The correlates of long-term weight loss: a group comparison study of obesity

Abstract

OBJECTIVE: Although the majority of weight loss attempts are unsuccessful, a small minority succeed in both weight loss and maintenance. The present study aimed to explore the correlates of this success.

METHOD: A group comparison design was used to examine differences between women who were classified as either weight loss maintainers (had been obese (body mass index, BMI=30+ kg/m2) and had lost weight to be considered non-obese (BMI<30 kg/m2) and maintained this weight loss for a minimum of 3 y; n=44), stable obese (maintained an obese weight (BMI=30+ kg/m2) for longer than 3 y; n=58), and weight loss regainers (been obese (BMI=30+ kg/m2), lost sufficient weight to be considered non-obese (BMI<30 kg/m2) and regained it (BMI=30+ kg/m2), n=40). In particular, the study examined differences in profile characteristics, historical factors, help-seeking behaviours and psychological factors.

RESULTS: The results showed that in terms of profile and historical factors, the weight loss maintainers had been lighter, were currently older and had dieted for longer than the other groups but were matched in terms of age, class and ethnic group. In terms of help-seeking behaviours, the weight loss maintainers reported having tried healthy eating more frequently but were comparable to the other subjects in terms of professionals contacted. Finally, for psychological factors the weight loss maintainers reported less endorsement for medical causes of obesity, greater endorsement for psychological consequences and indicated that they had been motivated to lose weight for psychological reasons.

CONCLUSIONS: Weight loss and maintenance is particularly correlated with a psychological model of obesity. This has implications for improving the effectiveness of interventions and the potential impact of current interest in medical approaches to obesity.

Introduction

Obesity is generally defined as a body mass index (BMI, kg/m2) of 30 or more. Using this definition, epidemiological studies in England reported that in 1994 13% of men and 16% of women aged 16–64 were obese and that the prevalence of obesity is increasing. Comparable figures for 1980 were 6% and 8%, and by the year 2005 it is predicted that 18% of men and 24% of women will be obese.1 Because of its association with physical problems such as heart disease, cancer and diabetes1,2 and its more controversial link with psychological problems such as depression and low self-esteem,1,2 psychologists, nutritionists, dieticians and endocrinologists have been involved in developing treatment programmes for obesity. The traditional treatment approach focused on encouraging the obese to eat ‘normally’ and this consistently involved putting them on a diet.3 In contrast, recent comprehensive, multidimensional cognitive–behavioural packages aim to broaden the perspective for obesity treatment and combine traditional self-monitoring methods with information, exercise, cognitive restructuring, attitude change and relapse prevention (eg Brownell4). In 1958, Stunkard concluded his review of the past 30 y of attempts to promote weight loss in the obese with the statement ‘Most obese persons will not stay in treatment for obesity. Of those who stay in treatment, most will not lose weight, and of those who do lose weight, most will regain it’.5 In 1993, Wadden updated this review and examined both the short- and long-term effectiveness of both moderate and severe caloric restriction on weight loss as determined by randomized controlled trials (RCTs).6 He examined all the studies involving RCTs in four behavioural journals and compared his findings to those of Stunkard.5 Wadden concluded that ‘Investigators have made significant progress in inducing weight loss in the 35 y since Stunkard's review’. He states that 80% of patients will now stay in treatment for 20 weeks and that 50% will achieve a weight loss of 20 lb or more. Therefore, modern methods of weight loss produce improved results in the short-term. However, Wadden also concludes that ‘most obese patients treated in research trials still regain their lost weight’. This conclusion has been further supported by a recent systematic review of interventions for the treatment and prevention of obesity which identified 92 studies which fitted the authors' inclusion criteria.7 The review examined the effectiveness of dietary, exercise, behavioural, pharmacological and surgical interventions for obesity and concluded that ‘the majority of the studies included in the present review demonstrate weight regain either during treatment or post intervention’. Accordingly, the picture for long-term weight loss is almost as pessimistic as it ever was.

