Journal of Psychosomatic Research 53 (2002) 1029 – 1036

Psychological aspects of weight maintenance and relapse in obesity

Susan M. Byrne*,1
Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford, OX3 7JX, UK

Abstract

Studies have shown that, among formerly obese individuals, important because a better understanding of these factors may
the continued practice of the behavioural strategies adopted lead to the development of improved psychological treatments
during weight loss is associated with the successful main- for obesity. The results of the studies suggest that a number
tenance of a new lower weight. Much less attention has been psychological factors, such as having unrealistic weight goals,
focused on the psychological factors that motivate the continued poor coping or problem-solving skills and low self-efficacy,
use or abandonment of these critical weight maintenance may have an important effect on the behaviours involved
behaviours. This paper reviews studies that have attempted to in weight maintenance and relapse in obesity, and further
identify psychological characteristics that may be associated research in this area is warranted. D 2002 Elsevier Science Inc.
with weight maintenance and relapse in obesity. This is All rights reserved.

Keywords: Obesity; Psychological factors; Weight maintenance; Weight regain

Introduction been based on a systematic analysis of the range of

processes that are involved in weight maintenance and
There is substantial evidence to suggest that the two most weight regain (as opposed to weight loss) in obesity.
widely available treatments for obesity (behavioural treat- Relapse in obesity appears to be attributable to individ-
ment and pharmacological treatment) can produce modest uals’ failure to adhere to the weight-control behaviours that
weight loss in the region of 5 –10% of initial body weight they adopted in order to lose weight, such as the continued
[1].2 This amount of weight loss is associated with signific- consumption of a low-fat diet, the use of regular physical
ant physical and psychosocial benefits for obese people so activity and the regular monitoring of weight [8,9,17,
long as it is maintained [2 –7]. The problem is that the 18,23 – 31]. However, little information has been obtained
weight lost as a result of these treatments is almost always regarding the psychological mechanisms that drive these
regained over time [8– 12]. behaviours. It is not clear how or why some formerly
The issue of how best to prevent, or at least minimise, the obese individuals (about 15 – 20% of successful weight
weight regain that follows treatment for obesity has become losers) are able to persevere with these forms of behaviour
a priority area for obesity research [13 –17]. Attempts to when the majority fail to do so. This paper will review
improve the maintenance of a new lower weight, by research that has attempted to identify the psychological
incorporating specific maintenance strategies into standard factors that may be associated with weight maintenance
behaviour therapy (BT) programmes, have, at best, delayed and weight regain in obesity. A better understanding of
weight regain [17 –22]. However, these strategies have been these factors may help to inform the development of new
largely atheoretical and pragmatic in nature, and have not psychological treatments for obesity that are designed to
address them.
* Tel.: +44-1865-226479; fax: +44-1865-226244.
E-mail address: [email protected] (S.M. Byrne).
1
Present address: University of Western Australia, Department of Research investigating psychological factors associated
Psychology, Nedlands, W. A. 6009. Tel.: +61-8938-03579; fax: +61-
with weight maintenance and relapse
8938-02655.
2
Surgical interventions result in much greater weight loss but are
only available, at present, for a small proportion of morbidly obese There have been two main approaches to obtaining
individuals [32]. information about the psychological variables that might

0022-3999/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved.
PII: S 0 0 2 2 - 3 9 9 9 ( 0 2 ) 0 0 4 8 7 - 7
1030 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036

