Exploring Pediatric Obesity Training, Perspectives, and Management Patterns Among Pediatric Primary Care Physicians
Abstract
Objective
Significant variability exists in the amount of formal obesity training obtained by physicians caring for pediatric patients. The study objective was to assess the relationship between formal obesity training and pediatrics physicians’ perceptions, practice patterns, overall knowledge, and confidence during management of pediatric obesity.
Methods
An anonymous survey was distributed via email from February 2020 through March 2020 at a large academic system. Internal medicine/pediatrics (46 total) and pediatrics (104 total) primary care providers were selected. Data were collected on the total number of obesity-related training hours by quartiles, demographics, physicians’ clinical practice patterns, and physicians’ knowledge of pediatric obesity management, along with their perceptions, attitudes, and beliefs.
Results
A total of 73 survey participants completed the survey: 69% were female, 77% were older than 40 years, and 74% were White. Physicians with the highest training were most likely to feel confident when managing pediatric obesity. However, only 20% of all physicians felt confident providing pre- and post-bariatric surgery care, and just 6% of physicians self-reported achieving management success.
Conclusions
Increased obesity training improves physicians’ confidence and leads to familiarity with management guidelines. Formal obesity training should be prioritized during residency and beyond so that physicians who care for pediatric patients are better equipped to offer unbiased and effective care.
Study Importance
What is already known?
- ► The amount of formal obesity training may limit obesity management and influence physicians’ perceptions, clinical practice patterns, familiarity with guidelines, and confidence.
- ► Little is known about how obesity-related training affects pediatrics physicians in their care for pediatric patients with obesity.
What does this study add?
- ► Physician perceptions, clinical practice patterns, and confidence in management of pediatric obesity are influenced by the amount of obesity-related training received as well as by physician specialty.
- ► General knowledge of bariatric surgery appears limited as well as pediatric pre- and postoperative management care confidence.
How might these results change the direction of research or the focus of clinical practice?
- ► Our results suggest that formal obesity training increases familiarity with guidelines as well as confidence in management, though specialty-related practice patterns also play a role.
- ► Further training on all aspects of obesity management is required to provide pediatric patients the best treatment options.
Introduction
Obesity is one of the most prevalent pediatric health conditions and it affects almost one-third of children and adolescents in the US ((1, 2)). Childhood obesity increases immediate and long-term comorbidities and mortality rates ((3-5)). A lack of formal obesity training and inadequate widespread availability of treatment and referral options are likely to limit efforts to effectively treat obesity ((6-8)). A review of the literature reveals a paucity of obesity education with only a small percentage of curricula in medical schools, residencies, and fellowship programs involving obesity-specific content ((7, 8)). It has been reported that medical students feel unfamiliar with obesity prevention and treatment. Self-reporting by pediatricians and primary care physicians demonstrated they also feel unprepared to successfully provide care to children with obesity ((9-11)).
Solutions to address this educational gap have been proposed. It was demonstrated that a lecture and video-based obesity curriculum explicitly targeting bias, diagnosis, and management of pediatric obesity improved postintervention measures of implicit bias ((12)). Interprofessional coaching and online simulation technology addressing motivational interviewing skills seem to improve clinician and patient satisfaction in primary care and pediatric obesity-focused settings ((13, 14)). Hands-on skills training in an established chronic weight management program was shown to improve physicians’ adherence to established pediatric weight management guidelines ((15)). Physicians also expressed interest in having clear obesity clinical decision tools embedded within their electronic health record (EHR) ((16)). Therefore, clinic-based, lecture-based, and online-based learning settings all seem useful.
Belay and colleagues reported that, compared with a cohort tested in 2006, pediatricians felt more prepared when counseling and discussing obesity-related issues with children and their families ((17)); however, only a small proportion felt their training had been effective. Moreover, there was limited information about perceived self-efficacy regarding their counseling efforts or whether or not their obesity knowledge and management tools were accurate and unbiased ((17, 18)). Therefore, this study aimed to assess the relationship between formal obesity training and pediatrics physicians’ perceptions, clinical practice patterns, overall knowledge, and confidence during management of pediatric obesity and to learn their personal habits and perspectives on the major contributors of obesity.
Methods
Participants and data collection
We designed a brief online survey that we distributed via the Research Electronic Data Capture (REDCap), a Web-based tool that supports customized data collection for research studies in a secure platform that complies with Health Insurance Portability and Accountability Act regulations ((19)). We targeted physicians who provide primary care to children and adolescents affiliated with Partners Healthcare System, a not-for-profit, large, unified health care system consisting of two large academic medical centers, three community teaching hospitals, and other affiliated specialty facilities and community health centers in the northeast US. We selected 46 internal medicine/pediatrics physicians and 104 general pediatrics primary care providers available through the Partners website. Data were collected anonymously, and 150 surveys were sent via email from February 18, 2020, through March 17, 2020. Participants who completed the survey were offered $20 in the form of a gift card as a remuneration for their time. A total of 8 participants did not provide electronic consent or expressed not feeling qualified to complete the survey. A total of 73 participants completed the survey, and these responses were used for analyses. As per the American Association for Public Opinion Research Outcome Rate Calculator, we obtained a 49% response rate, with a 5% refusal rate and 90% cooperation rate ((20)). The Institutional Review Board of the Massachusetts General Hospital approved this study.
