Estimating infectious diseases incidence: validity of capture–recapture analysis and truncated models for incomplete count data