be involved in weight maintenance. One approach has been Self-efficacy

prospective — subjects who have been recruited to take part
in treatment trials are followed over time in order to identify Jeffery et al. [25] followed a sample of 89 middle-aged
variables associated with the maintenance of a new lower men for 2 years after they had completed a 15-week group
weight. A second approach has been retrospective — indi- BT programme. The mean weight loss during the treatment
viduals who have lost weight in the past and who have been period was 13.5 kg (13.4% of initial body weight), but by
able to demonstrate successful long-term weight mainten- the end of follow-up (in which 91% of the original sample
ance are recruited and interviewed in order to determine participated) subjects had regained an average of 63% of
how they have achieved this success. the weight lost. As part of the assessment protocol,
These studies have been limited in several ways. The questionnaires (developed by the authors) to measure
prospective studies have been conducted primarily in uni- perceived psychological well-being, self-efficacy and self-
versity-based settings, which may not represent typical satisfaction were administered to the men at pretreatment,
methods for weight loss, and patients who volunteer for posttreatment and at 1-year follow-up. Pre- and posttreat-
clinical trials may not be representative of all individuals ment level of self-efficacy (defined as the confidence
seeking to lose weight. The retrospective studies have subjects felt in their ability to refrain from eating in the
usually drawn samples from the general population who context of different emotional states or situations involving
are likely to be more representative of obese people who food) was the only factor that was found to be positively
have lost weight. However, their results have relied on self- related to weight maintenance at both 1- and 2-year
reported retrospective data often collected many years after follow-up.
the initial weight loss had occurred. Many of the studies The results of another prospective study are also sug-
from both categories are limited by small sample sizes. gestive of a link between self-efficacy and successful weight
In addition, there is considerable variation in the way in maintenance. Gormally et al. [23] interviewed 33 of 40
which successful weight maintenance and weight regain female subjects 7 months after they had completed a
have been defined. In some studies, maintainers have been 4-month BT programme and lost an average of 5.5 kg. Of
defined as subjects who have maintained, for at least 1 year, the subjects interviewed, 7 (23%) were classified as suc-
a weight loss of 5 –10% of their maximum body weight cessful maintainers ( < 30% weight regain) and 13 (37%) as
[33], while in others, subjects have only been classified as regainers ( > 50% regain). The remaining 13 subjects had
maintainers if they have achieved and maintained a healthy shown little weight change throughout the study. At the
weight (BMI = 20 –25 kg/m2) for more than 2 years [34,35]. follow-up interview, maintainers reported significantly
Similarly, in some studies, only subjects who have returned greater confidence than the regainers (on a five-point
to their pretreatment weight have been classified as Likert-type scale) in being able to control both eating urges
regainers [34,36], whereas in other studies a regain of 5 lb and their weight. However, since these data are correla-
(2.2 kg) from posttreatment weight has warranted this tional, the results are clearly subject to the cause and effect
classification [33,37]. confound that was mentioned previously.
Finally, the majority of studies are limited by a cause
and effect confound. Even when the overall design has Adverse life events and coping skills
been prospective, the analyses involving psychological
factors have usually been correlational in nature because Three prospective studies have reported an association
the variables of interest have only been assessed at follow- between experiencing stressful life events and weight regain
up (rather than at pretreatment and/or posttreatment as [23,24,36]. Dubbert and Wilson [36], for instance, followed
well). Therefore, it cannot be determined whether the up (via mail and telephone) 58 of an original 62 subjects 30
characteristic is a cause, consequence or correlate of months after they had completed a 19-week BT programme.
weight maintenance. At follow-up, 43% of the subjects who had regained all of
Despite these limitations, some data have begun to the weight lost (n = 14), but only 15% of those who had
emerge about various psychological factors that may play maintained at least part of the weight lost (n = 44) reported
a role in weight maintenance. the occurrence of major illnesses, bereavements and other
environmental stressors after the end of treatment.
Other studies have shown that it is not the absence of
Prospective studies stress, but rather the ability to cope with stress, that appears
to be associated with successful weight maintenance
While a relatively large number of studies have followed- [23,24,38,39]. Gormally and Rardin [24] analysed data from
up patients who have participated in treatment trials for the interviews of 38 (of an initial 112) female subjects
obesity, only a small proportion of these have assessed 7 months after attending 16 group BT sessions. Sixteen of
psychological factors to determine their association with these subjects (42%) were classified as maintainers and 22
weight maintenance. Four factors have been identified as (58%) as regainers (had regained >50% of the weight lost).
potentially important. Both groups reported a similar proportion of unpredictable
 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036 1031

