There is an epidemic of obesity in the United States.1 In 1999–2002, 62% of U.S. women aged 20 years or older were overweight (defined as having a body mass index [BMI] greater than 25 kg/m2), one third were obese (BMI 30 or more),2 and 15% of girls aged 12–19 years were overweight (defined as having a BMI equal to or greater than the 95th percentile-age according to the Centers for Disease Control and Prevention [CDC] growth charts).2 Obesity is associated with many chronic diseases, including diabetes, hypertension, cardiovascular disease, and osteoarthritis, as well as with pathologies specific to the clinical practice of obstetrician–gynecologists, including menstrual irregularities, infertility, preeclampsia, and gestational diabetes.3,4 Direct U.S. medical costs associated with obesity are estimated to exceed $61 billion per year.1,5
Many people in the United States lack the skills needed to access, comprehend, and use nutritional and health information to make healthy diet and exercise choices.6 Physicians are a vital link between this information and the patients who can benefit from it. Obstetrician–gynecologists can do a great deal to prevent obesity-related morbidity and death by evaluating and offering appropriate counseling on diet and physical activity to all their patients and by offering appropriate weight-reduction treatment for those who are overweight or obese.4 Studies suggest that people who report receiving advice from their physician about weight loss are more likely to try to lose weight, eat fewer calories, and increase their physical activity than those who do not report receiving such advice.7,8 Results from the 1995 National Ambulatory Medical Care Survey indicated that visits to obstetrician–gynecologists accounted for a substantial proportion of office visits made by women aged 20 years or older but that obstetrician–gynecologists were less likely than cardiologists or general or family practitioners to counsel their patients about physical activity, diet, and losing weight.9
Results from a 2000 American College of Obstetricians and Gynecologists (ACOG) survey indicated that most U.S. obstetrician–gynecologists were well versed in the disease risks associated with obesity and considered weight management of patients to be an important part of their clinical practice but felt that their training in weight management was inadequate.10 The 2000 survey, however, did not address clinical practice patterns for weight management in detail. In recognition of the importance of obesity and the educational needs of obstetricians and gynecologists, ACOG conducted a baseline survey in 2005 designed to describe the knowledge, opinions, and clinical practice patterns of U.S. obstetrician–gynecologists with respect to obesity prevention and treatment.
PARTICIPANTS AND METHODS
Survey questionnaires were sent out to the 806 members of the Collaborative Ambulatory Research Network and to 1,000 randomly selected nonnetwork ACOG members. Network members have volunteered to participate in ACOG Research Department surveys and are typically sent 4–5 surveys per year. Membership in the network is actively maintained by ACOG’s Research Department, with the goal of constituting a national group of practicing obstetrician–gynecologists who are reflective of practicing ACOG members as a whole.11 We initially analyzed responses to this survey by respondents’ network status to determine whether the responses of the two groups differed.
Surveys were mailed to all potential participants in February 2005, with subsequent mailings to nonresponders in March and April. Responses were entered into a computer-based software package (SPSS 12.0, SPSS Inc, Chicago, IL) data file for analysis. The study identification number, sex, birth date, and geographic location for all physicians who were sent the questionnaire were entered into the database. This allowed us to compare the basic demographic characteristics of respondents with those of nonrespondents. The research was approved by the Institutional Review Board of Georgetown University.
The survey consisted of 33 questions about the demographic characteristics of respondents and their practices; the opinions of respondents about the risks, prevention, and treatment of obesity (scored on a 5-point Likert scale); the screening, treatment, and counseling practices of respondents in relation to national recommendations,12 and respondents’ perception of the adequacy of their training. Some of the screening and treatment practices included in the questionnaire were part of the recommended care for adolescents.13 The Likert scale we used and many of the survey questions were modeled after a survey of pediatric primary care practitioners on the management of child and adolescent obesity.14,15
Of the 1,806 surveys mailed, 16 (4 intended for network members and 12 for nonmembers) were returned as undeliverable. One randomly selected recipient had not yet finished her residency and was therefore considered ineligible, and an additional five surveys (four to network members and one to a nonmember) were excluded because they were determined to have been filled out by someone other than the intended recipient. A total of 900 usable surveys were returned and analyzed (437 from network members and 463 from nonmembers). Thus the overall response rate was 50.4% (54.8% for network members and 47.0% for nonmembers). Demographic characteristics of respondents and the limited demographic information available for nonrespondents are given in Table 1. Respondents differed significantly from nonrespondents in mean age (46.9±.3 years old versus 47.8±.3 years old, P<.001) and in sex distribution (51.3% female versus 38.8% female, P<.001).
