Objective: To describe the current knowledge, attitudes, and practices of French general practitioners (GPs) in the field of adult overweight and obesity management.
Research Methods and Procedures: A cross-sectional telephone survey interviewed a sample of 600 GPs, representative of the private GPs in southeastern France. A four-part questionnaire assessed personal and professional characteristics, attitudes and opinions about overweight and obesity, relevant knowledge and training, and practices (diagnostic methods, clinical assessments, weight loss objectives, types of counseling).
Results: Most GPs knew that weight problems are health-threatening, and 79% agreed that managing these problems is part of their role. Nevertheless, 58% did not feel they perform this role effectively, and one-third did not find it professionally gratifying. Approximately 30% had negative attitudes toward overweight and obese patients; 57% were pessimistic about patients’ ability to lose weight; 64% often set weight loss objectives more demanding than guidelines call for; and neither food diaries nor nutritional education were used systematically. GPs’ feelings of effectiveness and attitudes toward obese patients were associated with some professional (training) and personal (BMI, personal diet experience) characteristics.
Discussion: GPs’ feelings of ineffectiveness may stem from an underlying conflict between practitioners’ and patients’ representations of weight problems and the relationship problems this causes. Inadequate practices and health care system organization may also play a role.
The pandemic of overweight and obesity in developed and developing countries presents a challenge to public health and requires medical intervention, modifications of individual behavior, and environmental changes (1). Epidemiological studies in France, Spain, and Italy have shown this disease striking the Mediterranean area, where food habits have long seemed to protect against cardiovascular risks (2,3). General practitioners (GPs)1 have a significant role to play in preventing and diagnosing weight problems and in providing initial counseling (4,5). They are, after all, the health professionals consulted most often (6), and most patients believe that their GPs could—and wish they would—help them to lose weight (7). Nonetheless, GPs do not manage overweight and obesity satisfactorily, as they themselves recognize (8,9,10,11): they identify only about one-half of their overweight or obese patients (12,13,14) and advise only one-third of these patients to lose weight (15,16,17).
GPs’ attitudes toward and practices in the management of weight problems have been studied in various English-speaking countries (8,10,11,18), but we are aware of very few published studies from Mediterranean countries (9). We conducted a study of GPs in private practice in Provence (southeastern France) to document their knowledge, attitudes, and practices regarding adult overweight and obesity management.
Research Methods and Procedures
In March 2002, we set up a panel of GPs in private practice in Provence to evaluate their medical and prescribing practices over a 3-year period. Specialists, GPs with exclusive particular practices (e.g., homeopathy, acupuncture), and those planning to move outside the area or retire were excluded. We stratified the sampling base according to sex, age (<43, 43 to 52, and >52 years old), and size of practice area (<2000, 2000 to 200, 000, and >200, 000 inhabitants) and randomly selected 1200 GPs from the resulting strata to obtain a 600-GP sample representative of the 5435 private GPs in Provence. Of these 1200 GPs, 1076 (89.7%) met the inclusion criteria, and 600 of 1076 (55.8%) agreed to participate. Refusal rates were not significantly different between strata.
From May to July 2003, we conducted telephone interviews of these GPs to document their attitudes about and practices for overweight and obesity management: 580 of 600 (96.7%) participated, 12 of 600 (2.0%) withdrew, and 8 of 600 (1.3%) could not be contacted. We replaced these 20 GPs with new physicians randomly selected in corresponding strata. The distribution of sex, age, and size of practice area in our sample was similar to that of the regional GP population (Table 1).
Table 1. . Demographic characteristics of the sample of respondents and of the population of private general practitioners in Provence
|Sex|| || || || || |
| Male||4140||76.2||452||75.3||p = 0.65|
| Female||1295||23.8||148||24.7|| |
|Age (years)|| || || || || |
| <43||1398||25.7||158||26.3||p = 0.50|
| 43 to 52||2889||53.2||305||50.8|| |
| >52||1148||21.1||137||22.8|| |
|Size of practice area (number of inhabitants)|| || || || || |
| <2000||357||6.6||52||8.7||p = 0.13|
| 2000 to 200, 000||1334||24.5||150||25.0|| |
| >200, 000||3744||68.9||398||66.3|| |
To facilitate comparisons with previous studies, the questionnaire was based on a review of the international literature in overweight and obesity management in primary care and on U.S. and French guidelines for identifying, evaluating, and treating them (4,5). It was reviewed by a group of experts (four nutritionists, one endocrinologist, four physicians, and one sociologist) and pilot-tested with 17 GPs for length, clarity, and suitability. It included four main parts.
