Objective: To examine breakfast consumption in subjects maintaining a weight loss in the National Weight Control Registry (NWCR).
Research Methods and Procedures: A cross-sectional study in which 2959 subjects in the NWCR completed demographic and weight history questionnaires as well as questions about their current breakfast consumption. All subjects had maintained a weight loss of at least 13.6 kg (30 lb) for at least 1 year; on average these subjects had lost 32 kg and kept it off for 6 years.
Results: A large proportion of NWCR subjects (2313 or 78%) reported regularly eating breakfast every day of the week. Only 114 subjects (4%) reported never eating breakfast. There was no difference in reported energy intake between breakfast eaters and non-eaters, but breakfast eaters reported slightly more physical activity than non-breakfast eaters (p = 0.05).
Discussion: Eating breakfast is a characteristic common to successful weight loss maintainers and may be a factor in their success.
As the awareness of the negative health effects of obesity mounts, there is increasing focus on how to produce and maintain weight loss in obese individuals (1). Many overweight and obese individuals are able to lose significant amounts of weight, but the majority regain this weight over a period of 2 to 3 years or less (2). Because the improvements in health that occur with weight loss are, in most cases, lost on weight regain (3), it becomes important to learn how to help people maintain weight loss over the long-term. Unfortunately, most effort has been focused on identifying behaviors that contribute to short-term weight loss with less effort devoted to identifying behavior associated with weight-loss maintenance.
One obvious way to identify candidate behaviors for weight-loss maintenance is to study behaviors common to those who succeed in achieving and maintaining a weight loss over the long-term. The National Weight Control Registry (NWCR) was established in 1994 by Drs. James Hill and Rena Wing to investigate the characteristics and behaviors of individuals who have been successful at achieving their goal of losing weight and keeping it off long-term. The NWCR is the largest ongoing study of individual successful weight-loss maintainers. The registry consists of >2900 reduced-obese subjects across the United States. To qualify for NWCR membership, individuals must have maintained a weight loss of ≥13.6 kg (30 lb) for ≥1 year.
We have previously identified several behaviors that are common to successful weight-loss maintainers. These include: eating a diet low in fat and high in carbohydrate; regular self-monitoring of body weight and food intake; and high levels of daily physical activity (4).
A behavior that is often targeted in weight management programs and that may contribute to long-term success in weight maintenance is breakfast eating. Several large studies have examined the contribution of breakfast to general nutrition of the diets of adults and children (5, 6, 7). In general, individuals who eat breakfast regularly have more adequate micronutrient intakes (5, 6, 8) and lower percentages of calories from fat (8). These findings are more striking for those individuals whose breakfast includes a ready-to-eat cereal (7, 8). Despite these findings, the incidence of skipping breakfast has increased between 1965 and 1991 from 14% to 25% for U.S. adults (9). More Americans are routinely skipping breakfast perhaps because of a belief that by omitting breakfast they are reducing their total calorie intake and thereby helping their efforts at weight control (5, 10).
Although some studies have examined the role of breakfast consumption in short-term weight loss (11), there are no studies of the role of breakfast consumption in weight-loss maintenance. Short-term studies examining the role of breakfast and energy balance have had mixed results, with some studies showing lower total intakes in non-breakfast eaters (5, 10) and other studies supporting that breakfast skippers tend to eat more calorie-dense food later in the day (6, 12).
This study describes breakfast consumption in a large group of long-term obese-reduced individuals in the NWCR. Our objective was to see whether breakfast eating was a common characteristic among these individuals to evaluate whether breakfast eating may be a candidate behavior for successful weight-loss maintenance.
Research Methods and Procedures
Subjects and Procedures
The subjects in this study were 2959 members of the NWCR: 2350 were women (79.5%) and 607 were men (20.5%; 2 did not report gender). To be eligible for enrollment, an individual must be ≥18 years old, have lost ≥13.6 kg (30 lb), and maintained the weight loss for ≥1 year. Registry members were recruited over the past 7 years, in large part, through coverage of the NWCR by local and national media. Individuals contacting the registry were mailed an informed consent form. On return of the signed consent form, participants were sent questionnaires to be completed and returned in a postage-paid return envelope.
All data used in the present study were collected on initial entry into the NWCR. In particular, the following data were used for the analyses in this study.
Demographic and Weight History
Participants were asked to provide basic demographic information (age, gender, education level, and marital status) and to indicate their lifetime maximum adult weight, total weight loss, current body weight, and current height. Using this information current and maximum body mass indexes (BMIs; in kg/m2) were calculated. Participants also reported the dates they were at each weight so that the duration of the required 13.6-kg weight loss could be calculated.
Dietary Intake and Levels of Physical Activity
Dietary intake during maintenance was assessed by the Block Food-Frequency Questionnaire (13). Subjects were given a standardized list of 100 foods and asked to estimate their usual serving sizes and the frequency with which they consumed each food over the past year. To assess current levels of physical activity, participants were asked to complete the Paffenbarger Physical Activity Questionnaire, which asks questions about walking, stair climbing, and recreational activities engaged in during the past week (14). From this questionnaire an estimate of the energy expended through physical activity was calculated.