Survey instrument
We developed a survey to ascertain obesity training in pediatricians. In this survey, we collected self-identified demographic data, such as gender, racial/ethnic background, age, weight, height, country of origin, type of medical degree obtained, primary medical specialty, and year of medical school graduation. Additionally, data on known personal history of chronic illnesses such as overweight or obesity, asthma, diabetes mellitus, dyslipidemia, gastroesophageal reflux disease, coronary artery disease, hypertension, nonalcoholic fatty liver disease, obstructive sleep apnea, osteoarthritis, polycystic ovarian syndrome, and depression were collected. We gathered information on the pediatrician respondents’ clinical practice as we collected data on frequency of physicians’ recording a diagnosis of overweight or obesity in the EHR, physicians’ rapport with patients with overweight or obesity, and physicians’ knowledge about pediatric obesity management, along with their perceptions, attitudes, and beliefs. In order to ascertain their obesity education, we queried the duration and type of obesity training during their medical career and asked general knowledge questions regarding obesity management. Likewise, the pediatricians were asked to disclose whether they had a history of having overweight/obesity and their personal health habits to manage their weight. This included elements such as nutritional preferences, physical activity, stress reduction methods, and use of smartphone applications to manage weight. As per Centers for Disease Control and Prevention guidelines, BMI is a measure used to determine childhood overweight and obesity. In pediatrics, overweight is defined as growth chart percentile ≥ 85% and obesity as growth chart percentile ≥ 95% ((21)).
Statistical analysis
All statistical analyses were conducted using Stata software version 15.1 (StataCorp LLC, College Station, Texas). Univariate statistics were obtained using Student t test and χ2 analyses. Obesity training was characterized into four categorical variables based on the quartile (≤ 7.5, 7.5-15, 15-25, and > 25 ) of the total amount of hours dedicated to lectures, online training, group discussions, topic courses focused on weight management, and clinical hours of obesity-focused hospital rotation. Multivariable analyses were performed using ordered (i.e., ordinal) logistic regression models adjusted for training-hours quartiles, BMI class (normal, overweight, and obesity), personal history of having overweight or obesity in the past, and specialty (pediatrics or internal medicine/pediatrics).
Results
General characteristics of study sample
Table 1 summarizes the demographics of our cohort stratified by obesity training hours (quartiles). A total of 73 surveys (71% pediatrics and 29% internal medicine/pediatrics) were analyzed. All were primary care providers, and none of the responders reported having another medical specialty or subspecialty. Only one reported a master’s degree in public health as further graduate training. The majority were female (69%), older than 40 years old (77%), and White (74%). Most physicians were born in the US (84%). More than half (53%) reported having a normal BMI at the time of the survey, with 50% having experienced overweight/obesity in the past. Chronic diseases did not differ among quartiles and they were reported in 52% of the physicians. Adjusted analyses demonstrated that physicians who had graduated more recently were more likely to have received additional obesity-specific training compared with older colleagues (odds ratio [OR] 1.05 [95% CI: 1.0-1.1], P = 0.03). Male and pediatrics physicians were also more likely to have received increased obesity-related training overall (OR 2.8 [95% CI: 1.0-7.7], P = 0.05; and OR 3.5 [95% CI: 1.3-9.4], P = 0.02, respectively) (Table 2).
| Obesity training hours (quartiles) | Overall, N = 73 | ||||
|---|---|---|---|---|---|
| 1st, | 2nd, | 3rd, | 4th, | ||
| n = 22 | n = 18 | n = 16 | n = 17 | ||
| Age, n (%) | |||||
| 20-29 | (0.0) | (0.0) | (0.0) | 1 (5.9) | 1 (1.4) |
| 30-39 | 4 (18.2) | 4 (22.2) | 3 (18.8) | 3 (17.7) | 14 (19.2) |
| 40-49 | 4 (18.2) | 6 (33.3) | 7 (43.8) | 8 (47.1) | 25 (34.3) |
| 50-65 | 13 (59.1) | 7 (38.9) | 6 (37.5) | 5 (29.4) | 31 (42.5) |
| > 65 | 1 (4.6) | 1 (5.6) | (0.0) | (0.0) | 2 (2.7) |
| Sex, n (%) | |||||
| Male | 8 (36.4) | 1 (5.6) | 4 (25.0) | 10 (58.8) | 23 (31.5) |
| Female | 14 (63.6) | 17 (94.4) | 12 (75.0) | 7 (41.2) | 50 (68.5) |
| Race, n (%) | |||||
| Black | 2 (10.0) | (0.0) | (0.0) | (0.0) | 2 (2.9) |
| Asian/Pacific Islander | 3 (15.0) | 2 (11.1) | 2 (13.3) | 1 (5.9) | 8 (11.4) |
| White | 13 (65.0) | 15 (83.3) | 11 (73.3) | 13 (76.5) | 52 (74.3) |
