Determining the Effectiveness of Take Off Pounds Sensibly (TOPS), a Nationally Available Nonprofit Weight Loss Program

Authors

  • Nia S. Mitchell,

    Corresponding author
    1. Department of Internal Medicine, Division of General Internal Medicine, Colorado Health Outcomes, Center for Human Nutrition, University of Colorado Denver, Denver, Colorado, USA
    Search for more papers by this author
  • L. Miriam Dickinson,

    1. Department of Family Medicine and Colorado Health Outcomes Program, University of Colorado Denver, Denver, Colorado, USA
    Search for more papers by this author
  • Allison Kempe,

    1. Department of Pediatrics, Colorado Health Outcomes Program, University of Colorado Denver, Denver, Colorado, USA
    2. Children's Outcomes Research Program, The Children's Hospital, Aurora, Colorado, USA
    Search for more papers by this author
  • Adam G. Tsai

    1. Department of Internal Medicine, Division of General Internal Medicine & Center for Human Nutrition, University of Colorado Denver, Denver, Colorado, USA
    Search for more papers by this author

(Nia.Mitchell@ucdenver.edu)

Abstract

Obesity is a major public health issue in the United States. Many commercial weight loss programs are available, but their costs prohibit some people from participating. This study evaluated the effectiveness of Take Off Pounds Sensibly (TOPS), a low-cost, nonprofit weight loss program. Longitudinal mixed-effects repeated-measures modeling of the TOPS national database was used to model changes in weight for TOPS participants who joined in 2005, 2006, and 2007 and had at least one annual renewal between 2006 and 2008. Separate analyses were performed on individuals with consecutive annual renewal and those with nonconsecutive annual renewal. During the study period, 42,481 individuals renewed their membership at least once, including 2,427 individuals with nonconsecutive renewals. Individuals with consecutive renewals in TOPS lost 5.9–7.1% of their initial weight over a period of 1–3 years. People who remained in the program lost ∼6% of initial weight in the first year and maintained that weight loss for up to 3 years. TOPS participants with nonconsecutive renewal generally lost less weight than those with consecutive renewal. TOPS is associated with moderate weight loss among participants who remain in the program for at least 1 year. This degree of weight loss is likely to be clinically important for many individuals. TOPS is available at substantially lower cost than commercial weight loss programs, with similar results. Head-to-head trials of TOPS and popular commercial programs are needed.

Introduction

Obesity is perhaps the single most important health issue facing the United States today. According to the 2005–2006 National Health and Nutrition Examination Survey (NHANES), over 72 million US adults (34.3%) are obese (1). Obesity is associated with many common medical conditions including hypertension, diabetes, sleep apnea, coronary artery disease, stroke, and cancer (2,3). In addition, obesity disproportionately affects ethnic minorities (4) and lower socioeconomic status (SES) groups (5) and is associated with substantial economic costs (6,7). Weight losses of 5–10% of initial weight can result in significant health benefits including lower blood pressure, improved blood sugar control and lipid profiles (8,9,10), decreased progression of glucose intolerance to type 2 diabetes (11), and decreased severity of obstructive sleep apnea (12).

Take Off Pounds Sensibly (TOPS; Milwaukee, WI) is a national nonprofit organization that helps its members manage weight problems through group support and education (13). Although TOPS offers assistance to individuals with eating disorders, its major role is to help overweight and obese individuals achieve weight loss. The TOPS plan includes information about healthy eating, exercise, and behavior modification. Members receive a booklet with a 6-week lesson plan, a 1-year subscription to TOPS News, and membership in a local chapter (13). TOPS members are encouraged, but not required, to obtain a goal weight from a health-care professional. Members who attain this goal weight can become part of the maintenance arm of TOPS, known as Keep Off Pounds Sensibly. TOPS and Keep Off Pounds Sensibly members attend the same weekly weigh-in and support meetings. Four members are required to start a new TOPS chapter—a leader, a co-leader, a secretary, and a treasurer. The leader also appoints an individual to record weights.