Therefore, RCTs examining the effectiveness of interventions indicate that, although the majority of individuals may lose weight initially, the large majority eventually return to their baseline weight. However, within each trial a small minority not only lose weight initially but successfully maintain this loss. Research is therefore required to explore the differences between the majority of individuals who either fail to lose weight, or regain any weight losses and the minority of individuals who are successful at both weight loss and maintenance. What factors, therefore, may distinguish between the majority of failures and the minority of long-term successes? To date a few studies have specifically examined this minority group.8,9,10,11,12,13 This research together with data from the RCTs for obesity provide some preliminary insights. In particular, the literature highlights a role for a range of variables which, for the purpose of this study, have been conceptualized as profile characteristics, historical factors, help-seeking behaviours and psychological factors.

For profile characteristics, some studies suggest that baseline BMI14,15,16 aspects of class, including employment and income15,17 and gender,10,17 are important. However, the direction of such effects is not always consistent. For example, whilst some studies indicate that lower baseline weight is predictive of greater weight losses and maintenance,15 other studies indicate the reverse effect.16 In addition, many of these studies only examine successful weight loss and maintenance in the short-term as the numbers of weight loss maintainers in the longer-term are small. Secondly, research highlights a role for historical factors including an individual's previous dieting attempts and their weight history.11,12 In addition, it is possible that changes in their smoking behaviour18 and their reproductive history may be contributory factors.19 In terms of help-seeking behaviours, there appear to be several variables which are predictive of success. Primarily, research highlights a role for the types and intensity of weight loss methods used. For example, many studies emphasize the importance of dietary changes such as calorie controlled diets, low-fat diets, high fibre diets and healthy eating.8,9,11,13,15,20,21 In addition, several studies highlight the role of exercise and general increases in physical activity.8,9,20,21 Furthermore, research has highlighted the relative effectiveness of different interventions involving contact with a range of health professionals. These include psychological interventions such as cognitive–behavioural therapy (CBT), counselling, self-help groups and medical interventions involving drug therapy and surgery (see Ref. 7 for a review). The general conclusion from this research is that the more intense the intervention and the professional contact, the higher the probability of successful weight loss and maintenance. However, the relative impact of many of the different forms of intervention remains unexplored.

The final area highlighted by the literature involves psychological factors. However, there is very little research in this area. Rodin and colleagues22 reported the results from a study designed to assess the baseline psychological predictors of successful weight loss. Their results indicated a role for the individual's beliefs about the causes of obesity and their motivations for weight loss. A similar focus on motivations was also reported by Williams et al,23 whose results indicated that motivational style was predictive of weight loss and maintenance. Likewise, Kiernan et al12 indicated that individuals who were more dissatisfied with their body shape at baseline were more successful, suggesting that motivations for weight loss guided by a high value placed on attractiveness may also be important. Research examining other health problems such as chronic fatigue syndrome suggest a possible role for other psychological factors. In particular, research indicates a role for an individual's model of their problem including their beliefs about causes and consequences.24

In summary, RCTs indicate that interventions for obesity are unsuccessful. However, a small minority of individuals both lose weight and maintain this weight loss in the longer-term. The present study aimed to differentiate between the majority of individuals who are unsuccessful and the minority of individuals who successfully lose and maintain their weight loss. In particular, the study aimed to examine differences between individuals who have been obese and who have successfully lost weight and maintained this loss for at least 3 y, obese individuals who have lost weight and regained it and individuals who have remained obese for at least 3 y, in terms of profile characteristics, historical factors, help-seeking behaviours and psychological factors.

Method

Subjects

All subjects were currently a member of a nationwide slimming club. A total of 80 questionnaires were sent by post to those individuals considered to be weight loss maintainers. In addition, 151 questionnaires were distributed throughout the clubs as a means to access the stable obese and weight loss regainers. Of these, 51 of the postal questionnaires were returned giving a response rate of 63.7% and 111 of the hand-distributed questionnaires were returned giving a response rate of 73.5%. Overall, this is a response rate of 70.1%. However, nine of the hand-distributed questionnaires and seven of the postal questionnaires were rejected because the subjects did not fulfill the criteria for one of the three groups. In addition, four men completed the questionnaire and were excluded as their number was insufficient to draw any meaningful conclusions. Therefore, the final response rate was 61.5%. The responders were categorized into three groups as follows: (i) stable obese (n=58; BMI=30+for 3 y or more); (ii) weight loss regainers (n=40; individuals who had been obese (BMI=30+), lost weight in order to be non-obese (BMI<30) and regained weight in order to be currently considered obese (BMI=30+); (iii) weight loss maintainers (n=44; individuals who had been obese (BMI=30+; lost weight in order to be considered non-obese (BMI<30) and maintained this weight loss for 3 y or more).