| Hispanic | 1 (5.0) | 1 (5.6) | 1 (6.7) | 3 (17.7) | 6 (8.6) |
| Native American/Alaskan | (0.0) | (0.0) | 1 (6.7) | (0.0) | 1 (1.4) |
| Mixed | 1 (5.0) | (0.0) | (0.0) | (0.0) | 1 (1.4) |
| BMI, kg/m2, mean (95% CI) | 27.1 (24.3-29.9) | 26.2 (22.8-29.5) | 26.4 (24.1-28.6) | 25.2 (23.5-26.8) | 26.3 (24.9-27.6) |
| BMI class, n (%) | |||||
| Normal | 11 (50.0) | 10 (55.6) | 8 (50.0) | 10 (58.8) | 39 (53.4) |
| Overweight | 4 (18.2) | 4 (22.2) | 5 (31.3) | 5 (29.4) | 18 (24.7) |
| Obesity | 7 (31.8) | 4 (22.2) | 3 (18.8) | 2 (11.8) | 16 (21.9) |
| Prior overweight/obesity, n (%) | 14 (63.6) | 8 (44.4) | 7 (43.8) | 7 (43.8) | 36 (50.0) |
| Specialty, n (%) | |||||
| Pediatrics | 12 (54.6) | 13 (72.2) | 12 (75.0) | 15 (88.2) | 52 (71.2) |
| Internal medicine/pediatrics | 10 (45.5) | 5 (27.8) | 4 (25.0) | 2 (11.8) | 21 (28.8) |
| Chronic diseases, n (%) | |||||
| Asthma | 5 (22.7) | 4 (22.2) | 3 (18.8) | 6 (35.3) | 18 (24.7) |
| Type 2 diabetes mellitus | (0.0) | (0.0) | (0.0) | (0.0) | (0.0) |
| Dyslipidemia | 2 (9.1) | 1 (5.6) | 1 (6.3) | 1 (5.9) | 5 (6.9) |
| Gastroesophageal reflux disease | 2 (9.1) | 1 (5.6) | 5 (31.3) | 3 (17.7) | 11 (15.1) |
| Coronary artery disease | (0.0) | (0.0) | 1 (6.3) | (0.0) | 1 (1.4) |
| Hypertension | 1 (4.6) | 1 (5.6) | 2 (12.5) | 3 (17.7) | 7 (9.6) |
| Nonalcoholic fatty liver disease | (0.0) | (0.0) | 1 (6.3) | (0.0) | 1 (1.4) |
| Obstructive sleep apnea | (0.0) | (0.0) | 1 (6.3) | (0.0) | 1 (1.4) |
| Osteoarthritis | (0.0) | (0.0) | 1 (6.3) | 1 (5.9) | 2 (2.7) |
| Polycystic ovarian syndrome | (0.0) | (0.0) | 1 (6.3) | 1 (5.9) | 2 (2.7) |
| Depression | 5 (22.7) | 2 (11.1) | 2 (12.5) | (0.0) | 9 (12.3) |
| None | 10 (45.5) | 9 (50.0) | 8 (50.0) | 8 (47.1) | 35 (48.0) |
| Overall training quartile | Lectures | Online training | Group discussions | Hospital-based training | Special topic courses | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Male | 2.81 (1.0-7.7) | 0.05 | 2.04 (0.8-5.4) | > 0.1 | 3.01 (1.1-8.1) | 0.03 | 2.59 (1.0-6.7) | 0.05 | 0.58 (0.1-2.5) | > 0.1 | 3.19 (1.2-8.5) | 0.02 |
| Graduation year | 1.05 (1.0-1.1) | 0.03 | 1.05 (1.0-1.1) | 0.04 | 1.01 (1.0-1.1) | > 0.1 | 1.04 (1.0-1.1) | 0.08 | 1.14 (1.1-1.2) | 0.001 | 0.99 (0.9-1.0) | > 0.1 |
| Prior overweight/obesity | 0.63 (0.3-1.5) | > 0.1 | 0.57 (0.2-1.4) | > 0.1 | 1.11 (0.5-2.6) | > 0.1 | 0.35 (0.1-0.8) | 0.02 | 1.54 (0.4-5.3) | > 0.1 | 0.71 (0.3-1.7) | > 0.1 |
| Specialty | ||||||||||||
| Internal medicine/pediatrics | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| Pediatrics | 3.46 (1.3-9.4) | 0.02 | 1.72 (0.7-4.5) | > 0.1 | 1.42 (0.6-3.6) | > 0.1 | 2.39 (0.9-6.2) | 0.08 | 0.84 (0.2-3.0) | > 0.1 | 3.10 (1.1-8.6) | 0.03 |
- Significant P values (P < 0.05) are bolded. OR indicates odds ratio derived from ordered (i.e., ordinal) logistic regression model, which assumes proportional odds between each level of the dependent variable. “Ref” indicates a common reference point to which the other categories were compared for interpretation of results.
Clinical practice patterns, perceptions, and barriers of pediatric obesity management
Overall, 97% of the physicians agreed that obesity is a chronic disease. We assessed the physicians’ perceptions in each of the quartiles (Table 3). Most physicians disagreed with the perception that it is too difficult for children, adolescents, and young adults to change their behavior, and 73% disagreed that these patients are not generally interested in improving their weight status. Most physicians (77%) were likely to record a diagnosis of pediatric overweight or obesity in the EHR more than 75% of the time. Overall, 56% of physicians agreed that they do not receive adequate reimbursement for assessing patients with obesity. Among the major barriers to evaluating and managing patients, 82% of physicians agreed that there is a lack of adequate referral services for diet, physical activity, and weight management, and 62% identified long wait times for referrals to an obesity medicine specialist. Around 70% of the physicians agreed with feeling that there is a lack of effective tools and information to distribute to pediatric patients regarding obesity. More than two-thirds of the physicians agreed with feeling that there is a lack of effective pediatric obesity treatment options. More than half of all physicians (52%) had the perception that pediatric patients would be less likely to trust physicians if they had overweight or obesity; among these, 58% had a normal BMI, and 54% had overweight or obesity in the past. Only 38% of all physicians agreed that they feared offending the patient or parents when speaking about weight management. Finally, approximately 16% of the physicians thought that pediatric patients are generally not interested in improving their weight status.
