Association between influenza vaccination and mortality during influenza epidemic periods, adjusted for unmeasured confounding as estimated during summer periods
| Sets of confounders | Multivariable regression analysis | Propensity score matching | Propensity score regression analysis |
| Model 1# | 0.70 (0.54–0.91) | 0.73 (0.56–0.95) | 0.71 (0.53–0.93) |
| Model 2¶ | 0.68 (0.52–0.89) | 0.71 (0.54–0.93) | 0.69 (0.53–0.91) |
| Model 3+ | 0.69 (0.52–0.92) | 0.69 (0.52–0.92) | 0.70 (0.53–0.92) |
Data are presented as odds ratio (95% confidence interval). #: included observed demographics (age, sex); ¶: included age, sex and prior healthcare use (number of general practitioner (GP) visits); +: included age, sex, prior healthcare use, comorbidity status (cardiovascular and pulmonary comorbidity, diabetes mellitus and malignancies), and medication use. Prior healthcare use was classified into four categories: <6 GP visits, 6–10 visits, 11–15 visits and >15 visits.