The measurement of influenza vaccine coverage among health care workers

Background: Annual influenza vaccination is recommended for health care workers in both the United States and Canada. Estimations of vaccine coverage are commonly used to evaluate these vaccination programs. Purpose: We identify, discuss, and illustrate challenges including definitions of health care worker (HCW), selection of indicators, and data sources in the estimation of staff influenza vaccination coverage rates. Methods: To illustrate the impact of the factors we discuss, we created a database of a simulated pool of HCWs that included varying proportions of permanent, casual, and contract staff under differing scenarios of staff turnover and differing probabilities of individuals being vaccinated. The Excel 97 random number generator (Microsoft) was used to randomly allocate the HCW to different strata under differing staff turnover rates and to designate individuals as being vaccinated. Results: The nature of the staff targeted in the program policy has a large impact on the estimations of vaccine coverage. Different indicators provide data that might be useful for different purposes. The counts in the numerator and denominator of a period prevalence may be useful for estimation of the total workload required of the vaccination program. An incidence density might be useful as an indicator of the efficiency of the program in "capturing" staff for vaccination. The indicator that may be easiest is the point prevalence. Conclusion: Program evaluators must think carefully when planning to estimate staff vaccination coverage to avoid invalid comparisons of estimates over time and place. State or province-wide targets for health care worker (HCW) vaccination may be meaningless unless appropriate criteria for the calculation of influenza vaccination rates are developed and specified.