| Obesity training hours (quartiles) | |||||
|---|---|---|---|---|---|
| 1st, | 2nd, | 3rd, | 4th, | Overall, | |
| n = 22 | n = 18 | n = 16 | n = 17 | N = 73 | |
| I feel it is too difficult for children, adolescents, and young adults to change their behavior (n, %) | |||||
| Strongly disagree/disagree | 11 (50.0) | 11 (61.1) | 10 (62.5) | 10 (13.7) | 42 (57.5) |
| Neutral | 4 (18.2) | 3 (16.7) | 1 (6.3) | 1 (5.9) | 9 (12.3) |
| Strongly agree/agree | 7 (31.8) | 4 (22.2) | 5 (31.3) | 6 (35.3) | 22 (30.1) |
| Frequency of recording diagnosis of overweight or obesity in the EHR (n, %) | |||||
| < 50% of the time | 4 (18.2) | 1 (5.6) | 1 (6.3) | 0 | 6 (8.2) |
| 50%-75% of the time | 3 (13.6) | 6 (33.3) | 1 (6.3) | 1 (5.9) | 11 (15.1) |
| > 75% of the time | 15 (68.2) | 11 (61.1) | 14 (87.5) | 16 (94.1) | 56 (76.7) |
| I think patients are generally not interested in improving their weight status (n, %) | |||||
| Strongly disagree/disagree | 17 (77.3) | 16 (88.9) | 11 (68.8) | 9 (52.9) | 53 (72.6) |
| Neutral | 1 (4.6) | 1 (5.6) | 2 (12.5) | 4 (23.5) | 8 (11.0) |
| Strongly agree/agree | 4 (18.2) | 1 (5.6) | 3 (18.8) | 4 (23.5) | 12 (16.4) |
| I would treat obesity more regularly if there was reimbursement set aside for that purpose (n, %) | |||||
| Strongly disagree/disagree | 8 (36.4) | 5 (27.8) | 10 (62.5) | 11 (64.7) | 34 (46.6) |
| Neutral | 9 (40.9) | 11 (61.1) | 4 (25.0) | 3 (17.7) | 27 (37.0) |
| Strongly agree/agree | 5 (22.7) | 2 (11.1) | 2 (12.5) | 3 (17.7) | 12 (16.4) |
| Trust of weight loss advice from physicians with overweight/obesity (n, %) | |||||
| More likely to trust | 3 (13.6) | 1 (5.6) | 2 (12.5) | 1 (5.9) | 7 (9.6) |
| Equally likely to trust | 7 (31.8) | 10 (55.6) 7 (38.9) | 8 (50.0) | 3 (17.7) | 28 (38.4) |
| Less likely to trust | 12 (54.6) | 1 (5.6) | 6 (37.5) | 13 (76.5) | 38 (52.1) |
| I feel there is a lack of adequate referral services for diet, physical activity, and weight management (n, %) | |||||
| Strongly disagree/disagree | 0 | 2 (11.1) | 1 (6.3) | 2 (11.8) | 5 (6.9) |
| Neutral | 0 | 0 | 0 | 1 (5.9) | 1 (5.9) |
| Strongly agree/agree | 22 (100) | 16 (88.9) | 15 (93.8) | 14 (82.4) | 67 (82.4) |
| I think there are long wait times for referrals to obesity medicine specialists (n, %) | |||||
| Strongly disagree/disagree | 2 (9.1) | 0 | 2 (12.5) | 3 (17.7) | 7 (9.6) |
| Neutral | 9 (40.9) | 5 (27.8) | 2 (12.5) | 5 (29.4) | 21 (28.8) |
| Strongly agree/agree | 11 (50.0) | 13 (72.2) | 12 (75.0) | 9 (52.9) | 45 (61.6) |
| I feel there is a lack of effective tools and information to give to pediatric patients regarding obesity (n, %) | |||||
| Strongly disagree/disagree | 0 | 1 (5.6) | 2 (12.5) | 3 (17.7) | 6 (8.2) |
| Neutral | 3 (13.6) | 3 (16.7) | 4 (25.0) | 6 (35.3) | 16 (21.9) |
| Strongly agree/agree | 19 (86.4) | 14 (77.8) | 10 (62.5) | 8 (47.1) | 51 (69.9) |
| I feel there is a lack of effective treatment options in children, adolescents, and young adults with obesity (n, %) | |||||
| Strongly disagree/disagree | 3 (13.6) | 3 (16.7) | 1 (6.3) | 7 (41.2) | 14 (19.2) |
| Neutral | 4 (18.2) | 1 (5.6) | 2 (12.5) | 3 (17.7) | 10 (13.7) |
| Strongly agree/agree | 15 (68.2) | 14 (77.8) | 13 (81.3) | 7 (41.2) | 49 (67.1) |
| I feel bariatric surgery is a safe option for treating obesity in children, adolescents, and young adults (n, %) | |||||
| Strongly disagree/disagree | 4 (18.2) | 2 (11.1) | 4 (25.0) | 4 (25.5) | 14 (19.2) |
| Neutral | 11 (50.) | 6 (33.3) | 7 (43.8) | 5 (29.4) | 29 (39.7) |
| Strongly agree/agree | 7 (31.8) | 10 (55.6) | 5 (31.3) | 8 (47.1) | 30 (41.1) |
| I feel bariatric surgery is a useful tool for treating obesity in children, adolescents, and young adults (n, %) | 8 (11.0) | ||||
| Strongly disagree/disagree | 3 (13.6) | 1 (5.6) | 1 (6.3) | 3 (17.7) | 28 (38.4) |
| Neutral | 8 (36.4) | 9 (50.0) | 6 (37.5) | 5 (29.4) | 37 (50.7) |
| Strongly agree/agree | 11 (50.0) | 8 (44.4) | 9 (56.3) | 9 (52.9) | 8 (11.0) |
- EMR, electronic medical record.
Adjusted analyses revealed that, compared with internal medicine/pediatrics, pediatrics physicians were less likely to disagree with the statements “I feel it is too difficult for children, adolescents, and young adults to change their behavior” and “I think patients are generally not interested in improving their weight status” (OR 0.33 [95% CI: 0.1-0.9], P = 0.03; and OR 0.21 [95% CI: 0.1-0.7], P = 0.01, respectively) (Table 4). Pediatrics physicians were nearly five times more likely to record a diagnosis of obesity in the EHR than internal medicine/pediatrics (OR 4.9 [95% CI: 1.4-16.5], P = 0.01). Physicians in the fourth quartile of training were most likely to disagree with the statements that “there is a lack of information” (OR 4.93 [95% CI: 1.4-17.9], P = 0.02) or “[a lack of] treatment options to treat patients with obesity” (OR 4.14 [95% CI:1.1-15.3], P = 0.03) and were less likely to agree with the statement that they “would treat obesity more regularly if there was reimbursement set aside for that purpose” (OR 0.3 [95% CI: 0.1-1.0], P = 0.05).
