This study examined the prevalence, distribution and correlates of successful weight loss and successful weight maintenance over three years in a community-based sample of 854 subjects aged 20–45 at baseline. More than half (53.7%) of the participants in the study gained weight within the first twelve months, only one in four (24.5%) successfully avoided weight gain over three years, and less than one in twenty (4.6%) lost and maintained weight successfully. The findings underscore the importance of current public health efforts to prevent weight gain, and suggest that without much greater efforts to promote and support weight control the prevalence of obesity will continue to rise.
We are in the throes of an international obesity epidemic that has prompted health officials throughout the world to encourage people to avoid weight gain and, if overweight, to lose weight.1 Surprisingly little data is available about the extent to which the population is heeding this advice. Existing estimates of the prevalence and correlates of successful weight loss and/or weight gain prevention, based largely on retrospective self-report data, are widely variable.2,3,4,5 Thus, although the success of clinical intervention for weight loss, trends in weight change with age, and changes in the cross-sectional prevalence of obesity over time suggest that successful weight control in the general population is probably not very common,1 more objective data on this issue are needed.
We present data derived from a community-based study of weight gain prevention. The effects of that intervention on weight-related behaviours and mean change in body weight have been reported previously.6 The present study describes the prevalence, distribution and correlates of successful weight loss and successful weight maintenance over 3 y. This study is unique in providing data from a non-clinical sample in which success at weight control has been objectively assessed.
Participants were recruited by means of direct mailings and through media advertisements to the public and university employees. Since obesity and weight gain are more common among women of low socioeconomic status, additional recruitment efforts specifically targeted this group through the Supplemental Nutrition Program for Women, Infants and Children.
Subjects were randomly assigned to one of three treatment conditions. These have been described elsewhere in detail,6 and since this report is not concerned with the effects of the intervention, they are described only briefly here. Half of the subjects were assigned to a no-contact control group, one-quarter received education through monthly newsletters, and one-quarter received education plus participation incentives. The intervention groups received the same educational and behavioural messages that focused on monitoring weight, eating two servings of fruit and three servings of vegetables daily, reducing intake of high-fat foods, and walking three times/week for at least 20 min.
Subjects included in the analyses reported here are those aged 20–45 y at baseline, healthy, not pregnant in the year preceding the study or during the 3 y of follow-up, and for whom baseline, 1 and 3 y follow-up measures on weight are available. The analyses are based on 854 subjects, representing 70% (854/1226) of those initially enrolling for the weight gain prevention study. Most of those not included in the analyses presented here were excluded because one or more follow-up weight measures were not available (289/372), with the remaining 83 subjects excluded because they became pregnant during the study. Those excluded from the analyses were less likely to be White and more likely to be low-income women. At baseline, they also had a higher body mass index (BMI), consumed more fast food meals and more calories, watched more television and reported being more physically active.
At baseline, information was collected regarding age, marital status, education and ethnicity. Due the recruitment methods employed in this study, participants were classified as one of three ‘participant types': high-income women (n=431); low-income women (n=250); and men (n=173).
During baseline and three annual follow-up assessments, each subject's weight was measured in light street clothing without shoes on a calibrated balance beam scale. Height was measured using a wall-mounted ruler. Weight and height were used to compute BMI (BMI=weight in kilograms/ height in meters2).
Weight change at 1 y follow-up was calculated as the percentage difference between 1 y BMI and baseline BMI. Subjects were classified into the following weight change categories: (l) weight loss—lost ≥5% of baseline weight; (2) prevention of weight gain—maintained baseline weight or lost <5%; (3) small weight gain—gained <5% of baseline weight; and (4) large weight gain—gained ≥5% of baseline weight.
Successful weight control.
Weight loss success was defined as losing 5% or more of baseline body weight between baseline and 1 y follow-up (ie weight change category 1 and maintaining that weight or less for a further 2 y. Subjects who had not gained any weight at the 1 y follow-up (ie weight change categories 1 and 2) and did not gain weight for a further 2 y were considered to have successfully maintained their weight.
At baseline and each of the three annual follow-up visits, subjects were asked whether they had used any of 23 weight-control practices in the past year to lose or maintain weight. At baseline and each of the follow-ups, subjects were classified as having intentionally tried to lose or maintain weight if they reported trying any of these practices, or as not having intentionally tried to lose or maintain weight if they had not used any of the 23 weight-control practices.
Usual dietary intake was assessed at baseline and each of the three annual follow-up visits by means of the Block Food Frequency Questionnaire.7 For each of 60 foods, subjects were asked to report their usual serve size and frequency of consumption. This information was combined to provide an estimate of the total intake of energy, as well as percentage energy consumed as fat. Changes in energy and percentage energy as fat between baseline and 1 y and between 1 and 3 y were calculated.
