The overall goal of this project was to evaluate the distribution and determinants of hepatitis B vaccine use and compliance in community hospitals and health care facilities and among health care workers. This study was designed to identify institutional, occupational and behavioral variables among health care workers that impact their practice of effective preventive health measures, using receipt of the hepatitis B virus (HBV) vaccine as the paradigm. The health care facilities studied include the hospitals and chronic care facilities in Iowa in the Iowa Statewide Surveillance System (ISSS) and hospitals in Virginia. Two cross-sectional surveys of Iowa health care institutions were performed at baseline in early 1992 and 2.5 years later to define the rates and distribution of hepatitis B vaccine compliance prior to and following the implementation of OSHA's Bloodborne Pathogens Rule. A total of 141 ISSS facilities (94% participation) were surveyed just after OSHA's Draft Bloodborne Pathogens Rule (BPR) was published to determine mean institutional vaccination rates (IVR) in different occupational groups at baseline. Institutional HBV vaccine delivery program characteristics were categorized according to survey responses of infection control and occupational health personnel at the selected facilities. Data from the American Hospital Association's (AHA) Guide were incorporated to assess community and institutional health care organizational characteristics. Subsequently, data were collected from the same sample of facilities to evaluate changes in programs, institutional vaccination rates and examine the effect of OSHA's mandate. Additionally, identical data were collected from a sample of unstudied hospitals in Virginia. Multivariate statistical models were developed to identify organizational program characteristics of effective vaccine delivery programs. A random sample of 1,018 health care workers at risk of occupational exposure to blood and body fluids, stratified by occupation, were surveyed to identify factors independently related to HBV vaccine refusal or acceptance. Multivariate statistical models were developed and validated to identify occupational and behavioral characteristics associated with initiation or completion of the vaccine series. Vaccine implementation programs differed in terms of measured program characteristics by size of the institution and location (urban versus a rural county). Multiple linear regression analysis was used to identify factors associated with the institutional vaccination (IVR) at baseline in Iowa health care facilities. Several variables, including active encouragement and participation in vaccination by clinical leaders, the proportion of registered nurses in the workforce, and a requirement that employees receive the HBV vaccine for employment were independently associated with the IVR. The frequency of educational programs regarding HBV was of borderline significance. A multivariate linear regression model was developed to identify institutional and program factors associated with the IVR in Iowa facilities at follow-up. This model demonstrated the importance of workforce related factors (proportion of registered nurses, change in the proportion of laboratory workers), as well as program factors (provision of the vaccine at the worksite and at meetings, and the baseline rate of vaccination within the institution) in influencing the HBV IVR. A multiple logistic regression model was developed to identify the most important factors associated with achievement of a target summary rate of HBV vaccine acceptance of 90% or greater among all workers. Those factors independently related to the summary IVR included: the baseline vaccination rate within the institution, location in a rural area, and active encouragement of vaccination by the nursing director. At the individual worker level, HBV vaccine receipt was strongly related to social influence (physicians, supervisors, role models, friends and spouse) although perceived risks and knowledge of the disease were also important. Among those workers remaining unvaccinated, concern about side effects of the vaccine, knowledge of the disease and their own risk, and access to the vaccine were the major factors identified. Fifty-four percent (482 of 898) of previously non-immune workers had completed the series, while 70% (626) had received one or more doses. The most important independent predictors of HBV vaccination among individual health care workers included occupation, frequency of mucocutaneous blood exposure, prior influenza vaccination, and age. Independent predictors of initiating the vaccine series included younger age (OR = 0.98 per year, CI95 = 0.96-0.997, occupation (housestaff OR = 2.9, CI95 = 1.1-7.9 and nurses OR = 2.1, CI95 = 1.0-4.3 versus housekeepers), increased blood exposure frequency (OR = 2.4, CI95 = 1.6-3.5 for 1-6 vs 0 exposures in past year), and increased frequency of recent influenza vaccination (OR = 3.3, CI95 = 2.0-5.3 for 1 vs 0 doses in prior 3 years). Occupation (increased acceptance among housestaff, nurses, nursing assistants, laboratory technicians), increased frequency of blood exposure, and recent influenza vaccination were also independent predictors of HBV vaccine series completion.
The University of Iowa College of Medicine, Departments of Internal Medicine and Preventive Medicine and Environmental Health, 200 Hawkins Drive, Iowa City, IA 52242
Department of Internal Medicine, University of Iowa College of Medicine, Iowa City, Iowa