major life changes following treatment (such as a pro- (TFEQ [41]) or the Restraint Scale [42], predicted long-
motion, illness, bereavement, etc.), but only 20% of the term weight maintenance [43,44]. Westerterp-Plantenga
maintainers, compared to 81% of the regainers, reported a et al. [43] followed-up 27 obese women who had undergone
relapse in diet and exercise during the life change. In two 8-week bouts of a very-low-calorie-diet (VLCD). Two
addition, none of the maintainers, but almost all (82%) of years after the start of the first VLCD, the most successful
the regainers, reported eating in response to emotions subjects (seven women who had regained less than 50%
elicited by day-to-day life stressors. Nearly all of the of the weight lost) showed a significant increase in dietary
maintainers (80%) reported using effective problem solving restraint and this increase was found to be significantly
to cope with stress, either by generating new solutions or negatively correlated with weight regain. This finding,
applying concepts learned in treatment. In contrast, none of however, is not particularly informative. It probably
the regainers reported such behaviour. reflects the fact that maintainers were able to persist in
Drapkin et al. [38] found that the ability to generate the conscious restraint of food intake more successfully
coping responses to a number of hypothetical high risk than were regainers. In a subsequent study using data
scenarios predicted future outcome among 79 of an from two VLCD treatment groups, Pasman et al. [44]
original group of 93 obese subjects who had participated found that a high pretreatment score on the hunger scale
in a year-long BT programme. Before starting treatment, of the TFEQ (which measures the subjective feeling of
subjects were asked to listen to four scenarios describing hunger) significantly predicted weight regain at 14-month
high-risk situations, such as a family mealtime celebra- follow-up.
tion, an argument, watching television or a stressful
situation at work. The number of coping responses
generated in each situation was assessed. The results Retrospective studies
showed that subjects who generated coping responses to
more of the situations showed better maintenance of a The majority of retrospective studies in this field have
new lower weight at 12-month posttreatment. The ability been quasi-qualitative in nature, using open-ended questions
to generate any coping response, rather than the number or semistructured interviews to identify common features
or type of coping responses generated, appeared to be the among individuals recruited from the community who
important factor. report that they have successfully maintained a new lower
weight. These studies have taken several different forms.
Weight goals
Studies of successful weight maintainers
The results of one prospective study [40] have suggested
that individuals’ weight goals may play a role in determin- Colvin and Olson [37] interviewed a sample of 13 males
ing whether they will succeed or fail at maintaining a new and 41 females who had lost at least 20% of their initial
lower weight. In this study, nine men and 38 women, who body weight and maintained their new lower weight
had been 15% or more overweight but had reached their (to within a range of 2.2 kg) for 2 years. The amount of
self-determined goal weight in the preceding year, were weight lost ranged from 9 to 123 kg. Three psychological
recruited using newspaper advertisements.3 Questionnaires characteristics were observed to be common among the
developed by the authors were mailed to subjects every participants. First, all of the subjects reported that they had
4 months for 1 year. At the end of the year, over half (58%) achieved their weight goals and they expressed high levels
of the subjects were classified as maintainers (had regained of satisfaction with their present weight. Second, subjects
< 20% of the weight lost). The only questionnaire items to reported that losing weight had increased their concern with
discriminate significantly between maintainers and regainers and the attention they paid to, their weight, shape and
related to weight goals, with maintainers being more likely appearance. Third, the female subjects, in particular,
to have achieved their goal weight by the end of their weight reported that since losing weight they had become more
loss effort. confident, self-assured and autonomous, and more capable
of taking control of, and responsibility for, their lives.
Dietary restraint and subjective hunger
Studies of maintainers versus regainers
Two prospective studies have examined whether pretreat-
ment scores on measures of dietary restraint, disinhibition Rather than focusing exclusively on maintainers, a sub-
and hunger, such as Three Factor Eating Questionnaire set of retrospective studies have examined differences
between formerly obese individuals who have managed to
maintain a new lower weight and those who have lost
3
This study differed from the others in this category, because it weight but subsequently regained it [35,45]. Ferguson et al.
involved subjects who had lost weight on their own, rather than those who [45], for example, compared the characteristics of 41
had taken part in a treatment trial. women and 41 men who had lost at least 15% of their
1032 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036

