|Home | About | Journals | Submit | Contact Us | Français|
Although the primary care setting offers an innovative option for weight loss interventions, there is minimal research examining this type of intervention with low-income minority women. Further, there is a lack of research on the long-term effects of these programs. The purpose of this investigation was to examine the weight loss maintenance of low-income African-American women participating in a primary care weight management intervention. A randomized controlled trial was conducted with overweight and obese women (N = 144) enrolled at two primary care clinics. Women received a 6-month tailored weight loss intervention delivered by their primary care physician and completed follow-up assessments 9, 12, and 18 months following randomization. The weight loss maintenance of the tailored intervention was compared to a standard care comparison group. The weight loss of intervention participants (−1.52 ± 3.72kg) was significantly greater than that of standard care participants (0.61 ± 3.37kg) at month 9 (P = 0.01). However, there was no difference between the groups at the 12-month or 18-month follow-ups. Participants receiving a tailored weight loss intervention from their physician were able to maintain their modest weight loss up to 3–6 months following treatment. Women demonstrated weight regain at the 18-month follow-up assessment, suggesting that more intensive follow-up in the primary care setting may be needed to obtain successful long-term weight loss maintenance.
Recent epidemiological data indicate that ~66% of adults in the United States are overweight or obese (1). Furthermore, certain subgroups of the population, including women, ethnic minorities, and those from lower socioeconomic backgrounds, are at even greater risk for being overweight or obese (1,2). To combat this epidemic, there has been increasing interest in the role physicians can play in treating this condition (3-6), as some estimates indicate that 75% of adults in the United States visit a physician approximately five times per year (7). Among low-income primary care patients, rates of annual physician visits may be even higher (8). Therefore, physicians can play an important role in reducing the morbidity and mortality associated with over-weight and obesity, particularly among those most at risk, such as low-income individuals and ethnic minorities.
Research suggests that physician-directed weight loss treatments are feasible and that physician interventions (alone or in combination with specific dietary programs) can produce significant weight loss (5,9-12). One review of physician-supervised weight loss interventions showed weight losses ranging from 2.0 to 4.7 kg, while participants in control conditions generally gained or maintained weight (13). In another review, Harvey et al. (3) summarized the potential efficacy of physicians' management of obesity, although they also pointed out the limited number of studies on this topic and the need for additional research. One recent brief primary care intervention produced less promising weight loss results, although the behavioral counseling was administered by a nurse and/or dietician rather than physician (14).
Until recently, there has been a lack of research examining primary care interventions for obesity with low-income, minority women. Martin and colleagues (15) reported outcomes from a tailored weight loss program delivered by primary care physicians to low-income African-American women. In this study, a program consisting of six monthly visits with a primary care physician and directed at dietary modification and physical activity promotion resulted in significant but modest weight loss compared with a standard care condition. These results are particularly encouraging because African Americans tend to lose less weight and lose weight at a slower rate when compared to whites (16,17).
Regardless of the population or setting, one of the greatest challenges for weight loss interventions is successful weight maintenance following treatment because individuals regain much of the weight lost within 1 year following treatment (18,19). Therefore, although it is important to examine the efficacy of weight loss interventions implemented in primary care settings, it is equally imperative that the long-term effects of such interventions be evaluated. The purpose of this investigation was to follow up on previous outcomes (15) in order to examine the maintenance of weight loss among low-income African-American women participating in a primary care intervention. Participants returned to the primary care clinic for follow-up evaluations 9, 12, and 18 months from baseline. It was hypothesized that the tailored weight loss intervention would result in significantly greater weight loss or weight loss maintenance compared with a standard care condition.
