Background Adult vaccination coverage remains low in the United States (US), despite longstanding recommendations from the American Committee on Immunization Practices (ACIP) (CDC [Centers for Disease Control and Prevention], 2016). However, national vaccination coverage estimates may mask state-level heterogeneity (due to within-state variations in individual-level characteristics, as well as vaccination programs) and state-level coverage estimates are generally reported unadjusted (i.e., without identifying and adjusting for factors that vary by state and may influence uptake). Identifying factors associated with likelihood of vaccination may provide additional insights for states to adjust current strategies towards improving coverage and help to inform future research. The main objectives of this study were to identify basic characteristics associated with influenza; pneumococcal; tetanus, diphtheria, and acellular pertussis (Tdap); and herpes zoster (HZ) vaccination (individually and in compliance with age-appropriate recommendations) and to generate adjusted state-level vaccination coverage estimates.
Methods This study was a retrospective analysis of data from the 2011-2014 Behavioral Risk Factor Surveillance System (BRFSS), an annual surveillance system supported by the CDC that monitors behavioral risk factors for noninstitutionalized adults in the US. The study population included adults aged 18+ years who participated in the 2011, 2012, 2013, or 2014 BRFSS surveys, with subgroups used to estimate vaccination coverage based on age-specific ACIP recommendations for the general adult population. To assess compliance with recommendations, three age-specific subgroups were analyzed (i.e., individuals aged 18-59 years for influenza and Tdap vaccines; individuals aged 60-64 years for influenza, Tdap, and HZ vaccines; and individuals aged 65+ years for influenza, Tdap, HZ, and pneumococcal vaccines). Multivariable logistic regression models were employed to identify factors associated with the likelihood of vaccination and compliance with age-specific recommendations, including the effect of socio-demographics, health status, healthcare utilization, and state of residence. Models that used more than 1 year of BRFSS data also included control variables for survey year and the interaction of state × survey year. Model-adjusted vaccination coverage and compliance nationally and for each state were further estimated by using predicted marginal proportions from the regression models.
Results and Conclusion Several self-reported individual characteristics were associated with the likelihood of influenza vaccination. For example, characteristics associated with a higher likelihood of influenza vaccination included: female gender; increased age, level of education, and income; and presence of at least one chronic condition. Better health status, increased time since last health check-up, and difficulties accessing care were associated with a lower likelihood of influenza vaccination. Similar relationships between likelihood of vaccination and individual characteristics generally held true for receipt of pneumococcal, Tdap, and HZ vaccines, except that likelihood of Tdap vaccination decreased with age, while better health status was positively correlated with the likelihood of Tdap and HZ vaccination. After controlling for these characteristics, state of residence was further significantly associated with likelihood of vaccination. Model-adjusted vaccination coverage and compliance varied significantly by state, with ranges including the following: -- Influenza coverage (2014): 30.2-49.5% -- Pneumococcal coverage (2014): 64.0-74.7% -- Tdap coverage (2013): 18.7-46.6% -- HZ coverage (2014): 21.3-42.9% -- Compliance with age-appropriate influenza and Tdap vaccines among individuals aged 18-59 years (2013): 7.9-24.7% -- Compliance with age-appropriate influenza, Tdap, and HZ vaccines among individuals aged 60-64 years (2014): 4.1-14.4% -- Compliance with age-appropriate influenza, Tdap, HZ, and pneumococcal vaccines among individuals aged 65+ years (2014): 3.0-18.3%
Model-adjusted national estimates also varied by vaccine, with coverage ranging 28.8-69.4% across the four different vaccines and compliance ranging 6.6-14.3% across the three compliance measures. After adjusting for basic individual characteristics associated with vaccination, substantial heterogeneity across states remained, indicating that other factors (e.g., local vaccination attitudes/beliefs, vaccination policies) may be impacting adult vaccination coverage and compliance.