Professional and Personal Characteristics
We assessed the following characteristics: billing sector (in the controlled billing sector, the fee per consultation is set by the Health Insurance Administration; in the noncontrolled billing sector, GPs freely set their own fees according to market pressure and patient income), solo/group practice, subscription to medical journals, guidelines use, involvement in a health network (coordinated group of several health professionals organized to improve health care in a specific medical field), and postgraduate medical degrees. We also asked about their height and weight (for calculating their BMI), personal experience of dieting, behavior related to food intake, physical activity, and tobacco consumption.
Attitudes and Opinions about Overweight and Obesity
Eleven items allowed us to examine GPs’ perception of weight problems, involvement in their management, and perception of their training, effectiveness, and professional gratification in this field. Three items assessed their opinions about overweight and obese patients on a four-point Likert scale (from not at all to strongly) including: “do overweight (obese) people tend to be lazier and more self-indulgent than normal weight people?”; “are overweight and obese people able to lose weight and maintain that loss?” We included 17 items to assess GPs’ beliefs about risk factors for and consequences of obesity and their views of the importance of different outcomes as measures of success in its treatment. We used a six-point Likert scale for respondents to indicate the level of importance for each item (1 = not important, 6 = extremely important).
Knowledge and Training in the Field of Overweight and Obesity Management
GPs were asked 1) to estimate the prevalence of overweight and obesity among French adults; 2) about health risks associated with obesity in adults (eight risks, yes/no answers); 3) whether they were aware of the guidelines for obesity management; 4) their main source of relevant information [medical journals, continuing medical education (CME), obesity management guidelines, computer programs/Internet, experiences described by patients, dietitians/endocrinologists, mass media]; and 5) whether they had medical training about weight management and whether they felt they needed more training in various aspects (counseling about nutrition or physical activities, psychotherapy, behavioral therapy, drug treatment, surgical treatment).
Practices in the Management of Weight Problems
Participants were asked about their practices at different stages of weight management: diagnostic methods, clinical assessments, standard weight loss objectives and strategies, counseling about nutrition and exercise, follow-up, referral to other health professionals, and perceived barriers to overweight and obesity care. Participants responded on a four-point Likert scale (from never or rarely to always or almost always).
We used a computer-assisted telephone interview system to question participants. Each interview lasted ∼30 minutes.
Analysis began with simple frequency counts. We used χ2 tests to examine differences between the sample and the general population of GPs and to test differences in management for overweight and for obesity. We used the Fisher statistic to examine the differences between mean scores of various items (e.g., beliefs about causes of obesity). We conducted simple and backward multiple logistic regression analyses (entry threshold: p ≤ 0.15; exit threshold: p > 0.10) to study associations between GPs’ professional and personal characteristics, their opinions about their effectiveness in this field, and their attitudes toward obese people. We used the Hosmer-Lemeshow goodness-of-fit test to measure the fit of each model. Statistical analyses used SPSS software (version 11.0; SPSS Inc., Chicago, IL).
Most GPs (82.8%) practiced in the controlled billing sector, and 55.4% practiced in group practices. Only 44.8% subscribed to medical journals, 69.7% consulted guidelines, 47.8% had postgraduate medical degrees, and 15.7% were involved in a health network. The prevalence of overweight among GPs [30%; 95% confidence interval (CI) = 26.3 to 33.7] was close to that of the French adult population (29.4%; 95% CI = 28.9 to 29.9), but the prevalence of obesity was lower: 3% (95% CI = 1.8 to 4.6) vs. 9.6% (95% CI = 9.2 to 10.0) (19). Approximately one-third of GPs had dieted (84% lost weight); 71.5% considered their eating habits to be healthy; 76.9% reported exercising at least weekly; 69.9% were currently monitoring their food intake to lose or maintain weight; and 26% were current smokers.
