Factors associated with abandonment of infants born to HIV-positive women: results from a Ukrainian birth cohort

Abstract

Social marginalisation and other challenges facing HIV-positive pregnant women in Ukraine may put them at increased risk of relinquishing their infants to the state. We described rates of infant abandonment (exclusive non-parental care to most recent follow-up, censored at two years of age) and investigated associated factors using logistic regression models, in 4759 mother-infant pairs enrolled across six Ukrainian sites in the European Collaborative Study from 2000 to May 2009. Median maternal age was 26.0 years, 81.8% were married or cohabiting and 60.6% were nulliparous at enrolment. An injecting drug use (IDU) history was reported by 18.4%, 80.2% took antiretroviral therapy (ART) antenatally and most deliveries were vaginal. A small but significant proportion of infants had been cared for exclusively in institutions by their second birthday (2.1% overall), decreasing from 3.8% (15/393) in 2000-2002 to 1.6% (49/3136) in 2006-2009 (p<0.01), concurrent with prevention of mother-to-child transmission scale-up. A further 1% of infants spent some time in non-parental care. Antenatal ART was associated with an 88% reduced abandonment risk (adjusted odds ratio (AOR) 0.12), versus receipt of single dose nevirapine only, and this was reflected in HIV infection prevalence in the two groups (17.1% of abandoned infants versus 6.6% in parental care). Mothers without a cohabiting partner or husband were more likely to abandon (AOR 4.08), as were active IDUs (AOR 3.27) and those with ≥1 previous children (AOR 1.89 for second-born and AOR 2.56 for subsequent births). Women delivering by elective caesarean section were less likely to abandon (AOR 0.37 versus vaginal), as were those leaving full-time education later (AOR 0.61 for 17-18 years versus ≤16 years and AOR 0.23 for ≥19 years versus ≤16 years). Interventions to extend family planning and IDU harm reduction services along with non-stigmatising antenatal care to marginalised women are needed, and may reduce abandonment.

Figure 1

Definition of outcome variable ‘abandoned’ (ever or never in parental care during first two years of life), using follow up data censored at 24 months