body weight and maintained their new lower weight (to Studies based on the American National Weight Control
within 2.2 kg) for 1 year, with those of a group of Registry (NWCR)
‘‘unsuccessful dieters’’ (32 women and 28 men whose lack
of success was not defined). All of the subjects took part in The most detailed information so far about successful
a face-to-face interview. The maintainers were more likely weight maintainers has been provided by a series of studies
than their unsuccessful counterparts to report coping easily based on the NWCR. The NWCR is an ongoing registry that
with cravings for food and they attributed their suc- was set up in 1997 by Wing and colleagues, in order to
cess to characteristics such as determination, commitment identify individuals who are successfully maintaining a new
and patience. lower weight [46]. To enrol in the NWCR, participants must
Tinker and Tucker [35] interviewed a small sample have lost at least 13.6 kg (30 lb) and must have maintained
(n = 21) of previously obese men and women who had lost this weight loss for at least 1 year. Currently, there are over
weight to within the healthy range and maintained their new 3000 individuals in this registry [47].
lower weight for an average of 4.5 years. A group of 20 Participants are initially recruited through local and
currently obese adults who had lost, but regained, weight national television, radio, magazine and newspaper adver-
were also interviewed. Compared to regainers, maintainers tisements and then screened to verify that they fit the
reported experiencing fewer negative events during, and inclusion criteria. Registry members are sent a packet of
especially after, weight loss. This led the authors to suggest questionnaires, which include questions about demographic
that maintenance may depend on stable circumstances after and weight characteristics, weight-loss and weight mainten-
active behaviour change. However, this finding may well be ance strategies, and the effect of weight loss and mainten-
a consequence, rather than a cause, of successful weight ance on aspects of their lives. Additional self-report
maintenance. Regainers may be more likely to recall nega- questionnaires include the 20-item Centre for Epidemiologic
tive events after weight loss by way of explaining their Studies Depression scale (CES-D [48]), the TFEQ and
weight regain, or because of the negative impact of their questions taken from the self-report version of the Eating
failure to maintain weight. Disorders Examination [49].
In 1999, McGuire et al. [33] reported on the 1-year
Studies comparing maintainers, regainers and stable follow-up of 714 NWCR members (91% of the original
healthy weight individuals sample). They investigated whether any of the character-
istics measured at entry into the registry predicted weight
One limitation of studies, which have compared success- regain versus continued weight maintenance over the
ful maintainers to regainers, is that the maintainers generally subsequent year, and whether change in psychological
weigh less than the regainers, so any psychological differ- characteristics would occur in weight regainers compared
ences may be due to differences in current body weight. One with maintainers.
study [34] has compared maintainers to both regainers and Of the individuals assessed at 1-year follow-up (subjects
to stable healthy-weight subjects. In this study, 700 female were asked to report their present weight and to complete
volunteers recruited from a large health maintenance organ- again the self-report measures that were administered on
ization were asked to categorize themselves as maintainers, entry into the NWCR), 36% were classified as having
regainers or healthy weight. Fifty volunteers were then gained weight over the year (i.e., gained more than 5 lb
randomly selected from each of these categories for further [2.2 kg]); 57% were classified as having maintained weight
screening and a final sample of 108 women were selected (within 2.2 kg); and 6% were classified as having lost
for interview — 30 formerly obese women who had main- weight (i.e., more than 2.2 kg). It should be pointed out
tained a healthy weight for at least 2 years, 44 obese women here that a weight gain of 2.2 kg is a very narrow definition
who had previously lost at least 20% of their weight but had of regain, so the categorization described above may be
regained all of it and 34 healthy weight subjects with no problematic, especially given the study’s reliance on self-
history of obesity who had always remained within 3.6 kg of report data and the fact that weight fluctuation of this
their current weight. magnitude is common.4
The main features distinguishing maintainers from The regainers were found to differ significantly from the
regainers related to problem solving skills. Over 70% of maintainers on several of the initial assessment variables,
the regainers attributed their weight regain to eating in and a stepwise logistic regression model revealed four
response to stressful life events or to negative emotional factors that independently predicted which individuals
states. Few regainers (10%) reported using problem-solving would gain weight over the year versus those who would
or direct ways of coping with problems compared with
maintainers (95%) and normal weight subjects (60%). For
example, rather than dealing with problems directly, most 4
Moreover, despite an average weight gain of 7 kg, 81% of the
regainers reported using ‘‘escape-avoidance’’ ways of cop- regainers still met the entry criterion for the NWCR and 96% were still
ing such as eating, sleeping more or wishing the problem maintaining a 10% weight loss. None had returned to their maximum
would go away. lifetime weight.
 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036 1033