The difference in cost between TOPS and commercial programs is substantial. TOPS is a nonprofit organization. Most of its revenue is generated from annual dues, and its chapter leaders are elected volunteers. TOPS costs up to $90/year (13). In contrast, Weight Watchers costs between $480 and $625 annually, (14) and its group leaders are paid employees. Programs that include the cost of food (e.g., Nutrisystem, Jenny Craig) cost $3600–6500 annually (ref. 15 and K. Stephens, Jenny Craig Associate, personal communication, 5 August 2009).

Several studies have evaluated the effectiveness of commercial weight loss programs (16,17,18) and of self-initiated diet regimens (19,20,21). However, no recent studies have examined the effectiveness of nonprofit weight loss programs, such as TOPS. At least two large nonprofit weight loss programs operate in the United States, Overeaters Anonymous and TOPS (21). In 1974, Levitz and Stunkard used data from 16 TOPS chapters to compare behavior modification (provided by a professional therapist) with the TOPS program. They found that behavior modification was associated with greater weight loss and lower attrition at 1 year (22). Based on this study, TOPS incorporated behavior modification into its program. (21) Because of changes in the TOPS program and because the prevalence of obesity in the United States is nearly double what it was in 1974 (23), reevaluation of TOPS was indicated.

We conducted an analysis of the TOPS administrative database to evaluate the effectiveness of this low-cost weight loss program. Our primary aim was to determine the weight change among US members with consecutive annual renewal in the TOPS program. A secondary aim was to determine the weight change associated with nonconsecutive annual renewal.

Methods and Procedures

Study design

This study was a retrospective cohort analysis of longitudinal weight change of TOPS members.

Data sources

The data source is a 3-year subset of the national database of TOPS Club Inc. of members in the United States who joined TOPS in 2005, 2006, and 2007. The data contained information about renewal weights through 2008. The duration of the study was a function of the available data. Specifically, before this study, the data set was periodically purged of members who missed three consecutive annual renewals. Therefore, un-purged data were only available from 2005 through 2007.

The Information Technology manager at TOPS sent the investigators an Excel spreadsheet that contained the following descriptive categories: country in which chapter is located; membership status (active or inactive); state in which the chapter is located; gender; date member joined TOPS chapter; projected date of membership renewal; weight at program entry; highest weight recorded as a member of TOPS; weight at the most recent annual renewal; starting weight minus weight at the most recent annual renewal; highest weight minus weight at the most recent annual renewal; and weight at each annual renewal (2006, 2007, and 2008). The initial data also contained the first two letters of the members' first names, but these were removed by the investigators. Because the data source was de-identified and contained no protected health information, the analysis was designated as exempt by the Colorado Multiple institutional review board.

Members were weighed at weekly meetings by the weight recorder. The weight recorder entered members' weights on the renewal applications. Renewal applications were mailed to an area coordinator, who subsequently sent the renewal to the TOPS National Office. Renewal applications were processed at the national office and weights were entered into the TOPS database. Members' weights were sent to the national TOPS database only at the time of annual membership renewal. Therefore, follow-up weights were available only for people who renewed their membership at least once, and it was impossible to know the extent of individuals' participation in TOPS (e.g., number of meetings attended) between the times of renewals.

Study participants

Participants were those members who joined TOPS between 1 January 2005 and 31 December 2007. Exclusion criteria included the following: (i) Female members whose initial weights were <145 pounds (65.9 kg) and male members whose initial weights were <171 pounds (77.7 kg) and (ii) members whose annual membership was never renewed.

Female and male participants whose initial weights <145 pounds (65.9 kg) and <171 pounds (77.7 kg), respectively, were excluded from the analysis because these weights correspond with a BMI of 25 kg/m2 for women and men with heights of 5 feet 3.8 inches and 5 feet 9.4 inches, the average heights in the United States (24). Of the 111,908 members in the original data set, only 3,753 (3.3%) were eliminated based on this criterion. Another 48 people were excluded because there were no valid starting weight or gender data. There were 65,626 individuals who were eligible to be included in the study based on their starting weight who did not renew their membership; therefore they were excluded from the analysis. Thus, after exclusions, 42,481 TOPS participants who joined between 1 January 2005 and 31 December 2007 and who had at least one renewal between 2006 and 2008 were included in the final analysis. A diagram of study inclusion is outlined in Figure 1.

