Girls' Explanations for Being Unvaccinated or Under Vaccinated Against Human Papillomavirus

A Content Analysis of Survey Responses

Alice S. Forster; Jo Waller; Harriet L. Bowyer; Laura A. V. Marlow


BMC Public Health. 2015;15(1278) 

In This Article



Participants were girls aged 15 to 16 years old who completed a questionnaire three years after their year group were offered HPV vaccination in school (when they were 12 or 13 years). Participants were recruited through schools. To recruit the schools we created a sampling frame using Free School Meal Eligibility (FSME)[22] and General Certificate in Secondary Education (GCSE) attainment as indicators of socio-economic position. We categorised schools as being above or below the national average for FSME (23 %,[23]) and GCSE (53 % of pupils attaining 5 grades A*-C;[24]). All London government-funded schools with female pupils were entered into the sampling frame if they reported HPV uptake within 10 % of the national average[25] (89 schools). We randomly contacted schools in each cell of the sampling frame until the target sample was recruited (n = 13 schools). Data were collected over two waves in 2011/12 and 2012/13. One week before the research took place, parents received an information sheet about the study and an opt-out form. Consent was implied if the opt-out form was not returned. All girls in attendance when the research was conducted were given information about the study. Consent was presumed if the girls completed the questionnaire. We chose an opt-out consent approach to maximise the response rate, while still giving parents and girls the option of opting out (to enhance the validity of the findings). The study was assessed to be low risk (as it was a questionnaire study and was not asking questions that might invoke distress), asking 15 and 16 year old students about their attitudes to vaccination and vaccination behaviour. The schools had viewed the questionnaires and study protocol and assessed this approach to be appropriate. Researchers gave an introductory talk to the students emphasising that participation was voluntary and anonymous and that they did not need to complete any questions they felt uncomfortable answering. The approach was approved by the ethics committee. All girls received a debrief sheet after completing the questionnaire. More information about the sampling method is provided elsewhere.[11,26]


During a school lesson, respondents completed the questionnaire which asked them to report their HPV vaccine status. Vaccination status was assessed by asking girls which best applied to them ('I have had all three doses of the HPV vaccine'; 'I have had 1 or 2 doses of the HPV vaccine'; 'I have been offered the HPV vaccine, but I haven't had it'; 'I have not been offered the HPV vaccine' or 'I don't know'). Those who had not accepted the vaccine or who had started, but not completed the course were asked to report their reasons using free-text ('If you have not had the HPV vaccine, or you didn't have all 3 doses, why was this?'). Girls were 12 or 13 when they were offered the HPV vaccine and parental consent would have been requested for vaccination (although not mandatory), so girls' responses are likely to be a combination of their own beliefs about the vaccine and their parents' reasons for not consenting. Participants reported their ethnicity, religion, and whether they were practising that religion and completed the Family Affluence Scale,[27] a four item, objective measure of family affluence. The present study focuses on the reasons girls gave for being un-/under vaccinated, but the questionnaire also assessed HPV knowledge, attitudes to vaccination and sexual behaviour, the results of which are published elsewhere.[11,26] The study was approved by the University College London research ethics committee (0630/002).


Participants who had received one or two doses of the HPV vaccine (as opposed to the recommended 3 doses) were classified as under vaccinated and those who had been offered the vaccine, but who had not received any doses were classified as unvaccinated.

Chi-square (significance p < 0.05) was used to test for demographic differences in whether girls provided a reason to explain their vaccination status and to test for demographic differences between unvaccinated and under vaccinated girls who had provided a reason for their vaccination status. We used content analysis to analyse the girls' free-text responses. After reading through the responses LM developed a coding frame. Responses were then coded by LM and AF with an inter-rater reliability of 0.86 (kappa). Any discrepancies were discussed and resolved. Where multiple reasons were given multiple codes were allocated.

Girls' reasons are reported and discussed by vaccination status. We explored girls' reasons by ethnic group, but numbers were too small for statistical comparison.