maintain weight. First, those subjects who had a higher bias, because the subjects who chose not to answer
maximum weight and a greater initial weight loss were questionnaires after 1 year may have been more likely to
more likely to have relapsed. Second, subjects who had have regained weight.
maintained their new lower weight for a shorter period of
time were more likely to have regained weight at 1-year
follow-up. Third, subjects who, at the initial assessment, Conclusions
reported that they were still trying to lose weight were at
higher risk of weight regain than those who reported that The role of psychological processes in weight maintenance
they were aiming to maintain weight. And fourth, subjects and relapse
who reported higher dietary disinhibition levels at the
initial assessment were at greater risk of gaining weight It has been suggested that weight regain following
at 1-year follow-up. successful weight loss can be attributed almost entirely to
Over the follow-up period, maintainers showed no biological processes [51 – 54], and that behaviour modifica-
changes on any of the psychological measures taken at tion is simply inadequate to counteract physiological fac-
baseline, whereas the regainers, not surprisingly, reported a tors, such as a relatively low metabolic rate [55,56],
marked decrease in dietary restraint. In addition, regainers adaptive thermogenesis [57,58], decreased fat oxidation
not only reported higher levels of dietary disinhibition, [59,60], increased insulin sensitivity [61] or leptin resistance
binge eating and depressive symptoms than maintainers at [62], which may prime formerly obese people to regain the
the initial assessment, but also worsening on these param- weight they have lost.
eters over time. These findings suggest the possibility that Further information regarding the contribution of bio-
the regainers had already started to regain weight when logical factors to success or failure in long-term weight
they entered the registry, and then merely continued to maintenance should become available as research in this
regain weight over the following year. The reliability of area progresses. However, so far, studies have found little
the data relating to binge eating in this study is uncertain evidence that any of the factors listed above are consistently
since it was assessed by self-report questionnaire only predictive of subsequent weight regain in formerly obese
(rather than by interview). individuals [63,64]. Therefore, the role of other processes,
A subsequent study based on the NWCR [50] looked at including psychological processes that might account for the
the relationship between duration of weight maintenance phenomenon of weight regain, should not be overlooked.
and the effort, attention and pleasure associated with weight The findings from the studies reviewed here have been
maintenance. Registry members (758 women and 173 men) inconsistent and not always informative. Nevertheless, some
who had completed 1-year follow-up assessments were evidence has emerged to suggest that the following psycho-
asked to indicate, using Likert-type scales, firstly, how much logical factors may offer some potential explanations for
effort they devoted to weight maintenance; secondly, how why many formerly obese individuals fail to persist with
much attention they felt was required to maintain their new critical weight control behaviours following successful
lower weight; and, thirdly, how much pleasure they felt that weight loss.
they derived from successful weight maintenance.
Linear regression analysis indicated that as the duration Weight goals
of weight maintenance increased, a shift in the balance Several studies have suggested that those who achieve
between effort and pleasure seemed to occur. Subjects their predetermined goal weight, as compared with those
who had maintained a new lower weight for longer periods who do not, may be more successful at long-term weight
of time reported that significantly less effort and less maintenance [33,37,40]. Cooper and Fairburn [13,14] have
attention was needed to be applied to maintain their weight. suggested that the failure to achieve a desired weight may
In contrast, the pleasure derived from weight maintenance lead individuals to believe that any further attempts to
was not affected by duration. So the duration of weight control their weight will not be worthwhile, resulting in
maintenance was associated with greater pleasure relative to the abandonment of weight maintenance behaviours.
the effort required to maintain weight. The decrease in the
burden of maintaining the weight lost, coupled with no Coping strategies and problem-solving skills
change in the associated pleasure, may facilitate continued Studies from both categories have found that weight
maintenance of a new lower weight. maintainers tend to cope more successfully with adverse
Some caution should be used in generalising the find- life events than do regainers [23,24,34,38]. It appears that
ings from this group of studies based on the NWCR for maintainers may be able to use problem-solving skills to
two main reasons. Firstly, registry members may not be cope with stressful situations in a way that does not interfere
representative of all successful weight maintainers (97% of with their adherence to a weight maintenance regime. One
the participants were white and 54% had a college degree). study has also observed that regainers are more likely to
Secondly, although the attrition rate from the registry over report over-eating in response to negative emotional states
time (10%) was low, there may still have been a response than are maintainers [34]. This may reflect a tendency for
1034 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036

regainers to use food or eating to moderate negative mood the studies reviewed in this paper have provided some
states, rather than apply more appropriate coping strategies. limited information about the psychological factors that
may be important in this regard, however, the ways in
Self-efficacy which these factors work to influence weight maintenance
Two studies have found that successful weight main- is unknown. Further research, which focuses specifically on
tainers report greater confidence than regainers in their the psychological processes (particularly the cognitive pro-
ability to control their weight and their food intake cesses) involved in weight maintenance and relapse, is
[23,25]. It is possible that the belief that they can control clearly needed if we are to improve psychological interven-
their weight increases the likelihood that individuals will tions for obesity.
continue to engage in effective weight control behaviours A first step in this direction would be to generate some
following weight loss. There is also some evidence that, for initial hypotheses regarding the whole range of psycho-
maintainers, this increase in confidence can permeate other logical factors that might account for why, after losing
aspects of their lives [35,37]. weight, many individuals do not persist with the forms of
behaviour that are crucial for weight maintenance. These
Vigilance with regard to weight control hypotheses would then need to be tested in a series of
Formerly obese people probably have to impose con- prospective studies following-up formerly obese subjects for
tinual control on energy intake and energy expenditure in a period of time after they have lost weight. Such a
order to maintain a new lower weight, especially in the systematic analysis of the factors that influence successful
context of an environment that increasingly favours weight weight maintenance, as opposed to weight regain, in obesity
gain because of the ready availability of highly palatable, will lay the foundation for more focused psychological
high-fat foods and of a sedentary lifestyle [54]. Maintainers treatments, which may result in substantial improvements
appear to have recognised (and responded to) the need for in treatment response.
continued vigilance in this regard [37,43,45].

