Another commonly stated reason for non-immunization was the belief that vaccination weakens the natural immune system, which will be see more referred to as naturalistic beliefs.
Finally, prevention beliefs constitute the opinion that other means of prevention (i.e. regular hand disinfection, staying at home when ill) are more effective in preventing influenza than vaccination [26]. The aim of this longitudinal study was to test with a survey whether the intention to get vaccinated, as well as the measured social cognitive variables, are good predictors of the actual vaccination behaviour of HCP. The social cognitive variables that will be identified to predict actual vaccination uptake can serve as reference points for the systematic development of a program to increase influenza vaccination uptake of check details HCP. Dutch HCP belonging to an online panel (N = 1370) were invited in the last week of September 2013 to participate in a longitudinal survey about the factors that influence the decision to get vaccinated against influenza (baseline). HCP in the Netherlands commonly get offered influenza vaccination between October and November. Participants who got vaccinated before the last week of September were excluded from the sample (N = 23), as were HCP that indicated that they did not have direct patient contact (N = 199). In total, 556 participants were included in the baseline measure (response rate 40.6%). To
link intention to actual vaccination behaviour, participants who completed the first questionnaire were sent a second questionnaire in the last week of November 2013 (follow-up). The follow-up survey was completed by 458 (82%) participants. The first
online questionnaire consisted of 42 questions targeting social cognitive variables and additional beliefs about annual influenza vaccination, past behaviour, and socio-demographics. Variables were measured on 7-point Likert scales ranging from 1 = totally disagree to 7 = totally agree, unless otherwise indicated. Items measuring the same underlying theoretical construct were averaged into one single construct when internal consistency was sufficient (Cronbach’s alpha α > .60 isothipendyl or Pearson correlation coefficient r > .40). Table 1 provides an overview of the constructs and their internal consistency. In addition, past behaviour was measured with two questions (‘In past years I got vaccinated against influenza, when it was offered to me: 1 = always; 7 = never.’; ‘Did you get vaccinated against influenza this year (season 2012/2013)? yes/no.’). Past experience with influenza was measured with two questions (‘How often did you have influenza in the past? 1 = never; 7 = more than 10 times.’; ‘Did you have influenza last winter? no/yes, once/yes, more than once.’). These items measured own experiences of influenza-like illness (ILI) instead of laboratory confirmed influenza.