Factors influencing adolescent girls’ decision in initiation for human papillomavirus vaccination: a cross-sectional study in Hong Kong

This was a cross-sectional study conducted in Hong Kong during the period of October 2010 to November 2010. The participants were female secondary school students studying secondary two, S2 to secondary seven, S7 in age range 12 to 19. Invitation was sent to eleven secondary schools which had once participated in recent programs organized by Centre for Health Education and Health Promotion of The Chinese University of Hong Kong (CUHK). Eight schools located in different regions of HK consented to participate in the survey. One class of students from S2 to S5 was randomly selected in each school to participate while all female students from S6 to S7 were recruited to participate due to smaller class size (the class size of S6 and S7 is smaller and just over half of S5 students would be promoted to S6).

Written consent from parents and assent from adolescent girls were obtained prior to filling out the anonymous questionnaire. Ethical approval was granted by the Survey and Behavioural Research Ethics Committee of the Chinese University of Hong Kong.

The self-administrated questionnaire used in the study was developed with reference from previous relevant studies in Hong Kong [12, 13, 16] with inputs from experts in adolescent preventive medicine for content validation. It was then pilot tested for face validity on a convenient sample of 15 female students studying in S4 and S5. Questionnaires together with an instruction guide detailing the study procedures and points to note were sent to the teacher in charge of the participating schools. Under the arrangement of a school teacher, students were required to fill out the questionnaire in class and on their own, and no guidance or discussion of answers was allowed during the study. Upon receiving the completion notice from schools, researchers visited the participating schools to collect all the questionnaires sent, including both the completed and blank questionnaires. The questionnaire aimed to collect data reflecting the factors having impact on initiation of HPV vaccine such as students’ awareness and knowledge of cervical cancer, HPV infection and HPV vaccine; their attitudes and beliefs towards cervical cancer prevention; and other socio-demographic factors influencing the initiation of vaccination for adolescents. There were altogether 22 items and questionnaire design was based on theoretical framework of Health Belief Model focusing on perception of severity, susceptibility, barriers and benefits [20], the Theory of Reasoned Action focusing on attitudes and social norm [21], and social cognitive theory focusing on external determinants of behaviors apart from attitudes, values and belief [22]. There were three sections and the contents were described as follows:

Descriptive data from all variables in the questionnaire were computed. Each variable explored using Likert scale characterized the respondents’ attitude and was treated as attitude score during analyses. Knowledge scores were composited from the six questions reflecting participants’ understanding of cervical cancer and questions indicating their consideration of vaccination reflected the stage of change (Additional file 1). All demographic variables of respondents were used to investigate their association with students’ knowledge and attitude towards cancers, and initiation of HPV vaccination. Those variables not on Likert scale were treated as categorical variables. Chi-square test was used to analyze factors associated with initiation of HPV vaccination. T-test was applied to compare the knowledge and attitude score towards cancer and cancer prevention between those who were in pre-contemplation/contemplation stage and those who were in intention/action stage for HPV vaccination. As HPV vaccine is still not universally available, behavioural intention is an important outcome measurement as well as actual uptake rate. Logistic regression was conducted with action and intention as dependent variable reflecting proactive move while contemplation and pre-contemplation as reference category reflecting the passive motion to identify which independent variables predicted the initiation of HPV vaccination. The independent variables included demographic, personal and family factors, perception of vaccine and the scores of knowledge and attitudes. The scores of knowledge and attitudes were re-categorised as binary variables with answering half or more questions correctly as one category and less than half as reference. The multi-variate analysis was conducted with stepwise approach. All data analyses were performed using SPSS version 14.0. A p value of less than 0.05 was considered to be statistically significant.

Majority of the participants (95.9%) had heard of cervical cancer. Most respondents were aware that cervical cancer was a common cancer among women (92.9%), yet around half did not seem to have thorough understanding of its risk factors, including that HPV is a causative factor (41.1%), having multiple sex partners increases the likelihood of cervical cancer (48.2%), and cervical cancer can develop in any women with sexual activity (52.5%). Only half knew about Pap smear screening as a preventive measure (52.8%). About one third had misconception of prognosis with 37.3% of respondents stated that cervical cancer could hardly be treated. Regarding HPV vaccination to prevent cervical cancer, 71.9% of respondents were aware of it. The commonest channels of information were television (TV) (28.3%) followed by outdoor advertisement (12.5%), clinic/hospital (11.1%), newspaper (9.7%), schools (9.5%), magazine (7.1%) and internet (6.8%). Only 5.7% reported learning about it from parents or siblings.

Majority of respondents (87.4%) had the feeling that cancer is a terrible disease. Nearly 70% expressed their wishes to have more information on cervical cancer prevention, 71% respondents thought schools should provide more information to enhance their knowledge on prevention of cancer, and they were willing to change their lifestyles if they knew the ways of prevention (78.3%). Less than half of respondents (41.5%) felt that Hong Kong has provided adequate services to promote prevention of cancer.

About one third of respondents (33.5%) felt that all females should be vaccinated. Figure 1 shows 6.7% vaccination uptake rate (action), 34.8% intended to be vaccinated, and nearly 60% of respondents remained in contemplation (requesting more information before decision) and pre-contemplation stage. Cost of vaccination, side effects, being too young for vaccination, low perceived needs and fear of injection were raised as the most important concerns for not considering (Figure 1). To those respondents who intended to be vaccinated, perceived risk of developing cervical cancer, recommendation from doctors, support from parents, and effectiveness of the vaccine were reported to be important considering factors.

Students in higher grades were found to have higher score by both uni-variate and multi-variate analysis (p < 0.01) (Table 2). Family income, parents’ education level, and employment status were not associated with students’ knowledge on cervical cancer. Channels of information from newspaper, outdoor, internet and school were found to be associated with higher knowledge score with statistical significance by uni-variate analysis and information from schools and newspaper were found to be statistical significance by multi-variate analysis (Table 2). Significant difference in knowledge score was also found between those regarding cancer as terrifying disease compared with those who did not (p <0.01), and also those wishing to know more about cancer prevention in contrast to those not wanting to know more (p = 0.01).

Table 3 provides an uni-variate analysis of association between personal and family factors, and intention of vaccination. Those in intention and action stages had significantly higher percentages of mothers with education level secondary school or above and family income of HK$30,000 or above (P = 0.03). On the other hand, those considered reasonable cost of vaccine (P = 0.02) and seeking comments from relatives and close friends (P = 0.03) had higher percentage in contemplation and pre-contemplation stage (Table 3). The action and intention group had higher knowledge score than the contemplation and pre-contemplation group with statistical significance (P = 0.04) (Table 4). The students in action and intention stage had significant higher scores in rating cancer as terrible disease (P = 0.04) and more willing to make changes after knowing the ways of preventing cancer (P = 0.02) (Table 4).

Logistic regression showed that perceiving cancer as a terrifying disease, seeking comments from relatives and close friends, and school should provide information on cancer prevention were found to have significant correlation with initiation of HPV vaccination (Table 5).

The school should provide more information on prevention of cancer and also enhance the health education so students have correct understanding of the seriousness of cancer.

Primary healthcare providers especially general practitioners are credible sources of health information parents and students tend to trust. By providing reliable information regarding cervical cancer prevention and HPV vaccine via different media or even daily practices, primary healthcare providers can play a leading role in helping the public develop positive images about vaccination for cancer prevention, which would likely enhance the uptake rate of HPV vaccine.