Figure 1.

Study enrollment. The number in “low starting weights” includes 3,557 female participants with starting weights <65.9 kg (145 lbs), 196 male participants with weights <77.7 kg (171 lbs), and 48 participants for whom there were no valid starting weight or gender data.

Participants who were included in the analysis were divided into two groups based on whether their annual renewal was consecutive or nonconsecutive. Participants were classified as consecutive renewal if they renewed their memberships for consecutive years that they were eligible during the study period. Participants were defined as nonconsecutive renewal if they missed one renewal for which they were eligible, but subsequently renewed their membership. For instance, if a participant did not renew her membership in year 1, but did in year 2, she would be included in the nonconsecutive renewal category. We analyzed weight change separately for these two groups because we believed that participants with consecutive renewal might be more committed to weight management than those with nonconsecutive renewal and would therefore sustain greater weight losses.

Variables

Variables used were gender and weight. The primary study outcome was weight change during enrollment in the TOPS program. Weight was available at the time of program entry and annually at the time of membership renewal. Weight change from baseline was calculated in both kilograms and also as a percent of starting weight. The percentage of weight change was also categorized as a weight loss or weight gain of 0– <5%, ≥5– <10%, or ≥10% of initial weight.

The year that members joined is the cohort in which they were followed. Thus, there were three cohorts—2005, 2006, and 2007. The cohorts were combined to estimate average weight change based on duration of follow-up. For example, because each cohort was followed for at least 1 year, the average 1-year weight change includes data from all three cohorts. Similarly, the average 2-year weight change is based on participants who renewed their membership 2 years after they initially joined. Thus, 2-year weight change includes data from the 2005 and 2006 cohorts, but not the 2007 cohort (which had a maximum follow-up of only 1 year).

Statistical methods

Weight change analyses were stratified by gender because the starting weights of males and females were significantly different. Data were also separated for analyses based on whether the participant's membership renewal was consecutive or nonconsecutive.

The data were analyzed using mixed-effects repeated-measures models with individual random intercepts, thus allowing for different starting weights for each individual while adjusting for the correlation of multiple measures per individual (SAS Proc Mixed). Differences between baseline weight and subsequent renewal weights were estimated in the model and tested to determine whether weight change was significantly different from zero. Independent variables included year of follow-up and cohort year. Data were analyzed using SAS, version 9.2 (Cary, NC).

Individuals who renewed their membership were compared to those who did not renew their membership on two variables: (i) gender composition and (ii) average starting weight using χ2 analyses.

Results

Participant characteristics

Across all cohorts, the average starting weight for female participants was 98.2 kg and for male participants was 121.9 kg. More than 95% of the study population was female. This is similar to the gender distribution of all individuals who joined TOPS during the study period and is also consistent with the gender distribution in studies of commercial weight loss programs (21).

Comparison based on renewal status

Participants who did not renew their membership included a slightly higher percentage of women than those who renewed their membership (95.6 vs. 95.3%, P = 0.04). There was no statistically significant difference between the average starting weights of participants who renewed their membership vs. those who did not, either for women or for men. The average starting weights for females who renewed their membership and those who did not renew their membership were 98.2 and 98.1 kg, respectively (P = 0.35). The average starting weights for male participants who renewed their membership was 121.9 kg, compared to 123.1 kg for those who did not renew their membership (P = 0.13).

Weight change outcomes

The results of the repeated-measures model for individuals with consecutive renewal are shown in Tables 1 and 2. In Table 1, the weight change is shown in kilograms, and in Table 2, as the percentage change from baseline. The average weight change as a percentage of initial body weight was comparable for male and female participants throughout the 3 years of the study. Women and men who renewed their memberships at 1 year lost 5.9 and 6.3% of initial body weight, respectively. Female participants who renewed their TOPS memberships for two consecutive years lost 6.7% of initial weight during the first year and had a cumulative loss of 6.8% through the second year. Male participants who renewed their TOPS memberships for two consecutive years lost 6.6% of initial weight during the first year and had a cumulative loss of 6.6% through the second year. The cumulative weight loss as a percentage of initial body weight for female participants with three consecutive renewals was 6.8, 7.1, and 6.9% for years 1, 2, and 3, respectively. The cumulative weight loss for male participants with three consecutive renewals was 6.6, 6.6, and 6.4% for years 1, 2, and 3, respectively.

