Individual and family characteristics associated with health indicators at entry into multidisciplinary pediatric weight management: findings from the CANadian Pediatric Weight management Registry (CANPWR)

Objectives (1) To explore individual and family characteristics related to anthropometric and cardiometabolic health indicators and (2) examine whether characteristics that correlate with cardiometabolic health indicators differ across severity of obesity at time of entry to Canadian pediatric weight management clinics. Methods We conducted a cross-sectional analysis of 2-17 year olds with overweight or obesity who registered in the CANadian Pediatric Weight Management Registry (CANPWR) between May 2013 and October 2017 prior to their first clinic visit. Individual modifiable health behaviors included dietary intake, physical activity, screen time, and sleep. Family characteristics included parental BMI, family medical history, socioeconomic status and family structure. Linear mixed effects stepwise regression analysis was performed to determine which characteristics were related to each health indicator: BMI z-score; waist circumference; waist to height ratio; blood pressure; glycemia; HDL cholesterol; non-HDL cholesterol; triglycerides. Results This study included 1296 children (mean age +/- standard deviation: 12.1 +/- 3.5 years; BMI z-score: 3.55 +/- 1.29; 95.3% with obesity). Hours spent sleeping (estimated beta = -0.10; 95% CI [-0.15, -0.05], p = 0.0001), hours per week of organized physical activity (estimated beta = -0.32; 95% CI [-0.53, -0.11], p = 0.0026), daily sugared drink intake (estimated beta = 0.06; 95% CI [0.01, 0.10], p = 0.0136) and maternal BMI (estimated beta = 0.03; 95% CI [0.02, 0.04], p < 0.0001) were associated with BMI z-score (adj. R-2 = 0.2084), independent of other individual and family characteristics. Physical activity, total sugared drink intake and sleep duration were associated with glycemia and non-HDL cholesterol, independent of child BMI z-score. However, irrespective of obesity severity, little of the variance (0.86-11.1%) in cardiometabolic health indicators was explained by individual modifiable health behaviors. Conclusions Physical activity, total sugared drink intake and hours spent sleeping were related to anthropometric and some cardiometabolic health indicators in children entering pediatric weight management programs. This highlights the importance of these modifiable health behaviors on multiple health indicators in children with obesity.