At baseline, all registry participants (n = 2959) were asked how many days of a 7-day week they typically eat breakfast. Of the 2959 subjects, 2953 answered this question in the baseline questionnaire packet (6 subjects did not respond). In addition, ∼1 year after the creation of the registry, the Foods Habits Questionnaire (15) was added to the baseline registry packet, and individual items from this form were used to further determine registry members’ breakfast habits. First, subjects indicated (yes/no) if they had eaten breakfast in the last 3 months. Individuals responding yes to this question then indicated, using a four-point scale (1 = always and 4 = rarely or never), how often they ate a hot or cold cereal for breakfast, and how often they ate fruit (not fruit juice) for breakfast. Because this questionnaire was added after the start of the registry, data are only available for 1895 subjects. The terms breakfast and cereal were self-defined by the NWCR subject.
Data analyses were performed by using SAS (SAS/STAT User's Guide, Version 8, 2000; SAS Institute Inc, Cary, NC). Results are presented as mean ± SD or as frequencies. Subject characteristics were compared for men and women using independent-sample t tests for numerical variables and χ2 tests for frequencies.
Characteristics of Subjects
Selected characteristics of registry members are shown in Table 1. Of the registry members, 79% were women, 95% were white, and 64% were currently married. Although the minimum weight loss required for entry into the registry was 13.6 kg, many registry participants had lost more weight than the minimum requirement. On average, participants lost 32.4 ± 18.0 kg (31.9 ± 16.9 kg in women, 34.5 ± 21.8 kg in men, p = 0.006). Although the minimum duration of weight maintenance was 1 year, registry members on average maintained the minimum weight loss (13.6 kg) for 6 years. Registry members had a maximum BMI of 36.2 ± 8.6 kg/m2 before their successful weight loss attempt and a BMI of 24.8 ± 4.6 kg/m2 after their weight loss.
Table 1. Physical characteristics of NWCR subjects (mean ± SD)*
Women (n = 2350)
Men (n = 607)
Total sample (n = 2959)
Two subjects did not report gender.
44.8 ± 12.0
51.8 ± 13.2
46.3 ± 12.5
Maximum body mass index (BMI) (kg/m2)
36.1 ± 8.6
36.6 ± 8.9
36.2 ± 8.6
Current BMI (kg/m2)
24.5 ± 4.6
25.9 ± 4.2
24.8 ± 4.6
Change in BMI (kg/m2)
11.6 ± 6.1
10.6 ± 6.5
11.4 ± 6.2
Maximum weight (kg/m2)
99.1 ± 24.5
118.3 ± 30.9
103.0 ± 27.1
Current weight (kg)
67.2 ± 13.5
83.8 ± 15.2
70.6 ± 15.4
Weight loss (kg)
31.9 ± 16.9
34.5 ± 21.8
32.4 ± 18.0
Duration of weight loss (years)
5.7 ± 7.2
7.1 ± 9.1
6.0 ± 7.6
When NWCR members in the baseline questionnaire were asked how many days of a 7-day week they eat breakfast, only 114 (3.9%) subjects reported that they never eat breakfast (0 days per week) and 78.3% (n = 2313) reported eating breakfast 7 days per week. Almost 90% reported eating breakfast on most days of the week (4 days or more; Figure 1). Similar results were obtained with the Food Habits Questionnaire (15). Based on this questionnaire, 94.8% of the NWCR members (n = 1796) reported eating breakfast and 5.2% (n = 99) reported not eating breakfast in the last 3 months. The mean age (44.9 ± 10.4 years) for the subjects who did not report eating breakfast was slightly lower than the group who reported eating breakfast (47.4 ± 13.0 years, p = 0.024). Slightly more men responded that they did not eat breakfast (7.6%) than women (4.6%, p = 0.017).
Because two separate questions assessed breakfast consumption, we compared the congruency of answers between these two items. Of the 1796 subjects that responded that they had eaten breakfast in the last 3 months, only 12 then later reported that they eat breakfast 0 days per week. Of the 99 subjects who answered in the negative to eating breakfast in the last 3 months, 45 reported eating breakfast 1 to 7 days per week. Of the 45 subjects, 30 reported eating breakfast 1 to 2 days per week, whereas 10 reported that they eat breakfast 7 days per week. The κ-statistic for agreement between the question about breakfast eating in the past 3 months from the Food Habits Questionnaire and the question about eating breakfast 1 or more days per week from the baseline questionnaire was 0.64, indicating good agreement between the two questions (16). It is not clear why the 45 subjects who initially reported not eating breakfast reported some breakfast eating in a separate question. If, however, we classify these as breakfast eaters, it only decreases the proportion of NWCR subjects who skip breakfast.
