Introduction
Cervical cancer poses a significant health burden in low- and middle-income countries, including those in Southeastern Asia [
1,
2]. In Thailand, it ranks as the third most prevalent cancer. An age-standardized incidence rate, in 2020, was 16.4 per 100 000 women [
3]. Cervical cancer accounts for nearly 10% of all new cancer cases in Thailand [
3]. Moreover, the age-standardized mortality rate associated with the disease in Thailand (7.4 deaths per 100 000 women) [
3] is relatively high compared to that of high-income countries [
4]. The global strategy for cervical cancer elimination [
2] has three key goals. They are (1) vaccinate 90% of girls against human papillomavirus (HPV), (2) screen 70% of eligible women, and (3) provide treatment to 90% of women with positive screening results [
2]. The screening policy remains a crucial strategy for cervical cancer elimination in Thailand. However, Thailand’s national immunization program only began including the HPV vaccine for Thai girls in 2017 [
5]. The current HPV vaccine coverage is meager. The experts in the field mentioned that due to the vaccine supply shortage during the COVID pandemic, only approximately 10% of Thai girls received at least one dose of the vaccine.
Thailand’s national cervical cancer screening policy encompasses various screening methods, including the Papanicolaou test (Pap smear), visual inspection with acetic acid (VIA), and HPV DNA testing [
6]. These methods are provided free of charge to Thai women as part of the national screening program. Despite the implementation of this policy in 2005, the desired reduction in cervical cancer incidence and mortality has not been adequately achieved [
6]. The success of cervical cancer screening policies in low- and middle-income countries is hindered by several factors, including the performance of screening tests and the screening rate. The World Health Organization (WHO) recommends using HPV DNA testing as the primary screening test in the general population due to its superior screening accuracy. In 2020, Thailand transitioned from Pap smears and VIA to HPV DNA testing as the primary screening test for Thai women [
6].
The screening rate in the United States exceeds 80% among eligible women [
1]. In comparison, low rates have been reported in many low- and middle-income countries [
7]. The maximum adequate screening rate (i.e., screening at least once in 3 years) in Thailand has reached only approximately 40% [
5]. Previous studies have identified various factors that influence screening rates. For instance, women are more likely to undergo screening if the costs are low or free. Their knowledge levels regarding the disease and the importance of screening are also significant determinants [
8]. The nature of clinician-collected samples used in traditional screening methods, which lack privacy and can cause embarrassment, pain, and discomfort, also poses significant barriers to screening for Thai women [
8]. In this regard, we surmised that self-collected samples for HPV DNA testing would encourage more women to undergo screening.
The WHO guidelines [
2] and numerous studies [
9‐
11] have confirmed that self-collected samples for HPV DNA testing yield screening accuracies comparable to those of clinician-collected samples. In January 2022, Thailand initiated a pilot screening campaign that introduced self-collected samples for HPV DNA testing as an option for Thai women who prefer not to undergo clinician-performed screening. This pilot project covered the screening costs for approximately 80 000 women. However, in the long run, including this additional benefit package would incur higher costs and impact the country’s budget allocation strategy. Therefore, conducting an economic evaluation and budget impact analysis is crucial to confirm the feasibility and sustainability of this policy. Consequently, this study aimed to evaluate the cost-utility and budget impact of cervical cancer screening using self-collected samples for HPV DNA testing in Thailand.
Discussion
This study in Thailand focused on evaluating the cost-effectiveness and budget impact of a cervical cancer screening policy that utilizes self-collected samples for HPV DNA testing. Past research consistently demonstrates that HPV DNA testing offers a superior screening quality compared to the Pap smear and VIA methods [
2,
36]. However, traditional clinician-based screening methods pose obstacles such as embarrassment, inconvenience, pain, and discomfort [
37].
Our analyses revealed that implementing additional home-based self-collected samples for HPV DNA testing yielded the lowest total lifetime cost and the highest QALYs among the three policy options examined. Therefore, from a societal perspective, the policy involving additional home-based self-collected samples for HPV DNA testing proved to be cost-saving and the most favorable option.
Furthermore, our study demonstrated that screening women aged 25 and above resulted in more benefits than restricting screening to women aged 30 and above. Screening women aged 25–65 could prevent an additional 2.5 cervical cancer cases per 100 000 women while incurring an extra cost of only around 90 000 Thai baht annually. Moreover, this policy would save approximately 7.4 million Thai baht annually in cancer prevention costs (screening and CIN1–3 treatment costs) compared to limiting screening to women aged 30–65.
Our sensitivity analyses consistently indicated that the policy of additional home-based self-collected samples for HPV DNA testing remained the best option, even when considering the lowest expected screening rate achievable through self-sampling. These results emphasize the effectiveness and cost-saving potential of implementing the self-collected samples for HPV DNA testing for cervical cancer screening in Thailand.
Our findings align with previous studies that have examined the cost-effectiveness of self-collected samples for HPV DNA testing. For instance, a study in Switzerland found that self-collected samples for HPV DNA testing among nonattendees were cost-effective and reduced cervical cancer cases and related deaths [
38]. Additionally, a systematic review investigated cervical cancer screening in low- and middle-income countries, examining seven studies on self-collected samples for HPV DNA testing [
39]. The review findings revealed the cost-effectiveness of self-collected sample for HPV testing when a higher population coverage was achieved than with other screening methods.
While we did not have direct evidence of screening rates using self-collected sample kits for nonattendees, substantial evidence supports the idea that self-sampling increases population screening coverage [
38‐
40]. The utilization of self-collected samples for HPV DNA testing can help overcome barriers such as embarrassment, inconvenience, price, and test reliability. A study in Thailand showed that the self-sampling was widely accepted even among Muslim women, who constitute approximately 3% of the Thai female population [
8]. Moreover, multiple studies and guidelines have confirmed the reliability of HPV DNA testing using self-collected specimens.
