Successful lifestyle changes for weight reduction are heavily dependent on recognizing the importance of societal and cultural factors. Patients 13–19 years of age with a BMI ≥95th percentile are eligible for our multidisciplinary adolescent weight loss clinic. A behavioral questionnaire was administered at the initial visit. Patients were seen every 4–6 weeks. Bivariate analysis was used to identify sociodemographic factors associated with differences in weight loss. Overall, receiving reduced cost meals was associated with a lower likelihood of losing weight (kg) (
Childhood obesity is a growing problem in the 21st century. According to the 2015-2016 National Health and Nutrition Examination Survey report, 18.5% of the youth of US aged 2 to 19 were obese [
Apart from the significant comorbidities associated with childhood obesity, psychological, social, and behavioral consequences are also prevalent. Poor body image, low self-esteem, social isolation, discrimination, depression, and reduced quality of life are frequently observed in children with obesity [
Adolescent minorities in the US suffer obesity at a disproportionately higher rate than their Caucasian counterparts [
Efforts to address adolescent obesity over the last decade have resulted in an increased number of multidisciplinary and multicomponent interventions [
The purpose of this study was to identify which sociodemographic and behavioral factors were associated with weight loss among a diverse group of adolescents. To our knowledge, this is the first study assessing these specific categories—effort, goal setting, technology utilization, and self-perception—in a multidisciplinary adolescent weight loss clinic.
This was a post hoc analysis of data from the clinical registry of patients participating in the adolescent weight loss clinic at the University of Texas Medical Branch (UTMB Health). All patients were seen between 2016 and 2019. Patients were referred to the clinic by their primary care providers (PCPs) or other specialty providers and were instructed to follow-up every 4–6 weeks with the weight loss team composed of a pediatric surgeon, a pediatric gastroenterologist, a dietician, and a fitness instructor. Patients who had not presented within 10 weeks of their initial visit were considered lost to follow-up for visit 2, and patients who had not presented within 18 weeks of their initial visit were considered lost to follow-up for visit 3. Patients were screened for depression using the PHQ9 survey and referrals were made to a pediatric psychologist or psychiatrist when indicated. Anthropomorphic data was collected at each visit and laboratory data was collected when clinically indicated. Questionnaires were given separately to patients and caretakers on the first visit. The questionnaire was developed by faculty within the Departments of Preventive Medicine and Community Health, Nutrition and Metabolism, and Pediatrics and included past weight loss attempts, goal setting, dietary habits, activity level, family environment, and socioeconomic data. In the development of our questionnaire, several validated questionnaires were very influential, such as the body shape questionnaire (BSQ), body image assessment for obesity (BIA-O), multidimensional body-self relations questionnaire (MBSRQ), obesity related well-being (ORWELL-97), international physical activity questionnaire (IPAQ), and the diet satisfaction questionnaire (DSat-45). All data were deidentified and in accordance with our local IRB regulations for management of a clinical registry.
All patients enrolled in the adolescent weight loss clinic were evaluated. Those eligible for evaluation in the adolescent weight loss clinic were patients aged 13–18 with sex-specific BMI-for-age ≥95th percentile.
Primary outcomes measured were weight (kg) and BMI change (kg/m2). Anthropomorphic data and clinical data collected included height, weight, self-reported race, sex, age, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and presence of comorbidities. Comorbidities included HTN, gastroesophageal reflux disease (GERD), obstructive sleep apnea (OSA), DM2, and asthma. Height (cm) and weight (kg) were measured on the same scale and with the same stadiometer at each visit. Patients were measured in street clothes with no jackets, no shoes, and empty pockets.
A paired
A total of 189 children had an initial visit and were administered a survey questionnaire. Women and men were divided evenly at the initial visit, with 51% and 49%, respectively (Table
Cohort demographic composition.
Demographic | Visit 1 | Visit 2 | Visit 2 | Visit 3 | Visit 3 |
---|---|---|---|---|---|
|
|
OR (95% CI) |
|
OR (95% CI) | |
Total | 189 | 109 | 68 | ||
Sex | |||||
Female (ref.) | 96 (51%) | 55 (50%) | 37 (54%) | ||
Male | 93 (49%) | 54 (50%) | 1.22 (0.70–2.13) | 31 (46%) | 1.25 (0.9–2.28) |
Race | |||||
White (ref.) | 58 (31%) | 37 (34%) | 21 (31%) | ||
Black | 40 (21%) | 20 (18%) | 0.57 (0.25–1.29) | 9 (13%) | 0.51 (0.20–1.28) |
Hispanic | 89 (47%) | 51 (47%) | 0.76 (0.39–1.50) | 38 (56%) | 1.31 (0.66–2.59) |
Weight loss (kg) |
75 (69%) | 42 (62%) | |||
Reduced BMI (kg/m2) |
81 (74%) | 51 (75%) |
Cohort anthropomorphic composition.
