Oseltamivir and Risk of Serious Neuropsychiatric Events

Background: Reports of serious pediatric neuropsychiatric events during influenza treatment with oseltamivir have led to public health concerns. Whether oseltamivir is associated with neuropsychiatric events remains unclear

Objectives: To determine the association between oseltamivir use and serious neuropsychiatric events in children

Methods:

Methods: We assembled a retrospective cohort children aged 5-17 years enrolled in Tennessee Medicaid consecutive influenza seasons from 2016-2017 through 2019-2020. Each person-day of follow-up was assigned to one of five exposure cohorts with potential for varying risk of the outcome: 1) no exposure, 2) untreated influenza (10 day period following diagnosis), 3) treated influenza (influenza and oseltamivir dispensing) 4) post-treatment influenza (period between completion of oseltamivir and 10 days following influenza diagnosis) and 5) influenza prophylaxis (oseltamivir without influenza). The outcome was serious neuropsychiatric events, defined as an event resulting in hospitalization, identified using a validated algorithm (positive predictive value ~90%). The outcome included both neurologic (seizures, encephalitis, altered mental status, ataxia/movement disorders, vision changes, dizziness, headache) and psychiatric (self-harm behaviors, mood disorders, psychosis/hallucination) events. Incidence rates of neuropsychiatric events among exposure groups were compared using multivariable Poisson regression and presented as incidence-rate ratios (IRRs) with 95% CI after adjusting for age, sex, co-morbidities, concurrent medications, and measures of healthcare utilization. Sensitivity analyses were performed evaluating exposure and outcome definitions and unmeasured confounding.

Results: A total of 692,295 children contributed 19,688,320 person-weeks of follow-up, encompassing 1,230 neuropsychiatric events. Median age was 11.0 (IQR 5, 17) years (Table 1). Compared with episodes of untreated influenza, episodes of influenza treated with oseltamivir had lower risk of neuropsychiatric events (treated influenza IRR 0.54, 95% CI 0.33-0.89; post-treatment influenza IRR 0.44, 95% CI 0.26-0.76) (Table 2). Influenza prophylaxis was associated with a decreased risk of neuropsychiatric events compared to those with influenza (IRR 0.16, 95% CI 0.07-0.39). Negative control outcome, E-value, and sensitivity analyses altering exposure definitions suggest misclassification or unmeasured confounding would not explain the findings.

Conclusions: Oseltamivir use during influenza episodes was associated with a reduction in the risk of serious pediatric neuropsychiatric events. Our findings support oseltamivir use for prevention of these influenza-related complications.