Given that HPV DNA testing is covered by all health benefit coverage schemes in Thailand, cost should not hinder screening for Thai women. While some women prefer to collect samples at home, several studies have indicated that others opt to collect their samples in a healthcare setting. This preference stems from valuing the presence of healthcare workers who can offer information and address any concerns they may have [
8,
41]. Our study suggests that, from a societal perspective, the outcomes of home-based and healthcare setting-based specimen collection are comparable. Therefore, a revised policy should not be limited to a single strategy but should offer both options to maximize screening rates. However, it is essential to note that self-collected samples for HPV DNA testing cannot wholly replace clinician-based screening methods. The latter methods provide additional benefits by checking for other gynecological problems and catering to women willing to undergo clinician examinations.
Regarding the budget impact, our projections indicate that implementing a policy of self-collecting samples would result in an additional cost of approximately 661–681 million Thai baht per year. While this amount is slightly higher than the current budget for clinician-collected sampling, it is crucial to consider the potential benefits. If the policy permitting the self-collected samples for HPV DNA testing is implemented nationwide, an estimated additional 10 million women will undergo cervical cancer screening, over 10 years. Moreover, the policy will result in the prevention of at least 2000 cases of cervical cancer and 1500 cancer-related deaths per year in addition to what would be achieved by implementing clinician screening alone.
Furthermore, our study demonstrates that initiating screening at the age of 25 would prevent more cervical cancer cases than starting from 30. If there are budgetary, healthcare-workforce, or practicality constraints, the WHO suggests prioritizing cervical cancer screening for women who have never been screened, underscreened women, and women living with HIV [
2]. It is important to highlight that the incidence of cervical cancer differs between women aged 25–29 and those above 30. The rates are 6.9 cases per 100 000 women for women aged 25–29 and 12.6 cases per 100 000 women for those above 30 [
35]. Additionally, younger women have a higher probability of regression from HPV infection and CIN1 to the normal stage [
2,
21]. Policymakers should consider this information when making decisions about cervical cancer screening policies.
To our knowledge, this study represents the first investigation in Thailand to evaluate the cost-effectiveness and budget impact of self-collected samples for HPV DNA testing. A previous study in Thailand focused on comparing the cost-effectiveness of HPV DNA testing using clinician-collected samples to Pap smears [
36]. It was concluded that HPV DNA testing was Thailand’s optimal primary cervical cancer screening strategy. Despite the availability of coverage for all screening methods under health benefit schemes in Thailand, the current screening rate among Thai women still needs to be improved.
Our study supports the inclusion of self-collected samples for HPV DNA testing within health benefit coverage schemes, as this option offers greater benefits in cervical cancer prevention. Moreover, allowing self-collected samples for HPV DNA testing proves to be cost-saving compared to relying exclusively on clinician-collected samples.
Several factors contribute to the reliability and contextual relevance of our findings. First, the study framework involved obstetricians, gynecologists, and policymakers from the outset. Second, input parameters were primarily derived from systematic reviews and meta-analyses. Third, sensitivity analyses were conducted by varying parameters in the model, consistently confirming that using additional self-collected samples for testing was the optimal choice. Furthermore, the incremental cervical cancer prevention benefits associated with self-collected samples were observed at any additional screening rate achievable through self-sampling. Last, local data were incorporated into the analyses, ensuring that the results directly apply to policy decisions. Local data on screening rates, cancer incidence, and related costs were utilized whenever possible to ensure that the study accurately reflected the Thai context.
Several limitations to our study should be acknowledged. First, using an economic model with various assumptions have some drawback features. For an example, various individualized characteristic of target women in the virtual cohort could not be captured in the model. Moreover, due to limited data availability, we had to make assumptions about the additional screening rate that would be achieved using self-collected samples. However, sensitivity analyses demonstrated that incremental cervical cancer prevention benefits associated with self-collected samples were observed regardless of the specific additional screening rate.
Second, all costs related to adverse events were included in the treatment costs. Unfortunately, we could not separate these costs from the total treatment costs due to limitations in the database structure. The treatment costs were also based on data from only one university hospital. However, sensitivity analyses indicated that varying the treatment costs within plausible ranges did not change the interpretation that using self-collected samples for testing was the optimal policy option.
Third, our analyses focused on unvaccinated women only since the vaccine coverage rate among Thai girls and women is meager. As more information becomes available on the impact of HPV vaccines, it will be essential to update the model accordingly.
Fourth, our model did not consider the additional benefits of pelvic examinations, such as the detection of other genital disorders. Therefore, we recommend that self-collected samples for HPV DNA testing should be an option only for women who are unwilling to undergo screening by a clinician.
Fifth, various HPV tests with different sampling tools and analysis methods are available on the market, and their performance can vary. Meta-analyses have shown that the accuracy of self-collected samples for HPV testing is lower than that of clinician-collected samples, except when PCR-based DNA detection is used as the analysis method [
10]. Thereby, we suggested that PCR-based assay should be an only technique used for analyzing self-collected samples. A systematic review has also suggested that self-collected samples are more cost-effective than clinician-collected samples only when screening coverage increases [
39].
Additionally, there are challenges associated with self-collected samples for HPV DNA testing in our specific context. Our models did not consider the costs of public relations, provider training, and patient education. The average level of education in the Thai population is likely to be lower than in countries with a better economic status. Consequently, Thai women may require more guidance to understand the importance of cervical cancer screening. The quality of self-collected samples could also be affected by relatively lower levels of education, potentially leading to sampling errors, delivery errors, and contamination. As a result, our model may have underestimated certain costs associated with policy implementation. However, these fundamental costs related to policy implementation are expected to be short-term.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.