Effort versus benefits

In one study [50], maintainers reported that the benefits References
afforded by weight maintenance fully justified the effort
involved in adhering to a weight maintenance regime. It is [1] Wing RR. Behavioral approaches to the treatment of obesity. In:
Bray GA, Bouchard C, James WPT, editors. Handbook of obesity.
possible that the balance between the effort involved and the New York: Marcel Dekker, 1998. pp. 855 – 73.
benefits received from weight maintenance plays a role in [2] Andersen RE, Wadden TA, Bartlett SJ, Vogt RA, Weinstock RS.
either facilitating or hindering adherence to weight main- Relation of weight loss to changes in serum lipids and lipoproteins
tenance behaviours. in obese women. Am J Clin Nutr 1995;62(2):350 – 7.
[3] Blackburn GL. Benefits of weight loss in the treatment of obesity. Am
J Clin Nutr 1999;69(3):347 – 9.
Binge eating, dietary restraint, disinhibition and hunger [4] Foster GD, Wadden TA. The psychology of obesity, weight loss, and
A small number of studies have found that scores on weight regain: research and clinical findings. In: Blackburn GL, Kan-
various measures of binge eating, dietary restraint, disinhi- ders BS, editors. Obesity: pathophysiology, psychology and treatment.
bition and subjective feelings of hunger are related to weight New York: Chapman and Hall, 1994. pp. 140 – 66.
regain [33,43,44]. However, it is possible that some of these [5] Foster GD, Wadden TA. Social and psychological effects of weight
loss. In: Fairburn CG, Brownell KD, editors. Eating disorders and
results simply reflect subjects’ actual weight trajectories obesity: a comprehensive handbook. 2nd ed. New York: Guilford
over time. For instance, the finding that an increase in Press, 2002. pp. 500 – 4.
dietary restraint is associated with weight maintenance [6] Wadden TA, Sarwer DB, Arnold MA, Gruen D, O’Neil PM. Psycho-
(and a decrease with weight regain) is not at all surprising. social status of severely obese patients before and after bariatric sur-
gery. In: Sugerman HJ, Soper NJ, editors. Problems in general
High pretreatment levels of binge eating, disinhibition and
surgery: morbid obesity. New York: Guilford Press, 2000. pp. 13 – 22.
subjective hunger have been found to predict weight regain [7] Wing RR, Jeffery RW. Effect of modest weight loss on changes in
at follow-up [33,44]. However, the data relating to binge cardiovascular risk factors: are there differences between men and
eating in particular are unreliable, since they have been women or between weight loss and maintenance? Int J Obes
based solely on self-report questionnaire measures. 1995;19:67 – 73.
[8] Kramer FM, Jeffery RW, Forster JL, Snell MK. Long-term follow-up
of behavioural treatment for obesity: patterns of weight gain among
Directions for future research men and women. Int J Obes 1989;13:124 – 36.
[9] Stalonas PM, Perri MG, Kerzner AB. Do behavioral treatments of
Most pressing among the questions that remain obesity last? A five-year follow-up investigation. Addict Behav
unanswered in obesity research is: why do most people 1984;9(2):175 – 83.
who lose weight regain it? Relatively little research has been [10] Wadden TA, Frey DL. A multicenter evaluation of a proprietary
weight loss program for the treatment of marked obesity: a five-year
carried out on the psychological mechanisms that may be follow-up. Int J Eat Disord 1997;22(2):203 – 12.
associated with successful maintenance of a new lower [11] Wadden TA, Sternberg JA, Letizia KA, Stunkard AJ, Foster GD.
weight, as opposed to relapse, in obesity. The results of Treatment of obesity by very low calorie diet, behavior therapy,
 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036 1035

