Table 1.

Summary of findings

	No waning occurs (null hypothesis)	Waning occurs (alternative hypothesis)
Include persons vaccinated after influenza season begins	Biased away from the null: Waning erroneously inferred or overstated (claim 2(i))	Biased: Waning may be over- or under-estimated depending on balance of two effects (claim 2(ii))
Restrict analysis to persons vaccinated before start of season	Unbiased: No waning inferred (claim 2(iii))	Waning underestimated and potentially VE can appear to rise with time since vaccination (claim 2(iii))