Women's preferences for cervical cancer screening methods: a discrete choice experiment in Estonia and France.

Women’s preferences for cervical cancer screening methods: a discrete choice experiment in Estonia and France - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Womens Health . 2026 Apr 22;26:283. doi: 10.1186/s12905-026-04478-z Women’s preferences for cervical cancer screening methods: a discrete choice experiment in Estonia and France Violette Delisle 1, ✉ , Lise Rochaix 2 , Marc Bardou 3 , Anneli Uusküla 4 , Anna Tisler 4 , Carine Franc 1 Author information Article notes Copyright and License information 1 CESP, Inserm UMR-U1018, UVSQ, Université Paris-Saclay, 16 Av. Paul Vaillant Couturier, Villejuif, 94800 France 2 Paris School of Economics (Hospinnomics), Paris 1 Panthéon-Sorbonne, Paris, France 3 CIC1432, CHU Dijon Bourgogne and Université Bourgogne Europe, Dijon, France 4 Institute of Family Medicine and Public Health, University of Tartu, Tallinn, Estonia ✉ Corresponding author. Received 2025 Oct 29; Accepted 2026 Apr 13; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13235200  PMID: 42021337 Abstract Background This study explores women’s preferences for cervical cancer screening (CCS) in Estonia and France, particularly preferences for self-sampling (SS). Although many European countries have introduced organized CCS, screening behaviours still exhibit a strong social gradient, with CC disproportionately affecting populations facing barriers to access, such as low income, rural residence, migration status or low education. The WHO has set a goal of eliminating CC and has proposed different strategies of invitations, outreach campaigns and the introduction of SS as a potential lever to increase uptake. Methods To assess whether these CCS strategies align with women’s preferences, we conducted a discrete choice experiment (DCE) in Estonia and France. Pooling our data ( N = 4774), we used a two-stage decision-making framework; first, intervention preferences were assessed, and then the willingness to participate if preferred interventions were implemented was measured. We also conducted a heterogeneity analysis via latent class analysis. The DCE has three attributes: invitation methods, sampling methods (physician-based sampling or self-sampling), and awareness-raising campaigns. The results were analysed using a mixed multinomial logit model to capture heterogeneity in preferences and explore factors influencing CCS participation. Results The results show that the implementation of SS would: 1) increase uptake among women facing geographical barriers to CCS, 2) help engage first-time screeners in a screening pathway and 3) possibly deter participation from women already engaged in their screening pathway if implemented as sole mean of screening. Conclusions Our results thus advocate for tailored screening programs. Supplementary Information The online version contains supplementary material available at 10.1186/s12905-026-04478-z. Keywords: Discrete choice experiment, Prevention, Cervical cancer screening, Self-sampling Introduction Cervical cancer (CC) is the fourth most common cancer in women worldwide and ranks eleventh in the European Union (EU). However, significant disparities among EU countries exist, with evidence showing that cervical cancer incidence and mortality rates are socially stratified [ 1 , 2 ]. While highly educated women experience similar mortality risks across Europe, socioeconomic factors such as low education and low socio-economic status contribute to geographical differences both between and within countries [ 3 ] and these inequalities tend to grow [ 4 , 5 ]. These disparities translate into unequal availability, access and uptake of effective cervical cancer screening (CCS) programs. CC prevention relies both on primary prevention, Human Papillomavirus (HPV) vaccination, and secondary prevention via CCS, using either cytology-based or HPV-based-testing.CCS uptake is a multi-stage process, beginning with awareness and knowledge of the disease, its risk factors, and the benefits of screening. This is followed by the development of supportive beliefs and attitudes, including perceived susceptibility and severity, and a favorable perceived cost–benefit analysis. Accessible and inclusive information and access to screening services through effective communication and convenient options play an important role. Thus to improve CCS uptake, screening programs need to be tailored to address individual barriers to participation [ 6 ]. Organised screening programs have been shown to effectively reduce incidence of cervical cancer and its mortality, up to 80% [ 7 ]. This study is part of the CBIG-SCREEN project, which aims to improve CCS among vulnerable and underserved populations in Europe. The project seeks to design and implement strategies that align with women’s preferences in four countries- Estonia, France, Portugal and Romania. Our work contributes to this objective by informing the development of screening strategies in Estonia and France, targeting different groups: women who are distant from the healthcare system in Estonia, and women facing economic vulnerabilities in France. The value of this comparative approach lies in examining distinct populations within differing healthcare contexts, thereby identifying strategies that are either country-specific or potentially transferable across settings. CCS in Estonia Estonia has implemented national programs for cervical cancer prevention, including both HPV vaccination and organised CCS. HPV vaccination was introduced in 2018 and