Change in body image and psychological well-being during behavioral obesity treatment: Associations with weight loss and maintenance.

Abstract

This study reports on outcomes from a behavioral obesity treatment program, evaluating if treatment-related changes in body image and psychological well-being are predictors of weight change during treatment and after follow-up. Participants were 142 overweight/obese women (BMI=30.2+/-3.7kg/m(2); age=38.3+/-5.8 years) participants in a behavioral treatment program consisting of a 4-month treatment period and a 12-month follow-up. Psychosocial variables improved during treatment and these changes were correlated with 4-month weight reduction. Short-term changes in body size dissatisfaction (p=.002) and mood (p=.003) predicted long-term weight loss. Additional results suggest that there might be a predictive role of short-term changes in body size dissatisfaction and self-esteem on long-term weight loss after accounting for initial weight change (p<.028). We conclude that, along with weight changes, cognitive and affect-related processes influenced during obesity treatment may be related long-term success, in some cases independently of initial weight loss.

Full text

Page 1

Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance
Anto ´nio L. Palmeiraa,b,*, Teresa L. Brancob, Sandra C. Martinsb, Cla ´udia S. Mindericob, Marlene N. Silvab,
Paulo N. Vieirab, Jose ´ T. Baratab, Sido ´nio O. Serpab, Luı ´s B. Sardinhab, Pedro J. Teixeirab
aUniversity Luso ´fona de Humanidades e Tecnologias, Lisbon, Portugal
bFaculty of Human Kinetics, Technical University of Lisbon, Lisbon, Portugal
Introduction
Besides changes in weight, participantsin weight loss programs
often report improved life satisfaction and feeling more positive
about their bodies and their new-learned abilities, such as being
physically active and self-managing their weight. Research has
shown that these psychological improvements can be associated
with weight changes (Blaine, Rodman, & Newman, 2007;
Maciejewski, Patrick, & Williamson, 2005). While treatment-
related psychosocial changes (e.g., improved body image) are
considered valuable outcomes per se and a natural consequence of
losing weight, they are not necessarily viewed as mediators or
enabling factors for behavior change. Stice and Shaw (2002)
theorizedonthesepossible influences,showingthat twopathways
illustrate the role of body dissatisfaction on the development of
maladaptive eating behaviors, which may be related to poor
weight control and obesity: the dieting and the negative affect
path. For example, body dissatisfaction reductions might counter-
act extreme dieting patterns and negative affect that could arise
during a weight loss program. Additionally, the results from
project EAT have shown that, in adolescent females, baseline lower
body satisfaction predicted higher levels of dieting, unhealthy
weight control behaviors and binge eating after 5 years (Neumark-
Sztainer, Paxton, Hannan, Haines, & Story, 2006), leading the
authors to conclude that body image enhancement should protect
against the development of unhealthy behaviors linked to obesity.
Therefore, cognitive and affect-related changes that occur during
weight management may represent more than positive outcomes
and can in fact also influence (i.e., mediate) the effects of an
intervention (Palmeira et al., 2009). To the extent this occurs, these
psychosocial changes should be investigated not only as depen-
dent variables but also as behavior modification agents (Kahne-
man, Diener, & Schwarz, 1999). This line of inquiry appears
especially relevant for interventions that include regular physical
exercise, considering its well-known positive effects on emotions,
self-esteem, depression, and other psychological variables (Biddle
& Mutrie, 2001).
The Reciprocal Effects Model is a theoretical framework used in
educational psychology (Marsh & Craven, 2006), which could
provide valuable insight to the understanding of behavior change
in the context of weight management programs. This model
describes the causal relation between a specific component of
psychological well-being (e.g., fewer depressive symptoms) and a
performance indicator (e.g., behavior or weight change) as
dynamical and reciprocal. The reciprocal determinism proposed
in Bandura’s Social Cognitive Theory suggests similar processes,
Body Image xxx (2010) xxx–xxx
A R T I C L EI N F O
Article history:
Received 12 August 2009
Received in revised form 19 March 2010
Accepted 22 March 2010
Keywords:
Weight management
Body image
Psychological well-being
Women
A B S T R A C T
This study reports on outcomes from a behavioral obesity treatment program, evaluating if treatment-
related changes in body image and psychological well-being are predictors of weight change during
treatment and after follow-up. Subjects were 142 overweight/obese women (BMI = 30.2 ? 3.7 kg/m2;
age = 38.3 ? 5.8 years) participants in a behavioral treatment program consisting of a 4-month treatment
periodanda12-monthfollow-up.Psychosocialvariablesimprovedduringtreatmentandthesechangeswere
correlated with 4-month weight reduction. Short-term changes in body size dissatisfaction (p = .002) and
mood (p = .003) predicted long-term weight loss. Additional results suggest that there might be a predictive
role of short-term changes in body size dissatisfaction and self-esteem on long-term weight loss after
accounting for initial weight change (p < .028). We conclude that, along with weight changes, cognitive and
affect-relatedprocessesinfluencedduringobesitytreatmentmayberelatedlong-termsuccess,insomecases
independently of initial weight loss.
? 2010 Elsevier Ltd. All rights reserved.
* Corresponding author at: Universidade Luso ´fona de Humanidades e Tecnolo-
gias, A\C: Faculdade de Educac ¸a ˜o Fı ´sica e Desporto, 1749-028, Lisboa, Portugal.
Tel.: +351 217 515 587; fax: +351 217 515 557.
E-mail address: antonio.palmeira@ulusofona.pt (A.L. Palmeira).
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002
Contents lists available at ScienceDirect
Body Image
journal homepage: www.elsevier.com/locate/bodyimage
1740-1445/$ – see front matter ? 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bodyim.2010.03.002

