A Multi-center Survey of HPV Knowledge and Attitudes Toward HPV Vaccination among Women , Government Officials , and Medical Personnel in China

OBJECTIVES
To assess knowledge of HPV and attitudes towards HPV vaccination among the general female population, government officials, and healthcare providers in China to assist the development of an effective national HPV vaccination program.


METHODS
A cross-sectional epidemiologic survey was conducted across 21 urban and rural sites in China using a short questionnaire. 763 government officials, 760 healthcare providers, and 11,681 women aged 15-59 years were included in the final analysis. Data were analyzed using standard descriptive statistics and logistic regression.


RESULTS
Knowledge of HPV among the general female population was low; only 24% had heard of HPV. Less than 20% of healthcare providers recognized sexually naive women as the most appropriate population for HPV vaccination. There was high acceptance of the HPV vaccine for all categories of respondents. Only 6% of women were willing to pay more than US $300 for the vaccine.


CONCLUSIONS
Aggressive education is necessary to increase knowledge of HPV and its vaccine. Further proof of vaccine safety and efficacy and government subsidies combined with increased awareness could facilitate development and implementation of HPV vaccination in China.


Introduction
Cervical cancer is the second most common cancer among women worldwide, with an estimated 529,000 new cases and 275,000 deaths occurring every year (Ferlay et al., 2010).Over 85% of the cases occur in developing countries like China, where it accounts for 13% papillomavirus (HPV) strains 16 and 18 (Muñoz et al., 2003;Walboomers et al., 1999), and in China the percent is significantly higher at 84.5% (Chen et al., 2009).The recent development of two prophylactic vaccines against these strains, Gardasil ® , Merck & Co Inc., and Cervarix ® , GlaxoSmithKline Biologicals, holds the potential to drastically reduce the world burden of cervical cancer, and could have particular impact in China (Koutsky et al., 2002;Villa et al., 2005).
Both vaccines have been approved in more than 100 countries around the world, and Chinese government approval will likely be forthcoming after the completion of ongoing clinical trials.The HPV vaccines have potential for higher-than-average impact in China since cervical cancer mortality among young urban females has been recently increasing by 4.1% per year (Yang et al., 2003;Zhao et al., 2010).In order to develop an effective vaccination program, it is crucial to understand the population's knowledge of HPV and attitudes towards HPV vaccination, as they will ultimately make the decision for or against vaccination for themselves and their children.It is essential to examine socioeconomic and demographic factors, since some such as education and familiarity with the healthcare system, are more associated with HPV knowledge and vaccine receptivity than others (Brewer and Fazekas, 2007).
There has been only one study on this topic to date in mainland China, which reported that 15% of women in the general population had previously heard of HPV and that nearly 85% were willing to be vaccinated, but these statistics varied with education and region (urban versus rural) (Li et al., 2009).However, this study did not solicit information from healthcare providers or government officials, whose knowledge and attitudes are equally as important to understand, as they will be instrumental in vaccination program development and implementation.The only other Chinese studies in this area were conducted in Hong Kong (Lee et al., 2007;Kwan et al., 2008;Kwan et al., 2009), where differing societal and economic factors make the findings less applicable to mainland China.In this article, we report the results of a multi-center survey assessing knowledge of HPV and attitudes towards HPV vaccination among Chinese government officials and medical personnel as well as women from the general population in order to facilitate the development of a Chinese HPV vaccination strategy.

