Family Socioeconomic Status and Neurodevelopment Among Patients With Dextro-Transposition of the Great Arteries

Abstract

Importance: Data are limited on the longitudinal implications of socioeconomic status (SES) for neurodevelopmental outcomes among persons with complex congenital heart disease (CHD).

Objectives: To examine the association of family SES, maternal educational level, and maternal IQ with the neurodevelopment of individuals with dextro-transposition of the great arteries (d-TGA) from age 1 to 16 years and to identify how SES-related disparities change with age.

Design, setting, and participants: This cohort study analyzed data of participants enrolled in the Boston Circulatory Arrest Study, a randomized clinical trial conducted in Boston, Massachusetts, from 1988 to 1992. Participants were infants with d-TGA who underwent arterial switch operation and, after operation, underwent in-person neurodevelopmental status evaluations at ages 1, 4, 8, and 16 years. Analyses were conducted from April 2021 to August 2024.

Exposures: Mean Hollingshead scores at birth, age 1 year, and age 4 years were used to assign participants to SES tertiles (lowest, middle, or highest).

Main outcomes and measures: Age-appropriate neurodevelopmental outcomes assessed at 4 study time points (ages 1, 4, 8, and 16 years) via in-person administration of a range of well-validated measures. Standardized neurodevelopmental composite scores from each evaluation were derived from principal component analysis and compared across SES tertiles, adjusting for birth and medical characteristics. These scores were used to categorize the sample into latent classes; patient and medical factors for a 3-class model were used to estimate latent class using multinomial regression.

Results: The sample included 164 patients with d-TGA (123 males [75%]; mean [SD] gestational age at birth, 39.8 [1.2] weeks; 3 with Asian [2%], 6 with Black [4%], 5 with Hispanic [3%], and 146 with White [89%] race and ethnicity) and their mothers (mean [SD] age at birth, 28.5 [5.2] years). Lower SES tertile was associated with worse scores on most individual neurodevelopmental tests and worse neurodevelopmental composite scores at ages 4, 8, and 16 years. For example, mean (SD) neurodevelopmental composite scores at age 4 years were -0.49 [0.83] for lowest, 0.00 [0.81] for middle, and 0.47 [1.10] for highest SES tertile (F2 = 15.5; P < .001). When measured at consecutive time points, differences between SES tertiles were of similar magnitude. A latent class analysis produced 2- and 3-class models representing patients with stable (103 [64%] and 85 [53%]), improving (20 [13%]), and declining (57 [36%] and 55 [34%]) neurodevelopmental status. Those experiencing declines in neurodevelopmental status were more likely to have younger maternal age at childbirth (26.6 [5.1] vs 29.6 [4.9] and 29.1 [5.1] years; P = .002), lower maternal IQ (91.0 [14.1] vs 100.1 [11.1] and 96.2 [11.0]; P < .001), and lower SES (35.2 [10.8] vs 40.9 [9.9] and 35.8 [10.1]; P = .003) compared with those with stable or improving status.

Conclusions and relevance: This cohort study of individuals with d-TGA found an association between lower family SES and worse neurodevelopmental outcomes in childhood and continuing throughout adolescence as well as greater decline in neurodevelopmental status over time. Effective strategies are needed to improve access to neurodevelopmental monitoring and intervention services for children with CHD from lower socioeconomic backgrounds.

Figure 1.. Selected Neurodevelopmental Outcomes by Socioeconomic Status (SES) Tertile

The total number of observations was 144 for Psychomotor Development Index (PDI) (A), 155 for Wechsler Preschool and Primary Scale of Intelligence-Revised (WPPSI-R) and 154 for Wechsler Intelligence Scale for Children-Third Edition (WISC-III) Full Scale Intelligence Quotient (FSIQ) (B), 154 for Wechsler Individual Achievement Test (WIAT) and 140 WIAT-Second Edition II (WIAT-II) Mathematics (C), and 153 for Wide Range Assessment of Memory and Learning (WRAML) and 139 for Children’s Memory Scale (CMS) General Memory (D). eTable 1 in Supplement 1 provides the score ranges for these and other measures. Trends across SES tertiles were similar in age groups in panels B, C, and D (P > .32 on 2 df for each tertile, adjusting for birth and medical characteristics). Upper and lower ends of the boxes represent the 25th and 75th percentiles, respectively; horizontal line inside boxes represents the median; whiskers represent the IQR below the 25th percentile and above the 75th percentile; and circles above or below boxes represent data points outside the range of the whiskers.

Figure 2.. Neurodevelopmental Composite Scores by Socioeconomic Status (SES) Tertile and by Age Group

Thick lines depict mean values of neurodevelopmental composite scores by SES tertile (53 observations in the lowest, 52 in the middle, and 55 in the highest) and age group. Thin lines connect values of neurodevelopmental composite scores for individuals. Interactions between age groups and SES tertiles are significant across ages 1, 4, 8, and 16 years (P = .02 on 6 df) but not across ages 4, 8, and 16 years (P = .50 on 4 df), adjusting for birth and medical characteristics due to the middle and highest SES tertiles being similar at age 1 year.

Figure 3.. Neurodevelopmental Composite Scores by Latent Class Analysis

The number of observations was 103 in class 1 (stable) and 57 in class 2 (declining) (A) and 85 in class 1 (stable), 20 in class 2 (improving), and 55 in class 3 (declining) (B). Upper and lower ends of the boxes represent the 25th and 75th percentiles, respectively; horizontal line inside boxes represent the median; whiskers represent the IQR below the 25th percentile and above the 75th percentile; and circles above or below boxes represent data points outside the range of the whiskers.