Background:Previous cohort studies have investigated whether stimulants reduce injury risk in narcolepsy, but they faced potential biases such as confounding by indication and immortal time bias. We conducted a self‐controlled case series study to address these biases and assess the effect of stimulant use on the risk of fracture and head injury in patients with narcolepsy.
Methods:We used a Japanese health insurance claims database to identify patients with narcolepsy who were newly exposed to stimulants and who experienced fracture or head injury. We then applied multivariable conditional Poisson regression models to estimate adjusted incidence rate ratios (aIRRs) and their 95% confidence intervals (CIs) for outcomes, comparing the focal window with the referent window. Subgroup analyses assessed individual stimulants (methylphenidate, modafinil, and pemoline).
Results:Among 1790 new stimulant users, 233 had fracture and 189 had head injury. Overall stimulant use was not associated with fracture (aIRR, 0.89; 95% CI, 0.57–1.39) or head injury (aIRR, 1.40; 95% CI, 0.87–2.25). In subgroup analyses, methylphenidate use was associated with a lower fracture risk (aIRR, 0.28; 95% CI, 0.08–0.94) but did not affect head injury (aIRR, 1.16; 95% CI, 0.34–3.92). We observed no association between modafinil or pemoline and either outcome.
Conclusion:Our results indicate that overall stimulant use did not decrease the risk of fracture or head injury in patients with narcolepsy. However, methylphenidate may help prevent fractures. Clinicians should balance the safety and effectiveness of each stimulant with additional preventive strategies for injury.