Experience and acceptability for HPV self-sampling among women in Jiangsu province, China: a cross-sectional survey

Abstract This aim of this study was to investigate women’s knowledge about HPV along with their experience and acceptability of self-sampling in Jiangsu province, China. A total of 862 women aged 25–63 years old from Jiangsu province who purchased an HPV self-sampling test kit were invited to complete a questionnaire designed by the authors. Participants had high acceptability for HPV self-sampling with a mean score of 4.2 (95% [CI], 4.1–4.22) out of 5 points. 27% of participants preferred clinician-sampling, 33% preferred self-sampling, other 40% expressed no preference. Women with good knowledge about HPV and with a good experience with HPV self-sampling were more acceptable for self-sampling (P < 0.05). The biggest concern about HPV self-sampling of the participants includes ‘specimens’ spoilage’, ‘incorrect sampling’, ‘can’t get results in time’, and so on. HPV self-sampling can be used to improve cervical cancer screening coverage and participation rates in China. IMPACT STATEMENT Cancer screening and can be an alternative primary screening for cervical cancer. •What the results of this study add? This study adds new findings about Chinese women’s experience and acceptability of HPV self-sampling. We found that most women had high acceptability for HPV self-sampling in Jiangsu province, China, and high knowledge about HPV as well as good •What is already known on this subject? HPV self-sampling testing was proven to be useful for improving the uptake rate of cervical experience of self-sampling can improve the acceptability for self-sampling in women. •What the implications are of these findings for clinical practice and/or further research? Further research should assess the acceptability of women with less education or who never screened.


Background
Cervical cancer is one of the most common cancers in women, and its incidence and mortality both account for the fourth place of female malignant tumours, with 604,000 cases and 342,000 deaths recorded worldwide and 110,000 cases and 59,000 deaths in China in 2020 (Liu et al. 2021).According to World Health Organisation (WHO), about 90% of cervical cancer can be reduced by regular screening (WHO 2020).China launched a free cervical cancer cytology screening program covering women living in rural areas in 2009, which improved the cervical cancer screening rate to the same extent.But because of the large population, women's lack of cognition about cervical cancer prevention and treatment as well as embarrassment and discomfort during the gynaecological examination, the cervical cancer screening coverage and participation rate in China are at a low level (Huang et al. 2020).the rate of cervical cancer screening in China was about 26.7% (31.8% in urban areas and 22.3 in rural areas) in 2013 (Bao et al. 2018).
It has been established that cervical cancer has a great association with Human Papillomavirus (HPV) (WHO 2022).WHO published the Guideline for Screening and Treatment of Cervical Pre-cancer Lesions for Cervical Cancer Prevention in 2021, recommending HPV DNA testing as the first choice for cervical cancer screening and including self-sampling testing as a part of cervical cancer screening programs (WHO 2021).HPV self-sampling testing is reported to have similar sensitivity and specificity for detecting pre-cancer lesions to samples collected by physicians and is considered a safe and acceptable screening method by women (Racey and Gesink 2016).A large, randomised, non-inferiority trial conducted in the Netherlands found that there was no statistical difference in sensitivity and specificity between self-sampling and clinician-based sampling for CIN2þ lesions, with relative values of 0.96 and 1.00, as well as for CIN3þ lesions, with relative values of0.99and1.00,respectively (Polman et al. 2019b).A study in China shows that the coincidence rate between selfsampling and clinician-sampling was 97.28% (1071/1101), and the sensitivity of the detection CIN2/3 in the two samplings was 100%, the specificity was 83.33% and 90.4%, respectively (Li et al. 2018).
HPV self-sampling testing allows women more autonomy and privacy and is reported to facilitate screening participation rates, especially in under-screened populations.Many studies proved that self-sampling testing can improve the uptake rate of cervical cancer screening and can be an alternative primary screening for cervical cancer (Fokom et al. 2020, Polman et al. 2019a).A cross-sectional study showed that most participants (>80% of 725) thought vaginal self-sampling was easy, convenient, not embarrassing, and comfortable, and were confident in achieving the test, with 83% preferring self-sampling testing to clinician-sampling testing (Khoo et al. 2021).According to an investigation of the cervical cancer screening population in China, 83.27% (1145/1375) of the participants were willing to choose self-sampling for cervical cancer screening (Zhao et al. 2019).A survey in Yunnan province, China showed that more than 90% of the 600 rural women surveyed understood the self-sampling instructions and felt comfortable carrying out the self-sampling procedure, and 64.8% of them preferred self-sampling to have the clinician obtain the sample (Goldstein et al. 2020).While some studies were on the contrary, a survey of 1810 women in China found that only 29.94% of them had heard of HPV, and 42.32% accepted HPV self-sampling as a method of cervical cancer screening (He and He 2020).Another study investigated 544 women about awareness of HPV selfsampling in Shenzhen, China, showing that only 12.5% of them had heard of and done self-sampling for cervical cancer screening and 81.6% was apparently inclined to be sampled by doctors (Zhou et al. 2016).It's not clear whether HPV self-sampling testing is acceptable for women in Jiangsu province.In this study, we aim to investigate women who have done self-sampling about knowledge about HPV along with their experience and acceptability of self-sampling, analysing factors that influence the acceptability.