Page 2

i.e., that behavior both influences and is influenced by individual
factors and the environment (Bandura, 1997).
In weight management, the psychological variables that
potentially could be involved in these causal associations include
self- and body-esteem variables, mood, and depression (Kolotkin,
Crosby, Williams, Hartley, & Nicol, 2001; Wadden, Brownell, &
Foster,2002).Thisstatementissupportedbyseveralfindings,which
show that changes in psychosocial variables are not necessarily
associated with weight loss. For example, improvements in body
image (Foster, Wadden, & Vogt, 1997; Ramirez & Rosen, 2001), self-
esteem (Maciejewski et al., 2005), and depression (Wadden et al.,
1997) are inconsistently related to weight loss during obesity
treatment (Maciejewski et al., 2005). Nevertheless, these improve-
ments could be interpreted as influencing weight-related voluntary
behaviors, at least for some individuals, for example by strengthen-
ing behavioral adherence to the program (Baker & Brownell, 2000),
or possibly by the increment of psychosocial resources to cope with
the demands of the treatment tasks. Schwartz and Brownell
summarizedthisrationale whentheystated that ‘‘one could readily
imaginethattheabilitytoloseweightwouldbeimprovedbyreliefof
depression, anxiety, poor self-esteem, or body image distress’’
(Schwartz & Brownell, 2004, p. 53).
Toourknowledge,thishasnotbeensystematicallyevaluatedand
more studies are clearly needed, especially to ascertain whether
treatment-related psychosocial changes predict long-term weight
loss and maintenance.Teixeira etal. (2006)isoneofthe few studies
that provided results for this question, reporting that changes in
psychosocial variables related to exercise, eating and body image
during a 4-month weight management program were predictive of
16-month weight loss. In this study, exercise-related motivational
variables appeared to play a more important role in the long-term
results, while some other psychosocial variables (e.g., eating
behavior) were more predictive of 4-month weight changes.
Therefore, the focus of this study is to explore the hypothesis
that improvements in psychological variables during obesity
treatment are not only consequences but also an integral part of
the causal behavioral chain that ultimately leads to long-term
weight reduction. Specifically, the aim of this study is to analyze if
short-term changes in body image and psychological well-being
(self-esteem, mood, and depression) predict short and long-term
weight change, in overweight and moderately obese women
participating in a University-based weight management program.
First, we evaluated if improvements in body image and psycho-
logical well-being occurred during the first 4 months of the
program. Secondly, we assessed whether these psychosocial
changes were associated with treatment-related weight change.
Finally, we tested if short-term changes in psychosocial variables
influenced long-term weight change, before and after accounting
for treatment-related weight change.
Methods
Participants
Participants were recruited from the community for a weight
management program through newspaper ads, a website, email
messages, and flyers. Subjects were required to be older than 24
years, pre-menopausal, not pregnant, have a BMI between 25 and
40 kg/m2, and free from major disease to be eligible for the study.
One hundred and fifty-five women volunteered to participate. After
baseline measurements, some women were excluded (n = 6 due
to exclusion criteria detected during testing) and others (n = 7)
decidednottoparticipateduetopersonalincompatibilities.Hence,a
sample of 142 women (BMI = 30.2 ? 3.7 kg/m2; age = 38.3 ? 5.8
years; 47.7% attended college) started the 16-month University-based
behavioral obesity treatment program. During the first 4 months, all
participants received the same intervention, after which they
were randomized into two maintenance programs: (a) monthly
meetings;(b)monthlymeetingsplustwostructuredweekendexercise
sessions; or to a control group with no further contact. The duration
of this second phase was 12 months. Maintenance conditions and
control were pooled together for the current analysis since weight
change exclusively during maintenance was not different among
groups (percent body weight change: control = 1.3 ? 4.3%; monthly
meetings group = ?0.7 ? 5.3%; monthly meetings plus exercise
group = ?1.0 ? 5.2%; p = .192). Attrition was 6% at 4 months and
33% at 16 months. Data were analyzed for completers-only and also
using data imputation methods (e.g., Last Observation Carried
Forward), to control for attrition-related bias. However, since
psychosocial changes during treatment were not different between
completers and dropouts (p > .168) and similar magnitude in
correlations with 0–4 and 16 months’ weight change was observed
in both groups, we followed the more conservative completers-only
procedure. Therefore, the sample under analysis was reduced to 96
participants (BMI = 30.1 ? 3.6 kg/m2; age = 38.9 ? 5.7 years), who
completed the 16 months duration of the program. All participants
agreed to refrain from participating in any other weight loss program
and gave written informed consent prior to participation in the study.
The Faculty of Human Kinetics’ Ethics Committee approved the study.
Intervention
The main intervention included fifteen weekly meetings, which
lasted 120 min. Attendance averaged 83% and each group included
32–35 women. Theintervention was generally basedon the LEARN
weight management program (Brownell, 1997), and included
educational content and practical applications in the areas of
physical activity and exercise, diet and eating behavior, behavior
modification, and have been partially described before (Palmeira
et al., 2007; Teixeira et al., 2004). Physical activity topics included
learning the energy cost associated with typical activities,
increasing daily walking and lifestyle physical activity, planning
and implementing a structured exercise plan, and choosing the
right type of exercise, among many others. Examples of covered
nutritiontopics were learningthe caloric,fat, and fiber content and
the energydensityof commonfoods,the roleof breakfastand meal
frequencyforweightcontrol,reducingportionsize,andpreventing
binge and emotional eating. Cognitive and behavioral skills
including self-monitoring, self-efficacy enhancement, dealing with
lapses and relapses, enhancing body image, using contingency
management strategies, and eliciting social support were also part
of the curriculum.
The sessions were conducted by the same team to all treatment
groups, and was composed by two Ph.D.- and six M.S.-level
exercise physiologists, psychologists, and dieticians. Participants
were informed that weight reduction should be understood as a
long-term goal, and that a 5–10% weight loss was an appropriate
goal to be sought at the end of the program.
Instruments
Psychosocial variables
Data were collected at baseline and at the end of treatment.
Participants were required to attend two laboratory sessions in
order to complete all psychosocial assessments, in each evaluation
period. The instruments were Portuguese validated versions of
some of the most commonly used psychosocial instruments in
obesity research.
Body image
Body image was evaluated by three questionnaires, considering
its multidimensional nature. Body size dissatisfaction was
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
2
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002

