Stunting Among HIV-Exposed and HIV-Free Children in eSwatini: A Retrospective Evaluation of Associations with Birthweight, Feeding, and Caregiving Practices.

Background/Objectives. Associations between stunting and dietary practices have been understudied among HIV-exposed and HIV-free children. We investigated associations between birthweight, socio-demographics, and dietary and feeding practices with stunting at 9 and 18 months among children in eSwatini. Methods. We used generalized linear mixed models (GLMs) with a logit link to characterize associations between stunting and birthweight quartiles, socio-demographics, maternal characteristics, and infant dietary diversity scores. We examined the moderating effects of dietary factors on relations between birthweight and stunting. Generalized structural equation models characterized direct and indirect associations between exposures and stunting at 18 months, mediated through stunting at 9 months. Results. We included 367 HIV-exposed and HIV-free children. Infants in the third and fourth birthweight quartiles had reduced odds of stunting at 9 months [adjusted odds ratio (adj OR) 0.24 (IQR 0.11, 0.55), p < 0.001; 0.10 (0.03, 0.33), p < 0.001, respectively]. Moderation by dietary diversity was limited to a relative decline in the second birthweight quartile. Stunting prevalence significantly increased from 9 months (21%) to 18 months (37%). Mediated by stunting at 9 months, there were significant direct and indirect effects of birthweight and dietary diversity on stunting at 18 months. Conclusions. Among HIV-exposed and HIV-free children, stunting prevalence increased with age and was associated with lower birthweight. Dietary diversity attenuated stunting risk among children in the second birthweight quartile. Prenatal strategies to reduce low birth weight (LBW) and additional attention to the social determinants of health, particularly dietary diversity, are warranted in programs and policies to reduce stunting.