Fast food consumption.
At baseline and each of the three annual follow-up visits, subjects were asked to report how many meals per week they ate from fast food restaurants. Changes in the number of fast food meals consumed/week between baseline and 1 y, and between 1 and 3 y were calculated.
Physical activity was assessed at baseline and the three annual follow-ups using a self-administered version of an instrument that assesses frequency of participation in 13 physical activities over the previous year.8 A physical activity score was calculated by multiplying the frequency per week of each activity by its estimated intensity in metabolic equivalents, and summing these scores across the 13 activities. Changes in physical activity score between baseline and 1 y, and between 1 and 3 y were calculated.
At baseline and each of the three annual follow-up visits, subjects were asked to report how many hours of television they watched on an average day. Changes in the average hours of television viewing/week between baseline and 1 y and between 1 and 3 y were calculated.
Data analyses were performed using SAS statistical software.9 Descriptive statistics for 1 y weight change, successful weight loss and successful weight maintenance were examined using the means and frequencies procedures.
We examined the relationship between 1 y weight change and treatment condition, education, marital status, participant type, ethnicity and weight-control behaviour using descriptive statistics and the chi-square statistic, and between 1 y weight change and age, baseline BMI, changes between baseline and 1 y follow-up in the number of hours/week of television watched, physical activity, the number of fast food meals consumed, percentage energy consumed as fat, and total calorie intake (controlling for baseline levels of these) using the general linear modeling procedure.
Due to the small number of subjects who successfully lost weight, we did not examine the relationship between weight loss and other variables. We did examine the relationship between successful weight maintenance and treatment condition, education, marital status, participant type, ethnicity and weight-control behaviour using descriptive statistics and the chi-square statistic, and between successful weight maintenance and age, BMI at 1 y follow-up, changes between 1 y and 3 y follow-up in the number of hours/week of television watched, physical activity, the number of fast food meals consumed, percentage energy consumed as fat, and total calorie intake (controlling for baseline levels of these) using the general linear modeling procedure.
Table 1 describes the baseline characteristics of the study population. The average age of high-income participants was about 37 y. Almost all were White, and most were well educated and married. The low-income women were younger, less educated, less likely to be married and less likely to be White. The low-income women also reported a higher fat diet and more hours of television viewing. All three groups had a mean baseline BMI above 25 kg/m2. Most subjects reported they had attempted weight control during the previous year.
Weight change over one year
At 1 y follow-up, subjects had gained on average 0.7% of their baseline BMI. Slightly over half of the study population had gained weight at 1 y. Thirty-eight percent (n=325) had gained up to 5% of their baseline BMI; 15.7% (n=134) gained 5% or more of their baseline BMI. Just over a third of the sample (35.1%; n=300) maintained their BMI or lost up to 5%; 11.2% (n=96) lost 5% or more of their baseline BMI.
Univariate analyses revealed that 1 y weight change was not associated with treatment condition, age, education, marital status, participant type, ethnicity, baseline BMI, or whether subjects had intentionally tried to lose or maintain weight in the previous year. Changes between baseline and 1 y follow-up in total calorie intake and the number of hours/week of television watched were also not significantly related to weight change. However, 1 y weight change was inversely related to 1 y change in physical activity, and positively related to 1 y change in the number of fast food meals consumed (Table 2).
Successful weight control
Of the 96 subjects who had lost 5% or more of their baseline BMI at 1 y follow-up, 39 (40.6%) successfully maintained their weight loss for a further 2 y. In other words, only 4.6% of all subjects in this study (39/854) lost 5% or more of their baseline BMI and were able to maintain that weight loss for 2 y. Among the 396 subjects who did not gain any weight at 1 y follow-up, 209 (52.8%) successfully maintained their weight for a further 2 y (ie they were still at or below their baseline weight at 3 y follow-up). Overall, 24.5% of subjects in this study (209/854) were successful in preventing weight gain at 1 y and also at 3 y.
Univariate analyses revealed that successful weight maintenance was not associated with age, education, marital status, participant type, ethnicity, BMI at 1 y follow-up, whether subjects had intentionally tried to lose or maintain weight, or changes between 1 and 3 y follow-up in total calorie intake, percentage energy as fat and the amount of television watched. Successful weight maintenance was associated with changes between 1 and 3 y follow-up in physical activity and the number of fast food meals consumed. The mean change in physical activity score (3 y follow-up −1 y follow-up) for those who successfully maintained their weight was −1.2 (s.e.=1.8), compared to a mean change of −9.1 (s.e.=1.9) for those who gained weight (P=0.002). The mean change in the number of fast food meals/week was −0.2 (s.e.=0.1) for those who maintained weight, and 0.1 (s.e.=0.1) for those who gained (P=0.04).