| How often do you add overweight or obesity to the EHR? (more often) | I would treat obesity more regularly if there was reimbursement set aside for that purpose (agreement) | I feel there is a lack of effective tools and information to give to pediatric patients regarding obesity (disagreement) | I feel there is a lack of effective treatment options in children, adolescents, and young adults with obesity (disagreement) | I feel it is too difficult for children, adolescents, and young adults to change their behavior (disagreement) | I think patients are generally not interested in improving their weight status (disagreement) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Training hours (quartile) | ||||||||||||
| 1st | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| 2nd | 0.76 (0.2-3.0) | > 0.1 | 0.90 (0.3-2.8) | > 0.1 | 0.93 (0.3-3.1) | > 0.1 | 0.82 (0.2-2.7) | > 0.1 | 1.89 (0.6-6.0) | > 0.1 | 2.53 (0.6-10.0) | > 0.1 |
| 3rd | 2.83 (0.4-18.6) | > 0.1 | 0.45 (0.1-1.5) | > 0.1 | 2.18 (0.6-7.9) | > 0.1 | 0.64 (0.2-2.2) | > 0.1 | 2.03 (0.6-7.1) | > 0.1 | 1.39 (0.3-5.8) | > 0.1 |
| 4th | 5.46 (0.5-57.0) | > 0.1 | 0.26 (0.1-1.0) | 0.05 | 4.93 (1.4-17.9) | 0.02 | 4.16 (1.1-15.3) | 0.03 | 1.63 (0.4-6.1) | > 0.1 | 0.73 (0.2-2.9) | > 0.1 |
| BMI class | ||||||||||||
| Normal | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| Overweight | 0.87 (0.1-6.0) | > 0.1 | 0.35 (0.1-1.3) | > 0.1 | 0.85 (0.2-3.3) | > 0.1 | 0.98 (0.2-3.9) | > 0.1 | 2.56 (0.6-10.3) | > 0.1 | 0.57 (0.1-2.6) | > 0.1 |
| Obesity | 0.93 (0.2-5.4) | > 0.1 | 0.40 (0.1-1.8) | > 0.1 | 1.00 (0.2-4.3) | > 0.1 | 3.00 (0.7-12.2) | > 0.1 | 1.21 (0.3-5.2) | > 0.1 | 0.28 (0.1-1.5) | > 0.1 |
| Prior overweight/obesity | 0.51 (0.1-2.7) | > 0.1 | 1.82 (0.5-6.4) | > 0.1 | 0.95 (0.3-3.4) | > 0.1 | 0.50 (0.1-1.7) | > 0.1 | 1.95 (0.5-6.9) | > 0.1 | 2.89 (0.7-12.4) | > 0.1 |
| Specialty | ||||||||||||
| Internal medicine/pediatrics | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| Pediatrics | 4.88 (1.4-16.5) | 0.01 | 1.95 (0.7-5.2) | > 0.1 | 0.84 (0.3-2.3) | > 0.1 | 0.62 (0.2-1.7) | > 0.1 | 0.33 (0.1-0.9) | 0.03 | 0.21 (0.1-0.7) | 0.01 |
- Significant P values (P < 0.05) are bolded.
- OR indicates odds ratio derived from ordered (i.e., ordinal) logistic regression model, which assumes proportional odds between each level of the dependent variable. “Ref” indicates a common reference point to which the other categories were compared for interpretation of results.
Confidence in treatment of obesity
Overall confidence levels were low for the eight categories in our survey that evaluated confidence during obesity management. Specifically, only 60% of physicians felt well trained to provide exercise counseling, 51% felt they were well trained to provide nutrition counseling, and 37% felt they could provide motivational interviewing to patients. Notably, only 14% of physicians felt confident discussing weight loss medications as a treatment option, and an even smaller proportion (12%) felt confident discussing potential eligibility for bariatric surgery. Lastly, 20% of all physicians felt confident providing pre- and post-bariatric surgery care. Overall, only 6% of all the respondents felt generally successful in treating pediatric patients with obesity (Supporting Information Table S1).
Adjusted analyses demonstrated that physicians in the fourth quartile of obesity training were most likely to feel confident with most modalities of treatment, and that pediatrics physicians were significantly less likely to be confident when utilizing weight loss medications and when discussing bariatric surgery and follow-up care compared with internal medicine/pediatrics (Table 5).
