 However, subclinical levels of
depressive symptoms have been shown to predict weight
gain.28,30,31 Previous research on anxiety and weight change
is limited. Studies examining weight and anxiety have
focused primarily on eating disorders,32 making it difficult
to draw conclusions about the effects of anxiety on weight
gain vs weight loss. Previous research has found higher levels
of stress were associated with weight gain over 6- and 15-y
periods.9 Other research found that caregiver stress and
depression were associated with equal numbers of weight
gainers and weight losers compared to those with stable
weight.10 Additional studies with other naturalistic stressors
are needed to determine whether these stressful situations
are associated with weight losses, gains, or possibly both.
In conclusion, we found that parents of children recently
diagnosed with cancer reported significant psychological
distress and experienced weight gains averaging 1.76 kg over
3 months compared to weight stability in nonstressed
parents. Further research is needed to determine whether
weight gain occurs commonly in response to major life
stressors and the magnitude and duration of these weight
gains. Few studies have examined the effect of major
naturalistic stressors on physical activity. Differences were
most pronounced at Time 1, just after the cancer diagnosis.
In future studies it will be important to determine whether
there are differences prior to diagnosis and to investigate the
initial time period around the child’s diagnosis with
comparable controls. Finally, given the significant differences in activity levels of parents of healthy vs ill parents in
the present study and the impact of physical activity on
overall health,33 future research should focus more on the
effect of stress on physical activity as well as on eating and
International Journal of Obesity
Stress and weight
AW Smith et al
250
weight change. Better understanding of the effects of stress
on families of cancer patients may help practitioners
intervene during this vulnerable time.
17
18
References
1 Greeno CG, Wing RR. Stress-induced eating. Psychol Bull. 1994;
115: 444–464.
2 Lattimore PJ. Stress-induced eating: an alternative method for
inducing ego-threatening stress. Appetite 2001; 36: 187–188.
3 Epel E, Lapidus R, McEwen B, Brownell K. Stress may add bite to
appetite in women: a laboratory study of stress-induced cortisol
and eating behavior. Psychoneuroendocrinology 2001; 26: 37–49.
4 Rutledge T, Linden W. To eat or not to eat: affective and
physiological mechanisms in the stress-eating relationship.
J Behav Med 1998; 21: 221–240.
5 Herman CP, Polivy J, Lank CN, Heatherton TF. Anxiety, hunger,
and eating behavior. J Abnorm Psychol 1987; 96: 264–269.
6 Laitinen J, Ek E, Sovio U. Stress-related eating and drinking
behavior and body mass index and predictors of this behavior.
Prev Med 2002; 34: 29–239.
7 Ng DM, Jeffery RW. Relationships between perceived stress and
health behaviors in a sample of working adults. Health Psychol
2003; 22: 638–642.
8 Rosmond R, Dallman MF, Björntorp P. Stress-related cortisol
secretion in men: relationships with abdominal obesity and
endocrine, metabolic and hemodynamic abnormalities. J Clin
Endocr Metab 1998; 83: 1853–1859.
9 Korkeila M, Kaprio J, Rissanen A, Koshenvuo M, Sorensen TI.
Predictors of major weight gain in adult Finns: stress, life
satisfaction and personality traits. Int J Obes Relat Metab Disord
1998; 22: 949–957.
10 Fredman L, Daly MP. Weight change: an indicator of caregiver
stress. J Aging Health 1997; 9: 43–69.
11 Vitaliano PP, Russo J, Scanlan JM, Greeno CG. Weight changes in
caregivers of Alzheimer’s care recipients: psychobehavioral predictors. Psychol Aging 1996; 11: 155–163.
12 Dahlquist LM, Czyzewski DI, Jones CL. Parents of children with
cancer: a longitudinal study of emotional distress, coping style,
and marital adjustment two and twenty months after diagnosis.
J Pediatr Psychol 1996; 21: 541–554.
13 Grootenhuis MA, Last BF. Adjustment and coping by parents of
children with cancer: a review of the literature. Support Care
Cancer 1997; 5: 466–484.