Page 3

measured with the Body Image Assessment questionnaire (BIA;
Williamson, Davis, Bennett, Goreczny, & Gleaves, 1989), which
consists of nine silhouettes of increasing size, from which subjects
are asked to choose their current (i.e., perceived actual body size)
and ideal figures. The difference was used as a measure of
dissatisfaction. The Body Shape Questionnaire (BSQ; Cooper,
Taylor, Cooper, & Fairburn, 1987), a 34-item instrument with a
6-point Likert scale was used to measure affective, cognitive, and
behavioral dimensions of body image, especially regarding the
experience of, and preoccupation with ‘‘being fat’’. The total score
was used (a = .95), where higher values represent greater
preoccupation with body shape. The Physical Self-Perception
Profile (PSPP; Fox & Corbin, 1989) addresses five different
dimensions of the physical self-concept based on 30 items, rated
on a structured alternative scale. We used two of these
dimensions: Physical Self-Worth (the higher order construct of
this scale) and Body Attractiveness (a estimates of .75 and .80,
respectively).
Psychological well-being
Self-esteem was assessed with the Rosenberg Self-Esteem Scale
(RSES; Rosenberg, 1965), composed by 10 items answered on a 4-
point Likert scale. Higher scores of the RSES represent greater self-
esteem (a = .84). Mood disturbance was assessed with the Profile
of Mood States (POMS; McNair, Lorr, & Droppleman, 1971), which
measures the transient emotional state through 65 items on a 5-
point Likert scale. The questionnaire assesses 6 dimensions of
mood that can be used to calculate a Total Mood Disturbance score
(sum of the negative emotions subtracted by the positive Vigor
dimension, a = .92), which was used in the present study (higher
scores represent greater total mood disturbance). Questions
pertain to emotional states of the previous month. Depression
was evaluated with the Beck Depression Inventory (BDI; Beck &
Steer, 1987), a 21-item inventory measuring several symptoms of
depression. It uses a 4-point ordered scale and results in a total
score (a = .80), where higher scores represent greater level of
depressive symptoms.
Body habitus
Body weight was measured with a standardized procedure at
baseline, treatment’s end (4 months), and at follow-up (16
months), using an electronic scale (SECA model 770, Hamburg,
Germany). Three different variables were calculated to represent
weight outcomes: baseline to 4-monthweight change (short-term,
treatment weight change); baseline to 16-month weight change
(overall, long-term weight change); and 4–16-month weight
change (an indicator of weight stability after treatment).
Statistical procedures
All psychosocial variables were expressed as residuals of the 4-
month value regressed on the baseline value. Weight change was
calculated by the same procedure at 0–4 months and also for 16
months data. A weight maintenance variable was calculated using
the 16-month weight value regressed simultaneously on the
baseline and the 4-month values, to adjust for initial weight and
weightchangeduring treatment.Usingsuch‘‘residualized’’ change
scores creates a value that is orthogonal to the independent
variable score and represents a better measure of change, when
compared with pre–post subtraction procedure (Cohen, Cohen,
West, & Aiken, 2003).
Changes in weight and psychosocial variables were assessed by
paired t-tests and effect sizes, which were classified as small
(<0.30), medium (0.30–0.80), and large (>0.80) (Cohen et al.,
2003). To evaluate whether these changes were associated with
actual weight loss, Pearson correlation coefficients were calcu-
lated.Tostudythepredictivevalueoftreatment-relatedchangesin
psychosocial variables on long-term weight change (before and
after accounting for treatment-related weight change) we used
multiple linear regression (stepwise method, F to enter, p < .05)
and analysis of covariance (ANCOVA).
Results
Weight change from baseline to 4 months was ?3.7 ? 3.9%
(p < .001, ranging from ?17.7% to 6.3%), and from baseline to 16
months was ?4.5 ? 6.7% (p < .001, ranging from ?26.4% to 13.1%)
with large individual variability. These weight changes were not
associated with age (p < .857), baseline weight (p < .505), or
education level (p < .529). No significant change was observed in
the 4–16 months analysis (p = .622). Forty-five percent of the
participants reached the 5% weight loss goal at follow-up while
21.1% reached the 10% objective.
Results for psychosocial variables showed that all body image
variables improved during the intervention (medium effect sizes,
see Table 1). Body size dissatisfaction showed the greatest change,
with a reduction of more than one-half of a standard deviation
(d = ?0.52, p < .001). All psychological well-being measures
improved, especially depression (d = ?0.43, p < .001) and mood
disturbance (d = ?0.45, p < .001).
Associations between changes in psychosocial variables and
weight showed that changes in body attractiveness, body size
dissatisfaction, and total mood disturbance had the strongest
association with 4-month weight change (p < .001, see Table 2).
Changes in physical self-worth and depression showed weaker
Table 1
Change in psychosocial variables during treatment (Lisbon, Portugal, 2003–2005).
Baseline4 months
td
M?SDM?SD
Body image
Body size dissatisfaction (BIA)
Body shape concerns (BSQ)
Body attractiveness (PSPP)
Physical self-worth (PSPP)
2.26?0.84
92.32?27.49
10.94?2.78
11.95?3.77
1.84?0.77
84.04?22.90
12.14?3.11
13.03?3.45
5.43***
2.61**
?4.65***
?2.93**
?0.52
?0.33
0.41
0.30
Psychological well-being
Self-esteem (RSES)
Depression (BDI)
Total mood disturbance (POMS)
22.38?3.58
7.18?5.61
24.03?31.37
23.04?4.11
5.07?4.30
10.87?27.27
?1.73¥
4.96***
5.61***
?0.43
?0.45
Notes:n=96; pairedt-test; d,Cohen’s d: effectsize;BIA:Body ImageAssessment Questionnaire; BSQ: BodyShape Questionnaire;PSPP: Physical Self-PerceptionProfile;RSES:
Rosenberg Self-Esteem Scale; BDI: Beck Depression Inventory; POMS: Profile of Mood States.
**p<.01.
***p<.001.
¥p=.087.
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
3
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002