Some respondents did not answer every question. The number of missing responses per question (except for those related to counseling pregnant women) varied from 0 to 51 (5.7%). Among the three questions related to counseling pregnant women, the number of missing responses was 85–87; all respondents who did not answer those questions reported their specialty as gynecology. Excluding the questions related to counseling pregnant women, 84% of respondents answered every question reported here, and 98% left four or fewer questions blank. For those questions with missing responses, we excluded respondents who did not answer when calculating response frequencies.
We reported values for continuous variables as mean±standard error of the mean (SE). We tested differences in proportions between subgroups using χ2 tests and differences in means using F tests. We divided respondents into three approximately equal subgroups based on the year they completed residency: before 1986, 1986–1996, and after 1996. Reflecting the steady increase in the proportion of women obstetrician–gynecologists, the percentage of respondents who were women increased across these three subgroups (23.0%, 57.6%, and 72.4%, respectively). Accordingly, we stratified our analyses by respondents’ sex and residency completion subgroup.
Network members were, on average, 5 years older than nonmembers (P<.001) and likely to have been in practice longer as indicated by the median year in which they completed their residency training (P<.001). Despite these differences between Network members and nonmembers, there were no significant differences in their responses to this survey after we accounted for the year in which their residency was completed. We therefore aggregated the data.
The mean BMI of respondents based on self-reported height and weight was 25.1±0.1; 33.7% were overweight (BMI 25 or more, but less than 30), and 11.2% were obese (BMI 30 or more). One percent were underweight (BMI less than 18.5). About two of three respondents (64.7%) were concerned about their own weight, and 53.1% were trying to lose weight. Overweight and obese respondents were more likely to be concerned about their weight than were normal-weight respondents (80.2% and 95.0% versus 49.1%, respectively, P<.001) and with trying to lose weight (71.1% and 92.1% versus 34.2%, respectively, P<.001).
More than 90% of respondents indicated that primary care and preventive medicine was an important (53.7%) or very important (37.7%) part of their practice; 6.4% indicated that it was unimportant to their practice, and 2.2% expressed no opinion. All but one of the respondents rated weight as very important (68.2%) or important (31.7%) to the health of their patients. A majority of respondents strongly agreed with the statements “many of my patients would benefit from more physical exercise” (77.8%) and “poor eating habits and lack of physical activity are the main causes of obesity” (56.5%). A substantial proportion also strongly agreed with the statements “an increasing number of my adolescent patients are obese” (44.2%) and “prenatal care is an opportunity to modify maternal behaviors that will lessen the likelihood of maternal obesity” (35.6%). Although 44.7% agreed that “obesity has a strong genetic component,” 32.1% were neutral and 10.6% disagreed. The majority of respondents strongly disagreed with the statements “the health risks of obesity are overstated” (72.3%), “obesity in adolescence is less concerning than obesity in adulthood” (75.7%), and “weight reduction does not improve the health of obese patients” (79.9%). More recent year of residency completion was positively associated with strong agreement with the statements that “many of my patients would benefit from more physical activity” (P=.016) and “an increasing number of my adolescent patients are obese” (P=.039) and with strong disagreement with the statement that “the health risks of obesity are overstated” (P=.027) (data not shown). None of the other opinions were associated with year of residency completion, and there were no significant differences between men and women in the percentages that agreed or disagreed with any of these statements.
A majority of respondents (59.6%) reported that they weigh their nonpregnant patients “at every visit,” and an additional 36.4% reported they weigh patients “at most visits.” When asked to check all the methods that they would use to assess whether a patient is “obese,” 82.2% reported using BMI, 79.4% reported using weight, 47.0% reported using visual inspection, 42.1% reported using changes in weight, 13% reported using weight-for-height percentile, and less than 10% reported using other methods (eg, waist circumference, waist-hip ratio, skinfold thickness, weight-for-age percentile, or BMI-for-age percentile). Physicians who completed their residency after1996 were more likely to report that they used BMI to determine whether their patients were overweight (91.1% versus 78.0% for those who completed their residency in 1996 or before, P<.001), whereas those who completed their residency before 1986 were more likely than the other respondents to use such techniques as waist circumference, BMI-for-age percentile, skinfold thickness, and weight-for-age percentile (data not shown).
Most respondents reported that they screened their overweight patients for hypertension “most of the time,” and a substantial proportion reported that “most of the time” they screened their overweight patients for endocrine disorders and type 2 diabetes (Table 2). A majority of respondents did not frequently screen their overweight patients for the other conditions listed on the survey (Table 2). When taking a family history of their overweight patients, the majority of respondents reported asking “most of the time” about their family history of diabetes mellitus, cardiovascular disease, and hypertension but less often ask about other disorders (Table 3). Of the possible laboratory evaluations listed on the survey, the majority of respondents reported asking for a lipid or total cholesterol profile “most of the time,” a substantial proportion reported asking for serum glucose or thyroid function tests “most of the time,” and fewer than 15% reported ordering other tests “most of the time” (Table 4).