GPs’ Attitudes Toward Overweight and Obesity
Most GPs regarded obesity as a disease and agreed that their role includes weight problem management, but 57.5% felt that they do not manage it effectively. Approximately 30% considered overweight and obese patients lazier and more self-indulgent than normal weight people, and 57.2% were rather pessimistic about these patients’ ability to lose weight (Table 2).
Table 2. . GPs’ attitudes towards overweight and obesity (Provence, 2003)
|Obesity is a disease||595||2.5||7.2||33.4||56.8|
|Normal weight is important for health||599||0.7||0.2||16.9||82.3|
|For overweight and obese patients even small weight loss can produce health benefits||596||0.0||0.8||21.5||77.7|
|Most overweight patients should be treated for weight loss||598||0.8||5.9||47.5||45.8|
|Only obese patients should be treated for weight loss||595||40.2||39.8||14.6||5.4|
|Obesity management is necessary in the long term||598||0.2||0.3||15.9||83.6|
|GPs’ role is to refer overweight and obese patients to other professionals rather than attempt to treat them themselves||594||31.0||48.0||17.5||3.5|
|GPs should be models and maintain normal weight||595||7.7||8.7||53.1||30.4|
|I feel well prepared to manage overweight and obese patients||596||5.9||26.5||44.6||23.0|
|Treating overweight and obese patients is professionally gratifying||595||7.6||23.9||48.4||20.2|
|Obese people are lazier and more self-indulgent than normal weight people||597||27.3||41.9||26.1||4.7|
|Overweight people are lazier and more self-indulgent than normal weight people||594||25.9||45.5||24.9||3.7|
|Only a small percentage of overweight and obese people can lose weight and maintain this loss||597||11.9||30.8||48.2||9.0|
The multiple logistic regression (Table 3) showed that practice in the noncontrolled billing sector (p = 0.01), CME training about weight problems (p < 0.01), awareness of obesity management guidelines (p = 0.02), normal weight (p = 0.05), and personal success in losing weight (p = 0.01) were associated with a feeling of effectiveness in this field. Negative attitudes toward obese patients were associated with not subscribing to any medical journals (p = 0.03), awareness of obesity management guidelines (p = 0.03), and never having dieted themselves (p = 0.05; Table 3).
Table 3. . Determinants of GPs’ feelings of effectiveness and attitudes towards obese people (multiple logistic regression with sex and age forced; Provence, 2003)
|Billing sector|| || ||0.01|| || || |
| Controlled billing sector||1|| || || || || |
| Noncontrolled billing sector||1.94||1.22 to 3.08|| || || || |
|Medical journal subscription|| || || || || ||0.03|
| Yes|| || || ||0.67||0.46 to 0.96|| |
| No|| || || ||1|| || |
|Has taken a CME class about management of weight problems|| || ||0.00|| || || |
| Yes||1.66||1.17 to 2.36|| || || || |
| No||1|| || || || || |
|Knows guidelines for obesity management|| || ||0.02|| || ||0.03|
| Yes||2.35||1.13 to 4.86|| ||2.15||1.10 to 4.20|| |
| No||1|| || ||1|| || |
|Current BMI (kg/m2)|| || ||0.05|| || || |
| Underweight or normal range (<25)||1|| || || || || |
| Overweight (25 to 29.9)||0.59||0.39 to 0.90|| || || || |
| Obese (≥30)||0.71||0.24 to 2.13|| || || || |
|Has ever been on a diet|| || ||0.01|| || ||0.05|
| No||1|| || ||1|| || |
| Yes and it succeeded||1.91||1.25 to 2.91|| ||0.64||0.41 to 0.98|| |
| Yes but it failed||0.94||0.38 to 2.32|| ||0.55||0.16 to 1.17|| |
|Hosmer-Lemeshow test||p = 0.31|| || ||p = 0.95|| || |
GPs rated food intake as a significantly more important risk factor for obesity than stress, hormonal problems, or unemployment. They also rated the medical consequences of obesity as more important than its psychological and social consequences (Table 4).