Cereal and Fruit Consumption
NWCR subjects who answered that they did eat breakfast in the last 3 months (n = 1796) were then asked how often they ate a hot or cold cereal for breakfast: and how often they ate fruit (not fruit juice) for breakfast: 29.6% (n = 530) responded that they always eat cereal, 30.1% (n = 540) answered that they usually/often eat cereal, 20.6% (n = 369) responded that they sometimes eat cereal, whereas 19.7% (n = 354) responded that they rarely/never consumed cereal for breakfast (3 subjects did not respond). A total of 31.4% (n = 564) responded that they always eat fruit, 24.2% (n = 435) answered that they usually/often eat fruit, and 23.6% (n = 423) responded that they sometimes eat fruit, whereas 20.8% (n = 373) responded that they rarely/never consumed fruit for breakfast (1 subject did not respond).
Differences between Breakfast Eaters and Non-Eaters
We were interested in whether there were differences between breakfast eaters and non-eaters in terms of weight-loss maintenance or behaviors related to weight-loss maintenance. In this analysis, we defined breakfast eaters as those reporting eating breakfast four or more times per week and non-eaters as those eating breakfast three or fewer days per week. Using these criteria, we had 314 non-breakfast eaters and 2645 breakfast eaters. There was no difference between groups (non-breakfast eaters vs. breakfast eaters) in amount of weight lost (34 vs. 32 kg, p = 0.14) or in duration of weight-loss maintenance (7.7 vs. 7.9 years, p = 0.29). Furthermore, there was not a significant difference in self-reported energy intake between groups. Breakfast eaters reported an energy intake of 1394 kilocalories per day and non-breakfast eaters reported an intake of 1366 kilocalories (p = 0.50). Breakfast eaters did report engaging in slightly more physical activity than non-breakfast eaters (2657 vs. 2391 kcal/wk, p = 0.05)
We have previously defined some behaviors that are common to successful weight-loss maintainers in the NWCR (4). These include: eating a low-fat, high-carbohydrate diet; regular self-monitoring of body weight and food intake; and high levels of physical activity. We now add a fourth behavior common to these individuals—eating breakfast on a regular basis.
It is striking not only that breakfast eating is a frequent behavior among individuals within this group, but also that such a high proportion (78%) report eating breakfast every day of the week. It is also striking that very few (≤5%) of these successful weight-loss maintainers report never eating breakfast. Skipping breakfast seems to be far less prevalent among successful weight-loss maintainers than among the general public (9).
This was not a prospective study, so we cannot conclude that eating breakfast was a cause of the success in weight- loss maintenance. One purpose of establishing this registry was to identify characteristics of successful weight-loss maintainers to develop hypotheses about which behavioral factors may be important for long-term weight-loss maintenance. These results suggest that eating breakfast on a regular basis is a good candidate for such a behavioral factor.
There are several possible reasons why regular breakfast eating may be useful in weight-loss maintenance. First, eating breakfast may reduce the hunger seen later in the day that may in turn lead to overeating. We did not find differences in energy intake between groups, so it seems that those NWCR subjects who skip breakfast are not prone to overeating at subsequent meals. Second, breakfast eaters may choose less energy-dense foods during the remainder of the day. Finally, nutrients consumed at breakfast may leave the subject with a better ability to perform physical activity. We found some indication that breakfast eaters report engaging in more physical activity than non-breakfast eaters. It is also possible that eating breakfast regularly may simply be a marker for a low-calorie, low-fat eating style.
These results may be useful for individuals attempting to lose weight. Such individuals may be tempted to skip breakfast as a means of reducing total energy intake. However, our data suggest that skipping breakfast is a very uncommon behavior among individuals successful at long-term weight-loss maintenance, and results of several other studies have also suggested that eating a regular breakfast can have a positive impact on general nutrition indices. Thus, individuals attempting to lose weight may benefit from including regular, balanced breakfasts in their dieting regimens.
In previous studies of subjects in the NWCR, we have identified behaviors that are seen frequently in these successful weight-loss maintainers. As with other behaviors, a small proportion of NWCR subjects were able to achieve success in weight-loss maintenance without eating breakfast. In this study, these few individuals who skipped breakfast were as successful in weight-loss maintenance as breakfast eaters. There just were not many subjects in this group. In another study, for example, we found that 9% of NWCR subjects were able to maintain weight loss without regular physical activity (4). Although we cannot conclude that behaviors such as eating breakfast and regular physical activity are essential for success in weight-loss maintenance, they are so common among this group that we should carefully consider whether they can increase the chances of success in others trying to achieve and maintain a weight loss.
This was not a prospective study, so we cannot conclude that breakfast eating was a cause of success in this sample of highly successful weight-loss maintainers. Future studies, prospectively linking breakfast eating with success at weight maintenance, would help to clarify this relationship.
This study was supported in part by National Institutes of Health Grants DK02703, DK42549, and DK48520. Partial support was provided by an unrestricted grant from General Mills, Inc.