Respondents who completed their residency before 1986 were more likely than those who completed their residency later to screen their overweight patients “most of the time” for type 2 diabetes or orthopedic problems (P<.02). They also were more likely to ask about a family history of obesity and gallbladder disease (P≤.002) and to order serum glucose and thyroid function tests (P<.05). Although the differences were statistically significant, in most cases the absolute differences in the percentages were small except for the following: 53.3% of respondents who completed their residency before 1986 reported that they ordered thyroid function tests “most of the time,” compared with only 43.8% who completed their residency in 1986–1996 and 39.4% who completed their residency after 1996. There were no significant differences in assessment of obesity-related conditions by sex of the respondent.
For “nonpregnant patients who want to lose weight,” the most common reported recommendation was change in eating patterns (Table 5). Almost half of the respondents reported that they “most of the time” recommended that these patients limit their consumption of specific foods and control portion sizes (Table 5). Fewer reported recommending low-calorie diets, commercial diets, or low-fat diets (Table 5). Fewer than 16% reported that they recommended low-carbohydrate diets (eg, Atkins, Zone), and almost none recommend very-low-calorie diets “most of the time” (Table 5). Female respondents were significantly more likely to recommend portion size control (52.2% versus 36.9%, P<.001) and commercial diets (30.7% versus 21.1%, P<.001) “most of the time.” We found no significant associations between respondents’ likelihood of recommending any particular treatment for their overweight patients and the period in which they completed their residency (data not shown).
A majority (65.2%) of respondents reported that they “never” prescribed weight loss medications to their obese nonpregnant patients without an obesity-related medical condition. From a list of various medications they could prescribe to nonpregnant patients who wanted to lose weight (sibutramine, orlistat, phentermine, zonisamide, topiramate, bupropion, metformin, or “other”), the most commonly selected was metformin. Even for metformin, 56% of respondents reported that they never prescribed it, 13.7% “rarely,” 22.8% “sometimes,” 6.2% “often,” and 1.4% “most of the time.” In response to the question, “How often do you refer your obese patients to a behavior modification or therapy program for weight management?” 5% reported that they did so “most of the time,” 21.6% “often,” 37.9% “sometimes,” 25.3% “rarely,” and 3.3% “never.” Referral patterns did not differ by the age or sex of the respondent.
Most respondents reported that they counseled their patients about their physical activity and weight control “most of the time” or “often,” whereas a smaller proportion reported that they counseled their patients about sedentary activity, such as watching TV, “most of the time” or “often” (Table 6). Almost one of four respondents reported either rarely or never counseling their patients regarding their sedentary activity (Table 6). More than 85% of respondents reported counseling their pregnant patients about weight gain during pregnancy “most of the time” or “often” (Table 6), but fewer reported modifying their recommendations about weight gain during pregnancy on the basis of their patients’ prepregnancy BMI (35.7% “most of the time” and 28% “often”), and 25.2% reported that they rarely or never did so. About two of three respondents counseled their postpartum patients about weight maintenance or loss “most of the time” or “often” (Table 6).
When we stratified respondents’ reported counseling practices by their sex and period of residency completion, we found that period of residency completion, but not sex, was associated with the frequency with which respondents reported counseling all patients about sedentary activity (P=.001) and postpartum patients about weight maintenance or loss (P=.02), with respondents who completed residency by or before 1996 being more likely to counsel patients. In contrast, we found that those who completed their residency after 1996 were more likely to use their patients’ prepregnant BMI to modify their recommendations for weight gain during pregnancy (44.5% “most of the time”) (P<.001).
Among women, but not men, respondents, earlier completion of residency was associated with increased frequency of counseling patients about physical activity and weight control and with counseling pregnant patients about weight gain. Women respondents who completed their residency before 1986 were the most likely to counsel their patients: 57.6% reported counseling their patients about physical activity and 49.3% about weight control “most of the time,” and about 69.8% reported counseling their pregnant patients about weight gain “most of the time.”
There were no significant differences between men and women in how they viewed their medical training. A majority of respondents considered their medical training on the health consequences of obesity to be “adequate” (24.7%) or “barely adequate” (36.0%), whereas only 2.8% considered their training to be “comprehensive,” and 36.4% considered it to be “inadequate” or “nonexistent.” A majority (52.8%) considered their training on weight management to be “inadequate” or “nonexistent,” whereas only 0.6% considered it to be “comprehensive,” 11.6% “adequate,” and 35.0% “barely adequate.” Although respondents who completed their training after 1996 rated their training highest, 45.8% of them still considered their training on the weight management of patients to be “inadequate” or “nonexistent.”