Page 4

associations, albeit significant, with weight change during treat-
ment (p < .05). Changes in body shape concerns and self-esteem
were not associated with weight loss during treatment. All
previous associations were maintained for long-term weight
change (0–16 months), although with some reductions in the
strength of the correlations (Table 2). Changes in body size
dissatisfaction and self-esteem were associated with weight
change during the maintenance period (4–16 months).
The previous results showed that changes in body size
dissatisfaction, total mood disturbance, and self-esteem were
the associated with long-term weight management. For a more
detailed analysis of these associations, we have followed the
methods present in the paper by Teixeira et al. (2006), creating
three groups of similar size, based on a tertile-split of each of these
three psychosocial variables (‘‘residualized’’ 4-month score), and
weightchangecalculatedforeachgroup.Toreducetheinfluenceof
using arbitrary cut-offs, only the two extreme groups were
compared (Fig. 1). This procedure followed the methods present
in the Teixeira et al. (2006) paper; by showing the amount of
weight change obtained by the tertiles of change in the
psychosocial variables we have a clearer view of the impact of
these changes in the weight loss.
In the short-term, as expected, the group that decreased the
most in body size dissatisfaction and total mood disturbance had
thegreatestweightloss(bothp < .001).Thesamepatternwasseen
forlong-termweightlossesinthegroupswithstronger0–4-month
improvements in body size dissatisfaction (p < .001), total mood
disturbance (p = .005), and self-esteem (p = .025). Improvement in
body size dissatisfaction (p = .025) and self-esteem (p = .028) were
associated with group differences in long-term weight loss,
independently of short-term weight change. Contrarily, results
for total mood disturbance indicated that the association with
long-term weight change is dependent on initial weight loss
success; when adjusting for 0–4-month weight change, the
prospective associations were no longer significant.
The last set of analysis evaluated the multivariate association
between changes in psychosocial variables and weight change
(Table 3). Three stepwise multiple regression models were
computed with the same set of predictors, i.e., change in
psychosocial variables during treatment. These variables were
used to predict weight changes at 4, 16, and between 4 and 16
months, after controlling for the 0–4 month’s weight change.
The first model explained 29.2% of 0–4-month weight change
(p < .001), with total mood disturbance as the strongest predictor
(8.2% of unique variance explained – sr2). Changes in body
dissatisfaction (sr2= 7.6%) and body attractiveness (sr2= 7.3%) also
loaded significantly and independently on this model. The second
model explained 23.8% of weight change from baseline to follow-
up (p < .001). Of this, 11.3% was independently explained by
changes in body size dissatisfaction (p = .002) and 10.2% by
changes in total mood disturbance (p = .003). No predictor
significantlyentered the regression model for 4–16 months weight
change (R2= 2.8%, F(1,83) = 3.38, p = .069). However, considering
the results for the bivariate analysis (Table 2), we built a regression
modelwherethesignificantornear-significantvariables– changes
in body size dissatisfaction and self-esteem – were forced in, to
Table 2
Correlations between weight changes and psychosocial changes during treatment
(0–4 months; Lisbon, Portugal, 2003–2005).
Psychosocial variables Weight change
0–4 months 0–16 months 4–16 months
Body image
Body size dissatisfaction (BIA)
Body shape concerns (BSQ)
Body attractiveness (PSPP)
Physical self-worth (PSPP)
0.37***
0.17
?0.37***
?0.23*
0.42***
0.01
?0.25*
?0.22*
0.21*
?0.17
0.02
?0.08
Psychological well-being
Self-esteem (Rosenberg)
Depression (BDI)
Total mood disturbance (POMS)
?0.08
0.22*
0.39***
?0.19
0.21*
0.37***
?0.20*
0.09
0.14
Notes: n=96; paired t-test. BIA: Body Image Assessment Questionnaire; BSQ: Body
Shape Questionnaire; PSPP: Physical Self-Perception Profile; RSES: Rosenberg Self-
Esteem Scale; BDI: Beck Depression Inventory; POMS: Profile of Mood States.
*p<.05.
***p<.001.
Fig. 1. Weight loss after treatment (0–4 months) and after follow-up by treatment-
related changes in body size dissatisfaction, total mood disturbance, and self-
esteem. For each analysis, the sample was tertile-split into three groups based on
treatment changes in body size dissatisfaction, total mood disturbance, and self-
esteem. Weight change for the two extreme groups are displayed (mean and SEM).
16-Month weight change was compared between groups, after adjusting for 4-
month weight change (ANCOVA, indicated in each graph).
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
4
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002