A total of 144 African-American women were recruited from the office of their primary care physician and enrolled in the study. Participants were recruited from sequential clinic attendees who met inclusion criteria and provided written informed consent. To participate in the study, women had to be between the ages of 18 and 65 years, over-weight or obese (BMI ≥ 25 kg/m2), classified as low income (<$16,000 annual income), attendees of the primary care clinic for at least 1 year, and free of serious or uncontrolled medical conditions (e.g., renal or hepatic failure, cancer, immunological disease, uncontrolled hypertension). Women with well-controlled chronic diseases such as hypertension, diabetes, or hyperlipidemia were included. Exclusion criteria included the use of weight-altering medications, pregnancy, severe psychiatric illness, alcohol intake >14 drinks per week, or serious physical illness. Of the total women, 144 originally enrolled in the study, only 7 developed medical conditions that violated inclusion criteria (e.g., pregnancy, renal failure, hepatic failure, cancer), resulting in a sample size of 137.
Individual assessments with each participant were conducted at baseline and at the end of the 6-month treatment. Follow-up assessments were also conducted 9, 12, and 18 months following randomization. The baseline assessment provided information used for tailoring the monthly interventions and included the Block Food Frequency Questionnaire (20), the Food Preference Questionnaire (21), and a measure of physical activity (22). Assessed psychosocial variables included depression, Center for Epidemiological Studies Depression Scale (23); major life stress, Life Experiences Survey (24); minor life stress, Weekly Stress Inventory (25); and self-efficacy for eating behaviors, Weight Efficacy Lifestyle Questionnaire (26). Trained personnel measured participants' weight and height at each assessment in the physician's office using a standardized protocol with a calibrated scale and stadiometer. Weight and height were used to calculate BMI.
The first available clinic attendees meeting inclusion criteria and interested in participation provided written informed consent and were enrolled in the study. Eight primary care physicians were recruited from two family practice clinics involved in a state-wide primary care research network. All eight physicians from the clinics agreed to participate, following a detailed presentation of the proposed study during a clinic meeting. Physicians were randomly assigned to provide either a tailored weight loss intervention or standard care. This resulted in two physicians delivering tailored interventions at each clinic and two physicians delivering standard care at each clinic. Participants were assigned to receive one of the two interventions based on the random assignment of their physician. Recruitment continued until each physician had a maximum of 20 patients. At one clinic, 32 participants and 34 participants received the tailored intervention and standard care, respectively. At the other clinic, 36 participants and 35 participants received the intervention and standard care, respectively.
All physicians received 2h of instruction on general obesity treatment, as outlined by the National Heart, Lung, and Blood Institute's clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults (27). The four physicians providing tailored interventions received an additional 5 h of training, which addressed the assessment of stage of change, motivational interviewing, and techniques for the behavioral treatment of obesity. Training also included instruction on appropriate dietary recommendations, such as ways to reduce dietary fat intake, appropriate fruit and vegetable intake, how to read food labels, and how to modify recipes.
Participants were assigned to the standard care control group or the tailored intervention group based on the random assignment of their physician. Tailored interventions were derived from information provided by participants during the initial assessment. Participants in the tailored intervention group received five physician-counseled office visits on a monthly basis. Topics of the monthly meetings included introductory information on weight loss (month 1), ways to decrease dietary fat (month 2), ways to increase physical activity (month 3), dealing with barriers to weight loss (month 4), and healthy alternatives when eating out and shopping (month 5). They also received one maintenance session at month 6, which addressed ways to stay motivated during weight loss efforts. Each visit lasted ~15 min, resulting in a total of ~90 min of physician–patient contact. Physicians received protocols for each monthly visit, and participants received both oral recommendations from their physician as well as handouts summarizing the focus of each visit.
Tailored intervention participants received messages consistent with standard weight loss protocols, including gradual increases in physical activity with a goal of 150 min per week, decreased consumption of energy-dense foods, and increased consumption of fruits and vegetables. Both dietary and physical activity recommendations were personalized for participants based on their activity and food preferences, physical and environmental limitations, normal eating patterns, and caloric intake requirements needed to achieve weight loss. Participants received culturally specific menus and recipe books as well.
