Weight Loss Goals Among African-American Women With Type 2 Diabetes in a Behavioral Weight Control Program
African-American women with type 2 diabetes experience limited weight loss in behavioral weight control programs. Some research suggests that overly ambitious weight loss expectations may negatively affect weight losses achieved but it is unknown whether they affect weight loss among African-American women. The current study examined personal weight loss goals and expected satisfaction with a reasonable weight loss among African-American women with type 2 diabetes starting a behavioral obesity treatment. We also explored associations among these factors and weight loss treatment outcomes. Self-identified African-American women (N = 84) in a 24-session group program were assessed at baseline and 6-month follow-up. At baseline, women indicated weight loss goals of 14.1 ± 6.6 kg (14% of initial weight). They also reported relatively high expected satisfaction with a reasonable weight loss (7–10%). On average, participants lost 3.0 ± 3.9 kg (3% of initial weight) and attended 73 ± 21% of group sessions. Neither weight loss goals nor expected satisfaction with a reasonable weight loss was correlated with either actual weight loss outcome or attendance. Having higher personal weight loss goals was associated with lower expectations of satisfaction with a reasonable weight loss. This suggests that African-American women with type 2 diabetes enter treatment hoping to lose far more weight than they are likely to achieve. It is important to understand the psychosocial sequelae of failing to reach these goals on subsequent weight maintenance and future weight loss attempts within this population.
Obesity among individuals with type 2 diabetes presents serious health risks (1). Within this population, weight loss produces a cascade of significant health improvements, including improved metabolic control (2). However, individuals with type 2 diabetes can experience modest success in weight management compared with nondiabetic individuals (3). This dilemma is of marked significance for African-American women, who have particularly high rates of obesity (4) and diabetes (5). However, African-American women traditionally lose less weight in obesity treatment than other race/gender groups (6,7).
Individuals entering a behavioral weight control program report high, even unrealistic, weight loss goals for themselves. Concerns have been raised that unrealistic weight loss goals might undermine weight loss efforts (8); however, results have been mixed. Some studies have reported that weight loss goals have no relationship with treatment engagement (9,10) or weight loss outcomes (11) while others found that unrealistic goals were associated with smaller weight losses and reduced program engagement (12). Others report associations between higher goals and better long-term weight loss (10). Furthermore, high weight loss goals appear resistant to revision, even with a rational appeal meant to sway unrealistic expectations into a more reasonable range (11,13). However, little of the available research examining weight loss goals focuses on African-American women, and still less directly addresses expectations among African-American women with type 2 diabetes.
In one of the few studies to assess weight loss goals among African-American women entering a weight loss program, weight loss goals of 29% of initial body weight were reported (14). African-American women typically achieve an average of 3–5% weight loss in behavioral lifestyle programs (7). Thus, goals of 29% are unrealistically high and could present a challenge to sustained weight control efforts. This study did not examine associations between weight loss goals and actual weight loss, thus it remains unknown whether unrealistically high goals negatively impact weight loss outcomes in African-American women. Therefore, the current study examines personal weight loss goals and expected satisfaction with reasonable weight losses among African-American women with type 2 diabetes starting a behavioral weight control program. Further, it explores associations between these factors and engagement in the treatment program and weight loss outcomes.
Methods and Procedures
Participants were overweight and obese African-American women with type 2 diabetes enrolled in a trial examining methods to enhance a behavioral weight control program (15). All women were offered a 24-session group behavioral obesity intervention which met weekly over a 6-month period. Treatment focused on promoting dietary and physical activity habit changes and providing support strategies (15).
Women were eligible if they were overweight (BMI; kg/m2 = 27–50) and had type 2 diabetes but were otherwise generally healthy and could walk for exercise. Exclusion criteria included pregnancy, recent weight loss, use of insulin or inability to participate for the duration of the study. Individuals with uncontrolled diabetes (Glycated hemoglobin (HbA1c) >12%) or hypertension (diastolic blood pressure >90 mm Hg or systolic blood pressure > 140 mm Hg) were referred to their physician for treatment and invited to rescreen. All research procedures were approved by the University of Alabama at Birmingham institution review board for Human Use and the University of Arkansas for Medical Sciences institutional review board.
Demographics. Age, self-identified race, marital status, employment status, and educational status were ascertained at baseline by self-report questionnaires.
Height and weight. Participants were weighed without shoes on a calibrated balance beam scale and measured using a wall-mounted stadiometer. Percent weight change was calculated as (baseline weight − weight at 6-months)/baseline weight.
Weight loss goals. At baseline, participants were asked “How much weight would you like to lose in the first 6 months?” Participants responded with the number of pounds they wished to lose and their personal weight loss goal was calculated as the percent weight change desired (weight loss goal)/current weight. This approach follows the methods of others (10,11,14). Desired weight loss was converted into kg for reporting.
