The Centers for Disease Control and Prevention's Advisory Committee on Immunization practices (ACIP) recommends vaccinating populations at risk or health care workers (HCWs) who have direct patient contact as the best preventive measure against influenza.³ In response to recent vaccine shortages, the ACIP constructed guidelines that prioritize groups for vaccination during periods of low availability. Health care workers who have direct patient contact are identified as a prioritized group.⁵

Influenza vaccination reduces transmission of influenza. Evidence from several studies, including randomized-controlled trials, demonstrates that vaccinating HCWs against influenza reduces nosocomial infection, worker absenteeism, and increases provider recommendation of vaccinations.^6–14 Despite such evidence and ACIP recommendations, health care institutions report employee vaccination rates that range only from 15% to 40%.^12–14 Past literature that has reported vaccination rates for HCWs have not analyzed whether HCW demographic characteristics or occupational subtype are associated with vaccination rates.^12–14

We investigated influenza vaccination rates in a nationally representative sample, the sample adult core database of the 2000 National Health Interview Survey (NHIS), in order to estimate national HCW vaccination rates and determine whether these rates vary by demographic or occupational subgroups of HCWs.

The five occupational categories included in the analysis were: health-diagnosing professions; health-assessing and-treating professions; health aides; health technicians and technologists; and administrators and administrative support staff. Health-diagnosing professions and health-assessing and-treating professions are broad occupational categories that predominantly include the occupational categories of interest: physicians and registered nurses, respectively (T. Cosca, BS, written communication, January 3, 2005).

The overall HCW influenza vaccination rate was 35.8%. There were no significant differences between HCWs who did and did not receive influenza vaccination in educational attainment, geographic region, or the presence of a chronic condition.

In unadjusted analyses, health aides had the lowest odds of vaccination receipt (Table 2). Workers who were younger than 50, female, or black had the lowest odds of vaccination within their respective demographic categories. Weighted multivariable analysis revealed similar findings, although only the results for race (black vs. white) and age (<50 vs. 50 to 64) remained statistically significant (Table 2).

Although HCWs may have access to health care services through insurance or occupational health programs, younger, poorer, and black HCWs may not exhibit high utilization of preventive health care services such as immunization because of preconceptions about the lack of utility in being immunized, cultural barriers to health care, or competing priorities.¹⁶ Our results are consistent with prior studies that have shown low levels of influenza vaccination among blacks¹⁶ and disparities in access to care for poor black HCWs.¹⁷ Improving access to vaccination by incorporating culturally relevant and appropriately targeted educational interventions, improving outreach efforts, or increasing availability of preventive services may help to reduce this disparity.

There are potential limitations to this study. First, the receipt of influenza vaccination was determined by survey and is therefore subject to recall bias. Second, the NHIS categories of health-diagnosing and health-assessing occupations were not exclusively physicians or registered nurses, respectively. The Bureau of Labor and Statistics, which constructs the Standard Occupational Classifications, also conducts biennial employment occupation projections using a taxonomy that is consistent with the Standard Occupational and Industry Codes. In 1998, physicians were approximately 65% of the health-diagnosing professions and registered nurses were approximately 73% of the health-assessing and treatment professions category (T. Cosca, personal communication, January 3, 2005).¹⁸ Small differences among occupational categories such as dentists and physicians, nurses, or other health care providers may be obscured because of the broad occupational categories. However, if we assumed that only physicians and nurses were vaccinated within their respective occupational categories, at most 71% of physicians and 51% of nurses would have been vaccinated, rates that fall far short of optimal coverage. Finally, although the sample was restricted, with the exception of administrators, to workers likely to have direct patient contact, physicians, nurses, or other health care providers who do not have direct patient contact could have been included within the sample.

Low vaccination rates may occur either because HCWs do not believe in influenza vaccination efficacy, believe that their own immune system is sufficient, or are concerned about side effects such as Guillain–Barre and injection site pain.¹⁹ Misconceptions about influenza vaccinations have been correlated with HCW refusal of vaccinations in single institution studies.¹⁹ Information and policy regarding the benefits of influenza vaccination, including reduction of nosocomial infection, hospitalizations, and employee absenteeism, must be presented in a manner that is credible.

Active campaigns at the local level using occupational health programs or a vaccination team to provide education, distribute reminder notices, and schedule vaccination times could help increase vaccination rates.^{13, 20} Further work is needed to identify the best strategies for improving vaccination rates among all HCWs, but especially those who are young or black or work as health aides.