Despite the overall low assessment of their training, only a small percentage of respondents indicated that their own lack of “knowledge” or “treatment skills” were a barrier “most of the time” (2.1% and 3.4%, respectively) or “often” (8.5% and 14.6%, respectively). Eighty-two percent listed “lack of patient motivation” as a barrier to managing overweight and obesity “most of the time” or “often,” and almost half also indicated that “lack of visit time” (46.8%) and “lack of support services” (47.7%) were barriers “most of the time” or “often.” Among respondents who completed their residency in 1986 or later, women were more likely than their male counterparts to view lack of support services as a barrier “most of the time” or “often” (49.9% versus 36.5% for men, P=.001), and men were more likely to see “treatment futility” as a barrier “most of the time” or “often” (43.1% versus 32.1% for women, P=.014).
A majority of respondents reported that it was unlikely that they could help their patients lose weight (55.6%) or that their patients would follow their advice on diet (69.1%) or physical activity (67.5%). Physicians who believed that they could help their patients lose weight were significantly more likely to counsel patients regarding weight control (P<.001; Fig. 1), physical activity, sedentary activity, pregnancy weight gain, or postpartum weight loss (P<.001; data not shown). They also were more likely to refer patients to other health professionals, behavior modification, or therapy programs, and to prescribe weight loss medications (Table 7).
Previously, obstetrician–gynecologists were shown to be knowledgeable about the health risks of obesity.10 In this study, we found that most reported using patients’ BMI to screen for obesity in accordance with 1998 National Heart, Lung, and Blood Institute (NHLBI) recommendations.12 This represents a change since 2000; visual inspection (64.4%) was the most commonly reported method of assessing obesity for respondents to the 2000 survey, with 56.0% using BMI.10 In our survey the relationship was reversed, with 82.2% using BMI and 47.0% using visual inspection. Of some concern is that few used waist circumference or waist-to-hip ratio, which some have suggested may provide a better indication of cardiovascular risk.16 Also, although most respondents reported counseling their pregnant patients about weight gain during pregnancy, fewer reported modifying their recommendations for pregnancy weight gain by their patients’ prepregnancy BMI or counseling their postpartum patients about weight loss or maintenance.
Most respondents reported counseling their patients about physical activity, diet, and weight control “most of the time” or “often.” However, we also found that a substantial proportion do not counsel their patients about sedentary activity. Our data suggest that older physicians, especially women, are more likely to counsel their patients. The dietary counseling that respondents were most likely to provide focused more on strategies (changing eating patterns, limiting specific foods, and controlling portion size) rather than on specific (eg, low-calorie) diets. Although a low-calorie diet is the weight loss strategy recommended by NHLBI,12 results of a recent study suggest that certain eating patterns, such as a consistency in when one eats, have beneficial effects on energy metabolism and balance in obese patients.17
Only a minority of providers referred obese patients to a behavior modification or therapy program for weight management or prescribed weight loss medications to their obese nonpregnant patients without an obesity-related medical condition in contrast to NHLBI recommendations that behavioral therapy and pharmacotherapy be used as adjuncts to dietary modification and physical activity.12 We could not determine why these treatments were recommended infrequently, although physician skepticism regarding their ability to help patients lose weight could play a role (Table 7). Providers may be unfamiliar with the benefits of behavioral techniques, may use them in their own counseling and therefore do not feel a need to refer their patients, or may lack support services. Reasons for the infrequency of pharmacotherapy may include lack of knowledge or confidence in pharmacotherapy or success of patients with diet and physical activity alone. Although not directly comparable with this study, in the 2000 survey 36.4% of respondents reported they had prescribed weight loss medication within the previous year,10 implying that the proportion of obstetrician–gynecologists that employ pharmacotherapy has not increased.
Our study results are limited by potential reporting and selection biases. Both may lead to overestimating the rates of obesity screening and counseling practices. Respondents may overstate the frequency with which they engage in screening and treatment practices. Physicians who were more concerned about obesity in their patients may have been more likely to return surveys. Respondents were younger and more likely to be female than nonrespondents (Table 1).
