Effects of school closure on pandemic influenza incidence in Alberta, Canada

Normalization of case data 2 The bottom right of Figure 1 shows a bar plot of the cumulative sum, for each age, of all 3 laboratory-confirmed pH1N1 in 2009; it does not take the age structure of the population 4 into account. We investigated the potential effects of age structure on the shape of the bar 5 plot. Figure 5 compares the total incidence across age (panel a) with per capita incidence 6 (panel b). The distributions of raw incidence and incidence per 100,000 population are very 7 similar. Figure 5 also compares the age distributions of confirmed cases before and after 8 testing restrictions were implemented (on 30 October 2009), emphasizing the sampling bias 9 caused by testing restrictions. 10