61 - Association of Neighborhood Conditions with Hypertension in Children and Adolescents

Poster Number: 61
Publication Number: 61.1905

Background: Limited evidence has shown an association between neighborhood deprivation and hypertension in children using billing data. However, the mechanism of this association is unknown.

Objective: To evaluate the association between neighborhood conditions and hypertension and to determine the extent to which body mass index (BMI) mediates this association.

Design/Methods: This cohort study used electronic health record data from 47 primary care pediatric clinics in Connecticut. The study population included children aged 3-18 years that had ≥3 visits with blood pressure (BP) measurements recorded from 2014-2023. We extracted home zip code on the date of first BP measurement to calculate the Child Opportunity Index (COI), a measure of neighborhood conditions that impact health across three domains (education, health/environment, and socioeconomic). We categorized COI into quintiles from very low ( < 20th percentile) to very high opportunity neighborhoods (≥80th percentile). The primary outcome was hypertension defined as ≥3 visits with hypertensive BP during the study period according to the 2017 American Academy of Pediatrics guideline. We tested the association between COI and hypertension using multivariable logistic regression and computed the average marginal effect of COI adjusted for age and sex. We decomposed the total association between COI and hypertension into the indirect association mediated through BMI and the remaining direct association.

Results: Of 48,694 children (median 7.2 years [IQR, 4.0-11.6], 50.5% female), 32.0% lived in neighborhoods with very low COI and 27.4% very high COI (Table 1). The prevalence of hypertension was highest (5.3%) in children with very low COI and lowest (3.5%) in very high COI (P < 0.001 for trend; average marginal effect for very low vs very high COI: 2.0% [95% CI, 1.6-2.5%]) (Figure 1). Lower COI was associated with higher prevalence of obesity (P < 0.001 for trend). For children from high and moderate COI neighborhoods compared with the reference of very high COI, most of the association between COI and hypertension was mediated by BMI (75-83%), while the direct association was not significant (Table 2). For children from low and very low COI neighborhoods, BMI only partially mediated the association between COI and hypertension (39-49%), and the direct association remained significant.

Conclusion(s): Residing in lower-opportunity neighborhoods was associated with hypertension. This relationship may be partially mediated by BMI. Future research may explore other mechanisms through which neighborhoods affect BP in children.