An Analysis of Select Emerging Executive Skills in Perinatally HIV-1-Infected Children

Abstract

This study examined the effect of perinatal HIV-1 infection on emerging executive skills in children (n = 161) ages 8 to 12 years. HIV-positive (n = 76) and HIV-negative (n = 85) children were eligible to participate. The HIV-positive children included those who had experienced a CDC Class C event (greater severity, n = 22) and those who were HIV-positive but who had not experienced a CDC Class C event (less severity, n = 54). Measures of emerging executive functions completed by the children included subtests from the Developmental Neuropsychological Assessment (NEPSY), the Trail-Making Test-Part B, and a subtest from the Woodcock-Johnson Battery-Revised. Ratings of executive functions were obtained from caretakers using the Behavior Rating Inventory of Executive Functions. Generalized estimating equations methods, discriminate analyses, and global deficit score analyses were performed to determine whether differences emerged between the three clinical groups while using strict controls. The present results revealed significant group differences in unadjusted mean scores measuring executive functioning. However, such differences did not remain statistically significant when moderating variables were taken into consideration in the models. The apparent deficit in executive functioning for the HIV-positive children was found to be largely due to differential psychosocial and environmental factors rather than HIV disease and its severity, and in this cohort, the effects of HIV-1 infection on emerging executive functions appeared to be negligible when controlling for treatment and moderating psychosocial variables.

Key words:

• executive skills
• higher-order cognition
• HIV
• moderating variables
• pediatric
• psychosocial factors

Notes

*p-values are from an exact Chi-square or F-test statistic depending on the parametric or nonparametric classification comparing distributions among the three groups; values in (parenthesis) denote actual mean values not percentages (e.g., mean head circumference in inches).

**Missing values in select cells as a result of unavailable data (e.g., income not reported by three caretakers, data not collected, etc.). Not all values reach 100% because of rounding errors or missing data.

***Uninfected participants not included in group comparison for ART therapy and CDC Class.

****Includes alcohol. Cf. Nutt et al. (2007).

*p-value from an F-test statistic comparing distributions among the three groups.

**NEPSY subtest scores are scaled scores.

***NEPSY Overall Index is a T-score.

****WJ-R is a standard score.

⁺TMT mean scores are transformed reciprocal square root (higher scores represent better performance).

⁺⁺BRIEF scores are T-score.

*p-value from an exact F-test statistic comparing distributions. All values except TMT-B (seconds) are T-scores. d' refers to Cohen's effect size statistic of pairs of groups (e.g., HIV/C vs. HIV/No C, etc.).

*p-values are from an exact Chi-square or F-test statistic. % indicates the proportion of children scoring below one standard deviation beneath the mean.