Of the 4006 questionnaires distributed for the first survey, 1866 were completed and returned, giving a response rate of 46.6%. Of the 42 targeted units, 31 (73.8%) participated (including emergency medicine, internal medicine, and paediatrics units), representing 20% of all doctors and nurses working in these units in Hong Kong. Each geographical cluster had participating hospitals, and each hospital had at least one participating department. The number of paediatrics departments participating were significantly less than the other two specialties. Details of the response to the first survey are shown in figure 1. Nurses accounted for 71.1% of the respondents, and doctors accounted for 19.3%. The distribution of the doctors and nurses was not significantly different from the overall distribution of the human resources in these units. Most (75%) of the respondents were women, because of female dominance in the nursing profession.
Fig 1 Response rate to first questionnaire survey of healthcare workers in Hong Kong hospital departments
Of the 810 questionnaires distributed in the second survey, 389 were completed and returned. The response rate was 48.0%. Details of the second survey are shown in figure 2. All three invited departments had also participated in the first survey. The age and sex distribution of respondents were similar to those of the respondents in the same hospital in the first survey. Demographics of all respondents in both surveys are shown in table 1.
Fig 2 Response rate to second questionnaire survey of healthcare workers in Hong Kong hospital departments
Table 1 Characteristics of the respondents. Values are numbers (percentages) of respondents
The overall intention to accept pre-pandemic vaccination (H5N1 vaccine) was only 28.4% for the first survey, which was conducted at WHO influenza pandemic alert phase 3. The level of intention to accept increased to 34.8% in the second survey, when the WHO alert phase was level 5. Responses from three departments in the hospital where both surveys were conducted are shown in table 2. No significant changes in the level of intention to accept pre-pandemic vaccination (H5N1 vaccine) was observed, despite the escalation to phase 5 because of the wide spread of H1N1 virus (swine flu).
Table 2 Comparison of the intention to accept vaccination with H5N1 vaccine in one hospital at two different WHO influenza pandemic alert phases
The proportion of healthcare workers intending to accept pre-pandemic vaccination (H1N1 vaccine) was 47.9% when the WHO alert level was at phase 5. The respondents who were willing to accept H5N1 vaccines were more likely to accept H1N1 vaccines as well (91%); in contrast only 23.6% of those who declined H5N1 vaccination expressed an intention to accept H1N1 vaccination (P<0.0001).
The most common reasons for intending to accept vaccination were “wish to be protected” and “following Health Authority’s advice” (fig 3). The most common reason for refusal was “worry about side effects,” and other reasons included “query on the efficacy of the vaccine,” “not yet the right time to be vaccinated,” and “simply did not want the vaccine” (fig 4). More than half of the respondents thought that nurses should be the first priority group to receive the vaccines, followed by doctors and allied health professions, and then similar ratings for non-clinical and administrative staff. About half of the respondents (52.2% in the first survey and 56.0% in the second) wanted their family members to receive the vaccines as well. All the above responses remained constant in the different WHO alert phase levels.
Fig 3 Reasons of healthcare workers in Hong Kong hospital departments for intention to accept pre-pandemic influenza vaccination
Fig 4 Reasons of healthcare workers in Hong Kong hospital departments for intention to decline pre-pandemic influenza vaccination
Univariate associations between intention to accept H5N1 vaccination and other variables at WHO alert phase 3 are shown in table 3. Male sex, working in a specialty other than internal medicine, being a doctor, having fewer years of work in the health services, having received seasonal influenza vaccination in 2008-9, and perceptions that they were likely to contract the influenza and that a pandemic would seriously affect their lives were all significantly associated with greater intention to accept vaccination. In multiple logistic regression models for intention to accept vaccination (table 4), all of these variables remained significant except for specialty, which became marginally significant.
Table 3 Univariate association of variables affecting the intention to accept pre-pandemic influenza vaccination in the first survey of Hong Kong healthcare workers (phase 3 of WHO influenza pandemic alert). Values are numbers (percentages) of respondents (more ...)
Table 4 Multiple logistic regression model for intention to accept pre-pandemic influenza vaccination (H5N1 vaccine) in the first and second surveys of Hong Kong healthcare workers (at phases 3 and 5 of WHO influenza pandemic alert)
At WHO alert phase 5, only having received seasonal influenza vaccination in 2008-9 and younger age were found as significant associated factors for intention to accept H5N1 vaccination in multiple logistic regression (table 4).
For H1N1 vaccination, the factors showing a significant association with intention to accept at WHO alert phase 5 after adjustment by multiple logistic regression included younger age; having received seasonal influenza vaccination in 2008-9, and the perception that they were more likely to contract the pandemic influenza. The results are shown in table 5.
Table 5 Multiple logistic regression model for intention to accept pandemic influenza vaccination (H1N1 vaccine) in the second survey of Hong Kong healthcare workers (at phase 5 of WHO influenza pandemic alert)