Design

Because the target group (weight loss maintainers) is small the study accessed this group directly. This enabled sufficient numbers of individuals who had been obese and had both lost and maintained weight to be compared to the control groups.

Procedure

The weight loss maintainers were recruited via the database of a national slimming organization who keep records of those members who have lost and maintained weight and have the heights and weights of their current members, and were sent postal questionnaires for completion. The remaining subjects were recruited from individual slimming clubs in South East England and categorized into stable obese or weight loss regainers following questionnaire completion.

Measures

All subjects completed questionnaires consisting of the following items.

1. Profile characteristics.

Subjects recorded their age, sex, ethnic group (White/Asian/Black/other), class (working/middle/upper), height and present weight (to compute present BMI). In addition, subjects recorded their highest weight since 18 y, lowest weight since 18 y and completed a weight history for 2 y intervals since they were 18 y old. This was used to compute their weight change history as a means to categorize them into one of the three groups: stable obese, weight loss regainers, weight loss maintainers. Further, BMI difference scores were computed for the gaps between their highest ever and their present BMI, their lowest ever and their present BMI, their highest ever and their target BMI (BMI=29), and their lowest ever and their target BMI (BMI=29). In addition, the length of time in years since they crossed the boundary for group categorization (BMI=29) was computed. For the stable obese this reflected the number of years since they became obese since they were 18 y (BMI>29); for the weight regainers this reflected the number of years since they regained their weight to be considered obese (BMI>29) and for the weight loss maintainers this reflected the number of years since they could no longer be considered obese (BMI<30).

2. Historical factors.

(i) Diet history: subjects recorded their current dieting status (yes/no), age of first dieting attempt (y) and years of dieting (y). (ii) Clinical history: subjects recorded whether they had ever had the following: non-insulin-dependant diabetes mellitus (NIDDM), insulin-dependant diabetic mellitus (IDDM), heart disease, angina, cancer, TB, overactive thyroid, underactive thyroid or joint trauma (Have now? yes/no; Had in past? yes/no). These were summated to create a total number of health problems score. In addition, they recorded whether they had had any operations (yes/no). (iii) Health behaviours: smoking (Now? yes/no; ever? yes/no). (iv) Reproduction: number of children; weight gain and maintenance for first child (lb).

3. Help seeking behaviours.

(i) Weight loss methods: subjects rated the following weight loss methods for how often they had tried them ranging from ‘Never’ (1) to ‘Very often’ (5): calorie controlled diet, food avoidance, food weighing, low-fat diet, healthy eating, high fibre diet, exercise, weighing yourself, surgery, counselling, cognitive behaviour therapy (CBT) (with a psychologist), self-help group, jaw wiring. (ii) Professional contact: subjects rated the following professionals for how much contact they had had with them ranging from ‘None’ (1) to ‘Frequently’ (5): practice nurse, GP, hospital doctor, dietician, counsellor, psychologist, slimming club organizer.