Expected weight loss satisfaction. At baseline, participants were asked “How satisfied would you be if you lost 15–20 pounds in the first 6 months?” A weight loss of 15–20 pounds corresponds to ∼a 7–10% weight loss goal usually set in behavioral weight loss programs (2). Participants responded on a 5-point Likert scale (1 = extremely satisfied to 5 = extremely dissatisfied).
Session attendance. Attendance has been associated previously with both weight loss expectations and weight loss (12). Therefore, attendance was examined and is reported here as percentage of the total number of sessions offered.
We examined univariate associations among personal weight loss goals, expected satisfaction with reasonable weight loss and treatment outcome variables using Pearson's correlation coefficients. Regression models were used to examine the additive effects of weight goals and expected weight loss satisfaction on 6-month percent weight loss and treatment session attendance. Baseline BMI was included as a covariate in multivariate analyses. All statistical analyses were conducted using SAS (Version 9.2; SAS Institute, Cary, NC). A P value of 0.05 was established as the level of statistical significance.
All 84 self-identified African-American women enrolled in the study were included in the current analyses. Baseline characteristics are presented in Table 1. Women indicated that they desired to lose 31 pounds (14 kg) over the 6-month program (i.e., 14% of baseline body weight) and that they would be fairly satisfied (mean = 1.5 ± 0.8; scale: 1 = extremely satisfied and 5 = extremely dissatisfied) with a reasonable weight loss of 15–20 pounds. On average the sample lost 3.0 kg (3% of initial weight) and attended approximately three-quarters of weekly intervention sessions.
Table 1. Sample characteristics and outcomes of African-American women (N = 84)
Weight loss goals and expected satisfaction with an achievable weight loss were not significantly correlated with either weight loss outcome or attendance (see Supplementary Table S1 online). However, there was a significant positive association between personal weight loss goals and expected satisfaction (r = 0.374; P < 0.05) such that women who had higher weight loss goals reported lower expected satisfaction with a reasonable loss. Multivariate analyses indicated no independent associations of weight loss goals or expected weight loss satisfaction with actual weight loss outcome or treatment engagement.
Obese African-American women with type 2 diabetes reported high personal weight loss goals when starting a behavioral weight control program. They expressed a desire to lose 14% of their baseline body weight, and thus had somewhat more reasonable goals than the 29% weight loss desired by nondiabetic African-American women in the study by Dutton et al. (14). However, even the more reasonable 14% desired weight loss was unlikely to be achieved given that current behavioral obesity treatments typically produce weight losses of about 3–5 kg for African-American women (6,7). Both the current study and Dutton et al. (14) indicate that African-American women have ambitious weight loss goals and, as such, they resemble white women entering a weight loss program (8,9,10,11,12).
Entering a weight loss program with high (even unrealistic) personal weight loss goals neither appears to negatively impact weight loss among African-American women, nor does having high weight loss goals at the beginning of treatment appear to foster better weight loss. This is similar to the results among predominantly white women in which there was no association between weight loss goals and weight loss achieved (11) and contrasts with other studies that have observed significant correlations between higher personal weight loss goals and weight loss outcomes (10,12). The impact of these high weight loss ambitions and the modest weight losses achieved merit further exploration of potential psychosocial sequelae to failing to reach initial weight loss goals by such a large margin on subsequent weight maintenance or future weight loss attempts.
Limitations include a relatively small sample, although it represents the largest report to date of weight loss expectations and weight outcomes among African-American women with type 2 diabetes. Generalization to women without diabetes and women not presenting for weight loss must be made with caution. Further, women in the current sample were well-educated, most were employed and over half were married or living as married. These sociodemographic factors may curtail generalization to some subgroups of African-American women but represent a typical profile for African-American women who enter behavioral weight control studies (7). Finally, single items were used to measure weight loss goals and anticipated satisfaction; this follows the methods of others (10,11,14) but has limitations and likely does not fully capture the full range of nuances associated with what are likely multidimensional constructs.
The current study extends the existing literature on unrealistic weight loss expectations and weight outcomes to include African-American women with type 2 diabetes. The observation of high weight loss goals within this population contradicts speculations that African-American women may experience small weight losses in behavioral programs because they do not wish to lose substantial weight (14). Consideration of other psychosocial, cultural or environmental factors that may attenuate weight loss appears warranted.
Supplementary material is linked to the online version of the paper at http:www.nature.comoby
This study was supported by grant DK RO1-54041 and (in part) by the Intramural Research Program of the National Human Genome Research Institute, National Institutes of Health.
The authors declared no conflict of interest.