Pediatric Symptom Checklist-17: Factor Structure and Uniform Differential Item Functioning Across Gender and Age in HIV Orphans and Vulnerable Children in Zambia

Pediatric Symptom Checklist-17: Factor Structure and Uniform Differential Item Functioning Across Gender and Age in HIV Orphans and Vulnerable Children in Zambia

This study aimed to investigate the psychometric properties of the Pediatric Symptom Checklist-17 (PSC-17) in a sample of children orphaned or made vulnerable (OVC) by HIV in Zambia. Caregivers of 1,076 OVC (55.1% boys; M_age = 12.91 years) completed the PSC-17. Competing models, including confirmatory factor analysis (CFA), hierarchical CFA, bifactor CFA, exploratory structural equation modeling (ESEM), and bifactor ESEM, were tested to evaluate the optimal factor structure of the PSC-17. Results showed that the bifactor ESEM provided the best approximation of the PSC-17 data with a well-defined general psychosocial problems factor explaining 72% of the reliable variance in the total score and an internalizing factor containing 63% of reliable variance unique from the general factor. The observed overall psychosocial problems score was associated with lower academic achievement and working memory (with small effect sizes), supporting the discriminant validity of score interpretation. Results of multiple indicators multiple causes (MIMIC) analyses revealed that all items functioned equivalently across child gender and age.

Socioeconomic status and other factors associated with HIV status among OVC in Democratic Republic of Congo (DRC)

Background

Orphans and vulnerable children (OVC) are a high-risk group for HIV infection, particularly in Sub-Saharan Africa.

Purpose

This study aims to portray the socioeconomic profile of OVC and examine the association of household and parent/guardian characteristics with the HIV status of OVC.

Methods

For this quantitative retrospective study, we obtained data from ICAP/DRC for a total of 1,624 OVC from households enrolled for social, financial, and clinical services between January 2017 and April 2020 in two provinces of the Democratic Republic of Congo, Haut-Katanga and Kinshasa. We computed descriptive statistics for OVC and their parents' or guardians' characteristics. We used the chi-square test to determine bivariate associations of the predictor variables with the dichotomous dependent variable, HIV positivity status. To analyze the association between these independent variables and the dichotomous dependent variable HIV status after controlling for other covariates, we performed firth's logistic regression.

Results

Of the OVC included in this study, 18% were orphans, and 10.9% were HIV+. The chi-square analysis showed that among parents/guardians that were HIV+, a significantly lower proportion of OVC (11.7%) were HIV+ rather than HIV- (26.3%). In contrast, for parents/guardians with HIV- status, 9.0% of OVC were HIV-negative, and 11.7% of OVC were OVC+. The firth's logistic regression also showed the adjusted odds of HIV+ status were significantly lower for OVC with parents/guardians having HIV+ status themselves (AOR, 0.335; 95% CI, 0.171-0.656) compared with HIV-negative parents/guardians. The adjusted odds of HIV+ status were significantly lower for OVC with a monthly household income of < $30 (AOR, 0.421; 95% CI, 0.202-0.877) compared with OVC with a monthly household income > $30.

Conclusions

Our results suggest that, with the exception of a few household and parent/guardian characteristics, the risk of HIV+ status is prevalent across all groups of OVC within this study, which is consistent with the existing body of evidence showing that OVC are in general vulnerable to HIV infection. With a notable proportion of children who are single or double orphans in DRC, HIV+ OVC constitute a high-risk group that merits customized HIV services. The findings of this study provide data-driven scientific evidence to guide such customization of HIV services.