| I am generally successful in treating obesity in children, adolescents, and young adults (agreement) | I feel I am well trained to provide nutrition counseling to children, adolescent, and young adult patients (confident) | I feel I am well trained to provide exercise counseling to children, adolescents, and young adult patients (confident) | I feel confident about utilization of weight loss medications as a treatment of obesity (confident) | I feel confident when discussing potential eligibility for bariatric surgery with pediatric patients (confident) | I feel confident about providing patient care before bariatric surgery (confident) | I feel confident about providing patient care after bariatric surgery (confident) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Training hours (quartile) | ||||||||||||||
| 1st | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| 2nd | 1.38 (0.4-5.0) | > 0.1 | 1.20 (0.4-4.1) | > 0.1 | 1.03 (0.3-3.4) | > 0.1 | 0.41 (0.1-2.4) | > 0.1 | 1.36 (0.3-6.0) | > 0.1 | 1.90 (0.5-7.9) | > 0.1 | 0.96 (0.2-4.2) | > 0.1 |
| 3rd | 0.75 (0.2-2.8) | > 0.1 | 1.18 (0.3-4.3) | > 0.1 | 1.43 (0.4-5.0) | > 0.1 | 0.06 (0.0-1.0) | 0.05 | 1.86 (0.4-9.5) | > 0.1 | 1.17 (0.2-5.5) | > 0.1 | 1.37 (0.3-6.8) | > 0.1 |
| 4th | 4.65 (1.2-18.1) | 0.03 | 5.32 (1.4-20.9) | 0.02 | 3.92 (1.0-14.8) | 0.04 | 0.64 (0.1-5.1) | > 0.1 | 5.25 (1.2-23.5) | 0.03 | 7.06 (1.6-31.8) | 0.01 | 5.71 (1.2-26.9) | 0.03 |
| BMI class | ||||||||||||||
| Normal | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| Overweight | 0.70 (0.2-2.7) | > 0.1 | 1.58 (0.4-6.1) | > 0.1 | 1.70 (0.5-6.4) | > 0.1 | 5.21 (0.5-53.3) | > 0.1 | 1.04 (0.2-5.8) | > 0.1 | 4.69 (1.0-22.9) | 0.06 | 6.15 (1.2-30.7) | 0.03 |
| Obesity | 0.90 (0.2-4.4) | > 0.1 | 1.19 (0.3-5.0) | > 0.1 | 0.42 (0.1-1.7) | > 0.1 | 0.38 (0.0-4.4) | > 0.1 | 1.40 (0.3-7.6) | > 0.1 | 9.88 (1.7-56.7) | 0.01 | 3.18 (0.6-16.8) | > 0.1 |
| Prior overweight/obesity | 1.32 (0.4-4.7) | > 0.1 | 1.32 (0.4-4.7) | > 0.1 | 1.88 (0.5-6.4) | > 0.1 | 2.72 (0.3-22.0) | > 0.1 | 1.03 (0.2-4.8) | > 0.1 | 0.32 (0.1-1.4) | > 0.1 | 0.57 (0.1-2.4) | > 0.1 |
| Specialty | ||||||||||||||
| Internal medicine/pediatrics | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — | Ref | — |
| Pediatrics | 1.08 (0.4-3.1) | > 0.1 | 2.53 (0.9-7.5) | 0.09 | 1.63 (0.6-4.7) | > 0.1 | 0.03 (0.0-0.2) | < 0.001 | 0.26 (0.1-0.8) | 0.02 | 0.15 (0.0-0.5) | 0.002 | 0.05 (0.0-0.2) | < 0.001 |
- Significant P values (P < 0.05) are bolded. OR indicates odds ratio derived from ordered (i.e., ordinal) logistic regression model, which assumes proportional odds between each level of the dependent variable. “Ref” indicates a common reference point to which the other categories were compared for interpretation of results.
Metabolic and bariatric surgery knowledge in pediatric patients
While 60% of the survey respondents agreed that they would recommend evaluation by a bariatric surgeon if a patient met the standard criteria for metabolic and bariatric surgery (MBS), just 41% agreed that MBS is a safe option, and only 51% considered it a useful tool to treat pediatric obesity. A total of five questions were focused on MBS knowledge, and overall, only 59% were answered correctly (Table 6). Only 67% of all physicians were able to answer at least three out of five questions correctly, and pediatrics physicians were less likely to achieve this compared with internal medicine/pediatrics (OR 0.30 [95% CI: 0.1-1.0], P = 0.05) (Table 7).
| Obesity training hours (quartiles) | Overall, | ||||
|---|---|---|---|---|---|
| 1st, | 2nd, | 3rd, | 4th, | ||
| n = 22 | n = 18 | n = 16 | n = 17 | N = 73 | |
| Which BMI would typically qualify a patient for bariatric surgery? (correct answer: BMI 35+ with comorbidities) | |||||
| Correct, n (%) | 20 (90.9) | 16 (88.9) | 15 (93.8) | 17 (100.0) | 68 (93.2) |
| Incorrect, n (%) | 2 (9.1) | 2 (11.1) | 1 (6.3) | 0 | 5 (6.8) |
| What is the average expected excess body weight loss from Roux-en-Y gastric bypass? (correct answer: 50%-75%) | |||||
| Correct, n (%) | 16 (72.7) | 11 (61.1) | 8 (50.0) | 6 (35.3) | 41 (56.2) |
| Incorrect, n (%) | 6 (27.3) | 7 (38.9) | 8 (50.0) | 11 (64.7) | 32 (43.8) |
| What is the national 30-day mortality rate of patients who undergo Roux-en-Y bypass? (correct answer: < 1%) | |||||
| Correct, n (%) | 17 (77.3) | 14 (77.8) | 11 (68.8) | 11 (64.7) | 53 (72.6) |
| Incorrect, n (%) | 5 (22.7) | 4 (22.2) | 5 (31.3) | 6 (35.3) | 20 (27.4) |
| Which is the most commonly performed metabolic and bariatric surgery procedure in the US? (correct answer: sleeve gastrectomy) | |||||
| Correct, n (%) | 9 (40.9) | 7 (38.9) | 5 (31.3) | 3 (17.7) | 24 (32.9) |
| Incorrect, n (%) | 13 (59.1) | 11 (61.1) | 11 (68.8) | 14 (83.4) | 49 (67.1) |
| Patients who undergo bariatric surgery are expected to achieve their maximum weight loss within which time frame? (correct answer: 12-18 months) | |||||
| Correct, n (%) | 7 (31.8) | 8 (44.4) | 7 (43.8) | 8 (47.1) | 30 (41.1) |
| Incorrect, n (%) | 15 (68.2) | 10 (55.6) | 9 (56.3) | 9 (52.9) | 43 (58.9) |
| Bariatric surgery knowledge percentage of correct answers | |||||
| 3/5 correct, n (%) | 18 (81.8) | 12 (66.7) | 11 (68.8) | 8 (47.1) | 49 (67.1) |
| % of correct answers | 62.7 | 62.2 | 57.5 | 52.9 | 59.2 |
| At least 3 out of 5 questions correct | P | |
|---|---|---|
| OR (95% CI) | ||
| Training hours (quartile) | ||
| 1st | Ref | — |
| 2nd | 1.06 (0.3-4.2) | > 0.1 |
| 3rd | 0.63 (0.1-2.7) | > 0.1 |
| 4th | 0.43 (0.1-1.9) | > 0.1 |
| BMI class | ||
| Normal | Ref | — |
| Overweight | 3.00 (0.6-15.2) | > 0.1 |
| Obesity | 1.75 (0.3-10.1) | > 0.1 |
| Prior overweight/obesity | 0.87 (0.2-4.0) | > 0.1 |
| Specialty | ||
| Internal medicine/pediatrics | Ref | — |
| Pediatrics | 0.30 (0.1-1.0) | 0.05 |
- OR indicates odds ratio derived from ordered (i.e., ordinal) logistic regression model, which assumes proportional odds between each level of the dependent variable. “Ref” indicates a common reference point to which the other categories were compared for interpretation of results.
