Change in risk for narcolepsy over time and impact of definition of onset date following vaccination with AS03 adjuvanted pandemic A/H1N1 influenza vaccine (Pandemrix) during the 2009 H1N1 influenza pandemic

Purpose: To estimate risk for narcolepsy in defined time windows following exposure to adjuvanted A(H1N1) pandemic vaccine (Pandemrix) and impact of different definitions of index date for the narcolepsy diagnosis.
Methods: Vaccine exposure in approximately 30% of the Swedish population in 2009 was linked to information on narcolepsy diagnosis retrieved from the national patient registry. Cases were verified by a systematic chart review. Poisson regression was used to compare incidence in defined time windows following vaccination.
Results: Of 266 cases of narcolepsy identified, 25% (66/266) were prevalent cases with symptom onset documented before vaccine exposure. Incident cases had a median time interval between first symptom and the date recorded in the patient registry of 64 weeks (IQR 39-107) when vaccinated (N = 182) and 65 weeks (IQR 51-72) when unvaccinated (N = 16). With first symptom defining index date, the adjusted risk for narcolepsy in younger patients was increased 14 times during the first year after vaccination, three times elevated the second year, but with no detectable increased risk more than 2 years after vaccination exposure. Using the index date from the patient registry, the adjusted increase in risk was about seven times elevated for all three time intervals.
Conclusions: The magnitude of the estimated increased risk for narcolepsy following exposure to the A(H1N1) pandemic vaccine is highly dependent on the method used to determine the index date for disease onset. The sometimes very long and potentially variable interval from first symptom to a health care registry diagnosis complicates estimations of risk.