Page 5

predict 4–16 months’ weight change. This model did not reach
significance (p = .055), explaining 6.4% of weight change during
follow-up, with body size dissatisfaction as the strongest predictor
(sr2= 3.3%, p = .082), followed by self-esteem (sr2= 2.1%, p = .167).
Discussion
The present study analyzed the magnitude of changes in body
image and psychological well-being during a behavioral 4-month
weight loss program, evaluating the extent to which these changes
were predictive of short- and long-term success in weight loss, in
middle-aged women. Main findings showed that most psychoso-
cial variables improved significantly during the course of treat-
ment and psychosocial changes were generally correlated with
changes in body weight during treatment, except for self-esteem
and body shape concerns, and that improvements in body size
dissatisfaction and mood disturbance were significant predictors
of 0–16 months weight change. Additional results suggest that
there might be a predictive role of short-term changes in body size
dissatisfaction and self-esteem on long-term weight loss after
accounting for initial weight change.
During treatment, significant and positive changes were
observed on most putative psychosocial predictors. Together,
these constructs provide a broad characterization of participants’
well-being and psychological functioning in some of the most
relevant domains in obesity and health psychology research. The
greatest improvements were observed in body size dissatisfaction,
depression symptoms, and mood. Body image improvements are
in line with previous reports (Foster et al., 1997; Ramirez & Rosen,
2001) and can be considered a consistent outcome of weight loss
programs. Positive changes in depression symptoms and mood
disturbance also find support in previous studies. For example,
Wadden and Tanco reported lower depression symptoms after
treatment in obese women (Tanco, Linden, & Earle, 1998; Wadden
et al., 1997). Similarly, mood improvements in the present study
replicate findings by Rippe et al. (1998) and Wing, Epstein, Marcus,
and Kupfer (1984).
However, other studies have shown non-supportive data. For
example, a recent meta-analysis by Maciejewski et al., reported no
improvements in depression based on eight randomized con-
trolledtrials that used the BDI, and mixed results were reported for
mood changes in three studies that used the POMS (Maciejewski
et al., 2005). Importantly, these authors proposed that 5–10%
improvements in psychosocial measures are adopted as a marker
of success, mirroring established targets for weight loss (NHLBI,
1998). If we follow the most stringent 10% boundary, we observed
that the present treatment was successful for 39% of all
participants for depression, 52% for mood, and 54% for body size
dissatisfaction (data not shown). It is noteworthy that for the same
time period (i.e., 0–4 months), these psychological improvements
clearly surpassed those observed for weight loss: 38% of the
women reached the 5% goal while only 4% reached the 10%
objective.Theseresultssupportpreviousrecommendations,which
suggest a broader definition of success in weight management in
order to include psychosocial and behavioral, in addition to
physiological/clinical variables (Hill & Billington, 2002; Teixeira,
Going, Sardinha, & Lohman, 2005).
Since weight loss was an explicit goal of the program, one could
expect that changes in weight generally correlated with improve-
ments in psychological well-being. Bivariate associations showed
that most changes in psychosocial measures were in fact
associated with short- and also with long-term weight outcomes.
During the treatment phase, changes in body dissatisfaction, body
attractiveness and mood disturbance were especially associated
with weight change. To a lesser degree, a similar pattern of
association was observed for depressive symptoms and physical
self-worth. These results are similar to the ones recently reported
by Teixeira et al. (2006), who used a comparable methodology and
found that body shape concerns, physical self-worth, and body
attractiveness were significantly associated with treatment-
related weight outcomes. In the present study and using multi-
variate analysis changes in mood disturbance, body dissatisfaction
and body attractiveness emerged as independent significant
predictors, reflecting a closer association between changes in
these variables and short-term weight loss. Following the tenets of
the reciprocal determinism paradigm (Bandura, 1997) and the
Reciprocal Effects Model (Marsh & Craven, 2006), we hypothesize
that the dynamical reciprocity between weight outcomes and
these psychological variables is more predominant than for other
variables. That is, some of the resources used to lose weight were
likely motivated by improvements in those body image variables
and by the overall positive psychological state achieved (Palmeira
et al., 2009). Indeed, our results add empirical support the
proposition that the ability to reach meaningful weight reduction
should directly benefit from a healthier psychological profile
(Schwartz & Brownell, 2004).
Causality between weight and psychosocial changes is difficult
to assert during the treatment phase, as changes coexist
temporally. Conversely, the prospective associations between
short-termtreatmentpsychologicalchangesandlong-termweight
outcomes are more adequate models to explore potential cause–
effect relationships and a relatively unique feature of this study.
Bivariate analyses showed that most short-term predictors
maintained their associations with long-term weight change. To
an extent, this was expected as 0–4- and 0–16-month weight
changesharecommonvariance.However,aninterestingpatternof
association was noted for change in self-esteem, where no
correlations were found with short-term weight outcomes, while
for long-term results the associations became stronger (although
with marginally significant p values). This might partially be a
resultof the questionnaire used,since the RSESis a relativelybroad
measure of self-esteem. This could have limited our ability to
capture all dimensions of the construct that might be associated
with (and be relevant to) weight change. Self-esteem, a relatively
stable aspect of one’s personality, may take longer to positively
influence behavioral adherence aspects (e.g., improved eating and
exercise), which ultimately determine long-term weight control.
Self-esteem showed a particularly interesting set of results as it
proved to discriminate between the most and least successful
participants at the end of the program, independently of initial
weight change. A similar pattern was observed for body size
Table 3
Multiple regression analysis for the prediction of weight change during treatment
and after follow-up, from psychosocial changes during treatment (Lisbon, Portugal,
2003–2005).
Psychosocial variables0–4 months
b
psr2
Total mood disturbance (POMS)
Body size dissatisfaction (BIA)
Body attractiveness (PSPP)
R2(R2adj.)
Model’s F
0.27
0.25
?0.25
.008
.011
.013
8.2%
7.6%
7.3%
.29 (.26)
11.29<.001
0–16 months
b
psr2
Body size dissatisfaction (BIA)
Total mood disturbance (POMS)
R2(R2adj.)
Model’s F
0.32
0.30
.002
.003
.24 (.22)
12.96<.001
11.3%
10.2%
Notes: n=96; sr2: squared semi-partial correlation. BIA: Body Image Assessment
Questionnaire; PSPP:PhysicalSelf-PerceptionProfile;POMS: Profileof MoodStates.
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
5
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002