Initial visit | Visit 2 |
Visit 3 |
|
---|---|---|---|
Weight (kg ± SD) | 107.9 ± 25.9 | 110.4 ± 27.5 | 107.9 ± 29.8 |
Weight change (kg ± SD) | −0.91 ± 2.7 | −1.23 ± 4.1 | |
BMI (kg/m2 ± SD) | 39.2 ± 8.4 | 39.5 ± 8.6 | 38.9 ± 9.4 |
BMI change (kg/m2 ± SD) | −0.49 ± 0.91 | −0.81 ± 1.4 | |
Time from initial visit (weeks ± SD) | 8.2 ± 5.3 | 15.5 ± 7.2 |
A total of 109 patients had a second visit. The second visit occurred at a mean of 8.2 ± 5.3 weeks from the initial visit. Assuming a 10-week visit window from the initial visit, this represents a 68.9% retention rate. The average weight at visit 2 was 110.4 ± 27.5 kg with a BMI of 39.5 ± 8.6 kg/m2. The majority of patients at visit 2 lost weight (kg) (69%) and had a reduced BMI (74%). The average weight change at visit 2 was -0.91 ± 2.7 kg (
A total of 68 patients had a third visit. The third visit occurred at a mean of 15.5 ± 7.2 weeks from the initial visit. Assuming an 18-week visit window from the initial visit, this represents a 51.5% retention rate. The average weight at visit 3 was 107.9 ± 29.8 kg with a BMI of 38.9 ± 9.4 kg/m2. The majority of patients at visit 3 lost weight (kg) (62%) and had a reduced BMI (75%). The average weight change at visit 3 was −1.23 ± 4.1 kg (
Comorbidities in the patient population were rare (Table
Presence of comorbidities.
|
OR (95% CI) | |
---|---|---|
HTN | 9 (5%) | |
White (ref.) | 4 (8%) | |
Black | 2 (6%) | 0.82 (0.14–4.7) |
Hispanic | 3 (4%) | 0.47 (0.10–2.2) |
Female (ref.) | 3 (4%) | |
Male | 6 (7%) | 2.1 (0.51–8.8) |
OSA | ||
White (ref.) | 9 (17%) | |
Black | 10 (31%) | 2.22 (0.79–6.3) |
Hispanic | 6 (8%) | 0.40 (0.13–1.2) |
Female (ref.) | 13 (15%) | |
Male | 12 (15%) | 1.1 (0.45–2.5) |
Asthma | ||
White (ref.) | 8 (15%) | |
Black | 9 (28%) | 2.2 (0.75–6.5) |
Hispanic | 9 (11%) | 0.70 (0.25–2.0) |
Female (ref.) | 13 (15%) | |
Male | 13 (16%) | 1.1 (0.45–2.47) |
GERD | ||
White (ref.) | 1 (2%) | |
Black | 0 (0%) | 0.54 (0.021–13.6) |
Hispanic | 3 (4%) | 2.0 (0.20–19.8) |
Female | 2 (2%) | |
Male | 2 (2%) | 0.98 (0.13–7.1) |
DM2 | ||
White (ref.) | 2 (4%) | |
Black | 4 (13%) | 3.64 (0.63–21.2) |
Hispanic | 3 (4%) | 0.98 (0.16–6.1) |
Female (ref.) | 4 (5%) | |
Male | 5 (6%) | 0.77 (0.20–3.0) |
HTN—hypertension, OSA—obstructive sleep apnea, GERD—gastroesophageal reflux disease, and DM2—diabetes mellitus type 2.
When examining survey responses, statistically significant differences were noted between races and different ethnicities. Black patients were significantly more likely to report receiving free or reduced cost meals at school (
Racial, ethnic, and sex differences in survey responses. Results presented as odds ratio ±95% confidence interval. PE—physical education, DM2—diabetes mellitus type 2, TV—television, and PCP—primary care provider.
Hispanic patients were significantly more likely to report receiving free or reduced cost meals at school (
Female patients were significantly more likely to report feeling unhappy when looking in the mirror (
Weight loss (kg) was significantly less likely if patients or caretakers reported receiving free or reduced cost meals at school (
Survey responses predictive of weight and BMI reduction. Results presented as odds ratio ±95% confidence interval. PE—physical exercise, DM2—diabetes mellitus type 2, TV—television, and PCP—primary care provider.
