Effect Measure Modification by Birth Weight on the Association Between Overweight or Obesity and Hypertension in Children and Adolescents

Abstract

This cross-sectional study examines the prevalence of hypertension among children and adolescents with low, very low, normal, or high birth weight.

Children and adolescents with low birth weight (LBW) have a higher risk for hypertension and cardiovascular disease later in life.^1,2 Due to their lower number of nephrons and associated glomerular hyperfiltration, this population may be more sensitive to the adverse association of obesity with blood pressure (BP).¹ We aimed to determine the extent to which the association between overweight or obesity and hypertension in US children and adolescents is modified by birth weight (BW).

Methods

Data were obtained from the National Health and Nutrition Examination Survey (NHANES) for 2-year cycles from 1999 to 2000 and 2017 to 2018. We included participants aged 8 to 15 years because BP measurement started at age 8 years in NHANES, and BW was reported by parent or proxy recall until age 15 years. The Yale Institutional Review Board deemed this cross-sectional study exempt from review. Parents and participants 12 years or older provided written consent; children younger than 12 years provided assent. We followed the STROBE reporting guideline.

Overweight or obesity was defined as body mass index (BMI) at or above the 85th percentile.³ Birth weight was categorized as very low BW (VLBW; <1500 g), LBW (1500-2499 g), normal BW (2500-3999 g), and high BW (HBW; ≥4000 g). Three consecutive BP measurements were obtained per protocol. The mean of all BP measurements determined hypertension status based on American Academy of Pediatrics guidelines.⁴

Hypertension prevalence was compared between overweight or obesity and BW categories using the χ² test. We used modified Poisson regression⁵ to estimate prevalence ratio (PR) for the association of overweight or obesity with hypertension adjusted for age, sex, BW, and interaction between overweight or obesity and BW. The interaction was significant; thus, we investigated effect measure modification⁶ by estimating the PR of hypertension (1) for each combination of overweight or obesity and BW vs healthy BMI and normal BW as the reference and (2) for overweight or obesity within BW strata. We calculated the proportion of disease attributable to interaction among persons with both exposures.⁶ Additive interaction was present when the attributable proportion was more than 0.⁶

P < .05 indicated significance. Analyses accounted for the sampling design and were performed from June 2022 to January 2023 using Stata/SE 17.0 (StataCorp LLC).

Results

Of 14 615 participants (median [IQR] age, 12 [10-14] years; 49.1% girls and 50.9% boys [weighted percentages]) with BW, BMI, and BP data, 1.1% had VLBW, 7.2% had LBW, 80.6% had normal BW, and 11.1% had HBW. Overweight or obesity occurred in 35.2% of participants with VLBW, 36.1% with LBW, 38.1% with normal BW, and 43.8% with HBW.

Hypertension prevalence was higher in participants with overweight or obesity vs healthy BMI (7.1% [95% CI, 6.3%-8.1%] vs 3.3% [95% CI, 2.7%-4.0%]) and in participants with lower BW (VLBW: 12.1% [95% CI, 5.7%-24.0%]; LBW: 6.1% [95% CI, 4.6%-8.2%]; normal BW: 4.7% [95% CI, 4.1%-5.3%]; HBW 3.8% [95% CI, 2.7%-5.4%]). The association between overweight or obesity and hypertension was modified by BW (P for interaction = .02). Participants with overweight or obesity in the VLBW vs normal BW category had higher hypertension risk (PR, 7.73 [95% CI, 2.80-21.33] vs 2.06 [95% CI, 1.59-2.67]; P = .005). Compared with participants with healthy BMI and normal BW, those with overweight or obesity and VLBW had over a 9-fold higher hypertension risk (PR, 9.23; 95% CI, 4.11-20.72) (Table; Figure). The proportion attributable to interaction among children with overweight or obesity and VLBW was 77.2% (95% CI, 57.9%-96.5%; P < .001).

Table. Effect Measure Modification by Birth Weight (BW) on the Association Between Overweight or Obesity and Hypertension (N = 14 615).

BW categorya	No. of participants	Overweight or obesity	Hypertension prevalence (95% CI), %	PR (95% CI)	PR (95% CI), within BW strata	P value for interactionb
Normal BW	11 707	No	3.4 (2.7-4.2)	1 [Reference]	2.06 (1.59-2.67)	NA
		Yes	6.8 (5.9-7.9)	2.06 (1.59-2.67)
HBW	1493	No	1.7 (0.8-3.5)	0.50 (0.23-1.08)	3.85 (1.88-7.86)	.11
		Yes	6.6 (4.7-9.0)	1.92 (1.28-2.87)
LBW	1204	No	4.7 (3.0-7.3)	1.39 (0.88-2.18)	1.94 (1.04-3.60)	.83
		Yes	8.7 (5.9-12.7)	2.66 (1.75-4.04)
VLBW	211	No	3.5 (1.8-6.7)	1.06 (0.55-2.02)	7.73 (2.80-21.33)	.005
		Yes	28.0 (12.1-52.3)	9.23 (4.11-20.72)

Abbreviations: HBW, high birth weight; LBW, low birth weight; NA, not applicable; PR, prevalence ratio; VLBW, very low birth weight.

^a Weight was categorized as normal BW (2500-3999 g), HBW (≥4000 g), LBW (1500-2499 g), and VLBW (<1500 g).

^b P values are shown for the partial interaction term comparing the association between overweight or obesity and hypertension in each BW category with the association in the reference category of normal BW.

Figure. Estimated Probability of Hypertension by Overweight or Obesity Body Mass Index (BMI) and Birth Weight.

Estimated margins with 95% CIs were calculated using the modified Poisson regression model with the covariates set at female sex aged 12 years.

Discussion

We found modification of the association between overweight or obesity and hypertension by BW. This study provided evidence that the interaction of lower BW and obesity with cardiovascular outcomes reported in adults² may exist in childhood.

Study limitations included the reliance on BW recall, lack of additional pregnancy-related data, and hypertension diagnosis based on BP at 1 visit. Nevertheless, the results highlight the public health importance of preventing obesity in children and adolescents, especially those with VLBW.

Supplement.

Data Sharing Statement