Overall A(H1N1)2009 pandemic influenza vaccine uptake was low at 11.1% (CI95%: 9.8 - 12.4). The main result of the study is the observed low influenza vaccine uptake of 16.3%(95% CI 11.4-22.8) among individuals less than 65 years old who are at-risk for complications due to influenza infection. This IVC is considerably lower than the seasonal IVC in this same group (35.5%). Contrary to seasonal IVC, being at-risk for influenza was not associated with higher pandemic vaccine uptake.
In France, 93% of A(H1N1)2009 influenza severe cases were younger than 65 years, and 53% had an underlying disease putting them at high risk for seasonal influenza [7
Early in the pandemic, pregnancy was considered to be a risk factor for A(H1N1)2009 influenza complications [8
], and therefore pregnant women were included in the first priority group to be invited for pandemic vaccination. Although public health experts strongly recommended vaccination for pregnant women, pandemic IVC remained low in this population (12.8%). This was despite the results of a survey conducted in November 2009, that found that 37.9% of pregnant women and 34.8% of at-risk individuals for influenza complications intended to be vaccinated against pandemic influenza, i.e. a much higher figure than the IVC actually estimated through our survey [9
The highest pandemic IVC was observed among children younger than 5 years old, who were one of the first groups to be invited for pandemic vaccination. An increased pandemic influenza uptake was associated with the number of children younger than 5 years old in the household and with the number of persons in the household. These determinants are not found for seasonal influenza. This suggests that compliance with the pandemic vaccination campaign was higher in families with children.
Low pandemic vaccine uptake (10.4%) was observed in persons older than 64 years, which is much lower than the seasonal vaccine uptake (62.6%). Adults older than 64 years old without risk factors for influenza were the last group to receive their voucher in January 2010, when the epidemic wave was over in mainland France. As a consequence, less than half of the people in this age group reported having received the pandemic vaccination voucher in January. Fortunately, elderly people were relatively unaffected by the pandemic, presumably because of cross-protective antibodies [10
]. A personal voucher for free seasonal vaccination was sent to all persons older than 64 years, making it impossible to distinguish people with underlying diseases for this age group in our survey.
WHO recommended that "All countries should immunize their health-care workers as a first priority to protect the essential health infrastructure" [11
]. This recommendation was followed in France and HCW were the first group to be invited for pandemic vaccination. The vaccination campaign for this population began in health care settings (such as hospitals, clinics) before the opening of mass vaccination centres. Pandemic IVC for health care workers was estimated at around 30%. This result was dramatically lower than the expected rate (62%) estimated by a study on acceptability for A(H1N1) vaccination conducted among HCW between June and September 2009 [9
]. However, analyses of determinants for IVC showed that being an health care worker is predictive of a better influenza vaccination uptake both for seasonal and for pandemic influenza.
Our study allows to compare at a national level for the same influenza season, vaccination coverage of seasonal and pandemic influenza and determinants of these vaccinations. However, this study has several limitations. First, it has the usual limitations of retrospective surveys based on self-report, such as recall bias. Second, a better participation of persons favourable to vaccination can not be avoided. Third, our study excluded households without telephone and those with only mobile-telephones but, this was in part corrected by adjusting sampling weights by age, gender, region and town size. Because of small numbers in some groups, confidence intervals were sometimes wide, explaining that results should be interpreted with caution. We believe, however, that the impact of these possible biases or limitations is likely to be limited and that the data produced by this survey contribute significantly to the knowledge of IVC. The validity of our results was supported by comparison of estimates produced by other French data sources. Seasonal IVC estimated by this survey are close to those of previous seasons produced by the French general health insurance scheme for at-risk individuals [12
] and for the overall population [13
]. Pandemic IVC estimates are slightly higher than those produced by official statistics [14
] (5.3 millions individuals with at least one vaccination, around 8.3% of the population) and those estimated from data produced by the French general health insurance scheme (5.2 millions, around 8.0% of the population) [15
]. The small discrepancies observed may be explained by the participation and reporting biases known to occur in surveys based on self-reporting, the possibility that some vaccinations could not be recorded in the general health insurance vaccination database, or a combination of both mechanisms.
In most countries where the figures are known, pandemic IVC remained low. Data collected by the French National Assembly investigating committee into the influenza A(H1N1) vaccination campaign in France noted uptakes of 10% or less in Germany, England, Belgium, Spain and Italy [16
]. Higher IVC than those observed in France were reported in the United States, Sweden, Canada, the Netherlands and Japan [17
]. In England, uptake was reported to be 37.1% in risk groups (including pregnant women) [18
Analysis of determinants of IVC coverage for pandemic influenza in comparison with those for seasonal influenza shows differences and allows drawing some hypotheses explaining the low vaccination uptakes observed. Doubts about the severity of the A(H1N1)2009 epidemic, about the safety and effectiveness of adjuvanted pandemic vaccines approved through a fast-track procedure, and the lack of post-marketing surveillance data may explain why people were reluctant to be vaccinated. In France, as in other countries, these subjects were highly discussed in the media in the autumn of 2009. In November 2009, only one third of the French general population considered A(H1N1)2009 influenza illness to be a "severe" or "very severe disease", and respondents with a higher perception of severity of illness and a higher level of concern were significantly more likely to accept vaccination [9
]. Results of our study suggest that a higher level of concern about pandemic influenza was observed among children and in households with the presence of children. The fact that higher pandemic IVC was observed in households where the head of the family was a farmer may be related to the origin of this new virus first reported as of swine origin.
The mass vaccination centres set up for the pandemic immunization campaign may also be a key factor explaining the low vaccination coverage. Contrary to the pandemic vaccination campaign, the administration of seasonal influenza vaccines is done by general practitioners (GPs). The role of primary care physicians as a key factor for reaching high vaccination coverage has been clearly shown [20
]. Finally, the choice of a vaccination campaign communication targeting the whole French population, with few specific messages targeting at-risk individuals may also explain the low vaccination coverage found in our study. Uptake of seasonal influenza vaccine has been shown to be a strong predictor of vaccination intention against pandemic influenza in French and US populations [21
]. A positive relation between seasonal and pandemic vaccine uptake was shown in our survey: individuals vaccinated against seasonal influenza were more likely to be vaccinated against pandemic influenza. Offer of a free vaccination has been shown to be a positive determinant for increasing seasonal influenza uptake, both in the general population and among health care workers [22
]. However cost cannot explain the low pandemic IVC observed in France because the pandemic influenza vaccination was offered free of charge for everyone.
The reasons behind the higher pandemic IVC when the head of the family had a high education level, a managerial or intermediate occupation could reflect that these populations are more sensitive to public health recommendations or perhaps less sensitive to rumors about the lack of efficacy or safety of the vaccines. Results of our study allow to draw some hypotheses but reasons behind reluctance for pandemic vaccination needs to be further investigated in order to be better prepared for future health threats that may require mass immunisation campaigns.