Table 1.  Weight change of members with consecutive renewal
inline image
Table 2.  Percent weight change of members with consecutive annual renewal
inline image

Nonconsecutive renewal

The results for TOPS participants with nonconsecutive renewal are shown in Tables 3 and 4. Table 3 shows the weight change in kilograms; Table 4 shows the weight change as a percentage of initial weight.

Table 3.  Weight change for members with nonconsecutive annual renewal in kilograms
inline image
Table 4.  Percentage weight change for members with nonconsecutive annual renewal
inline image

TOPS participants with nonconsecutive renewal generally lost less weight than those with consecutive renewal. There were 1,724 female participants and 86 male participants who did not renew their memberships in the first year, but had at least one subsequent renewal. In this group, 1,724 women lost 1.8% of their baseline weight at year 2 and 377 women lost 3.3% of their baseline weight at year 3; while 86 men lost 5.2% of their baseline weight at year 2 and 18 men lost 10.7% at year 3.

There were 537 female participants and 36 male participants who renewed their memberships in years 1 and 3, but did not renew in year 2. In this group, women lost 5.4% of their baseline weight in the first year, and in year 3 their cumulative weight loss dropped to 2.6%, while men lost 3.4% of their baseline weight in the first year, with a cumulative weight loss of 0.8% in year 3. There were 44 female participants who did not renew their memberships in the first or second years, but renewed in the third year. These women lost 4.5% of their initial weight at the year-3 renewal. There were no males in this category.

The gender distribution of individuals with consecutive annual renewal and nonconsecutive annual renewal were similar—95.4% of those with consecutive renewal were female, compared with 95.0% of those with nonconsecutive renewal (P = 0.39).

Categorial weight loss

In Table 5, the weight change of participants is categorized into groups as the percentage change in initial weight, stratified by renewal year. This table includes all individuals who renewed their memberships during the study period, regardless of whether their renewal was consecutive or nonconsecutive. Within each renewal cohort, between 48 and 53% of individuals lost ≥5% of their initial weight during the study period.

Table 5.  Weight change as a percentage of initial weight
inline image

Discussion

Although obesity rates have stabilized (1), overall obesity rates remain high, and weight management remains a major public health challenge for the United States. This study evaluated weight change among people who joined TOPS, a nationally available nonprofit weight loss program, over 3 years. In this database analysis of TOPS, we found that people who stayed in the program lost 5.9–6.8% of initial weight in the first year and maintained that weight loss over the next 2 years. TOPS participants with consecutive renewal lost more weight while in the program, as compared to those with nonconsecutive renewal.

Weight losses among program completers in this study are similar to losses among program completers in a randomized trial of Weight Watchers (i.e., ∼5 kg after the first year; 25). Our results are also comparable with those of completers' analyses from randomized trials that evaluated popular self-help diets including the Atkins, Ornish, Weight Watchers, and Zone diets (17,18,25,26,27). If we had used a baseline carried forward analysis, as done in randomized trials, the average weight losses in TOPS would be significantly smaller. However, we did not believe such an analysis was appropriate for this data set.

The Institute of Medicine defines successful long-term weight loss as “losing at least 5% of body weight by the completion of a weight loss program…, and keeping it [off]… for at least 1 year” (28). By this criterion, ∼50% of TOPS participants with at least one annual renewal met the definition of successful long-term weight loss. Weight losses for TOPS participants with consecutive annual renewal were stable for up to 3 years. It is unclear why weight was so stable for this subset of participants. They may simply represent a select group of highly motivated individuals.

The average weight loss estimated in this study is modest, and most participants are unlikely to be categorized as normal weight after this amount of weight loss. However, the amount of weight loss achieved by TOPS participants has been associated with improved health outcomes in other studies. For example, the Diabetes Prevention Program showed that a 7% weight loss was associated with a 58% reduction in the incidence of type 2 diabetes among individuals with impaired glucose tolerance, as compared to usual care (11). Other studies have shown that weight losses of 5–10% are associated with lower blood pressure, decreased insulin resistance, improved lipid profiles (8,9), and improvements in sleep apnea (12). Although no comorbidity data were collected in TOPS participants, individuals with weight-related conditions could be expected to sustain the same improvements in health seen in other studies.