Methods
Women in Jiangsu province who completed HPV self-sampling testing by purchasing an HPV self-sampling kit through the HPV self-sampling testing platform 'Checknet' were invited to complete a questionnaire about knowledge about HPV and cervical cancer along with their experience and acceptability of self-sampling.We chose women who could read and understand the questionnaire.Women over 25 years of age were included as recommended by China's 2017 guidelines (Chinese Preventive Medicine Association Women's Health Branch 2017).Women who were diagnosed with cervical cancer or pregnant were excluded.
The sample size was calculated based on a previous study in China, 64.8% of women who had experienced HPV selfsampling screening preferred self-sampling to clinician sampling (Goldstein et al. 2020).Assuming a 20% non-response rate with a confidence level of 99% and 10% confidence interval width, the estimated sample size for this survey was calculated as 775 women.
A questionnaire was designed based on literature and expert consultation for this study.It consisted of four parts.
The first part was about demographic characteristics, marital status as well as the history of cervical cancer screening.The second part comprised questions on knowledge about HPV.Likert scale was used in this part, which falls into five response categories, scoring from 1 to 5 points for each question with a total score of 7-35 points.The third part also used a five-point Likert scale to survey their experience of HPV self-sampling procedure and the last two questions are scored in reverse, with a total score of 9-45 points.The last part was about the acceptability of HPV self-sampling screening.The questionnaire was given to the participants through the HPV self-sampling test platform, along with a cover letter explaining its purpose.All questionnaires were anonymous, ensuring confidentiality.
EXCEL and SPSS version 25.0 were used to manage and analyse the Date.Descriptive statistics were used to evaluate demographic characteristics, knowledge about HPV, self-sampling experience, and acceptability.The continuous independent variable was checked for normality assumption and described by the average and standard deviation or median and quartile.The categorical variable was described by number and percentage.Chi-square and ANOVA were used for categorical and continuous variables respectively to look for significant factors associated with willingness to choose HPV self-sampling for cervical cancer screening.P < 0.05 was considered statistically significant.
This study received ethics approval and consent from the Ethics Committee, Sir Run Run Shaw Hospital, Zhejiang University School of Medical, Hangzhou, China (approval number: Research20200415-38).

