JSTOR

Methods

Nurses in Hong Kong were targeted in a self‐administered questionnaire survey to assess their experience with SARS, their attitudes toward an impending avian influenza epidemic, and their views on getting vaccinated. Nurses who registered as members of the Hong Kong Nurses General Union (the Nurses Branch, the Enrolled Nurses Branch, and/or Hong Kong Chinese Civil Servants’ Association) were invited to participate in this study. Memberships of the 3 organizations are not mutually exclusive, and all nurses practicing in Hong Kong are eligible for registration. The questionnaire, which was attached to an introductory letter explaining the nature of the study and the protection of anonymity, was sent to a member's postal address. Because almost all nurses in Hong Kong are Chinese, the questionnaire was constructed in Chinese, following discussion with the representatives of the nursing associations, and was pilot‐tested before being administered. Ethical approval was obtained from the ethics committee of the Chinese University of Hong Kong.

The questionnaire was divided into 3 main parts. In the first part of the questionnaire, respondents were asked to provide their age, sex, and their years of service. They were also asked about the nature of their job during the SARS outbreak. In the second part, respondents were asked about their vaccination status in 2006 and their intention to get vaccination in 2007. Respondents were also asked to indicate the reasons for and barriers to receiving vaccination and their perception of the effectiveness of the vaccine. In the third part, respondents were asked about their attitudes toward an impending avian influenza epidemic in Hong Kong. The questions were adapted from the study by Imai et al.²⁰ A 6‐point Likert scale was used to assess respondents' perceptions of the likelihood of an avian influenza epidemic in Hong Kong, their perceptions of control of the infection, and their perceptions of the risk of contracting avian influenza in relation to their work.

Statistical analysis was performed with SPSS (version 13.0). The odds ratio (OR) was used to calculate group differences in categorical data. Data were further analyzed by multivariate analysis, with statistical significance defined as .

Results

Characteristics of Respondents

In the second week of September 2006, a total of 2,929 questionnaires were mailed; 104 of these questionnaires were not delivered owing to incorrect postal addresses. We collected 946 completed questionnaires over a 4‐week period in September and October 2006; however, not all respondents answered every question. Three of the respondents had retired before 2003, and 2 gave incomplete responses; these were excluded from final data analysis. In calculating the response rate, we assumed that 10% of the respondents had enrolled in 2 of the associations listed above and that about 2% of our sample was not actively practicing.²¹ The response rate of our study thus fell between 33.5% and 36.3%.

Of the 941 respondents who returned questionnaires available for data analysis, it was found that 831 (88.3%) were female and that 716 (76.1%) provided direct patient care. About half of the respondents (438 [46.5%]) were 36‐45 years of age. The majority (367 [83.7%]) of them had been working in the profession for more than 11 years (Table 1).

Table 1.  Characteristics of Nurses Who Responded to the Influenza Vaccination Questionnaire

Open New Window

Vaccination

Slightly more than half of the respondents (538 [57.2%] of 941) had received influenza vaccination in 2006. With respect to the intention to be vaccinated in 2007, there were 435 (46.2%) respondents who indicated that they would do so in the coming year, 348 (37.0%) who indicated that they would not, and 158 (16.8%) who were undecided. During the years of 2006 and 2007, the reasons for getting vaccinated were similar, and they are listed in Table 2.

Table 2.  Rates of and Reasons for Influenza Vaccination Among Nurses Who Responded to the Questionnaire

Open New Window

About half of the respondents (521 [55.7%] of 935) perceived the vaccine as an effective means to protect themselves from contracting influenza. This perception was significantly associated with vaccination acceptance in both 2006 (OR, 7.69 [95% confidence interval {CI}, 5.75‐10.31]) and 2007 (OR, 11.10 [95% CI, 8.09‐15.23]). Furthermore, respondents who were vaccinated in 2006 were more likely to be vaccinated in 2007 (OR, 98.64 [95% CI, 55.73‐174.60]).

Impact of SARS and the Threat of an Avian Influenza Epidemic

During the SARS epidemic, 448 (48.1%) of 931 respondents worked in wards or hospitals with SARS patients. About 300 (32%) of them did not perform any SARS‐related duties. The remaining 183 respondents (20%) carried out duties that did not involve direct interactions with SARS patients, such as public education, infection control, and quarantine of residents of Amoy Gardens (the housing estate in Hong Kong where a community outbreak of SARS took place).

Regarding the attitudes toward an impending avian influenza epidemic, less than half of the respondents (426 [45.4%] of 938) considered that it would take place in Hong Kong. Although the majority of them were afraid of contracting avian influenza (636 [68.0%] of 936) and perceived limited control over their risk of contracting the virus because of the nature of their work (681 [72.4%] of 941), most of them accepted the risk of infection as part of their work (678 [72.1%] of 940). A small proportion of them believed that they should not take care of patients with avian influenza (128 [13.6%] of 939) and would consider a job change to avoid the risk of infection (204 [21.7%] of 941).

We calculated the ORs for the associations between experience of SARS and/or concern about of an avian influenza epidemic and acceptance of influenza vaccination (Table 3). Concern about the risk of contracting avian influenza was associated with acceptance of vaccination in 2006. The intention to get vaccinated in 2007 was associated with neither SARS experience nor concern about an impending avian influenza epidemic.