In comparison, standard care participants received no special instructions regarding weight loss and were seen as needed for regular medical care. Physicians providing standard care had received training on current guidelines for the treatment of obesity (27) though no specific weight loss protocol was provided for physicians to use with study participants. Standard care physicians were instructed to provide their usual obesity management conducted during a typical office visit. It was ultimately up to the physician (in collaboration with his/her patients) to determine how much (if any) weight loss counseling was included in primary care encounters during the study.
All participants received a reimbursement of $35 per visit for completing the baseline, post-treatment, and three follow-up assessments. Tailored intervention participants also received reimbursement of $10 for each monthly treatment visit. To decrease attrition, reimbursement was provided to combat potential barriers to successful participation, such as transportation and childcare issues (28). Physicians received reimbursement for all office visits associated with the project, including tailored intervention and usual care visits, consistent with state Medicaid reimbursement practices. To assure the accurate presentation of the intervention protocol, the project coordinator was on site for each physician visit, and random site visits were conducted by the principal investigator to observe the physician-administered treatment. The institutional review boards of all participating sites approved the study.
Descriptive analyses were conducted to summarize baseline demographic and weight information: t tests and χ2 analyses were used to examine possible baseline differences between the two treatment groups. As weight outcomes immediately following treatment have been reported previously (15), current results focus on weight maintenance at the three follow-up visits. Analyses utilized an intention-to-treat model, and baseline values were carried forward to each of the follow-up assessments when data were missing. Thus, participants lost to attrition were included in the analyses, assuming no weight change had occurred. Secondary analyses including only participants who completed the intervention and follow-up assessments were conducted as well.
Because the physicians at each of the two clinics were the unit of randomization (and participants received the intervention corresponding with the random assignment of their physician), participants were nested within physicians, and physician assignment was accounted for in the model as a random effect. Physicians were nested within the two clinics. Analyses first tested for significant clinic × treatment interaction to examine the possibility of differential treatment effects at the two clinics. These interaction terms were nonsignificant at all time points, and the two clinics were pooled for final analyses. Mixed-model ANOVA was used to examine the differences in weight change between the intervention and standard care groups at each of the follow-ups, accounting for the variance due to physician and participant. A logistic regression model that incorporated physician as a random effect was used to compare treatment groups on proportion achieving 5% weight loss. The main analyses reported here do not adjust for baseline covariates. Given the good group balance on baseline covariates, such adjustment appeared unnecessary; additional analyses (not reported here) confirmed that the major conclusions regarding group differences over time were insensitive to covariate adjustment. Analyses were conducted using Stata 9.0 (StataCorp LP, College Station, TX) and SAS 9.1 (SAS Institute, Cary, NC).
The mean baseline weight of participants was 101.95 ± 19.37 kg, and the mean BMI was 38.85 ± 7.63 kg/m2. The mean age of participants was 41.8 ± 12.0 years; 28% of the sample was married, and 78% had completed high school. There were no baseline differences between the tailored intervention and standard care groups on these weight and demographic variables (see Table 1). In addition, there were no baseline differences between participants on the basis of physician providing care or the clinic in which they were receiving care. After accounting for medical exclusions and women lost to follow-up for other reasons, 105 completed the 6-month program, which resulted in an attrition rate of 27%. Another 42 participants discontinued the study by month 9, 51 discontinued by month 12, and 53 discontinued by month 18. Thus, the attrition rates at the 9-month, 12-month, and 18-month follow-up assessments were 29, 35, and 37%, respectively. Comparing between the two treatment groups, attrition was significantly greater (29%) in the intervention group immediately following the active treatment phase (i.e., month 6) as compared with standard care (12%), P < 0.01. Attrition was also greater among intervention participants (44%) at the final follow-up (i.e., month 18) as compared with standard care (23%), P < 0.02.
