Family based behavioral treatment in adolescents suffering from obesity: evolution through adulthood

This study showed a favorable long-term evolution of the BMI z-score for most of the young adults suffering from obesity who participated in a FBBT during their adolescence. These results might suggest the long term benefits of FBBT at this age, which is recognized to be effective in children, but with less evidence in adolescents [9]. However, effectiveness of the treatment can only be assumed, as there is no control group. Nevertheless, earlier findings showed that adolescents suffering from obesity are at risk for weight gain in adulthood with little chance of spontaneous favorable resolution [2, 16].

This study has identified two behaviors associated with a favorable evolution on the long term: to have a daily physical activity and to monitor the weight regularly. Other factors such as age, gender, or other socio-demographic determinants, as well as initial BMI, treatment response at the end of the one-year program or follow-up duration after the program were not associated with a favorable or unfavorable evolution. Dietary changes were not associated with a better outcome probably because the question was too vague, without specifying the quantity or quality of these changes leaving the participants to roughly estimate them.

The impacts of regular physical activity on weight for both adolescents and adults are well studied. In our study, being active every day predicted the best evolution of BMI-z-score as opposed to once a week or less, in line with the guidelines on pediatric obesity management recommending 30 min of moderate-to-vigorous physical activity per day, 5 days per week for adolescents [7, 9, 17, 18]. However, this study did not measure the intensity of the activity which would have been interesting to evaluate if our population met the recommended guidelines.

Monitoring weight regularly was associated with a favorable evolution on the long-term. This is consistent with earlier findings identifying self-weighing as a useful weight control strategy [19‐22]. However, conclusions differed on the psychological impact of this behavior. Zheng and al. reported no association between self-weighing and negative psychological outcome, while Pacanowski et al. had a more balanced conclusion, with negative psychological outcomes particularly in younger participants, and an association with the occurrence of disordered eating. Studies focusing on adolescents supported these conclusions and recommended a cautious use of regular weighing in that population that should be supported by a professional and in a particular setting (weight loss program versus normal weight adolescents) [19, 23‐26]. In our study, we did not evaluate psychological impacts of self-weighing and thus cannot draw any conclusions. However, our results supported regular self-weighing as a part of a weight management strategy for young adults who were included in an FBBT when they were younger [24, 27].

Interestingly, participants with a higher BMI were the least prone to self-weighing. This may be explained by the emergence of negative emotions such as body shame and the feeling of not being able to lose weight after the weighing, reinforcing the need for these patients to have some support by health professional for their weight monitoring [17‐20].

Regarding medical weight-management follow up during the transition to adulthood, this study showed that about half of our studied population has discontinued their medical follow-up, whereas most of them still had a BMI above 30. This result is concerning and shows the difficulties for long-term weight management at this age, even though they have a chronic disease leading to numerous somatic complications. In this study, a quarter of them explained their drop out by a lack of motivation or dissatisfaction with the program, which can be interpreted as a feeling of helplessness/lassitude regarding weight loss efforts. Transition of care is known to be problematic in this population [11]. It is argued by the fact that most of the time, adolescents do not stay in treatment long enough to transition, which is also the case in our study, with a mean follow up of 2.7 years. One of the reasons might be that they or their parents do not identify obesity as a health problem, but more as a behavioral issue or a lack of motivation that does not need to be supported by a health professional. Another reason might be that they have not succeeded in making lifestyle changes to lose weight, despite several years of weight management, leading to a feeling of ineffectiveness and lassitude, resulting in a demotivation for treatment both for the youth and their parents. Asking for assistance may also be viewed as difficult and embarrassing. Cost for the patient could be a reason for a drop out after 18, as the Swiss health insurance system can require substantial contributions to medical cost after majority, whereas before 18, costs are covered by the insurance. However, few of them reported this limitation in our questionnaire. Surprisingly, half of our participants would have been interested in restarting a follow-up if offered at the long-term visit. These results indicated that health professionals should be more proactive with these patients, offering them an appointment at regular intervals, even if they are often considered unnecessary by healthcare providers and/or parents. Ensuring that the patient has made the transition to adult care could also be a way of restarting an interrupted follow-up. Clinical guidance on transition care in adolescents with obesity should be developed as proposed by Shrewsbury and al in order to be more efficient with those patients [11].

This study has some limitations. First, the sample was small, with a high number of adolescents who did the program but didn’t participate in the study, even if the main reason for not participating was that they were unreachable. Therefore, a recruitment bias cannot be excluded. Those not included were analyzed at baseline (T0) as well as their evolution at the end of their participation in the group therapy. At baseline, the mean age was similar to the study group (14.1 ± 1.3 vs. 14.3 ± 1.8 years). They were a somewhat more overweighted at baseline (BMI 31.61 ± 4.8 vs.30.85 ± 4.4; BMI-z score 2.67 ± 0.64 vs. 2.43 ± 0.69). The evolution of the BMI z-score between T0 and T1 was favorable in 92.8% of them versus 94.7% in the study group. Only few adolescents increased their BMI z-score as in the study group (7.3% vs. 5.3). The main difference was the gender distribution, with substantially more boys (60% vs. 34%) in the group who didn’t participate in the study.

Among those who could be reached, 84% did participate, suggesting that the study was well perceived. Furthermore, the adolescents did not know at the end of the program that they would be contacted for a long-term clinical evaluation, and thus we were not able to refer to drop-out rates per se. Finally, drop-out rate is a common problem in adolescent obesity studies, which probably explains the few studies on the long-term evolution (more than 2 years) in this population emphasizing our results.

Another limitation is the use of BMI z-score, reported to be a poor indicator of adiposity among youths in some recent publications, particularly those with severe obesity [9]. However, no standard definition of weight loss success exists for the pediatric populations [7, 9].

Finally, the changes in the intensity of physical activity and diet habits have not been fully addressed in this study, preventing us from making any robust conclusions. However, we would have probably lost a certain number of participants if they had been asked to wear a device to measure their amount of physical activity and intensity. For diet habits, it is known that even with food recall questionnaire, this issue is difficult to address.

Despite these limitations, this study is one of the few on the long-term evolution of adolescents suffering from obesity.