Polypharmacy and adverse drug reactions among hospitalised children: determinants and clinical and economic impact in a multicentre prospective cohort and nested matched case-control study in Northwest Ethiopia.

BACKGROUND: Medication-related harm in children is a major global public health concern, with adverse drug reactions (ADRs) contributing substantially to morbidity and healthcare costs. This study aimed to estimate ADR incidence, identify determinants of polypharmacy, examine its association with ADR risk and assess the clinical and economic impact of ADRs. METHODS: A multicentre prospective cohort study with a nested case-control component was conducted among paediatric inpatients in four comprehensive specialised hospitals in Northwest Ethiopia. Cox proportional hazards regression identified predictors of ADR occurrence, and logistic regression assessed determinants of polypharmacy. For the nested case-control analysis, categorical outcomes were compared using the McNemar test, while continuous outcomes were compared using a generalised linear model. Statistical significance was set at p<0.05 with 95% CIs. RESULTS: Among 1247 paediatric inpatients, 320 (25.7%) experienced at least one ADR, totalling 347 events (22.5 per 1000 patient-days). Independent predictors of ADRs included off-label drug use (adjusted HR (AHR) 1.3, 95% CI 1.0 to 1.6), immunocompromised status (AHR 1.6, 95% CI 1.2 to 1.9), three disease conditions (AHR 1.4, 95% CI 1.0 to 1.9), four or more disease conditions (AHR 1.9, 95% CI 1.2 to 2.4), systemic anti-infective use (AHR 1.4, 95% CI 1.1 to 1.8) and excessive polypharmacy (AHR 1.4, 95% CI 1.0 to 1.9). Polypharmacy (≥5 medications) was observed in 56.7% of patients and was associated with multiple prescribers (AOR 1.48, 95% CI 1.1 to 2.0), hospital stay ≥14 days (AOR 1.60, 95% CI 1.1 to 2.6), each additional antibiotic (AOR 2.5, 95% CI 2.2 to 2.9) and each additional intravenous medication (AOR 2.6, 95% CI 2.1 to 3.0). ADR cases had longer hospital stays (15.7±5.6 vs 12.2±3.0 days; β=0.213, 95% CI 0.171 to 0.256) and higher medical costs (6027±1037.6 vs 4483.8±703.9 Ethiopian birr; β=0.296, 95% CI 0.270 to 0.322) than controls. CONCLUSIONS: ADRs are common among hospitalised children in Northwest Ethiopia and are associated with prolonged hospital stay and increased healthcare costs. Polypharmacy is also highly prevalent, driven by the complexity of care, and independently increases the risk of ADRs. Strengthening pharmacovigilance and promoting rational prescribing with targeted monitoring of high-risk children may help to reduce medication-related harm and economic burden in paediatric patients.