and their combination: a five-year perspective. Int J Obes 1989;13: [32] Kral JG. Surgical treatment of obesity. In: Bray GA, Bouchard C,
239 – 46 (Suppl.). James WPT, editors. Handbook of obesity. New York: Marcel Dekker,
[12] Walsh MF, Flynn TJ. A 54-month evaluation of a popular very low 1998. pp. 977 – 93.
calorie diet program. J Fam Pract 1995;41(3):231 – 6. [33] McGuire MT, Wing RR, Klem ML, Lang W, Hill JO. What predicts
[13] Cooper Z, Fairburn CG. A new cognitive behavioural approach to the weight regain in a group of successful weight losers? J Consult Clin
treatment of obesity. Behav Res Ther 2001;39(5):499 – 511. Psychol 1999;67(2):177 – 85.
[14] Cooper Z, Fairburn CG. Cognitive behavioral treatment of obesity. In: [34] Kayman S, Bruvold W, Stern JS. Maintenance and relapse after
Wadden TA, Stunkard AJ, editors. Obesity: theory and therapy. 3rd ed. weight loss in women: behavioral aspects. Am J Clin Nutr 1990;
New York: Guilford Press, 2002. pp. 465 – 79. 52(5):800 – 7.
[15] Perri MG. The maintenance of treatment effects in the long-term [35] Tinker JE, Tucker JA. Environmental events surrounding natural re-
management of obesity. Clin Psychol: Sci Pract 1998;5:526 – 43. covery from obesity. Addict Behav 1997;22(4):571 – 5.
[16] Perri MG, Corsica JA. Improving the maintenance of weight lost in [36] Dubbert PM, Wilson GT. Goal setting and spouse involvement in the
behavioural treatment of obesity. In: Wadden TA, Stunkard AJ, edi- treatment of obesity. Behav Res Ther 1984;22:227 – 42.
tors. Obesity: theory and therapy. 3rd ed. New York: Guilford Press, [37] Colvin RH, Olson SB. A descriptive analysis of men and women who
2002. pp. 368 – 89. have lost significant weight and are highly successful at maintaining
[17] Jeffery RW, Epstein LH, Wilson GT, Drewnowski A, Stunkard AJ, the loss. Addict Behav 1983;8(3):287 – 95.
Wilson GT, Wing RR. Long-term maintenance of weight loss: current [38] Drapkin RG, Wing RR, Shiffman S. Responses to hypothetical high risk
status. Health Psychol 2000;19:5 – 16 (Suppl.). situations: do they predict weight loss in a behavioral treatment program
[18] Baum JG, Clark HB, Sandler J. Preventing relapse in obesity or the context of dietary lapses? Health Psychol 1995;14(5):427 – 34.
through posttreatment maintenance systems: comparing the relative [39] Grilo CM, Shiffman S, Wing RR. Relapse crises and coping among
efficacy of two levels of therapist support. J Behav Med 1991;14: dieters. J Consult Clin Psychol 1989;57(4):488 – 95.
287 – 302. [40] Marston AR, Criss J. Maintenance of successful weight loss: inci-
[19] Perri MG, Shapiro RM, Ludwig WW, Twentyman CT, McAdoo WG. dence and prediction. Int J Obes 1984;8(5):435 – 9.
Maintenance strategies for the treatment of obesity: an evaluation of [41] Stunkard AJ, Messick S. The Three Factor Eating Questionnaire to
relapse prevention training and posttreatment contact by mail and measure dietary restraint, disinhibition and hunger. J Psychosom Res
telephone. J Consult Clin Psychol 1984;52(3):404 – 13. 1985;29:71 – 83.
[20] Perri MG, McAdoo WG, Spevak PA, Newlin DB. Effect of a multi- [42] Herman CP, Polivy J. Anxiety, restraint, and eating behavior. J Ab-
component maintenance program on long-term weight loss. J Consult norm Psychol 1975;84(6):66 – 72.
Clin Psychol 1984;52(3):480 – 1. [43] Westerterp-Plantenga MS, Kempen KP, Saris WH. Determinants of
[21] Perri MG, McAdoo WG, McAllister DA, Lauer JB, Jordan RC, Yan- weight maintenance in women after diet-induced weight reduction. Int
cey DZ, Nezu AM. Effects of peer support and therapist contact on J Obes Relat Metab Disord 1998;22(1):1 – 6.
long term weight loss. J Consult Clin Psychol 1987;55:615 – 7. [44] Pasman WJ, Saris WH, Westerterp-Plantenga MS. Predictors of
[22] Perri MG, McAllister DA, Gange JJ, Jordan RC, McAdoo WG, Nezu weight maintenance. Obes Res 1999;7(1):43 – 50.
AM. Effects of four maintenance programs on the long-term manage- [45] Ferguson KJ, Brink PJ, Wood M, Koop PM. Characteristics of suc-
ment of obesity. J Consult Clin Psychol 1988;56:529 – 34. cessful dieters as measured by guided interview responses and Re-
[23] Gormally J, Rardin D, Black S. Correlates of successful response to straint Scale scores. J Am Diet Assoc 1992;92(9):1119 – 21.
a behavioral weight control clinic. J Couns Psychol 1980;27(2): [46] Klem ML, Wing RR, McGuire MT, Seagle HM, Hill JO. A descriptive
179 – 91. study of individuals successful at long-term maintenance of substan-
[24] Gormally J, Rardin D. Weight loss and maintenance and changes in tial weight loss. Am J Clin Nutr 1997;66(2):239 – 46.
diet and exercise for behavioral counseling and nutrition education. [47] Wing RR, Klem ML. Characteristics of successful weight maintainers.
J Couns Psychol 1981;28(4):295 – 304. In: Fairburn CG, Brownell KD, editors. Eating disorders and obesity:
[25] Jeffery RW, Bjornson-Benson WM, Rosenthal BS, Lindquist RA, a comprehensive handbook. 2nd ed. New York: Guilford Press, 2002.
Kurth CL, Johnson SL. Correlates of weight loss and its maintenance pp. 588 – 92.
over two years of follow-up among middle-aged men. Prev Med [48] Radloff LS. The CES-D Scale: a self-report depression scale for research
1984;13(2):155 – 68. in the general population. Appl Psychol Meas 1977;1(3):385 – 401.
[26] Jeffery RW, Wing RR. Long-term effects of interventions for weight [49] Beglin SJ, Fairburn CG. Evaluation of a new instrument for the de-
loss using food provision and monetary incentives. J Consult Clin tection of eating disorders in community samples. Psychiatry Res
Psychol 1995;63:793 – 6. 1992;44(3):191 – 201.
[27] Perri MG, McAdoo WG, McAllister DA, Lauer JB, Yancey DZ. [50] Klem ML, Wing RR, Lang W, McGuire MT, Hill JO. Does weight loss
Enhancing the efficacy of behavior therapy for obesity: effects of maintenance become easier over time? Obes Res 2000;8(6):438 – 44.
aerobic exercise and a multicomponent maintenance program. J Con- [51] Bray GA, York B, DeLany J. A survey of the opinions of obesity
sult Clin Psychol 1986;54(5):670 – 5. experts on the causes and treatment of obesity. Am J Clin Nutr 1992;
[28] Schoeller DA, Shay K, Kushner RF. How much physical activity is 55(Suppl. 1):151S – 4S.
needed to minimize weight gain in previously obese women? Am J [52] Goodrick GK, Poston WS, Kimball KT, Reeves RS, Foreyt JP. Non-
Clin Nutr 1997;66(3):551 – 6. dieting versus dieting treatment for overweight binge-eating women.
[29] Wadden TA, Letizia KA. Predictors of attrition and weight loss in J Consult Clin Psychol 1998;66(2):363 – 8.
patients treated by moderate and severe caloric restriction. In: Wadden [53] Weintraub M, Sundaresan PR, Schuster B, Ginsberg G, Madan M,
TA, Itallie TB, editors. Treatment of the seriously obese patient. New Balder A, Stein EC, Byrne L. Long-term weight control study: II
York: Guilford Press, 1992. pp. 383 – 410. (weeks 34 to 104). An open-label study of continuous fenfluramine
[30] Wadden TA, Vogt RA, Andersen RE, Bartlett SJ, Foster GD, Kuehnel plus phentermine versus targeted intermittent medication as adjuncts
RH, Wilk JE, Weinstock RS, Buckenmeyer P, Berkowitz RI, Steen to behavior modification, caloric restriction, and exercise. Clin Phar-
SN. Exercise in the treatment of obesity: effects of four interventions macol Ther 1992; 51(5):595 – 601.
on body composition, resting energy expenditure, appetite, and mood. [54] Wilson GT. Behavioural treatment for obesity: thirty years and count-
J Consult Clin Psychol 1997;65(2):269 – 77. ing. Adv Behav Res Ther 1994;16:31 – 75.
[31] Wadden TA, Vogt RA, Foster GD, Anderson DA. Exercise and the [55] Buemann B, Astrup A, Christensen NJ, Madsen J. Effect of moderate
maintenance of weight loss: 1-year follow-up of a controlled clinical cold exposure on 24-h energy expenditure: similar response in postob-
trial. J Consult Clin Psychol 1998;66(2):429 – 33. ese and nonobese women. Am J Physiol 1992;263(6 Pt. 1):E1040 – 5.
1036 S.M. Byrne / Journal of Psychosomatic Research 53 (2002) 1029–1036