Table 4. . GPs’ beliefs about obesity risk factors and consequences and their perception of success in the management of weight problems (in decreasing rank order; Provence, 2003)
|Risk factors†|| || |
| Eats too much fat||5.2||0.93|
| Eats too much||5.1||1.02|
| Eats too much sugar||4.9||1.01|
| Insufficient physical activity||4.7||1.08|
| Genetic factors||4.5||1.25|
| Repeated dieting||4.2||1.26|
| Stress. anxiety, and depression||4.1||1.07|
| Hormonal problems||3.7||1.28|
| Low income, unemployment||3.3||1.19|
|Consequences‡|| || |
| Medical problems||4.9||1.11|
| Psychological problems||4.3||1.04|
| Social problems||3.8||1.09|
|Success indicators in weight problems management§|| || |
| Adoption of healthier diet and exercise habits||5.3||0.75|
| Weight loss to the normal BMI range||5.1||0.96|
| Improvement of body image and self confidence||5.0||0.89|
| Even small weight loss but long-lasting||4.7||0.94|
| Improvement in clinical indicators||4.6||1.07|
GPs’ Knowledge and Training in the Field of Weight Control
Most GPs (51.2%) underestimated the prevalence of overweight in the French adult population, whereas one-half overestimated the prevalence of obesity. Nearly all recognized most health consequences of obesity (premature mortality, type II diabetes, sleep apnea, hypertension, increased surgical risks, phlebitis), but 53% were unaware of the risks of infertility, and 45.5% were unaware of the risks of some cancers. Only 6.7% were aware of the guidelines for obesity management. One-half reported that their main source of information in this field was medical journals, 25.2% was CME, and 10.4% was experience described by patients. Just over one-half of the GPs (54.2%) had taken a CME class in weight management, and 80% acknowledged they needed more training, especially about nutrition counseling and behavioral therapy.
Most GPs (88.5%) often or always used BMI as a diagnostic method, whereas only 41% often or always measured the waist (Table 5).
Table 5. . GPs’ practices in the field of adult overweight and obesity management (Provence, 2003)
|Diagnosis methods|| || || || || |
| Weight without reference to height||600||77.0||7.5||7.8||7.7|
| Waist/hips ratio||599||44.2||20.9||19.2||15.7|
| Waist measurement||599||38.7||20.2||23.5||17.5|
| Comparison with ideal weight (according to Lorentz formula) (39)||598||46.8||19.7||18.4||15.1|
|Weight management advice and tools|| || || || || |
| Eat less during meals||595||10.4||13.9||25.0||50.6|
| Eat less fat||598||0.5||1.8||17.9||79.8|
| Don't eat between meals||597||6.0||3.0||16.1||74.9|
| Eat less sugar||597||0.7||4.5||21.4||73.4|
| Eat more fruits and vegetables||597||0.7||3.9||17.6||77.9|
| Consume fewer caloric drinks||598||0.0||0.3||8.2||91.5|
| Definitely avoid specific foods||599||53.4||11.2||14.4||21.0|
| Follow personalized low-calorie diet (1200 to 2200 kcal/day)||597||11.6||14.6||37.7||36.2|
| Follow very-low-calorie diet (<1200 kcal/day)||598||55.0||22.6||14.7||7.7|
| Follow commercial diet||598||53.8||33.8||8.7||3.7|
| Exercise (sports)||598||3.8||9.4||30.1||56.7|
| Do more exercise in everyday life (e.g., walking, gardening)||587||2.0||2.7||18.6||76.7|
| Leaflets on healthy behavior||597||12.6||21.4||31.0||35.0|
| Food diary||597||37.4||23.3||20.4||18.9|
| Nutritional education||598||17.6||18.6||29.9||33.9|
More than 90% of the participants often or always assessed individual risk factors, physical activity, dietary habits, patients’ expectations and motivation, patients’ psychological state, existence of food behavioral problems, and weight history; a lower but still high proportion (70% to 90%) often or always assessed respiratory problems, venous and lymphatic conditions, joint diseases, calory intake, energy expenditure, and social status; 60% to 70% often or always checked for hepatomegaly or steatosis, sleep disorders, and pain; and 30% to 40% often or always assessed skin condition and looked for cancerous breast nodules.