Frank and colleagues18 found that women physicians who had training, had self-confidence in their abilities, or felt that counseling was relevant to their practice were more likely to counsel their patients about nutrition and weight than those without these characteristics. In this study, we found that obstetrician–gynecologists’ belief that they “very likely” could help their patients lose weight was significantly associated with their likelihood of counseling them about weight control (Fig. 1), physical activity, sedentary activity, pregnancy weight gain, and weight loss postpartum (data not shown), as well as with their likelihood of referring patients to behavioral therapy or prescribing weight loss medications (Table 7).
Despite a general assessment that their prepractice training in weight management was inadequate, most obstetrician–gynecologists appear to screen their patients for obesity appropriately and to counsel them about weight control, diet, and physical activity. However, a substantial proportion of obstetrician–gynecologists do not counsel their patients about their sedentary activity, prescribe weight loss medications for overweight patients, or refer patients to behavioral therapy. We found that a significant barrier to appropriate treatment and counseling was the physicians’ skepticism that their advice will be heeded or that they can actually help patients lose weight. We cannot determine the source of this skepticism, although published studies do show that the long-term success rate for most weight loss methods is modest, at best.19 We also found that those with more experience were more likely to counsel their patients about their weight. These findings suggest that providing obstetrician–gynecologists with additional training in how to effectively counsel their patients about their weight may increase the frequency with which they provide obesity prevention and management services, as well as the effectiveness of the services they do provide.
1. Stein CJ, Colditz GA. The epidemic of obesity. J Clin Endocrinol Metab 2004;89:2522–5.
2. Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA 2004;291:2847–50.
3. Nuthalapaty FS, Rouse DJ. The impact of obesity on obstetrical practice and outcome. Clin Obstet Gynecol 2004;47:898–913.
4. Cogswell ME, Perry GS, Schieve LA, Dietz WH. Obesity in women of childbearing age: risks, prevention, and treatment. Prim Care Update Ob Gyns 2001;8:89–105.
5. U.S. Department of Health and Human Services. The Surgeon General’s call to action to prevent and decrease overweight and obesity. Rockville (MD): U.S. Department of Health and Human Services, Public Health Service, Office of the Surgeon General; 2001.
6. Blackburn GL. Teaching, learning, doing: best practices in education. Am J Clin Nutr 2005;82 suppl:218S–21S.
7. Galuska DA, Will JC, Serdula MK, Ford ES. Are health care professionals advising obese patients to lose weight? JAMA 1999;282:1576–8.
8. Nawaz H, Adams ML, Katz DL. Weight loss counseling by health care providers. Am J Public Health 1999;89:764–7.
9. Woodwell DA. National Ambulatory Medical Care Survey: 1995 summary. Adv Data 1997;286:1–25.
10. Power ML, Holzman GB, Schulkin J. Obstetrician–gynecologists’ views on the health risks of obesity. J Reprod Med 2001;46:941–6.
11. Coleman VH, Power ML, Zinberg S, Schulkin J. Contemporary clinical issues in outpatient obstetrics and gynecology: findings of the Collaborative Ambulatory Research Network, 2001–2004: Part I. Obstet Gynecol Surv 2004;59:780–6.
12. National Institutes of Health, National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity: the evidence report. Rockville (MD): U.S. Department of Health and Human Services; 1998.
13. Barlow SE, Dietz WH. Obesity evaluation and treatment: expert committee recommendations. Pediatrics 1998;102 (3):E29. Available at: http://www.pediatrics.org/cgi/content/full/102/3/e29
. Retrieved June 23, 2006.
14. Barlow SE, Dietz WH, Klish WJ, Trowbridge FL. Medical evaluation of overweight children and adolescents: Reports from pediatricians, nurse practitioners, and registered dieticians. Pediatrics 2002;110:222–8.
15. Barlow SE, Trowbridge FL, Klish WJ, Dietz WH. Treatment of child and adolescent obesity: Reports from pediatricians, pediatric nurse practitioners, and registered dieticians. Pediatrics 2002;110:229–35.
16. Yusuf S, Hawken S, Ounpuu S, Bautista L, Franzosi MG, Commerford P, et al. Obesity and the risk of myocardial infarction in 27,000 participants from 52 countries: a case-control study. Lancet 2005;366:1640–9.
17. Farshchi HR, Taylor MA, Macdonald IA. Beneficial metabolic effects of regular meal frequency on dietary thermogenesis, insulin sensitivity, and fasting lipid profiles in healthy obese women. Am J Clin Nutr 2005;81:16–24.
18. Frank E, Wright EH, Serdula MK, Elon LK, Baldwin G. Personal and professional nutrition-related practices of US female physicians. Am J Clin Nutr 2002;75:326–32.
19. Anderson JW, Konz EC, Frederich RC, Wood CL. Long-term weight-loss maintenance: a meta-analysis of US studies. Am J Clin Nutr 2001;74:579–84.