Page 6

dissatisfaction. In other words, those individuals who improved
self-esteem and body dissatisfaction the most had a higher
likelihood of finishing among the most successful weight loss
group, to an extent independently of their initial weight loss. The
present findings can be viewed as endorsing the development of
interventions that more proactively aim at improving psycholo-
gical well-being, not merely weight loss, something which has
been recommended before but with little supportive empirical
evidence(Hill&Billington,2002;Waddenetal.,2002). Specifically,
the treatments should include body image-related contents, in line
with the suggestions by Rosen and colleagues (Ramirez & Rosen,
2001; Rosen, 2003; Wadden et al., 2002), because body image
improvements could facilitate the use of psychosocial resources
and lead to better adherence to the weight management tasks
(Palmeira et al., 2009; Schwartz & Brownell, 2004).
Conclusions
Clearly, there is a need to further investigate the interaction
between weight loss and psychological well-being. To our
knowledge, only one previous study analyzed the association of
short-term psychological change with long-term weight outcomes
(Teixeira et al., 2006). It showed that initial changes in exercise
intrinsic motivation were predictive of long-term weight results,
above and beyond change in variables related to eating behavior
(e.g., cognitive restraint and disinhibition) and some body image
variables.The presentresultssuggestthat positivechangesinbody
dissatisfaction, mood and depression, physical self-worth and
body attractiveness can, to different degrees, predict long-term
weight change in women undergoing behavioral weight manage-
ment. Future studies should evaluate if some of these variablesadd
explained variance (of key health behaviors) to that reached by
more established health behavior change models (Bougart &
Delahanty, 2004). From a methodological point of view, benefit
could be gained from the evaluation of reciprocal determinism
(among individual, behavioral and environmental factors) using
the methods developed to analyze the Reciprocal Effects Model;
this could be performed with multilevel techniques, mediation
analyses, and/or using a larger number of data points during
intervention and/or follow-up, for both psychosocial and weight
variables (Marsh & Craven, 2006).
The average weight loss observed in obesity treatment
programs is generally short-lived, which highlights the need to
investigate predictors of long-term success (Elfhag & Rossner,
2005; Teixeira et al., 2009). The current findings suggest that
changes in psychological well-being taking place during weight
management programs might independently contribute to long-
term success. Because causal paths between psychosocial and
behavioral/weight changes are most likely closely intertwined
(Palmeira et al., 2009), concepts such as reciprocity and dynamics
(Marsh & Craven, 2006) could be useful to better understand how
change occurs during behavioral obesity treatment management
and other health behavior change interventions.
Conflict of interest statement
The authors state that there is no conflict of interest regarding
the present manuscript.
Acknowledgements
This study was funded by the Portuguese Science and
Technology Foundation and by the Oeiras City Council. The
investigators are grateful to Roche Pharmaceuticals Portugal, Becel
Portugal, and Compal Portugal for small grants and donations. We
also wish to thank all women who participated in the trial for their
commitment to this research project.
References
Baker, C. W.,& Brownell, K. D. (2000). Physical activity and maintenance of weight loss:
Physiological and psychological mechanisms. In C. Bouchard (Ed.), Physical activity
and obesity (pp. 311–328). Champaing, IL: Human Kinetics.
Bandura, A. (1997). Self-efficacy. The exercise of control. New York: WH Freeman and
Company.
Beck, A. T., & Steer, R. A. (1987). Manual for the Beck depression inventory. New York:
Psychological Corporation.
Biddle, S. J., & Mutrie, N. (2001). Psychology of physical activity: Determinants, well-being
and interventions. London: Routledge.
Blaine, B. E., Rodman, J., & Newman, J. M. (2007). Weight loss treatment and psycho-
logical well-being: A review and meta-analysis. Journal of Health Psychology, 12,
66–82.
Bougart, L. M., & Delahanty, D. L. (2004). Psychosocial models. In Boll, T. J., Franck, R. G.,
