Hospital admission among HIV-exposed uninfected children compared to HIV-unexposed children - a nationwide cohort study

OBJECTIVE: The main objective of this study was, on a national level, to investigate the risk of in-hospital admissions and use of antibiotics during the first four years of life among HIV-exposed uninfected (HEU) children compared to a matched control group of HIV-unexposed (HU) children.

DESIGN: A nationwide register-based cohort study.

METHODS: All HEU children born in Denmark from 2000-2012 were individually matched to five HU controls. Outcomes were risk of hospital admission (any, due to an infectious disease, observation/non-specific diagnosis) and use of antibiotics during the first four years of life. Incidence rate ratios were estimated using Poisson regression analysis.

RESULTS: In total, 317 HEU children and 1581 matched controls were included. HEU children had a three-fold increased risk of overall admissions (IRR 3.49 (95% CI 2.98-4.08)). There was no difference in risk of admission due to infectious diseases (IRR 1.11 (95% CI 0.73-1.70)) and no difference in use of antibiotics (IRR 0.88 (95% CI 0.73 - 1.04). The excess risk/100 PY of admission was primarily caused by an increased risk of admission due to observation/non-specific diagnosis (excess IR 22.6 (95% CI 18.2 - 27.0), IRR 6.06 (95% CI 4.84-7.61)).

CONCLUSION: HEU children had an increased risk of overall hospital admission mainly due to an increased risk of admission due to observation/non-specific diagnosis. There was no increased risk of admission due to infectious disease. The excess risk of admission among HEU may be related to prophylactic treatment and/or HIV-testing rather than somatic disease related to HIV or exposure to antiretroviral therapy.