4. Psychological factors.

Subjects rated a series of individual items relating to psychological factors which were summated to create total scores and examined for internal reliability using Cronbach's alpha as follows. (i) Beliefs about the causes of obesity: subjects rated a series of items on a scale ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5) to reflect the extent that they believed the following were causes of obesity which were summated to provide total scores: (a) exercise (lack of exercise, not being physically active, inactive lifestyle; alpha=0.8); (b) diet (eating fat, eating sweet foods, eating high calorie foods, eating when not hungry; alpha=0.82); (c) medical (genetics, hormone imbalance, taking medication, eg HRT or the pill, slow metabolism, chemical imbalance in the brain; alpha=0.73); (d) psychological state (depression, lack of will power, laziness, lack of motivation; alpha=0.85). (ii) Beliefs about the consequence of obesity: subjects rated a series of items to reflect the extent to which they believed they were consequences of obesity which were summated to produce total scores as follows: (a) medical (joint problems, heart disease, stomach cancer, bowel cancer, diabetes; alpha=0.87); (b) psychological (depression, anxiety, phobias, low self-esteem, lack of confidence; alpha=0.81). (iii) Motivations for weight loss: subjects rated a series of items on a scale ranging from ‘totally disagree’ (1) to ‘totally agree’ (5) to reflect their motivations for weight loss relating to the following: (a) health (be healthier, live longer, be fitter; alpha=0.65); (b) attractiveness (be more attractive, be able to wear nice clothes, feel more confident about the way I look; alpha=0.59), (c) confidence (increase my self-esteem, like myself more, feel better about myself; alpha=0.77); (c) symptom relief (feel less breathless, feel more energetic, feel more agile; alpha=0.8); (d) external pressure (please my family/partner, please my friends, please my doctor; alpha=0.81).

Results

The results were analysed to examine differences between the three groups (stable obese, weight loss regainers, weight loss maintainers) in terms of their profile characteristics, historical factors, help-seeking factors and psychological factors using ANOVA and post hoc tests for parametric data and Kruskal– Wallis/Mann–Whitney U/Chi squared for non-parametric data. Due to the number of comparisons being made, alpha was set at 0.01.

1. Profile characteristics

Subjects' profile characteristics are shown in Table 1. Overall, the results indicated that the stable obese had been obese (BMI>29) for 12.26 y (s.d.=9.25), that the weight regainers had regained their weight in order to be reconsidered obese (BMI>29) for 5.76 y (s.d.=6.98) and the weight loss maintainers had been non-obese for 8.98 y (s.d. 5.61). Further, the weight loss regainers had spent on average 4.13 y (s.d.=4.15) below the obese threshold (BMI<30). In terms of differences between the groups, the results indicated that the three groups were comparable in terms of their ethnic group, class and height. However, they were different in terms of their age and present BMI. Post hoc tests indicated that the stable obese were younger than both the weight regainers and the weight loss maintainers. In addition, the weight loss maintainers reported a lower present BMI than the other two groups. In terms of changes in weight, the results indicated that the three groups were different in their highest BMI ever, their lowest BMI ever and the gaps between their present BMI and their lowest ever, highest ever and target BMIs. In particular, the weight loss maintainers reported a lower lowest BMI ever and a larger gap between their lowest BMI ever and their target BMI than both the stable obese and the weight loss regainers. Further, the weight loss maintainers reported a lower highest BMI ever, and a larger gap between this variable and their target BMI than the stable obese but were comparable to the weight loss regainers on these latter two variables.

Table 1 Profile characteristics

2. Historical factors

The subjects' diet, clinical, health behaviour and reproductive histories are shown in Table 2.

Table 2 Historical factors

Diet history.

The results indicated that the groups were comparable in terms of the age they first started to diet. However, the weight loss maintainers were less likely to be currently dieting, and reported having tried to lose weight for longer than both the stable obese and the weight loss regainers.

Clinical history.

Subjects were comparable in terms of their number of health problems and whether they had had any operations.

Health behaviours.

The three groups were comparable in terms of their smoking behaviour (now/ever).

Reproductive history.

The results indicate that the stable obese were less likely to have had children. However, of those that had had children, the three groups had had comparable numbers of children. Further, the three groups were comparable in terms of weight gained and maintained for their first child.

3. Help-seeking behaviours

The subjects use of different weight loss methods and their professional contact are shown in Table 3.

Table 3 Help seeking behaviours: methods used and professional contact

Methods used.