Of the nearly 112,000 individuals who enrolled in TOPS over a 3-year period whose starting weights met the inclusion criteria, 37% renewed their membership at 1 year. While only a few published studies of this size have reported retention rates at 1 year, 37% is substantially better than the 9.0% retention rate calculated from a similar study of the Jenny Craig program (29). Although the dropout rate of 63% at 1 year is high, TOPS is a low-cost intervention. Thus participants bear little financial risk by joining. We speculate that the lower dropout rate in TOPS could be due to the camaraderie that develops among participants and/or the ownership some members may feel because they hold offices in the chapter.

TOPS is similar to Weight Watchers in its format, but there are important differences. Both programs stress behavior modification including diet and exercise to achieve weight loss, and both offer live group meetings and an online option. One important difference is that TOPS chapters are led by volunteer leaders who are elected by their peers, whereas Weight Watchers group leaders are paid employees. Another important difference is that TOPS is significantly less expensive than Weight Watchers, as described in the introduction.

The availability of a national nonprofit weight loss program could potentially make structured weight loss treatment accessible to people with lower incomes. Rates of overweight and obesity are higher among lower SES groups in the United States, including overrepresentation from African Americans, Latinos, and Native Americans (5). Nonsurgical weight loss programs are typically not covered by private insurance or by Medicare or Medicaid, and people of lower SES are less likely to use commercial weight loss programs than higher SES individuals (30). For these reasons, a program like TOPS could make weight loss available for these vulnerable populations. At least two state Medicaid programs have reimbursed the cost of Weight Watchers for their beneficiaries (31,32). If prospective trials confirm that TOPS is as effective as Weight Watchers, health-care payers may consider reimbursing this lower-cost option.

Several limitations apply to this study. First, TOPS collects data for administrative, as opposed to research, purposes. Thus, the analysis is limited by a lack of demographic information about program participants. Specifically, no data were available regarding race/ethnicity, age, or SES. A second limitation is that, until recently, TOPS did not record the heights of its members. Thus, we could not calculate BMI and were unable to categorize individuals as underweight, normal weight, overweight, or obese. Because of the lack of BMI data, some individuals could have been included in the data set for whom weight loss would not be clinically indicated. However, we attempted to limit this by excluding individuals who might have had a BMI <25 kg/m2 (based on average heights for adults in the United States). Therefore, this is believed to be a relatively small number of individuals. Another limitation is the possibility of withdrawal bias due to the lack of data on the weight trajectory of individuals that dropped out of TOPS. Published data suggest that these individuals are likely to gain ∼2 pounds/year.(33) Finally, co-intervention may also have occurred if TOPS members were participating in other weight loss programs simultaneously.

Despite the limitations, this study has important strengths. To our knowledge, this is the largest study of a weight loss program in the United States. The large sample size and the national availability of TOPS enhance the generalizability of the findings. Thus, this study provides important information about the real world effectiveness of TOPS.

In conclusion, this study demonstrates that TOPS, a nationally available nonprofit weight loss program, is associated with clinically significant weight loss among participants who remain in the program for at least 1 year. Furthermore, TOPS is available at substantially lower cost than commercial weight loss programs. Future studies should examine the impact of nonprofit programs on populations with high obesity rates and should include randomized trials to directly compare the effectiveness and relative cost-effectiveness of TOPS, as compared to for-profit commercial programs. Future studies of TOPS will be aided by the inclusion of more demographic information, including age and height, which the program has recently begun to collect.

ACKNOWLEDGEMENT

We acknowledge TOPS Club, Inc. for providing access to their data, including administrative and logistical support. We also thank Brenda Beaty, MSPH, from the Colorado Health Outcomes Program, for her assistance with data analysis. This study was funded, in part, by grant # T32-HP10006 (N.S.M.). None of the authors have any financial conflict of interest related to the contents of this article.

DISCLOSURE

The authors declared no conflict of interest.

Ancillary