Study population and sampling method
This multi-center, cross-sectional, epidemiologic study included three categories of respondents drawn from multiple urban and rural regions of China: women from the general population, government officials, and healthcare providers.For the general population category, the large national census and unbalanced economic development necessitated four-stage sampling to obtain a representative study population from seven different geographic regions of China, each of which was stratified into rural or urban based on economic development and population density.One urban site was selected from each geographic region from the 23 provincial general and women and children's hospitals that had participated in a previous cervical cancer study through the National Cervical Cancer Consortium of China.Two rural sites were selected from each geographic region from the 43 county hospitals enrolled in a national rural cervical cancer screening project supported by the Ministries of Finance and Health (MOF/H).In total, this study included seven urban and fourteen rural sites.In urban areas, subjects were selected by systematic sampling using a pre-assigned sampling interval from outpatients, inpatients, and their female relatives who were physically present at a study hospital on the day of the study.In rural areas, subjects aged 30-59 years were systematically selected from the MOF/H registry in each of the 14 counties using a pre-assigned sampling interval.Subjects aged 15-29 years were selected from the same counties as their older counterparts using cluster sampling; villages within each county were randomly selected, and then every woman between the ages of 15 and 29 years in the village was asked to participate in the study.Surveys were discontinued in each age group when the target sample size was reached.The ratio of urban to rural women and the proportion of subjects per age group approximated the Chinese female population aged 15-59 years in the year 2000 (Ministry of Health, 2007), which had an urban: rural ratio of 2:3 with 12.5% aged 15-19 years, 11.6% aged 20-24 years, 14.3% aged 25-29 years, 15.4% aged 30-34 years, 13.2% aged 35-39 years, 9.7% aged 40-44 years, 10.3% aged 45-49 years, 7.6% aged 50-54 years and 5.5% aged 55-59 years.
Government officials were stratified into two groups, those working in Culture, Education, and/or Health, and those working in Agriculture, Industry, Transportation, or other departments and were drawn from the same urban and rural areas selected for the general population.In rural areas, officials were stratifed by level of government (county, commune, or village) and selected by cluster sampling, while in urban areas, convenience sampling was used due to the large number of government officials from which to select.All government officials were selected from officials who were either hospital inpatients, outpatients, or visiting another patient in the hospital.Healthcare providers were also drawn from the same urban and rural areas as the women from the general population and government officials and stratified sampling was used to select participants for the study.In urban areas, subjects were stratified by profession: gynecologists/obstetricians, pediatricians, and others, and then randomly selected.In rural areas, most doctors were general practitioners, so they were stratified by the level of their home community (county, commune, or village) and randomly selected.

Informed Consent
Before enrollment, all participants were given information about the objectives of the study, benefits and risks associated with participation, and assurance of the confidentiality of the information provided.All participants were given the opportunity to ask questions and receive answers.All participants signed an informed consent form and the study was approved by the Institutional Review Board of the Cancer Institute, DOI:http://dx.doi.org/10.7314/APJCP.2012.13.5.2369 Multi-center Survey of Knowledge andAttitudes to HPV Vaccination in China Chinese Academy of Medical Sciences (CICAMS).

Data Collection
Local interviewers were trained to orally administer a questionnaire to survey demographic characteristics and knowledge and attitudes about HPV and the HPV vaccine.All women from the general population were interviewed by a trained interviewer in a private area to avoid interruptions, and the survey information was recorded anonymously to encourage openness and honesty.The questionnaire used has been validated in several previous studies in China (Rong et al., 2002;Li et al., 2009).Questions were originally written in Mandarin Chinese in simple and clear language and knowledge of HPV and cervical cancer was assessed before further questions were asked on these topics.Many questions were multiple choice, and respondents could pick one or multiple answers.The survey for the general population also assessed sexual debut age and sexual behaviors, data from which are being published elsewhere (Zhao et al., 2012).The questionnaires for government officials and healthcare providers used many of the same questions as the general population survey but also included some questions specific to their profession, such as asking about the cervical screening techniques they recommended to patients or what venue they considered best for a future vaccination program.Some government officials and healthcare providers received the questionnaire to complete themselves, while others were interviewed by trained interviewers.All completed questionnaires underwent regular quality control to check for missing or invalid responses.If data were found missing, the respondant was asked to answer the missing questions again when the respondant had not left the survey place, otherwise, the respondant could not be included in the analysis when the relative question was analyzed.