[56] Dulloo AG, Jacquet J. Adaptive reduction in basal metabolic rate in lism in weight-stable postobese individuals. Am J Clin Nutr 1995;
response to food deprivation in humans: a role for feedback signals 62(4):735 – 9.
from fat stores. Am J Clin Nutr 1998;68(3):599 – 606. [61] Yost TJ, Jensen DR, Eckel RH. Weight regain following sustained
[57] Leibel RL, Rosenbaum M, Hirsch J. Changes in energy expenditure weight reduction is predicted by relative insulin sensitivity. Obes Res
resulting from altered body weight. N Engl J Med 1995;332(10): 1995;3(6):583 – 7.
621 – 8. [62] Folsom AR, Jensen MD, Jacobs DR, Hilner JE, Tsai AW, Schreiner
[58] Stock MJ. Gluttony and thermogenesis revisited. Int J Obes Relat PJ. Serum leptin and weight gain over 8 years in African American
Metab Disord 1999;23(11):1105 – 17. and Caucasian young adults. Obes Res 1999;7(1):1 – 8.
[59] Astrup A, Buemann B, Christensen NJ, Toubro S. Failure to in- [63] Campfield LA, Smith FJ. The pathogenesis of obesity. Best Pract Res
crease lipid oxidation in response to increasing dietary fat content Clin Endocrinol Metab 1999;13(1):13 – 30.
in formerly obese women. Am J Physiol 1994;266(4 Pt. 1): [64] Wing RR, Sinha MK, Considine RV, Lang W, Caro JF. Relationship
E592 – 9. between weight loss maintenance and changes in serum leptin levels.
[60] Larson DE, Ferraro RT, Robertson DS, Ravussin E. Energy metabo- Horm Metab Res 1996;28:698 – 703.