When stratified by race, additional predictors were found. White patients were significantly less likely to lose weight (kg) if they received free or reduced cost meals at school (
Racial and ethnic survey responses predictive of weight and BMI reduction. Results presented as odds ratio ±95% confidence interval. PE—physical exercise, DM2—diabetes mellitus type 2, TV—television, and PCP—primary care provider.
Black patients were more likely to lose weight (kg) if patients reported eating 2-3 servings of fruits or vegetables per day (
Hispanic patients were less likely to lose weight (kg) if patients or caretakers reported that patients had been tested for obstructive sleep apnea (
Female patients were less likely to have lost weight if patients or caretakers reported receiving free or reduce d cost meals at school (
Sex survey responses predictive of weight and BMI reduction. Results presented as odds ratio ±95% confidence interval. PE—physical exercise, DM2—diabetes mellitus type 2, TV—television, and PCP—primary care provider.
Male patients were less likely to lose weight if patients or caretakers reported receiving free or reduced cost meals at school (
Our study successfully identified several statistically significant predictors of weight loss in adolescents. As expected, responses to our initial visit questionnaire were significantly different between demographic groups, indicating that the barriers to weight loss are varied and unique to each group. In addition, these results highlight areas for improvement in tailoring behavioral, dietary, and activity changes to specific demographic groups of adolescents. Example areas for improvement include reformation of the National Student Lunch Program (NSLP), emphasizing healthy body images for male adolescents, providing culturally competent healthy recipes for minority families, and an increased emphasis on proper sleep hygiene. Critically examining the results of our clinic after implementing these changes will provide an opportunity for improvement on a continuous basis.
Receiving free or reduced cost school lunches was associated with significantly lower odds of weight and BMI reduction. This association was present in both males and females. Racial and ethnic stratification showed that Hispanic and Black patients were significantly more likely to report receiving free or reduced cost school lunches than White patients; but receiving free or reduced cost school lunches was only significantly associated with reduced odds of weight and BMI reduction in White and Hispanic patients. Notably, surrogate indicators of socioeconomic status, such as family income, amount spent on groceries weekly, and amount spent on restaurants weekly, were not found to be significant predictors of weight or BMI reduction. The NSLP has long been known to have no protective effects for childhood obesity, even after addressing the selection bias caused by differences in NSLP participants and nonparticipants [
Females were significantly more likely than males to report feeling unhappy when looking in the mirror or comparing their looks to those they see on television or in magazines. When stratified by sex, these feelings were associated with a significantly higher likelihood of having a reduced BMI. In males, however, feeling unhappy with their looks was associated with a significantly reduced likelihood of reduced BMI. This evidence suggests that body image issues may represent a larger roadblock to weight loss in adolescent males than females. Body image issues may also be underrecognized in the adolescent male community. In adults, men are as likely as women to seek care for body dysmorphic disorder, but they are less likely to receive treatment [
Caretaker-reported ability to afford healthy food was a significant predictor of weight loss among White patients; however, in Black patients, it was associated with a significantly reduced likelihood of weight loss. Interestingly, if Black patients reported eating several servings of fruits and vegetables per day, they had increased odds of weight loss. The ability to afford healthy food in our population may not be a deciding factor for weight loss in Black patients, and weight loss may instead be dependent on cultural factors in the Black American population. Studies have demonstrated that, in general, Black Americans are more accepting of larger body sizes [
Black patients and Hispanic patients were less likely to lose weight if caretakers reported that they had been tested for OSA. Similarly, male patients were less likely to lose weight or have a reduced BMI if patients or caretakers reported testing for OSA. Black patients and male patients were also less likely to lose weight if they reported being too tired for weight loss. These survey questions are interlocked since patients with obesity who reported being tired are those who are most likely referred for sleep apnea testing [
This study had several limitations. The attrition rate for our adolescent weight loss clinic, although similar to that reported in other weight loss clinics [
The deidentified survey and clinical data used to support the findings of this study are available from the corresponding author upon request.
An earlier version of this MS was presented as an abstract presentation in Obesity week 2019, Vegas, 2019.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Dr. Bowen-Jallow was supported by a research career development award (K12HD052023: Building Interdisciplinary Research Careers in Women’s Health Program-BIRCWH; PI: A. B. Berenson) from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and T32DK007639 (to B. D. Hughes) from the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.