Major contributors of obesity from physicians’ perspective
The major contributors to obesity according to the survey respondents are the overconsumption of food (96%), followed by physical inactivity (86%), poor nutritional habits (84%), consumption of sugar/sweetened beverages (82%), eating at restaurant/fast food (74%), genetic factors (70%), psychological problems (55%), repeated dieting with weight cycling (22%), lack of willpower (21%), osteoarthritis (21%), medications (19%), and metabolic defect or an endocrine disorder (15%). There were no significant differences between training quartiles (Supporting Information Table S2).
Personal habits of physicians
Physicians were asked four questions regarding their personal health habits, and their responses were divided into frequently/almost always, sometimes, and infrequently/almost never. Of the physicians, 84% disclosed frequently/almost always having good nutritional habits, and 71% disclosed that they performed regular physical activity, of at least 150 minutes of moderate intensity or 75 minutes of vigorous activity, weekly; 47% disclosed infrequently/almost never using stress-reduction techniques, and only 27% disclosed using them almost always/frequently. Just 26% sometimes used any technique like yoga, meditation, or prayer. Finally, 78% disclosed almost never/infrequently using a smartphone application to help manage their weight, while 12% disclosed sometimes using it and only 10% disclosed frequently/almost always using it (Supporting Information Table S3). Those with overweight were less likely to consider themselves to have good nutritional habits or to perform regular exercise compared with those in the other BMI groups (OR 0.17 [95% CI: 0.0-0.8], P = 0.03; and OR 0.09 [95% CI: 0.0-0.5], P = 0.004, respectively). Pediatrics physicians were also less likely to perform regular exercise compared with internal medicine/pediatrics (OR 0.32 [95% CI: 0.1-0.9], P = 0.04). Finally, physicians with obesity were six times more likely to use a smartphone application to manage their weight (OR 6.03 [95% CI: 1.4-25.8], P = 0.02) (Table 8).
| Do you perform regular physical activity of at least 150 minutes of moderate intensity or 75 minutes of vigorous activity weekly? (more often) | Do you consider yourself to have good nutritional habits? (more often) | Do you use smartphone applications to help you manage your weight? (more often) | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Training hours (quartile) | ||||||
| 1st | Ref | — | Ref | — | Ref | — |
| 2nd | 1.45 (0.4-5.1) | > 0.1 | 2.21 (0.6-8.2) | > 0.1 | 1.12 (0.3-4.8) | > 0.1 |
| 3rd | 0.66 (0.2-2.3) | > 0.1 | 0.36 (0.1-1.3) | > 0.1 | 2.69 (0.7-10.9) | > 0.1 |
| 4th | 1.12 (0.3-4.4) | > 0.1 | 1.86 (0.5-7.0) | > 0.1 | 1.52 (0.3-6.8) | > 0.1 |
| BMI class | ||||||
| Normal | Ref | — | Ref | — | Ref | — |
| Overweight | 0.63 (0.1-2.8) | > 0.1 | 0.59 (0.1-2.4) | > 0.1 | 0.76 (0.2-3.4) | > 0.1 |
| Obesity | 0.09 (0.0-0.5) | 0.004 | 0.17 (0.0-0.8) | 0.03 | 0.92 (0.2-4.4) | > 0.1 |
| Prior overweight/obesity | 2.69 (0.7-10.6) | > 0.1 | 1.17 (0.3-4.3) | > 0.1 | 6.03 (1.4-25.8) | 0.02 |
| Specialty | ||||||
| Internal medicine/pediatrics | Ref | — | Ref | — | Ref | — |
| Pediatrics | 0.32 (0.1-0.9) | 0.04 | 1.02 (0.4-2.9) | > 0.1 | 0.44 (0.1-1.4) | > 0.1 |
- Significant P values (P < 0.05) are bolded. OR indicates odds ratio derived from ordered (i.e., ordinal) logistic regression model, which assumes proportional odds between each level of the dependent variable. “Ref” indicates a common reference point to which the other categories were compared for interpretation of results.
Discussion
In this study, we elucidate the relationships between obesity training and physician perceptions, clinical patterns, and confidence when managing pediatric obesity. We assessed physician knowledge regarding MBS and major contributors to obesity and explored whether their perspectives were significantly influenced by their specialty training and current or prior weight status.
Our study found that pediatrics physicians with more obesity training felt more successful and confident when treating obesity and discussing treatment options that include nutrition, physical activity, and MBS. While a majority of physicians perceive a lack of effective tools and educational obesity information for pediatric patients ((22-26)), our study found that having more obesity-specific training made physicians more likely to disagree with such statements and less likely to expect monetary compensation for obesity management. These findings suggest that more obesity training could decrease knowledge gaps and increase pediatrics physicians’ confidence in treating pediatric overweight and obesity.
As the prevalence of pediatric obesity rises, the need for studies that appropriately address its management has become critical ((8, 27)). Similar to results from adult studies ((6, 28)), our study found that pediatrics physicians who graduated more recently were more likely to have received increased obesity-related training. Our data also suggest that primary care physicians focused on the pediatric population perceive that a lack of adequate referral services for diet, physical activity, and weight management is a barrier to achieving successful obesity management. These perceptions align with the current literature that has reported limited geographic availability of obesity-trained pediatrics physicians, as well as other concerns that come both with the lack of affiliation with teaching hospitals and with serving low-income families ((7, 29-31)). A study focused on the American Board of Obesity Medicine found that only 38% of physicians who held this certification had a pediatric background ((7)), which is insufficient to meet the high demand of pediatric patients with overweight and obesity.