Data Management and Analysis
Two local staff members separately entered survey information into an electronic database that was periodically sent to the data manager at CICAMS in Beijing.The data were double-checked so that both sets of data agreed.Contingency tables using Pearson Chisquared or Fisher's exact tests and tests of trend were used to establish crude associations of socio-demographic and behavior measures.There were five interview questions in all three surveys which directly measured knowledge of HPV and cervical cancer, and eight correct answers (two questions had multiple correct answers).A "knowledge score" on a scale from 1 to 8 was assigned to each respondent corresponding with the number of correct answers they selected.Analyses stratified by the three surveyed population categories and geographic region were used to identify independent determinants of HPV knowledge, with the knowledge score dichotomized at the halfway mark, with higher knowledge considered as 4 or more points on the knowledge scale.Multivariable logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) adjusted for relevant parameters as identified by preliminary analysis, which included demographic variables, genital infection history, knowledge to vaccines, anxiety on cervical cancer, acceptance of the vaccine for her daughter or herself and so on.Age was analyzed using ordinal regression analysis.Some data were analyzed using only descriptive statistics and were stratified by population and geographic region.Statisticians at CICAMS analyzed the data using the Statistical Package for the Social Sciences (SPSS) version 13.0 (SPSS, Chicago, IL).

Results
This study was carried out from May to December 2009 with a total of 13,375 people surveyed.For the general female population, 11,852 women were interviewed from seven urban and 14 rural areas of China and 171 ineligible cases were excluded (four were less than 15 years old, and 167 lived outside the selected study sites), for a total of 11,681 women (98.6%) aged 15 to 59 years in the final analysis.763 government officials and 760 healthcare providers were also interviewed for this study from across the country.Characteristics of the study population are shown in Table 1.The vast majority of those surveyed were Han Chinese.

HPV Knowledge
Only 3% of the general female population correctly answered all five knowledge questions for 8 points, while the majority of women scored 2 out of 8 points.Thirtyfour percent of urban women and 16% of rural women had previously heard of HPV, and numbers were similar for having heard of the HPV vaccine.Almost 91% of urban and 84% of rural women had heard of cervical cancer.Table 2 shows a multivariate logistic regression analysis of sociodemographic variables and their association with HPV knowledge and vaccine acceptance.Compared to women with less education, women with at least a senior high education were significantly more likely to have an HPV knowledge score of 4 or higher (OR: 1.84, 95% CI: 1.63-2.08).The age of the women was subjected to an ordinal regression analysis, and each successively older age group was found to have a statistically significant higher likelihood of having higher HPV knowledge than the next younger age group (OR: 1.12, 95% CI:1.1-1.2).
Among government officials, 7% scored the maximum 8 points on the HPV knowledge scale, and the majority scored 2 out of 8 points.Approximately 44% of all officials had heard of HPV before the survey, 95% had heard of cervical cancer, and 33% had heard of the HPV vaccine.Healthcare providers scored the highest on the HPV knowledge scale, with at least 20% of both urban and rural practitioners scoring the maximum 8 points.The majority of urban providers scored 7 out of 8 points, and the majority of rural providers scored 6 out of 8 points.96% and 80% of urban and rural providers, respectively, had heard of HPV, 95% and 89% associated it with cervical cancer, and 68% and 58% had heard of the vaccine.Gender, geographic level of the hospital, and experience level were not associated with their HPV knowledge level.The survey for healthcare providers also included several additional questions concerning high risk HPV types; over 68% of all providers were aware of the difference between high and low risk HPV.Tables 3 and 4 show the multivariate logistic regression for government officials and healthcare providers, respectively.
Table 5 presents reasons cited by respondents for and against vaccination.In both urban and rural areas, health benefits and fear of suffering from cervical cancer were cited as the top reasons for getting the vaccine among the general population.Rationales against HPV vaccination varied by geographic area and type of respondent.The majority of both urban and rural medical and government personnel were worried about the safety of the vaccine.Approximately 89% of all medical personnel would have recommended the HPV vaccine to their patients, and more than 70% of all healthcare providers thought the vaccine should be available in China.76% of government officials thought the HPV vaccine should be included in the National Immunization Program.