The results showed that the stable obese, weight loss regainers and weight loss maintainers were comparable in their reported use of food avoidance, food weighing, self-weighing, low-fat diets, high-fibre diets, exercise and self-help groups (no subjects had used surgery, counselling, cognitive–behavioural therapy or jaw wiring and these were removed from the analysis). However, they differed in terms of their reported use of calorie controlled diets and healthy eating. The results indicated that the weight loss maintainers reported greater use of healthy eating than both the stable obese and weight loss regainers. Further, the weight loss regainers reported more frequent use of calorie controlled diets than the stable obese.

Professional contact.

The results showed that the three groups were comparable in terms of their contact with practice nurses, GPs, hospital doctors, dieticians and slimming club organizers (insufficient subjects had contacted a counsellor or psychologist).

Psychological factors.

Subjects' beliefs about the causes, consequences and their motivations for weight loss are shown in Table 4. For causes, the three groups were comparable in their ratings of diet, exercise and psychological factors as causes of obesity. However, the weight loss maintainers reported lower agreement that it was caused by medical factors compared to both the stable obese and the weight loss regainers. For consequences, the three groups were comparable in their ratings of medical consequences. However, both the weight loss maintainers and the weight loss regainers reported a greater agreement with psychological consequences than the stable obese. Finally for motivations, the three groups were comparable for health, symptoms, attractiveness and external pressure as motivations for weight loss. However, the weight loss maintainers reported a greater belief in confidence as a motivation compared to the stable obese.

Table 4 Psychological factors

Discussion

The present study aimed to examine differences between weight loss maintainers and those individuals who had either maintained a stable obese weight or lost sufficient weight to be considered non-obese but regained it. In particular, the study focused on the potential importance of profile characteristics, historical factors, help-seeking behaviours and psychological factors. However, there are some methodological problems which need to be considered. Firstly, the sample was not selected to be representative of all individuals who either are or have been obese. In fact, the sample consisted of women who belonged to one weight loss organization. Therefore, the extent to which the results can be generalized to other samples is limited. Secondly, the data was collected retrospectively—that is the subjects were asked to complete the questionnaire after they had either lost weight and regained or maintained this loss. In terms of the profile characteristics, historical factors and help-seeking behaviours, this has implications for the accuracy of the subjects' recall. For the psychological factors it raises questions concerning the direction of causality. Research using a representative sample and a longitudinal design is needed to address these issues. However, the results do provide some insights into this under studied group.

The results indicated that the three groups were comparable in terms of ethnic group, social class and height, suggesting that such profile characteristics were unrelated to the degree of weight loss success. These results are in contradiction to some previous studies.10,15,17 However, most previous studies have only explored successful dieting in the short-term. It is possible that these profile characteristics do not relate to longer-term maintenance of weight loss. The results, however, did indicate a role for BMI, with the weight loss maintainers indicating a lower lowest BMI and a lower highest BMI than the other groups. This provides support for previous research15 and suggests that weight loss maintenance may be more likely in those individuals who, although are obese, have been less obese than the more heavy subjects. In addition, the results indicate that the weight loss maintainers were older than the stable obese. This may either reflect the methods of sampling used or indicate that weight loss and maintenance is the result of many years of perseverance. This suggestion is supported by the findings for historical factors. In particular, the results indicated that although the subjects' smoking, reproductive and clinical histories were unrelated to successful weight loss and maintenance, those who had both lost weight and kept it off reported a greater number of years spent dieting.

In terms of help-seeking behaviours, the results showed no differences between the three groups in terms of the contact they had had with health professionals. Furthermore, the results indicated that using food avoidance, food weighing, self-weighing, a low-fat diet, a high-fibre diet, exercise and attending a self-help group as methods of weight loss were also unrelated to successful weight loss maintenance. However, the weight loss maintainers reported greater use of healthy eating. Further, the results also showed a difference between weight loss regainers and the stable obese in terms of their use of calorie controlled diets. However, the results suggested that although calorie controlled diets may relate to initial weight loss they also relate to weight regain. These results find reflection in some previous research which has highlighted the importance of dietary changes.8,9,11,15 However, they suggest that successful weight loss maintenance may be related to a much narrower range of behaviours than often believed. Therefore, although research has argued that the greater the intensity of the intervention the more likely the chance of longer-term success, the results from the present study suggest that the actual type of method used is also important, with healthy eating being beneficial and calorie controlled diets being detrimental.