14 Hoekstra-Weebers JE, Jaspers JP, Kamps WA, Klip, EC. Risk factors
for psychological maladjustment of parents of children with
cancer. J Am Acad Child Adolesc Psychiatry 1999; 38: 1526–1535.
15 Sawyer MG, Antoniou G, Toogood I, Rice M, Baghurst PA. A
prospective study of the psychological adjustment of parents and
families of children with cancer. J Paediatr Child Health 1993; 29:
352–356.
16 Sawyer MG, Streiner DL, Antoniou G, Toogood I, Rice M.
Influence of parental and family adjustment on the later
International Journal of Obesity
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
psychological adjustment of children treated for cancer. J Am
Acad Child Adolesc Psychiatry 1998; 37: 815–822.
Speechly KN, Noh S. Surviving childhood cancer, social support, and
parents’ psychological adjustment. J Pediatr Psychol 1992; 17: 15–31.
Buzzard IM, Faucett CL, Jeffery RW, McBane L, McGovern P,
Baxter JS, Shapiro AC, Blackburn GL, Chlebowski RT, Elashoff
RM, Wynder EL. Monitoring dietary change in a low-fat
intervention study: advantages of using 24-hour dietary recalls
vs. food records. J Am Diet Assoc 1996; 96: 574–579.
Stunkard AJ, Messick S. The three-factor eating questionnaire to
measure dietary restraint, disinhibition, and hunger. J Psychosom
Res 1985; 29: 71–83.
Paffenbarger RS, Wing AL, Hyde RT. Physical activity as an index
of heart attack risk in college alumni. Am J Epidemiol 1978; 108:
161–175.
Wing RR, Epstein LH, Patternostro-Bayles M, Kriska A, Nowalk
MP, Gooding W. Exercise in a behavioral weight control
programme for obese patients with Type 2 (non-insulin-dependent) diabetes. Diabetologia 1988; 31: 902–909.
Harris JK, French SA, Jeffery RW, McGovern PG, Wing RR. Dietary
and physical activity correlates of long-term weight loss. Obes Res
1994; 2: 307–313.
Cohen S, Kamarck T, Mermelstein R. A global measure of
perceived stress. J Health Soc Behav 1983; 24: 385–396.
Rahe RH. Epidemiological studies of life changes and illness. Int J
Psychiatry Med 1975; 6: 131–146.
McNair PM, Lorr M, Droppleman LF. POMS manual, 2nd edn. San
Diego; Educational and Industrial Testing Service; 1981.
Cassileth BR, Lusk EJ, Brown LL, Cross PA. Psychosocial status of
cancer patients and next of kin: normative data from the Profile
of Mood States. J Psychol Oncol 1985; 3: 99–105.
Lichtman SW, Pisarska K, Berman ER, Pestone M, Dowling H,
Offenbacher E, Weisel H, Heshka S, Matthews DE, Heymsfield SB.
Discrepancy between self-reported and actual caloric intake and
exercise in obese subjects. N Engl J Med 1992; 327: 1893–1898.
Haukkala A, Uutela A, Salomaa V. Depressive symptoms, cynical
hostility, and weight change: a 3-year follow-up among middle
aged men and women. Int J Behav Med 2001; 8: 116–133.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn. American Psychiatric Association:
Washington, DC; 1994.
Barefoot JC, Heitmann BL, Helms MJ, Williams RB, Surwit RS,
Siegler I. Symptoms of depression and changes in body weight
from adolescents to mid-life. Int J Obes Relat Metab Disord 1998;
22: 688–694.
DiPietro L, Anda RF, Williamson DF, Stunkard AJ. Depressive
symptoms and weight change in a national cohort of adults. Int J
Obes Relat Metab Disord 1992; 16: 745–753.
Godart NT, Flament MF, Perdereau F, Jeammet P. Comorbidity
between eating disorders and anxiety disorders: a review. Int J Eat
Disord 2002; 32: 253–270.
U.S.D.H.H.S.. Physical activity and health: a report of the Surgeon
General, US Department of Health and Human Services, Centers
for Disease Control and Prevention, National Center for Chronic
Disease Prevention and Health Promotion, Atlanta, GA, 1996.