Characteristics of women who completed each assessment were compared to those lost to attrition. At each of the three follow-ups, women who completed the follow-up were significantly older than noncompleters, 43.81 ± 11.46 years vs. 35.94 ± 11.80 years at the 9-month follow-up, t(142) = 3.54, P < 0.01; 44.11 ± 11.26 years vs. 37.02 ± 12.21 years at the 12-month follow-up, t(142) = 3.43, P < 0.01; and 44.55 ± 11.04 years vs. 36.61 ± 12.16 years at the 18-month follow-up, t(142) = 3.95, P < 0.01. Women who completed the 9-month and 18-month follow-up assessments also had lower educational levels than participants lost to attrition, although these differences were very small in magnitude. At the first follow-up, the average last grade completed by assessment completers was 11.41 ± 1.29 vs. 11.97 ± 0.88 for noncompleters, t(140) = 2.45, P < 0.02. At the 18-month follow-up, the average last grade completed by assessment completers was 11.35 ± 1.33 vs. 11.92 ± 0.86 for study noncompleters, t(140) = 2.68, P < 0.01. No other baseline variables differentiated study completers with those lost to attrition.
Results indicated that the weight change from baseline of the tailored intervention group (−1.52 ± 3.72kg) differed significantly from the weight change of the standard care group (0.61 ± 3.37kg) at 9 months postrandomization, F(1, 6) = 12.32, P = 0.01 (see Figure 1). Therefore, the intervention group was more successful in maintaining a lower weight from baseline than the standard care group. Figure 1 illustrates that although the standard care group gained a slight amount of weight between the end of treatment and the first follow-up, the tailored intervention group demonstrated small but continued weight loss during this period. Only 13% of intervention participants maintained at least 5% weight loss from their baseline weights at month 9, compared with 7% of the standard care group, P = 0.39.
Twelve months after baseline, the weight change of the tailored intervention group (−1.38 ± 3.69 kg) was no longer significantly greater than the weight change of the standard care group (−0.16 ± 3.63 kg), F(1, 6) = 3.80, P = 0.10. As seen in Figure 1, the tailored intervention group maintained an average weight nearly identical to that achieved immediately following treatment 6 months earlier. Somewhat surprisingly, the standard care group demonstrated a decrease in weight between the 9-month and 12-month follow-ups. The proportion of participants in each group who maintained at least 5% weight loss at month 12 was nearly identical (10% in the tailored intervention, 11% in standard care), P = 0.81.
Eighteen months after baseline, the weight change of the tailored intervention group (−0.49 ± 3.33 kg) was not significantly greater than the weight change of the standard care group (+0.07 ± 3.75 kg), F(1, 6) = 0.85, P = 0.39. Although the intervention group demonstrated weight below their baseline levels, they had regained most of the weight lost during the active-treatment phase (see Figure 1). At month 18, only 7% of the intervention group demonstrated at least 5% weight loss, whereas 12% of standard care participants achieved this level of success, P = 0.40.
Supplemental analyses were conducted including only those women who completed the study protocol (i.e., had available data at the final follow-up). These outcomes were very consistent with those reported using an intent-to-treat approach. At month 9, participants in the tailored intervention demonstrated significantly greater weight loss (−2.21 kg) than participants receiving standard care (0.68 kg), P < 0.01. At month 12, the intervention group lost an average of 1.97 kg, whereas the standard care condition lost 0.19 kg, although this difference only approached significance, P = 0.07. At month 18, there was no significant difference in the weight loss of the two groups (intervention = −0.88 kg; standard care = 0.09 kg), P = 0.28.
Previous research has suggested that African-American women show less success with weight loss efforts (16). Most individuals, regardless of ethnicity, have difficulty maintaining weight losses (18). Therefore, the goal of the current study was to examine the weight loss maintenance of overweight and obese African-American women who participated in a physician-delivered weight management intervention. Participants lost a very modest amount of weight (<2 kg) after the brief intervention, and current results suggest that women who received tailored weight loss interventions from their primary care physician were able to maintain their weight loss 3–6 months following treatment. There was a significant difference in weight loss maintenance between the intervention and standard care groups at the first follow-up (i.e., month 9).
