Positive Outliers: A Mixed Methods Study of Resiliency to Childhood Obesity in High-Risk Neighborhoods.

OBJECTIVE: Despite the high prevalence of obesity and the clustering of risk by neighborhood, few studies have examined characteristics which promote healthy child weight in neighborhoods with high obesity risk. We aimed to identify protective factors for children living in neighborhoods with high obesity risk. METHODS: We identified neighborhoods with high obesity risk using geolocated electronic health record data with measured body mass index (BMI) from well-child visits (2012-2017). We then recruited caregivers with children aged 5 to 13 years who lived in census tracts with mean child BMI percentile ≥72 (February 2020-August 2021). We used sequential mixed methods (quantitative surveys, qualitative interviews) to compare individual, interpersonal, and perceived neighborhood factors among families with children at a healthy weight (positive outliers [PO]) versus families with ≥1 child with overweight or obesity (controls). Regression models and comparative qualitative analysis were used to identify protective characteristics. RESULTS: Seventy-three caregivers participated in the quantitative phase (41% PO; 34% preferred Spanish) and twenty in the qualitative phase (50% PO; 50% preferred Spanish). The frequency of healthy caregiver behaviors was associated with being a PO (Family Health Behavior Scale Parent Score adjusted β 3.67; 95% CI 0.52-6.81 and qualitative data). Protective factors also included caregivers' ability to minimize the negative health influences of family members and adhere to family routines. CONCLUSIONS: There were few differences between PO and control families. Support for caregiver healthy habits and adherence to healthy family routines emerged as opportunities for childhood obesity prevention in neighborhoods with high obesity risk.

Duke Scholars

Author Michelle Joette White Pediatrics, Hospital Medicine

Author Naomi Nichele Duke Pediatrics, General Pediatrics and Adolescent Health

Author Cynthia Lea Green Biostatistics & Bioinformatics, Division of Biostatistics

Author Tracy Truong