Table 3.  Associations Between Experience of Severe Acute Respiratory Syndrome (SARS), Concern About Avian Influenza, and Acceptance of Influenza Vaccination

Open New Window

Predictors of Vaccination

Two logistic regression analyses were performed to explore the predictors of the rates of vaccination in 2006 and 2007, controlling for age, sex, and work experience (Table 4). Vaccination in 2006 was the outcome measure in the first regression model. Results show that the fear of contracting avian influenza, the perceived limited control over avian influenza infection, and the perceived effectiveness of influenza vaccine were associated with acceptance of vaccination in 2006. The second regression model measured the predictors of the intention to be vaccinated in 2007. The perceived effectiveness of the vaccine and vaccination in 2006 were also significant predictors in this model. Interestingly, nurses who did not perform direct patient care duties had the highest intention to get vaccinated.

Table 4.  Factors Associated With Acceptance of Vaccination Among Nurses Who Responded to the Questionnaire

Open New Window

Discussion

To our knowledge, this is the first study of the acceptance by nurses of yearly influenza vaccination in Hong Kong. We found a vaccination rate of approximately 50%, which was higher than that reported in UK, France, Spain, and Australia.^8,10,22,23 Despite the encouraging finding, we are concerned about the decrease in vaccination acceptance in 2007, compared with the previous year. We have thus explored whether the factors associated with vaccination acceptance were different for these 2 years.

In 2005, the threat of an impending human‐to‐human avian influenza epidemic in Hong Kong had driven the government and the society at large to get prepared for the challenge.²⁴ Nurses who were afraid of the infection and who perceived the risk of infection as uncontrollable were more likely to get vaccinated in 2006. By the end of 2006 influenza season, an avian influenza epidemic had not happened. Therefore, the impact of this impending threat on rates of influenza vaccination was greatly diminished in 2007, as reflected in the present study.

The SARS outbreak of 2003 heightened public awareness of personal hygiene and increased health‐seeking behavior in Hong Kong.²⁵ We hypothesized that it might affect the rates of influenza vaccination in the following years. However, our data did not support this hypothesis. Three reasons might explain our findings. First, the outbreak of SARS happened 3 years ago. Compared with the more current threat of an impending avian influenza epidemic, SARS was very unlikely to be of immediate concern. Second, Hong Kong had maintained a high level of vigilance in preparing for another SARS epidemic,²⁶ but this vigilance has diminished. The public might have become desensitized, and the impact of SARS on the rate of influenza vaccination is therefore much lower than anticipated. Third, although both avian influenza and SARS are infectious diseases, the former is associated with a virus that belongs to the same influenza family. It is therefore not surprising to find that the threat of avian influenza had a stonger influence on rates of vaccination than the threat of SARS. We could not exclude the possibility that SARS might have had an impact on the acceptance of influenza vaccination in 2004 and 2005. However, this could not be confirmed in the present study.

We might consider the increase in rates of influenza vaccination as a positive effect of the threat of an impending avian influenza epidemic, but we should not rely on such threats alone to maintain a high vaccination rate in healthcare settings. In developing strategies to increase and maintain vaccination rates, we need to look at determinants that are modifiable—in this case, the attitudes of nurses. Some nurses refused vaccination because it could only protect them from selected virus strains over a limited time period. On the other hand, the majority of nurses did not identify with the primary objective of providing influenza vaccination in the healthcare setting. Only about 30% of our sample regarded “the need to protect others” as a major determinant of their getting vaccinated. In our study, nurses who worked in the field had a lower rate of vaccination, which may reflect their perceived risk of contracting the infection and their doubts about the efficacy of the vaccine. This attitude contrasts with the more positive attitude of nurses who did not work in the field and who may have in fact exhibited behavior similar to that of the general public.

A vaccination campaign has been shown to be effective in increasing the immediate rate of vaccination. In addition to such a campaign, use of an information‐processing model is particularly appropriate for sustaining an increase in the vaccination rate in the long run.²⁷ Healthcare workers should be informed of the efficacy of influenza vaccine and the objectives of vaccination so that they can adjust their attitudes toward influenza vaccination. Furthermore, in agreement with previous studies, we have found that the experience of getting vaccinated has had a positive influence on the likelihood of getting vaccinated again subsequently.^3,6,23 If vaccination becomes a mandatory requirement for healthcare workers or is made a professional standard, it should be effective in maintaining an increase in the vaccination rate in the healthcare setting.

Our study has 3 limitations. First, the data on vaccination in 2007 were derived from the responses of nurses to a survey question on one’s inclination to be vaccinated. We were unable to collect data on the actual vaccination rate before the influenza vaccination program began. Second, we failed to demonstrate the impact of experience of SARS on the seasonal influenza vaccination rate over a 3‐year period, because we were unable to gather data on vaccination rates in 2004 and 2005. Third, a self‐reporting questionnaire may not reflect the behavior of respondents in real‐life situations. Also, a majority of our respondents were currently in practice. We were unable to get responses from nurses who had resigned because of their concern about contracting SARS or avian influenza.

In conclusion, Hong Kong has a satisfactory influenza vaccination rate within the nursing profession. Acceptance of vaccination was associated with the occurrence of a preceding health‐related crisis. Developing a long‐term monitoring system is important, to help keep track of the vaccination rates in the health profession. This system would enable us to determine whether a concurrent or impending epidemic would affect the rate of vaccination, and it would also help inform policy makers on how to design effective strategies to improve the rate of vaccination of healthcare workers.