Results
A total of 1000 questionnaires were sent out and 862 valid questionnaires were collected.The age of the 862 participants was between 25 to 63 years old, with an average age of 41 ± 6.9.Of the 862 participants, 71% (613) were in normal BMI, and 29% (249) had a bachelor's degree or above.Most of these women were married (92%).Workers (468, 54%) and company employees (237, 27%) made up the majority of the participants.A majority of women (720, 84%) participated in cervical screening within one year.Still 3.8% of women (33/862) had never participated in cervical cancer before.These demographic characteristics were shown in Table 1.Participants had high acceptability for HPV self-sampling with a mean score of 4.2 (95% confidence interval [CI], 4.1-4.2) out of 5 points.As for the preference for cervical cancer screening methods of the 862 participants, 231 (27%) preferred clinician-sampling, 286 (33%) preferred self-sampling, other 345 (40%) expressed no preference.Reasons for choosing clinician sampling include more accurate sampling (398/576), can see more information about the disease (435/576), more reliable results (256/576), and traditional concepts (85/576).Protecting privacy (424/631), convenience, and ease of operation (509/631) as well as avoiding pain and fear of female inspection of a speculum (327/631) were the main reasons for choosing self-sampling.Most participants (641, 74%) were willing to recommend HPV self-sampling to others.The biggest concern about HPV self-sampling of the participants included specimens' spoilage (600, 70%), incorrect sampling (573, 67%), can't get results in time (191, 22%), can't understand results (145, 17%), unreliable results (228, 27%) and so on.
Knowledge about HPV and experience of HPV selfsampling among the study participants were shown in Table 2.A total of 862 participants got an average score of 25 (95% confidence interval [CI], 24.6-25.4),ranging from 7 to 35 points.The question 'Do you know the procedures for handling HPV positivity?' got the lowest score of 3.1 (95% confidence interval [CI], 3.1-3.2),while 'Do you know the HPV self-sampling procedure?' got the highest 3.8(95% confidence interval [CI], 3.7-3.9).The average score of the experience of HPV self-sampling among the 862 participants was 35.2 (95% confidence interval [CI], 34.8-35.6),with a range of 17 to 45. Acceptability of self-sampling was associated with income (P ¼ 0.020, Table 3), time of last cervical cancer screening (P < 0.001), the result of latest cervical cancer screening (P ¼ 0.003), knowledge about HPV (P < 0.001) and experience of self-sampling (P < 0.001).Preference for the method of cervical cancer screening was associated with age (P ¼ 0.048, Table 3), education background (P < 0.001), the result of the latest cervical cancer screening (P ¼ 0.047), as well as experience of HPV self-sampling (P < 0.001).BMI, marital status, childbearing history, income, occupations along with knowledge about HPV and cervical cancer hadn't statistically significant effects on preference for self-sampling (P <0.05).

