The immense burden of diabetes underscores the utility and necessity of prevention. The findings of 2 landmark randomized clinical trials, the Diabetes Prevention Program (DPP), and the Look AHEAD study, provided evidence for the benefit of intensive modifications in lifestyle behaviors among people at high risk for diabetes as well as people with diabetes (1-3). In the DPP, weight loss was shown to be highly effective in reducing the incidence of diabetes among people at high risk for developing the disease. In the Look AHEAD study, weight loss was associated with improvements in diabetes control and reductions in cardiovascular disease risk factors. Thus, diabetes prevention efforts have a 2-pronged approach with a common strategy: primary prevention among people at risk for diabetes and secondary prevention among people with diabetes to prevent or delay the onset of complications, via weight loss (4).

Although clinical trials have demonstrated the benefits of weight loss among overweight and obese people either at high risk for diabetes or with diagnosed diabetes, few reports have described weight-loss behavior among these groups in the general population. Additionally, population-based reports have shown that the desire to achieve weight loss and the desire to create healthful changes overall occur at varying levels in the United States (5,6). Similarly, the literature has shown that among obese people, weight management behavior varies and that physician recommendation for weight loss is minimal, but few studies have focused on people with or at risk for developing diabetes (7,8). The objectives of this study were to 1) identify the prevalence of attempts to control or lose weight, reduce the amount of fat or calories in the diet, or increase physical activity reported and 2) examine physician recommendations to control or lose weight, reduce the amount of fat or calories in the diet, or increase physical activity, reported by overweight and obese people with prediabetes or diabetes.

The mean age for both prediabetic and diabetic men and women was 57 years (Table 1). The mean BMI was more than 30.0 kg/m², the obese range, for all groups. Most respondents had a high school education or more. Reported health status was low; the majority of the men and women with prediabetes or diabetes identified their health as good, fair, or poor. Among people with diabetes, 20% of men and 30% of women reported the use of insulin (data not shown).

Overall, among people with prediabetes in this sample, 82% reported trying to control or lose weight, 62% reduced the amount of fat or calories in their diet, and 53% increased physical activity (data not shown). Among people with diabetes in the sample, 75% reported trying to control or lose weight, 71% reduced the amount of fat or calories in their diet, and 57% increased physical activity (data not shown). For both prediabetes and diabetes, a higher proportion of women than men reported trying to control or lose weight (range, 70%-86%) (Table 2). Diabetic women had a higher reported prevalence of reducing the amount of fat or calories in their diet than diabetic men (range, 55%-75%). Increasing physical activity was reported by fewer respondents (range, 53%-59%), and no sex difference was observed in either group. Approximately two-thirds of the respondents indicated that a physician had advised them to control or lose weight, reduce the amount of fat or calories in their diet, or increase physical activity (Table 2).

Among women with prediabetes, women reporting physician advice to increase physical activity were more likely to report an increase in physical activity compared to women with prediabetes who did not report receiving physician advice (Table 3). Men with prediabetes were more likely to report an increase in physical activity with physician advice, but no significant association was found between self-report of reducing the amount of fat or calories in their diet and physician advice for this behavior in this group. In addition, the association between physician advice for lifestyle behavior and the reported lifestyle behaviors of the respondents was significant for both men and women with diabetes for all 3 behaviors. The magnitude of association between advice and report of behavior change among men with diabetes was stronger than among women with diabetes.

In this cross-sectional analysis, approximately 82% of prediabetic and 75% of diabetic overweight and obese people reported trying to control or lose weight, but fewer reported reducing the amount of fat or calories in their diet (approximately 62% for prediabetic and 70% for diabetic respondents) or increasing physical activity (approximately 53% for prediabetic and 56% for diabetic respondents). In this study, which presents some of the first population-level estimates of lifestyle behavior and recommendations for behavior for these groups, 58% to 68% of the sample reported receiving a physician recommendation for weight loss, increasing physical activity, or reducing the amount of fat or calories in their diet. This research highlights that despite the advantages of weight loss that can be achieved through lifestyle modification, many overweight and obese people with prediabetes and diabetes do not report behaviors conducive to weight loss or physician advice for such behaviors.

