Associations between pharmacy choice and influenza vaccination: Mail order vs community pharmacy users

BACKGROUND: Despite the effectiveness of vaccines, US adult vaccination rates remain low. This is especially true for the influenza vaccine, which is recommended annually and widely available. The accessibility of community pharmacies as convenient places to receive influenza vaccines has been shown to increase uptake. However, use of mail order pharmacies may reduce in-person pharmacist encounters and reduce the likelihood that users receive annual influenza vaccines. OBJECTIVE: To determine the association between the type of pharmacy a patient uses and their likelihood of receiving an influenza vaccine. METHODS: This cross-sectional cohort study used the 2018 Medical Expenditure Panel Survey to observe noninstitutionalized US adult pharmacy users. Pharmacy type was dichotomized into community use only vs any mail order pharmacy use. Multivariable weighted logistic regression was used to identify associations between the type of pharmacy used and influenza vaccination, adjusting for sociodemographic, health status, and health care access and utilization confounders. All analyses were stratified by age (< 65 and ≥ 65 years). RESULTS: The aged younger than 65 years and aged 65 years and older samples had 8,074 and 4,037 respondents who represented 95,930,349 and 40,163,276 weighted observations, respectively. Compared with community pharmacy users, mail order users were more likely to be aged 65 and older, be White, have high income, and have a usual source of care (P < 0.0001). Adjusted odds ratios (AORs) for influenza vaccination were significantly lower among community pharmacy users than mail order users among individuals aged younger than 65 years (AOR=0.71; 95% CI = 0.580.87) but was not significant among those aged 65 years and older (AOR = 0.87; 95% CI = 0.69-1.09). CONCLUSIONS: Community pharmacy users aged younger than 65 years are less likely to receive the influenza vaccine than their mail order pharmacy user counterparts. These counterintuitive results could be caused by residual confounding due to differences in factors that influence pharmacy use type and vaccination likelihood. Further exploration is needed to account for differences between these populations that independently drive vaccination choice.