The final set of factors examined related to the subjects' beliefs about the causes and consequences of obesity and their motivations for weight loss. The results indicated that the weight loss maintainers reported a lower belief that it was caused by medical factors and a greater belief that psychological changes were consequences of obesity. Further, the results indicated that they had been motivated to lose weight for reasons relating to confidence rather than pressure from others or medical reasons such as health and symptom relief. This supports previous research which has indicated a role for an individual's model of the illness and their motivations for change.12,23 However, the results from the present study indicate that it is not their model in general which relates to successful weight loss maintenance but the endorsement of a psychological model. Further, it is not just having higher motivations per se, but motivations which relate to the individual's psychological state. Perhaps these results indicate the importance of short-term rewards both in terms of their experience and a belief that they can be achieved. In particular, weight loss may only be both attained and maintained if obesity is perceived as a problem which can be modified and if any modifications brings changes in the short-term which are valued by the individual concerned.

In summary, the present study provides some insights into the factors which may relate to long-term weight loss and maintenance. In particular, the weight loss maintainers appeared to have been lighter, to be older and have dieted for longer, to have tried healthy eating more frequently, to endorse a psychological model of obesity and to have been motivated to lose weight for psychological reasons. Primarily, these results have implications for understanding the processes involved in weight loss maintenance. In particular, they suggest that it is not only what an individual does which is predictive of success, but also what they believe not only about obesity but also about what they do. Accordingly, for an obese person to lose weight and keep this weight off it would seem that they need both to change their behaviour and believe that their own behaviour is important. Further, they need to perceive the consequences of their behaviour change as valuable. Secondly the results also have implications for improving the success of weight loss interventions. In line with this, it could be argued that treatment should only be targeted at individuals who possess those characteristics which are associated with success. Weight-related interventions could be specifically aimed at younger, lighter individuals who have a psychological model of obesity and are motivated to lose weight by psychological reasons. Such selective treatment could improve the success rates of the interventions. However, rather than regarding such predictors as immutable, it is possible that some of these predictive factors themselves could be modified as part of the intervention. For example, although age and BMI may be givens, perhaps interventions could not only encourage individuals to adopt the behaviours which appear to be predictive of success (ie healthy eating), but also to adopt the ‘successful’ beliefs. Weight loss interventions could include an emphasis on obesity as a modifiable problem and encourage obese individuals to focus on the short-term psychological benefits of weight loss. Such an approach would be considerably different to most current interventions with their emphasis on health, symptom relief and their use of group pressure to increase motivation. However, more importantly such an approach would also conflict with the current medicalization of obesity and the preoccupation with a medical solution.25

Therefore, the results from the present study indicate that a psychological model of obesity may be most predictive of weight loss. Yet, the search for a new wonder drug encourages a medical model. If such a drug is found to be successful then all well and good. If it is not, then not only will many obese people's hopes have been raised unnecessarily, but their beliefs may have been changed in a way that may actually be detrimental to their future attempts at weight loss. Successful weight loss and maintenance may be predicted by an individuals' belief system. But more importantly it may be predicted by the belief that this is the case.

References

  1. 1

    Department of Health . Obesity: reversing the increasing problem of obesity in England. A report from the Nutrition and Physical Activity Task Forces HMSO: London 1995.

  2. 2

    Bray GA . Effects of obesity on health and happiness.In Brownell KD, Foreyt JP (eds) Handbook of eating disorders: physiology, psychology and treatment of obesity, anorexia and bulimia. Basic Books New York 1986.

  3. 3

    Stuart RB, Davis B . Slim chance in a fat world: behavioural control of obesity. Research Press: Champaign, IL 1972.

  4. 4

    Brownell KD . The LEARN programme for weight control. American Health: Dallas, TX 1990.

  5. 5

    Stunkard AJ . The management of obesity NY State J Med 1958 58: 79–87.