Discussion
In this study, participants who had no preference (40%) for the methods of cervical cancer screening accounted for the largest, the second was self-sampling (33%), and the last was clinician-sampling (27%).73% (631/862) participants of in our study were willing to choose HPV self-sampling for cervical cancer screening.This is middling compared to other surveys.69% of women were acceptable for HPV self-sampling in Hong Kong, China (Wong et al. 2020), and more than 90% of women in a study in Italy (Abdullah et al. 2018).A meta-analysis showed that 97% (95% CI: 95-98%) women considered HPV self-sampling to be an acceptable method for cervical cancer screening, and 59% (95% CI: 48-69%) women reported preferring self-sampling for cervical cancer screening (Nelson et al. 2017).In general, women who have experienced self-sampling have a high acceptance of self-sampling.About 89% of participants in this study got a score of �28 in the experience of HPV self-sampling, and 49% of them got a score of �36.Participants were satisfied with the instructional videos and user guide and report query method and expressed that self-sampling was convenient, private, and less painful and embarrassing.Similar to the results of relevant studies, a study showed that compared with clinician-sampling, women reported significantly less discomfort, pain, tension and shame in self-sampling (Polman et al. 2019a).A survey in Bolivia showed that 86.9% of women said that the operation of cervical HPV self-sampling was simple, comfortable and painless (Allende et al. 2019).Surveys in Thailand, the United States, Australia (Sultana et al. 2015, Winer et al. 2016, Phoolcharoen et al. 2018) also showed that compared with clinician-sampling, more than 80% of women said that the overall experience of self-sampling was good, and self-sampling was more convenient, less embarrassing, no discomfort or little discomfort.All above, the cervical HPV self-sampling operation is simple, the participants have a good experience, and can avoid the embarrassment of clinician-sampling and the pain and discomfort experience of insertion of a speculum, which provides a good basis for the promotion of cervical HPV self-sampling in China in the future.
According to this survey, 576 women chose clinician sampling in that 'more accurate sampling', 'doctors can see more information about diseases', 'more reliable results' and 'traditional concept'.And the biggest concern about HPV self-sampling of the participants includes 'specimen spoilage', 'incorrect sampling', 'can't get the result in time', 'causing bodily harm', as well as 'unreliable results' and so on.Women in other studies express similar concerns about self-sampling.They cared about whether they carried out self-sampling correctly and about the accuracy of the test result (Williams et al. 2017, Nishimura et al. 2021).A review (Marshall et al. 2019) indicated that elaborating the procedure of cervical HPV self-sampling to participants in advance and discussing and resolving problems that they might encounter could minimise fear and uncertainty and improve their acceptance of HPV self-sampling.Therefore, more detailed and easier operation instructions such as illustrations and video tutorials would be needed to increase the acceptability of HPV selfsampling in Chinese women.Meanwhile, the medical staff participate in providing guidance and consultation services to solve the problems and doubts of the participants and improve their confidence in self-sampling.
This survey found that women who participated in cervical cancer screening within one year had higher acceptability than others.This is different from other studies, in which women who were under-screened or had never been screened were reported to prefer HPV self-sampling more than those screened regularly (Chaw et al. 2022).This may be because most participants in this study had screened for cervical cancer in two years and few participants had never done it before.We also found that women with a high level of knowledge about HPV and with a good experience with HPV self-sampling were more acceptable for self-sampling.But in this study, still, 30% of participants had a score of �21 in knowledge about HPV and cervical cancer, and only 24% got a score of >28.A study investigated 6576 Chinese women who were screened for cervical cancer and found that 38.14% of them had no knowledge of cervical cancer and only 13.73% had good knowledge of cervical cancer (Liu et al. 2022).To increase self-sampling acceptability and participation in China, it's important to improve knowledge about HPV and raise public awareness of cervical cancer screening as well as optimise self-sampling testing procedures (Madzima et al. 2017, Zhang et al. 2021).
More than 90% of participants in this study had screened for cervical cancer in two years.The screening rate was higher than the average rate in China.That may be because we recruited women who had voluntarily purchased the selfsampling kit for HPV testing.These women may be more motivated to participate in cervical cancer screening.As people pay more and more attention to health, the participation rate cervical cancer screening in China is increasing year by year.However, due to the large population and the huge workload of cervical cancer screening, the current model of screening by doctor sampling in China can't meet the demand.Meanwhile, some rural women in China were affected by their educational background, age, traditional concepts, poor transportation services and lacked cognition of cervical cancer screening, the initiative and participation rate of cervical cancer screening low (Huang et al. 2020).And the coverage rate of cervical cancer screening in rural women of childbearing age was 52.6% in 2017 in China (Committee 2019).HPV self-sampling was reported to have good specificity and sensitivity compared with clinician sampling.It can be used for cervical cancer screening, improving cervical cancer screening rate, especially among women in remote or those who have never screened.HPV self-sampling is not widely known in China, and some self-sampling testing platforms only provide testing services, lacking consultation and follow-up management.To promote the application of HPV self-sampling in China, it's necessary to improve the public's awareness of HPV self-sampling, optimise the self-sampling procedure and improve the function of the testing platform.

Limitations
This study had some limitations.We recruited women who had voluntarily purchased the self-sampling kit for cervical cancer screening.Only a handful of participants were more than 5 years or had never attended a cervical cancer screening.Second, most of the participants we recruit had a high academic qualification, restricting us to evaluate the attitude to HPV self-sampling of women with less education.

Conclusions
This study found that most women had high acceptability for HPV self-sampling in in Jiangsu province, China.High knowledge about HPV as well as good experience of self-sampling can improve the acceptability for self-sampling in women.HPV self-sampling can be used to improve cervical cancer screening coverage and participation rates in China.

Table 1 .
Demographic characteristics and acceptability of HPV self-sampling.M 25 : the first quartiles, M 75 : the third quartiles.
a Body Mass Index.b

Table 2 .
Knowledge and experience about HPV self-sampling.

Table 3 .
Factors associated with acceptability and preference for cervical cancer screening methods.
a v 2 test.b Fisher's exact test.c ANOVA test.d Non-parametric test.