Baum, A., & Wallander, J. L. (Eds.), Handbook of health psychology: Models and
perspectives in health psychology (vol. 3). Washington, DC: American Psychological
Association. pp. 201–248.
Brownell, K. D. (1997). The LEARN program for weight control. Dallas, TX: American
Health Publishing Company.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/
correlation analysis for the behavioral sciences (3rd edition). Mahwah, NJ: Laurence
Erlbaum Associates.
Cooper, P. J., Taylor, M. J., Cooper, Z., & Fairburn, C. G. (1987). The development and
validation of a Body Shape Questionnaire. International Journal of Eating Disorders,
6, 485–494.
Elfhag,K., & Rossner,S. (2005).Who succeedsinmaintainingweight loss?Aconceptual
review of factors associated with weight loss maintenance and weight regain.
Obesity Reviews, 6, 67–85.
Foster, G. D., Wadden, T. A., & Vogt, R. A. (1997). Body image in obese women before,
during, and after weight loss treatment. Health Psychology, 16, 226–229.
Fox, K. R., & Corbin, C. B. (1989). The physical self-perception profile: Development and
preliminary validation. Journal of Sport and Exercise Psychology, 11(4), 408–430.
Hill, J. O., & Billington, C. J. (2002). It’s time to start treating obesity. American Journal of
Cardiology, 89, 969–970.
Kahneman, D., Diener, E., & Schwarz, N. (Eds.). (1999). Well-being: Foundations of
hedonic psychology. New York: Russel Sage Foudantion.
Kolotkin, R. L., Crosby, R. D., Williams, G. R., Hartley, G. G., & Nicol, S. (2001). The
relationship between health-related quality of life and weight loss. Obesity
Research, 9, 564–571.
Maciejewski, M. L., Patrick, D. L., & Williamson, D. F. (2005). A structured review of
randomized controlled trials of weight loss showed little improvement in health-
related quality of life. Journal Clinical Epidemiology, 58, 568–578.
Marsh, H.W.,& Craven, R. G. (2006). Reciprocal effects ofself-concept andperformance
from a multidimensional perspective. Perspectives on Psychological Science, 1, 133–
163.
McNair, D. M., Lorr, N., & Droppleman, L. F. (1971). Manual for the Profile of Mood States.
San Diego: EdITS.
Neumark-Sztainer, D., Paxton, S. J., Hannan, P. J., Haines, J., & Story, M. (2006). Does
body satisfaction matter? Five-year longitudinal associations between body satis-
faction and health behaviors in adolescent females and males. Journal of Adolescent
Health, 39, 244–251.
NHLBI. (1998). Clinical guidelines on the identification, evaluation, and treatment of
overweight and obesity in adults: The evidence report. Washington, DC: NIH—
National Heart, Lung, and Blood Institute.
Palmeira, A. L., Teixeira, P. J., Branco, T. L., Martins, S. S., Minderico, C. S., Barata, J. T.,
et al. (2007). Predicting short-term weight loss using four leading health
behavior change theories. International Journal Behavior Nutrition and Physical
Activity, 4, 14.
Palmeira, A. L., Markland, D. A., Silva, M. N., Branco, T. L., Martins, S. C., Minderico, C. S.,
et al. (2009). Reciprocal effects among changes in weight, body image, and other
psychological factors during behavioral obesity treatment: A mediation analysis.
International Journal Behavior Nutrition and Physical Activity, 6, 9.
Ramirez, E. M., & Rosen, J. C. (2001). A comparison of weight control and weight control
plus body image therapy for obese men and women. Journal of Consulting and
Clinical Psychology, 69, 440–446.
Rippe, J. M., Price, J. M., Hess, S. A., Kline, G., DeMers, K. A., Damitz, S., et al. (1998).
Improved psychological well-being, quality of life, and health practices in mod-
erately overweight women participating in a 12-week structured weight loss
program. Obesity Research, 6, 208–218.
Rosen, J. C. (2003). Improving body image in obesity. In J. K. Thompson (Ed.), Body
image,eatingdisorders,andobesity:Anintegrativeguideforassessmentandtreatment
(pp. 425–440). Washington, DC: APA.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.
Schwartz,M.B.,&Brownell,K.D.(2004).Obesityandbodyimage.BodyImage,1,43–56.
Stice, E.,& Shaw,H. E.(2002). Roleofbodydissatisfactionintheonset and maintenance
of eating pathology: A synthesis of research findings. Journal of Psychosomatic
Research, 53, 985–993.
Tanco, S., Linden, W., & Earle, T. (1998). Well-being and morbid obesity in women:
A controlled therapy evaluation. International Journal of Eating Disorders, 23,
325–339.
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
6
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002

