Sensitivity Analysis
Imputation of outliers in both influenza cohorts and their effect on the
sample are shown in Supplemental Table 3. After model adjustment with
the new clusters, there were no statistically significant differences in
the odds or model-adjusted means of outcomes for the 2017/2018 influenza
season (Supplemental Figure 1). For the 2018/2019 influenza season,
after adjustment those in new cluster 1 had significantly higher odds of
an intensive care unit stay compared with those in new cluster 2 (OR
4.62, 95%CI:1.34,15.97; Supplemental Figure 1).