  6. 6

    Wadden TA . Treatment of obesity by moderate and severe calorie restriction: results of clinical research trials Ann Intern Med 1993 119: 688–693.

  7. 7

    Systematic review of interventions in the treatment and prevention of obesity. NHS Centre for Reviews and Dissemination, University of York 1997.

  8. 8

    McGuire MT, Wing RR, Klem ML, Hill JO . Behavioral strategies of individuals who have maintained long term weight losses Obes Res 1999 7: 334–341.

  9. 9

    Kayman S, Bruvold W, Stern JS . Maintenance and relapse after weight loss in women: behavioural aspects Am J Clin Nutr 1990 52: 800–807.

  10. 10

    Colvin RH, Olson SB . A descriptive analysis of men and women who have lost significant weight and are highly successful at maintaining the loss Addictive Behav 1983 8: 287–295.

  11. 11

    Hoiberg A, Berard S, Watten RH, Caine C . Correlates of weight loss in treatment and at follow up Int J Obes 1984 8: 457–465.

  12. 12

    Kiernan M, King AC, Kraemer HC, Stefanick ML, Killen JD . Characteristics of successful and unsuccessful dieters: a application of signal detection methodology Ann Behav Med 1998 20: 1–6.

  13. 13

    Klem ML, Wing RR, McGuire MT, Seagle HM, Hill JO . A descriptive study of individuals successful at long term maintenance of substantial weight loss Am J Clin Nutr 1997 66: 239–246.

  14. 14

    Stuart RB, Guire K . Some correlates of the maintenance of weight lost through behaviour modification Int J Obes 1978 2: 225–235.

  15. 15

    Neumark-Sztainer D, Kaufmann NA, Berry EM . Physical activity within community based weight control programme: programme evaluation and predictors of stress Public Health Rev 1995 23: 237–251.

  16. 16

    Wadden TA, Foster GD, Wang J, Pierson RN, Yang MU, Moreland K, Stunkard AJ, VanItallie TB . Clinical correlates of short and long term weight loss Am J Clin Nutr 1992 56: 271S–274S.

  17. 17

    Wong ML, Koh D, Lee MH, Fong YT . Two year follow up of a behavioural weight control programme for adolescents in Singapore: predictors of long term weight loss Ann Acad Med Singapore 1997 26: 147–153.

  18. 18

    Klesges RC, Klesges L . Cigarette smoking as a dieting strategy in a University population Int J Eating Disord 1988 7: 413–419.

  19. 19

    Ohlin A, Rossner S . Maternal body weight development after pregnancy Int J Obes 1990 14: 159–173.

  20. 20

    Haus G, Hoerr SL, Mavis B, Robison J . Key modifiable factors in weight maintenance: fat intake, exercise and weight cycling J Am Diet Assoc 1994 94: 409–413.

  21. 21

    French SA, Jeffrey RW . Current dieting, weight loss history and weight suppression: behavioural correlates of three dimensions of dieting Addict Behav 1997 22: 31–44.

  22. 22

    Rodin J, Bray GA, Attkinson RL, Dahms WT, Greenwa FL, Hamilton K, Molitch M . Predictors of successful weight loss in an outpatient obesity clinic Int J Obes 1977 1: 79–87.

  23. 23

    Williams GC, Grow VM, Freedman ZR, Ryan RM, Deci EL . Motivational predictors of weight loss and weight loss maintenance J Pers Soc Psychol 1996 70: 115–126.

  24. 24

    Heijmans MJ . Coping and adaptive outcome in chronic fatigue syndrome: the importance of illness cognitions J Psychosom Res 1998 45: 39–51.

  25. 25

    Sjostrom L, Rissanen A, Andersen T, Boldrin M, Golay A, Koppeschaar HPF, Krempf M . Ramdomised placebo controlled trial of orlistat for weight loss and prevention of weight regain in obese patients The Lancet 1998 352: 167–172.

Download references

Acknowledgements

The author gratefully acknowledges the help of Cheryl Pitt and Dr Elizabeth Evans with data collection.

Author information

Correspondence to J Ogden.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • obesity
  • successful weight loss maintenance
  • psychological models

Further reading