In this large and heterogeneous sample of adults, relatively few men or women were successful in either losing weight or avoiding weight gain over a 3 y period. Among the participants in this study, all of whom were interested in preventing weight gain and half of whom participated in a low intensity intervention, more than half gained weight within the first 12 months, only one in four successfully avoided weight gain over 3 y, and less than one in 20 lost and maintained weight successfully. In addition, reported weight control efforts were unrelated to success. As far as we are aware, these findings are unique, being based on objective data derived from a population-based sample.
The rate of successful weight loss we observed was lower than previous estimates.2,3,4,5 This might be due to several factors. Unlike the earlier studies, we used objective rather than self-reported weights to assess success at weight control. We also defined success as weight loss or maintenance over the course of a year that was maintained for a further 2 y, whereas earlier studies have used a variety of other criteria. In addition, while drawn from the general population, it may be that, compared to the rest of the population, our volunteers are people who have greater problems controlling their weight. Finally, the data reported here were collected recently, while the earlier research is at least 10 y old (some dating back more than 20 y). It may be that the environment is more ‘obesogenic’ now than it was a decade ago, and therefore it is now more difficult for individuals to successfully control their weight, even for those with an interest in doing so.
The findings presented here underscore the importance of current public health efforts to prevent weight gain. Although attempts at dieting and weight control are promoted by health authorities and are common in the general popualtion,10 our findings suggest that, without much greater efforts to promote and support weight control, most people will be unable to avoid weight gain and very few will manage to lose weight, and as a consequence the prevalence of obesity will continue to rise.
World Health Organisation . Prevention and management of the global epidemic of obesity. Report of the WHO consultation on obesity. World Health Organization: Geneva, 1998.
Schachter S . Recividism and self-cure of smoking and obesity. Am Psychol 1982 37: 436–444.
Jeffery RW, Wing RR . Recividism and self-cure of smoking and obesity: data from population studies Am Psychol 1983 38: 852.
Williamson DF, Serdula MK, Anda KF, Levy A, Byers T . Weight loss attempts in adults: goals, duration, and rate of weight loss Am J Public Health 1992 82: 1251–1257.
French SA, Jeffery RW, Folsom AR, McGovern P, Williamson DF . Weight loss maintenance in young adulthood: prevalence and correlations with health behavior and disease in a population-based sample of women aged 55–69 y Int J Obes Relat Metab Disord 1996 20: 303–310.
Jeffery RW, French SA . Preventing weight gain in adults: the pound of prevention study Am J Public Health 1999 89: 747–751.
Block G, Hartman AM, Dresser CM, Carroll MD, Gannon J, Gardner L . A data-based approach to diet questionnaire design and testing Am J Epidemiol 1986 124: 453–469.
Jacobs DR, Hahn LP, Pirie P, Sidney S . Validity and reliability of a short physical activity history: CARDIA and the Minnesota Heart Health Program J Cardiopulmonary Rehab 1989 9: 448–459.
SAS Institute . SAS language and procedure: usage version 6. SAS Institute: Cary, NC, 1989.
French SA, Jeffery RW . Consequences of dieting to lose weight: effects on physical and mental health Health Psychol 1994 13: 195–212.
This research was supported by grant DK45361 from the National Institute of Diabetes and Digestive and Kidney Diseases, with additional funding from the Centers for Disease Control and Prevention. David Crawford was supported by a National Health and Medical Research Council Public Health Fellowship and a Heart Foundation Nutrition Research Fellowship.
About this article
Cite this article
Crawford, D., Jeffery, R. & French, S. Can anyone successfully control their weight? Findings of a three year community-based study of men and women. Int J Obes 24, 1107–1110 (2000). https://doi.org/10.1038/sj.ijo.0801374
- weight maintenance
Altered cross-talk between the hypothalamus and non-homeostatic regions linked to obesity and difficulty to lose weight
Scientific Reports (2017)
Improving Cancer Risk Awareness Including Obesity as a Risk Factor for Cancer in a Small U.S. Community
Journal of Cancer Education (2017)
Meeting physical activity guidelines is associated with reduced risk for cardiovascular disease in black South African women; a 5.5-year follow-up study
BMC Public Health (2014)
Long term effects of a health promotion intervention in low socioeconomic Arab- Israeli kindergartens
BMC Pediatrics (2013)
Journal of Bioethical Inquiry (2013)