In order to explore longitudinal changes, we assessed QOL, depressive symptoms, and BMI z-score indicators for overweight youth over time. Overall, across three visits (V1-V2-V3), our results indicated youth’s BMI z-score decreased slightly, their QOL significantly increased, and teen depression level improved. Likewise, caregivers’ perception of their youth’s QOL increased across three visits.
Interestingly, youth from V1 to V2 and V1 toV3 had significant improvements in their QOL, despite their BMI z-score and the majority of our sample being either obese or severely obese. This is especially important to consider, given that other researchers have reported that quality of life is inversely related to weight; as a youth’s weight increased, his/her QOL decreased, suggesting that the most overweight youth have the most significantly impaired QOL . In a cross-sectional study, Williams et al.  compared youth of different BMI categories (normal, overweight, obese). In that research, obese youth were found to have a lower QOL than their normal and overweight peers. However, previous longitudinal research focusing on differences between youth in different severity categories of obesity (i.e., obese vs. severely obese) has not been done; our results indicate that even those who are most obese (severely obese or ≥ 99th percentile) had positive results in QOL with treatment, even with only modest improvements in BMI and no change in weight category.
After adjusting for time, improvement in teen depression (PHQ9A score) was strongly related to QOL improvement. This result emphasizes potential value in assessing for depression (in the teen and caregiver) in tandem with a QOL inventory. While past researchers have assessed for youth or caregiver depression or for QOL , none published to date have assessed for youth depression longitudinally in tandem with a QOL inventory. In light of our observed significant association between QOL and PHQ9A, our results suggest that with treatment using an integrated model emphasizing both physical and psychosocial factors, in obese youth both QOL and depression can improve even when BMI change is modest. In childhood obesity treatment, BMI improvement is ultimately the goal; however after youth make improvements in QOL and depression they may be more able and confident to adopt and work toward goals that result in weight-loss.
This study has several limitations. Previously researchers have demonstrated differences in treatment seeking vs. non-treatment seeking youth in QOL and on other measures of psychosocial health [13, 15]. Thus, our results are best applied to those youth and families who are seeking treatment. Secondly, our sample was collected from one site in one specific location. Based on our population, the generalizabilty of our findings may only apply to those youth similar to our sample who are English speaking, and obese or severely obese, rather than overweight.
Although it was not our purpose, we looked for predictors of attending (or not attending) a follow-up and did not find any significant predictor based on race, BMI category, weight loss, gender, or age. Additionally, we only assessed depression in teens, not in youth under 12 years of age. We believe that our convenience sample, which was followed for up to two years, accurately reflects childhood obesity treatment attrition, whereby national attrition rates are estimated to be between 27% and 73% and half of participating families in treatment programs drop out of treatment [31, 32].
We also only analyzed youth-caregiver dyads, which may neglect other important family members that are important in the youth’s daily life (e.g., grandparents, teachers, siblings, etc.). Finally, we only report data for up to three visits with relatively long time periods between each visit, in part due to the rural community, and the frequency (typically monthly) at which visits can be offered at the clinic. Thus different results might be seen in settings that can see patients more frequently (such as stage three treatment facilities that are recommended to see patients weekly for “intensive” treatment). Given that our results suggest improvement with subsequent visits, it is important for childhood obesity treatment centers, including our own, to determine ways our healthcare providers, administration, clinic procedures and policies and financial issues can be addressed to ensure the availability of consistent follow-up appointments.
Although there are several important limitations described above, our research does offer new insight into longitudinal changes in QOL and PHQ9A for obese and severely obese youth in treatment. First, we chose to track both physical and psychosocial outcomes over time. Many longitudinal efforts only measure weight-based variables (BMI, BMI z-score, or nutrition/physical activity behaviors), and neglect psychosocial variables. When researchers do include psychosocial variables, they often only use one variable, which is most commonly depression. Secondly, published longitudinal research has typically been conducted in structured large-scale settings which allow for the controlling of appointment scheduling at specific time intervals either monthly as indicated in the expert recommendations for Stages 3 and 4 treatment  or based on specific behavioral treatment protocols (i.e., 3 months, 6 months, 12 months, etc.). Our study could not control for these factors and instead represents a real-world rural outpatient clinical retention effort for overweight youth and their families. Even without being able to schedule appoints on a routine basis, we found that QOL, PHQ9A and BMI z-score were positively affected over time for youth in our sample. This brings up an important point about clinical versus statistical significance. Clinically, for youth who have been gaining weight over time, decreases in weight gain, leveling of current weight, and subsequent weight loss are all considered successes. The results of our study show that those participants who continued through a second and third visit had leveling of BMI z-score or slight declines, which is clinically significant, even though it only just had statistically significance.
In order to address the different ways obesity affects both physical and psychosocial variables for youth and their family, these findings suggest benefit to using brief validated measures, such as the PedsQL to explore youth and caregiver perceptions of youth’s QOL and the PHQ9A to assess depression in teens and caregivers. In addition, there may be additional benefit to clinicians and researchers tracking youth longitudinally throughout treatment to investigate the relationship between youth’s QOL and those who level or decline in BMI verses those who increase or gain; specifically, to help determine if there is a certain QOL threshold that youth may reach before they begin to show signs of weight loss. Addressing both physical and psychosocial variables within medical treatment, as now recommended in care recommendations, is one way to incorporate these factors into the complex care of obese youth and their families and hopefully enhance overall success of treatment.