The findings of this investigation are consistent with prior research. One study using data from the National Health and Nutrition Examination Survey reported weight-loss attempts for more than half of overweight and obese adults, and the Behavioral Risk Factor Surveillance System (BRFSS) reported estimates of weight-loss attempts of 60% to 72% for people with diabetes (11,12). The reported prevalence of physician advice also corresponds to prevalence estimates found in the general US population (8,13-15). Another analysis using data from BRFSS found that 42% of obese respondents in the general population reported physician advice for weight loss, but that prevalence rose to 61% among people with diabetes (13). This report, coupled with previous studies, reinforces the need for continued efforts to promote the adoption of and physician advice for lifestyle behavior change for overweight and obese people with prediabetes and diabetes.

Self-reported receipt of physician advice for lifestyle behaviors was strongly correlated with reported attempts to change lifestyle behaviors in this sample. Cross-sectional correlations were noted between reported behaviors (attempting to control or lose weight, reduce fat or calories in the diet, increase activity) and physician recommendations to do so; however it was not possible to identify whether physician advice precipitated behavior. Previous reports from the general population have shown a positive association between receipt of advice for lifestyle behaviors among the obese and adoption of the behavior; these are consistent with the findings of our study, which focused on people with prediabetes and diabetes (8,16,17). Likewise, although physician advice is significant, successful weight loss and weight control and sustained lifestyle behaviors are predicated upon many factors, including behavioral interventions, social support, and patient education. Advice from physicians for lifestyle behavior is a key first step of provider involvement to promote weight loss, but physician involvement is also important in tailored weight-management interventions, and physicians can help patients reduce barriers to achieving weight loss (18-23).

Physician counseling can be a vital part of weight management therapy (24,25). Each encounter physicians have with obese and overweight patients is a potential occasion to counsel weight loss (24). Approximately one-third of the overweight and obese respondents in our study did not report receiving recommendations to adopt these behaviors from their physicians. This suggests that physicians may be missing opportunities to offer lifestyle change advice. A few studies have identified barriers to physicians' providing advice for weight loss and lifestyle behavior change among their overweight and obese patients, which include lack of time during the office visit, training, and attitude toward obesity (23,26). Further studies may need to investigate strategies to reduce barriers to physicians providing lifestyle advice to their patients.

This study has a few limitations. Study variables were all self-reported; thus, recall bias is possible. Physician recommendation for lifestyle behaviors were reported by study respondents and not their physician; therefore, respondents may have reported not receiving advice when they actually did receive it. Data from a nationally representative provider-based survey have reported estimates of physician advice for physical activity and diet to obese patients somewhat lower than the current study; thus it is also possible for our respondents to have overreported receipt of physician advice (27,28). Additionally, because this is a cross-sectional study, it is not possible to assign direct causation between reported advice and reported behavior changes. Other factors may influence our findings that we were unable to measure. For instance, it was not possible to assess the nature and frequency of advice for lifestyle behavior change respondents received from physicians or the respondents' exposure to other lifestyle change strategies. BMI based on self-reported height and weight was used to classify the measured weight status of respondents. Self-reported height and weight data are often misreported (29). Finally, NHIS is limited to civilian, noninstitutionalized adults and does not include people living in nursing homes or other institutions. Thus, the findings of this study are generalizable to the overweight and obese noninstitutionalized civilian population with prediabetes and diabetes in the United States.

Appropriate lifestyle behaviors (which are important in both preventing diabetes and improving disease management and outcomes) and physician advice to adopt these behaviors are not reported by all overweight and obese adults with prediabetes and diabetes. Respondents who noted physician advice for these important behavioral changes also reported attempts to improve lifestyle behaviors more often than those not receiving this advice. Active physician involvement in obesity management for people with prediabetes and diabetes may be an overlooked strategy for population approaches to prevent diabetes and improve outcomes for people with diabetes.