Moreover, it is alarming that only 6% of all respondents felt generally successful in treating pediatric patients with obesity. Overall confidence levels were low, especially with motivational interviewing, discussing weight loss medications, and potential eligibility for MBS. Furthermore, less than 60% of the questions regarding MBS were answered correctly. Motivational interviewing has a positive impact in clinical practice and helps to reduce BMI in children ((18, 32)). There are medications available to treat childhood obesity that should be considered in concurrence with other multidisciplinary approaches ((24-26)). Additionally, MBS is an effective and safe tool to achieve sustained weight loss in adolescents ((26, 33, 34)). Given that obesity-related morbidities are more intractable and persistent after adolescence ((35)), a greater depth of education on how to utilize these tools to treat it in the pediatric population is needed.
In our study, we also adjusted our data to compare between two common specialties that manage pediatric obesity in the primary care setting: pediatrics and internal medicine/pediatrics. We found that pediatrics physicians were more likely to have obesity-related training than internal medicine/pediatrics physicians. It has been reported that pediatricians are better than other physicians from other specialties at recording a diagnosis of obesity in the medical record ((17)), which held true in our study. However, our findings also suggest that pediatricians may more likely perceive that pediatric patients are not interested in improving their weight status and that it is too difficult for them to change their behaviors compared with the internal medicine/pediatrics specialty. Though it is possible that this latter perception may vary based on the age of the patient, it is necessary to be aware that a physician’s personal biases, assumptions, and expectations might influence pediatric weight management ((12, 36)). Awareness of weight stigma is essential at all stages of training to encourage the best, unbiased clinical practices.
Furthermore, pediatricians as a group seem less confident with managing weight loss medications and the stages of MBS and seem less likely to have accurate MBS knowledge. This might explain the low rates of MBS utilization in the pediatric population ((33, 37)), though other factors such as the lack of insurance coverage, scarcity of qualified multidisciplinary centers ((34)), and racial and socioeconomic disparities ((38)) may also influence its low use. It is crucial for all primary care physicians to comprehend and be updated regarding the benefits and availability of such tools for obesity management in order to improve their confidence, especially in such a vulnerable population. We also found that pediatricians seem less likely to regularly engage in physical activity. Whether or not disclosing their personal habits influences their clinical practice and the behaviors of their patients remains unknown.
Obesity during childhood not only increases comorbidities and leads to higher mortality rates in adulthood but it is often associated with weight stigma that increases the likelihood of depression, anxiety, eating disorders, decreased physical activity, and avoidance of medical services ((39, 40)). It was encouraging that our findings placed “lack of willpower” at the lower end of the major contributors of obesity, suggesting that physicians may be aware that obesity is the result of multiple factors. It is critical to recognize obesity as a chronic disease and prioritize education of pediatrics physicians in managing obesity to provide an unbiased medical evaluation. Further efforts are necessary to address the gaps in obesity training received by residents and fellows who provide care throughout childhood.
To our knowledge, this is the first study to formally assess pediatrics physicians’ perceptions, practice patterns, confidence, and knowledge while adjusting for obesity-related training hours in a not-for-profit large academic system. This study revealed objective areas that may contribute to the educational gaps that currently exist in the treatment of pediatric obesity. One primary limitation in this study is its modest sample size. Also, it was not possible to estimate the percentage of pediatric patients that internal medicine/pediatrics physicians see in their practice or the age range of pediatric patients under the care of the surveyed primary care physicians in both specialty groups. It would be shortsighted to generalize these results to all pediatrics primary care providers given that our respondents have access to a large, tertiary, multidisciplinary weight center and their perceptions did not take into consideration age groups. Moreover, our survey did not provide a definition for “success,” which could be interpreted in multiple ways such as weight loss, psychological benefit, waist circumference changes, improved cardiometabolic function outside of weight loss, etc.; this could explain why such a low percentage of clinicians reported feeling successful in treating pediatric patients for obesity. However, our survey ascertained several interesting findings, and although some of the responses might not be fully representative of those of pediatrics primary care physicians throughout the US, the large amount of ground covered in our manuscript presents the opportunity for further research on pediatric obesity caregivers in order to elaborate on these findings.
In conclusion, pediatrics physicians have some confidence in counseling and discussing obesity-related issues with children and their families; however, there exists an overall lack of confidence in managing pediatric obesity supported by the low success rates that were self-reported in our study. This appears more marked among pediatricians, who also seem to hold more negative beliefs about the prospect of changing a child’s behavior given a lack of interest in improving his or her weight status, which could lead to intrinsically biased management. Nevertheless, increased obesity training seems to improve physician confidence and leads to better familiarity with guidelines and management options. Formal obesity training should be prioritized during residency and beyond so that physicians who care for pediatric patients are better equipped and able to offer unbiased and effective obesity-specific care.
Funding agencies
This work was funded by the Physician/Scientist Development Award (PSDA) granted by the Executive Committee on Research (ECOR) at Massachusetts General Hospital (MGH) (to FCS); National Institutes of Health (NIH) P30 DK040561 (to FCS) and L30 DK118710 (to FCS); and the MGH Healthcare Transformation Lab Healthcare Innovation Research Fellowship (to NPP). The content is solely the responsibility of the authors and does not necessarily represent the official views of the MGH Executive Committee on Research or the NIH.
Disclosure
The authors declared no conflict of interest.
Author contributions
KJCR and NPP contributed equally to all aspects of this manuscript and should be considered co-first authors. KSC and AYS contributed significantly to data interpretation and drafting of the manuscript. FCS contributed to all aspects of idea conceptualization, study design, data analysis and interpretation, and drafting of the final manuscript.
