Effectiveness of influenza vaccination on hospitalization outcomes among older patients with diabetes

Objective: In older populations admitted for diabetes, limited evidence suggests that influenza vaccination protects against hospitalization outcomes. Methods: This study pooled 27,620 hospitalizations recorded for elderly diabetes patients from the Beijing Elderly Influenza Vaccination Information Registration Database (2013-2018) and the Beijing Urban Employee Basic Medical Insurance Database (2013-2018). Generalized linear regression and propensity score matching were conducted to estimate the effects of influenza vaccination on hospitalization outcomes (in-hospital all-cause mortality, readmission, length and costs of hospitalization), adjusting for measurable confounding factors. The low influenza period (May-July) was used as a reference period to adjust for unmeasured confounding factors during the peak influenza period (November-January). Results: In propensity score matching, influenza vaccination in peak influenza period could reduce the risk of inhospital death (OR: 0.47[0.22,0.97]) and readmission (OR: 0.70[0.60,0.81]), length of hospitalization (beta: -1.32 [-1.47, -1.17]) and medical costs (GMR: 0.90[0.88,0.92]). After adjusting for unmeasured confounding factors, influenza vaccination was associated with 17% (ratio of ORs: 0.83 [0.69, 1.02]) lower risk of readmission and shorter length of hospitalization (difference in beta: -0.23 [-0.62, 0.16]). The subgroup analyses showed that male patients with older age and poorer health conditions could benefit more after influenza vaccination. Conclusion: Influenza vaccination could significantly improve hospitalization outcomes in elderly diabetic patients. This provides evidence supporting free influenza vaccination policies for vulnerable populations in lowand middle-income countries.