H1N1 Vaccine Acceptability and Correlates of Vaccine Refusal Among Physicians in the Emory Residency Training Programs

Background: Influenza vaccination is recommended for all health care workers (HCWs). HCWs are also one of the high priority target groups for pandemic (2009 novel H1N1) influenza vaccination. Vaccine refusal among this population has important implications for the potential spread of disease.

Objective: Study aims were to elucidate attitudes toward H1N1 vaccination acceptance among physicians in the Internal Medicine, Pediatrics, Emergency Medicine, and OB/GYN training programs at Emory University, to understand correlates of refusal, and to identify intervention points to increase vaccine coverage for H1N1 vaccination in a setting where seasonal influenza vaccination was mandatory and H1N1 vaccination was encouraged but not mandated.

Methods: A 34-item questionnaire was administered in November 2009. Statistical analyses included item frequencies, bivariate correlations, exploratory factor analyses, internal consistency assessment, and binary logistic regressions.

Results: A total of 210 HCWs including housestaff from the Departments of Internal Medicine, Emergency Medicine, OB/GYN and Pediatrics, attending physicians, and medical students were surveyed. Ninety-five percent (n=199) had received vaccination for seasonal influenza and 52% (n=110) reported H1N1 vaccination. Eighty-one percent of participants reported they had been immunized or were likely to accept the H1N1 vaccine. In multivariate analysis, significant factors predicting acceptance included lower perceived risk of vaccine-induced illness (including Guillain-Barre syndrome) [OR=2.16, CI (3.45, 1.33)] and greater perceived vaccine-associated protective benefit [OR=2.19, CI (1.32, 3.65)]. Predictors were consistent across specialty, gender, and ethnicity. Vaccine refusal most strongly correlated with familial non-acceptance of vaccination (r=-.316, p<.001), non-recommendation of H1N1 vaccine to others (r=-.342, p<.001), and perceived vaccine-induced health problems (r=.298, p<.001). Those who believed the vaccine could cause active influenza infection were less likely to accept the vaccine [OR=.526, CI (0.37, 0.75)].

Conclusions: Physicians surveyed were significantly more likely to have received a mandatory vaccine (seasonal influenza) than the H1N1 vaccine, which was encouraged but not required. Most indicated they were likely to get vaccinated for H1N1 but many had not done so already although H1N1 continues to circulate in the Atlanta area. Perceived risk and perceived benefit were strongest predictors of plans to accept the vaccine. The results suggest H1N1 vaccination coverage may be increased by making vaccination mandatory or by focusing on vaccine safety messages that address the perceived risk of active influenza infection.