Page 68 - Preventing pertussis in early infancy - Visser

Chapter 5
Chapter 5
pertussis cocooning vaccination would most likely include these professionals because they care for infants in the vulnerable first half year of their lives.
Questionnaire and variables
Our Dutch questionnaire is based on a theoretical framework (Figure 1). The framework, which describes the various determinants for accepting pertussis vaccination, originated from the results of a literature review, our previous qualitative study (Visser et al. 2016), and application of the Reasoned Action Approach (Fishbein et al. 2010). The questionnaire included 123 items and was developed in parallel with a study of parental intention to accept pertussis vaccination (published elsewhere) (Visser et al. 2016).
The primary outcome measure of the questionnaire was the HCW intention of accepting a pertussis vaccination if it were offered in a cocooning strategy. We measured both personal and psychosocial determinants that potentially influence the intention to accept vaccination, as well as potential determinants of attitude (behavioural beliefs). The personal determinants of intention included personal characteristics, vaccination history, and pertussis experience.
The psychosocial determinants included attitude, social norm, perceived behavioural control, anticipated negative affect, and decisional uncertainty. The first three are main psychosocial determinants of behaviour derived from the Reasoned Action Approach. The remaining two determinants were relevant in the qualitative study. The behavioural beliefs underlying attitude were selected from the qualitative study findings and the literature review. Table 1 and Appendix 1 show the psychosocial measures and behavioural beliefs.
We used a seven-point Likert scale to measure the primary outcome (intention), the psychosocial determinants of intention, and the behavioural beliefs underlying attitude. We combined items that were theoretically linked, but only when there was enough internal consistency (Cronbach’s Alpha α > 0.70 or Pearson correlation r > 0.60). Items for each participant were then averaged into one single measure.
Data collection
In November and December of 2012, we invited maternity assistants, midwives, and paediatric nurses to participate. We used a pragmatic approach for collecting data for each group. Potential participants had a month to complete the questionnaire. Based on a power calculation, we needed at least 200 respondents in each group. To increase the response rates, the eligible populations were informed that every 20th participant would receive a small gift voucher.
68
66