Discussion
To our knowledge, this is the first study to evaluate HPV knowledge and attitudes towards HPV vaccination among government officials and medical personnel, in addition to women from the general population, in mainland China.Although mainland and Hong Kong Chinese populations are not necessarily comparable, the low levels of HPV knowledge found in our study among the general female population were consistent with the knowledge levels found in the studies from Hong Kong (Lee et al., 2007;Kwan et al., 2008;Kwan et al., 2009), as well as with low knowledge levels noted in a review of 28 vaccine acceptability studies in the U.S. (Brewer and Fazekas, 2007).
Although higher than among the general female population, the relatively low knowledge level of government officials and medical personnel found in our study is troubling.Many surveyed healthcare providers did not recognize that HPV vaccination should occur before sexual debut, after which the risk of infection with HPV increases dramatically (Panatto et al., 2009).The majority of all healthcare providers selected females older than age 16 years as prime targets for vaccination, when in fact a recent study on the sexual debut age in China suggests that girls should be vaccinated for HPV between the ages of 13-15 years, before the conclusion of mandatory schooling (Zhao et al., 2012).Lower knowledge levels among rural healthcare providers compared to urban providers highlights the importance of focusing education efforts for healthcare professionals in less populated areas.Non-obstetrician/gynecologists should also be targets of increased education, as specialists in this field were 8.8 times more likely to have higher knowledge of HPV than other types of providers.In addition, education of government officials should target the higher levels (county and city officials) since they had less HPV knowledge than more local officials but will have more oversight over HPV vaccination programs.
The factors found in logistic regression analysis to be associated with willingness to be vaccinated against HPV or consent for vaccination of their daughters are important to consider.First and foremost, women with higher levels of HPV knowledge were significantly more open to vaccination for both themselves and their daughters than women with knowledge scores under 4 out of 8 on the knowledge scale.This finding reinforces the importance of public education in the success of a HPV vaccination program.In addition, many of the other factors which were associated with openness to HPV vaccination, such as exposure to other vaccines and frequent gynecological exams, all represent contact with the medical system in some way.Further education of providers will likely influence even more women to get the vaccine.It is unclear why professors and associate professors of medicine were significantly less likely to consent to HPV vaccination for their daughters than lower levels of providers; one possible explanation is that professors are more removed from clinical work than residents and attendings, and thus have less exposure to new clinical guildelines than their younger colleagues.
Despite low HPV knowledge levels, the strong approval for HPV vaccination for both themselves and their daughters among the general population is consistent with that recently found in China (Li et al., 2009) as well as in other countries (Brabin et al., 2006;Woodhall et al., 2007).The most common arguments against vaccination in our study included worries about the safety of the vaccine and low perceived personal risk of cervical cancer, attitudes which should both be responsive to education efforts.The HPV vaccine might be best introducted into China through government-sponsored programs to reassure people as to its safety, something also suggested by Li et al. (2009).For rural women, the cost of the vaccine was also a prominent barrier to vaccination, and indeed, the high cost (advertised retail price US $360 for the 3-shot series) bodes ill for its affordability to any Chinese women except the most wealthy.The fact that only 21% of rural women in our study were willing to pay more than RMB 500, or about US $80, for the entire series shows the need for drastic government subsidization of the vaccination program.Although many government officials cited worry about cost as a major concern for the success of HPV vaccination programs, the majority felt that the government should play some role in financing vaccinations, which is a positive indicator for future policies.
Women from the general population indicated high levels of trust in doctors and nurses as well as hospital lectures as their source of information about the HPV

FigureFigure 2 .
Figure 1.Opinions on the Appropriate Population to Receive HPV Vaccination

Figure 3 .
Figure 3. Responses of Government Officials to the Question, "What is the appropriate action of the Chinese government when the HPV vaccine appear on the market?".1=No direct government intervention; 2=Encouragement plus subsidization; 3=Active education campaigns; 4=Encouragement of vaccination for women in highrisk areas; 5=Wait-and-see due to unclear long-term safety and efficacy; 6=Mandate all eligible women get vaccinated