Page 7

Teixeira, P. J., Palmeira, A. L., Branco, T. L., Martins, S. S., Minderico, C. M., Barata, J. T.,
et al. (2004). Who will lose weight? A cross-cultural reexamination of predictors of
weight loss in women. International Journal Behavior Nutrition and Physical Activity,
1, 12.
Teixeira, P. J., Going, S. B., Sardinha, L. B., & Lohman, T. G. (2005). A review of
psychosocial pre-treatment predictorsofweight control.Obesity Reviews, 6,43–65.
Teixeira, P. J., Going, S. B., Houtkooper, L. B., Cussler, E. C., Metcalfe, L. L., Blew, R. M.,
et al. (2006). Exercise motivation, eating, and body image variables as predictors of
weight control. Medicine Science in Sports and Exercise, 38, 179–188.
Teixeira,P.J.,Silva,M.N.,Coutinho,S.R.,Palmeira,A.L.,Mata,J., Vieira,P.N.,etal.(2009).
Mediators of weight loss and weight loss maintenance in middle-aged women.
Obesity (Silver Spring) Advance online publication. doi:10.1038/oby.2009.281.
Wadden, T. A., Vogt,R. A., Andersen, R.E., Bartlett, S. J.,Foster, G. D., Kuehnel, R. H.,et al.
(1997). Exercise in the treatment of obesity: Effects of four interventions on body
composition, resting energy expenditure, appetite, and mood. Journal of Consulting
and Clinical Psychology, 65(2), 269–277.
Wadden, T. A., Brownell, K. D., & Foster, G. D. (2002). Obesity: Responding to the global
epidemic. Journal of Consulting and Clinical Psychology, 70, 510–525.
Williamson, D. A., Davis, C. J., Bennett, S. M., Goreczny, A. J., & Gleaves, D. H. (1989).
Development of a simple procedure for assessing body image disturbances.
Behavioral Assessment, 11, 433–446.
Wing, R. R., Epstein, L. H., Marcus, M. D., & Kupfer, D. J. (1984). Mood changes
in behavioral weight loss programs. Journal of Psychosomatic Research, 28,
189–196.
A.L. Palmeira et al./Body Image xxx (2010) xxx–xxx
7
G Model
BODYIM-305; No. of Pages 7
Please cite this article in press as: Palmeira, A. L., et al. Change in body image and psychological well-being during behavioral obesity
treatment: Associations with weight loss and maintenance. Body Image (2010), doi:10.1016/j.bodyim.2010.03.002