One-half the sample (50.8%) set a loss of 5% to 15% of initial weight as the objective for overweight patients. To reach this goal, 23.2% often or always prescribed drug treatment, 31.6% recommended psychotherapy, 14.8% recommended a behavioral therapy, and almost one-half included a spouse or a close relative in the treatment. Too stringent treatment objectives for obese patients were set by 64.3% of GPs (weight loss to normal BMI or >15%; Table 6).
Table 6. . Objectives and strategies in the field of overweight and obesity management (Provence, 2003)
|Usual weight loss objective|| || |
| No weight gain||2.2||1.4|
| Weight loss of 5% to 15% of initial weight and its maintenance||50.8||34.3*|
| Weight loss of >15% of initial weight and its maintenance||14.0||45.0*|
| Weight loss to the normal BMI range||33.0||19.3*|
|Management strategies (% of often or always responses)|| || |
| Drug treatment||23.2||40.5*|
| Behavioral therapy||14.8||24.2*|
| Inclusion of a spouse or a close relative in the management||48.9||60.2*|
Nearly all participants (90% to 99%) gave traditional nutritional advice (Table 5). Only 35% often or always counseled avoiding specific foods, and 22% suggested very low-calorie diets; 36.2% never, rarely, or sometimes offered nutritional education, and 60.7% never, rarely, or sometimes recommended that patients use food diaries.
Nearly all of the GPs saw these patients at least monthly, and 40% often or always proposed a telephone follow-up between consultations. Less than one-third referred patients to other professionals, and only 31% often or always referred their overweight or obese patients to a dietitian.
The most common problems experienced by GPs in treating overweight or obese patients were lack of patient motivation (often or always encountered by 60.1% of participants), lack of support from patients’ relatives (57.1%), lack of time (53.3%), nonreimbursement of consultations with dietitians (51.3%), and patient's nutritional knowledge (39.2%).
We found that most GPs believed their role in overweight and obesity management is important but did not feel that they performed it effectively. This observation is consistent with results of other studies of GPs (8,9,11) and also, interestingly, of endocrinologists and internists (20). Several obstacles may explain this apparent contradiction.
Physicians and patients have different perceptive and attitudinal models of weight problems, and one major obstacle may be the problem this creates in their relationship. Our results suggest that GPs’ perceptions of overweight and obesity are shaped by a model that blames the victim (21): they perceive behavioral factors (food habits and physical activity) as greater risk factors for obesity than genetic factors or stress and unemployment; behavioral factors are generally considered more controllable by the individual than unemployment, for example (18). This model clashes with patients’ views: they attribute more importance to risk factors over which they have little or no control (21). Moreover, one-third of the GPs had stereotypical and negative attitudes toward overweight and obese patients. Although these attitudes seem less prevalent among health professionals than they were 30 years ago (22,23), they are still held by 30% of GPs, internists, and cardiologists and by a lower fraction of endocrinologists (9,10,11,20), and their prevalence tends to increase with patient BMI (18,24). Studies suggest that this basic disagreement may be associated with poorer patient outcomes (25,26) and that patients’ negative responses to these attitudes creates a vicious circle that reinforces the doctors’ attitudes (24). Support for this observation comes from the less negative attitudes and greater feelings of effectiveness of GPs who have successfully lost weight themselves: personal experience may reduce the discrepancies between GPs’ and patients’ representations of the disease.
Because negative attitudes toward the obese and feelings of effectiveness were not associated with age, it is unlikely that either initial training or experience affected them substantially. Our results suggest, however, that appropriate information may improve GPs’ attitudes toward and opinions about obese patients. GPs who subscribed to medical journals were less likely to think that obese people tend to be lazier than normal weight people, perhaps because they were more aware that environmental obesity risk factors, not controllable by patients, exist. Although guidelines stress that GPs’ attitudes may affect the quality of their patient treatment (4,27), awareness of the guidelines was, surprisingly, associated with more negative attitudes, perhaps because awareness of the guidelines does not necessarily entail their use.