By month 12, however, the group difference was not significant. At month 18, the intervention group had regained 64% of their achieved weight loss, and there were no group differences in weight at this final follow-up. Furthermore, only a small minority (7-13%) of intervention participants achieved and maintained a weight loss of at least 5% at any follow-up, which did not differ from standard care values. These results suggest that a more intensive primary care intervention may be required in order to achieve clinically meaningful weight loss for low-income African-American women. More regular (e.g., monthly) appointments may be useful for weight loss maintenance, or adjunctive follow-up programs may be beneficial for improving clinical outcomes.
Several characteristics of the current program and the study sample are important to consider when interpreting these results. First, the patients' primary care physicians delivered the interventions, and primary care physicians are viewed as important sources of health-risk advice (29,30). The majority of primary care patients reported the usefulness of physician advice for weight loss and would like regular review of this topic by their physician (31). Also, one study found that African-American women reported greater need for individual counseling from their physician on weight loss as compared with white women's preferences (32).
The current program was also designed to address weight loss issues salient to the individual and, therefore, incorporated dietary and physical activity recommendations on the basis of information provided by participants regarding food preferences, physical limitations, and other relevant factors. The program was also culturally sensitive and designed for African-American women specifically. The tailored nature of this program may have played a role in the initial successful weight loss maintenance because tailored interventions may produce greater behavioral changes relative to nontailored programs (33). The finding that regain eventually occurred is consistent with the intervention literature and suggests that this tailored primary care intervention may have limited utility without continued treatment contacts.
Although the current program was successful in helping participants maintain weight loss 3–6 months following treatment, it is also important to note that the amount of weight lost with this program was modest. Several points should be kept in mind when considering this issue. First, the treatment was very brief in nature (i.e., 6 monthly physician visits lasting ~15 min each). Second, these results were achieved with African American women, a group previously shown to have less success with weight loss efforts (16). Third, even small decreases in body weight have been associated with improved chronic disease outcomes such as hypertension and diabetes (34-38). Finally, the simple prevention of weight gain (even if no weight loss occurs) can be helpful because individuals tend to experience gradual weight gain over time in the absence of treatment (39,40).
Several limitations of the current study should be noted. Although the intervention was relatively brief compared with many behavioral programs, the time constraints and reimbursement policies associated with primary care could make such an intervention impractical for many clinical settings. The current sample included low-income African-American women aged between 18 and 65 years, which limits the generalizability of these findings. For instance, the effects of such a program with obese adolescents, higher -SES groups, or other ethnic groups cannot be ascertained from the current results. Also, providing participants with incentives may limit the generalizability of this primary care program. Attrition is another potential problem of the current study, as 37% of the original sample was lost to follow-up by the 18-month assessment visit. However, replacing missing follow-up data with baseline weights provided a conservative estimate of treatment effects. This study would be strengthened by even more extended follow-up intervals to determine the course of weight change during a lengthier maintenance period. Given the weight regain demonstrated by the intervention group at the final follow-up, one might assume that a complete return to baseline weight would likely occur. Further work is clearly needed to improve the effects of primary care weight loss interventions with this high-risk population.
The current results supplement the literature on weight loss maintenance and suggest a brief, physician-delivered weight loss intervention can be initially effective in achieving modest weight loss among African-American women. However, this tailored intervention did not result in significant weight loss maintenance 12–18 months later, as the intervention group demonstrated gradual weight regain over this extended follow-up interval. These results provide unique information regarding the extended efficacy of a primary care model of weight loss treatment implemented with a group of obese, minority women, and results suggest continued contact in the primary care setting may be needed to achieve further weight loss and weight maintenance over extended periods.
This study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (Grant R01 DK57476) and cosponsored by the Centers for Disease Control and Prevention, the Center for Chronic Disease Prevention and Health Promotion, Division of Nutrition and Physical Activity, and the Office of Research on Women's Health.
The authors declared no conflict of interest.