One striking finding was that the GPs substantially underestimated the prevalence of overweight and overestimated that of obesity. This suggests that they may rely mainly on a therapeutic rather than preventive approach to weight problems, an attitude that may delay management of weight problems. A U.S. study showed that health care providers advise relatively few overweight patients to lose or even to not gain weight (16), although maintaining current weight is known to be easier than losing weight (4).
Some more specific practices may also impede satisfactory outcomes: >60% of GPs set stricter weight loss objectives for obese patients than recommended (4,5). This finding seems to be new: previously published results showed that GPs have reasonable expectations about weight loss (8,28). This result may be consequential because it may reinforce unrealistic and unachievable weight loss goals that may be a significant source of failure for patients (29,30,31). One of the things patients want most from a primary care physician is help in setting realistic weight goals (7). GPs should also help them to improve their body image and self-esteem and make them aware that a small weight loss can produce important medical benefits (30,31).
Additionally, participants’ nutritional counseling practices did not meet guidelines for successful dietary therapy: 36.2% rarely provided nutritional education, and 60.7% rarely suggested use of a food diary (4,27). Thus, nutritional management often seems limited to one-shot advice and neglects tools that could help induce long-term behavior modifications.
Beyond these barriers, other problems related to the health care system must be considered, including time constraints, modes of reimbursement, and training content. Most GPs in this study, as in others (32,33), reported frequently lacking time, a problem known to be a significant barrier to preventive care in general practice (34,35). Explanations for patients, discussion of treatment and prevention, and health education require longer consultations. We found GPs in the “fixed-fee” billing sector were less likely to feel effective in overweight and obesity management. The type of remuneration influences length of consultation (36); therefore, GPs in this sector may have shorter consultations than the others.
GPs who had taken a CME course and were aware of weight control guidelines felt more effective, probably because this training increased their self-confidence and convinced them that success is possible. The need for better training in the field of weight control was recognized by 80% of the GPs in our panel and has been pointed out by internists, endocrinologists (20), and gynecologists (37). Analyses of potential associations between GPs’ age and various practices and knowledge did not provide convincing evidence that young GPs, who should have received better initial training about nutrition and obesity management, have better practices and are better-informed than their older colleagues (results not shown).
These self-reported data may not exactly reflect the reality of respondents’ attitudes and practices. GPs may be reluctant to declare that obese patients are lazier than normal weight people, that they do not feel that they provide effective treatment, or that they lack confidence in their ability to counsel patients in this area. They may also have overreported practices recommended by the guidelines, such as BMI use for diagnosing weight problems. Moreover, because this study is cross-sectional, we cannot draw causal inferences from the associations observed.
In summary, GPs felt that management of weight problems was one of their responsibilities. Although their diagnostic practices require improvement, most of their reported practices follow the guidelines relatively closely. Nevertheless, like specialists, including endocrinologists and internists, most believed their treatment was of limited effectiveness, and one-third reported dissatisfaction with it. Obstacles to satisfactory management seem to include GPs’ attitudes and opinions and to some extent inadequate practices. In particular, we found, in a relatively new observation, that GPs set weight loss objectives for obesity management that were frequently too stringent.
A current reform of the French health care system will give GPs a central role in health education and prevention. Their attitudes and knowledge may thus determine whether they can maintain this role and perform it effectively. However, organizational aspects, especially remuneration for the time needed to implement prevention and education, must also be considered: for example, the institution of incentive fees for consultations for preventive and education purposes might be useful. Assessment of these measures would then be necessary (38). Facilitating work cooperation between GPs and medical auxiliaries (such as dietician) might also improve prevention and management of weight problems.
This work received technical and financial support from the Southeastern France Regional Union of Private Practitioners (URML PACA) through the Southeastern France Regional Union of Health Insurance (URCAM PACA) Funds for Quality in Ambulatory Health Care (FAQSV). We thank Drs. C. Colette, C. Fischler, M. L. Frelut, P. Garandeau, M. Gerber, J. C. Gourheux, Jouret, P. Y. Lussault, M. Pellae, M. Rousseaux-Romon, M. Tauber, and H. Thibault for invaluable help and advice and Jo Ann Cahn for editorial assistance.