has achieved 57% in the target adolescent population [ 8 ]. For secondary prevention, Estonia has adopted a universal, fully organized CCS program targeting women aged 30–65. HPV testing is recommended every five years, and the screening is provided free of charge to all women in the target group, irrespective of their insurance status. Invitations are issued electronically through the Estonian health portal by the Cancer Screening Registry or sent by postal mail for individuals not registered on the health portal. Although screening participation has been increasing, it remains moderate, with only 50.6% of eligible women attending in 2021 and Estonia continues to record one of the highest cervical cancer mortality rates in the EU [ 9 ]. Fear of screening, often related to anticipated pain or discomfort, concerns about the procedure itself, and fear of receiving a cancer diagnosis, as well as lack of awareness of cervical cancer risk factors continue to be important barriers to participation [ 10 ]. CCS in France France introduced HPV vaccination for girls, aged 12 to 14 years, in 2007, and extended the recommendation to boys in the same age group in 2021. Vaccination covers around 37% of the target female population in 2023 [ 11 ]. CCS is organised around a hybrid system that combines opportunistic and organised screening. Opportunistic screening occurs during routine medical visits, while the nationwide organised program specifically targets women who are overdue for screening and relies on the regional cancer screening coordination centres to identify them (based on medical administrative data) and send them invitation letters through mail. Since January 2024, this responsibility has transitioned to the National Health Insurance. Under the organised screening program, both the sample collection and the laboratory analysis are fully covered, whereas in opportunistic screening, out-of-pocket costs may arise depending on the type of health professional consulted and the individual’s insurance coverage. However, financially vulnerable individuals in France generally benefit from Complémentaire Santé Solidaire (CSS), i.e., supplementary public universal health insurance coverage that also prohibits professionals from charging extra fees. Cytology is recommended at ages 25, 26 and 29, followed by HPV testing every five years from ages 30 to 65. During the period 2020–2022, 59.5% of women performed the test within the recommended intervals, representing an increase compared to 2017–2019 when coverage reached 56.6% [ 12 ]. Screening in both countries remains below the WHO’s recommended 70% participation threshold for CCS [ 13 ]. Across settings, participation in cervical cancer screening is known to be influenced by individual beliefs and barriers as well as structural characteristics of screening programmes. The Health Belief Model [ 14 ] highlights the role of awareness, knowledge, perceived susceptibility, and perceived barriers, while Andersen’s behavioural model emphasises how enabling features of health services shape access to preventive care [ 15 ]. These approaches support examining how modifiable programme characteristics may act as barriers or facilitators of screening participation . Numerous studies highlight persistent barriers to CCS, particularly influenced by socioeconomic factors such as low income and education, which are associated with lower CCS uptake [ 16 ]. Lack of awareness and low health literacy also hinder screening participation [ 10 , 17 ]. These challenges are exacerbated by geographical inequalities, especially in rural areas, where limited healthcare access and a shortage of medical facilities further reduce CCS rates [ 18 , 19 ]. The WHO [ 13 ] has identified several key strategies to boost participation and effectively reach women who do not currently take part in CCS: Self-sampling (SS) allows women to collect their own biological sample for HPV testing, using a swab or brush to collect cells from the cervix, which can then be tested for HPV. This method appears to be the most innovative lever offering greater flexibility to reach populations living in rural/remote areas [ 20 ] and to reduce travel and time costs [ 21 ]. Compared with physician-based sampling (PBS), SS offers advantages such as ease of use, privacy [ 22 , 23 ] and reduced stress [ 24 ]. While SS has been shown to be effective in increasing the uptake of CCS among never- or underscreened populations [ 25 , 26 ], this is highly dependent on implementation modalities [ 22 , 24 , 27 ]. Qualitative studies also highlight the perceived risk that the sample might be collected incorrectly [ 23 , 28 ] that the results might not be accurate. Invitation to screening: Personalized invitation letters with scheduled appointment details are known to significantly increase screening participation rates, with further gains achieved by adding a reminder one-year later [ 29 ]. A three-step invitation process (automated messages/calls, manual calls, face-to-face interviews) in Portugal increased uptake by 17% compared to a standard letter [ 30 ]. The method of invitation to screen has been shown not only to influence PBS but also SS uptake [ 27 ]. Health information and education also represent key dimensions for increasing CCS uptake [ 6 , 31 ]. To increase the impact of CCS policies by increasing awareness, community campaigns such as door-to-door campaigns, community outreach, and media support are useful [ 20 ]. Patient navigators, which aim to guide individuals through their health care journey, facilitate the delivery of essential and personalized prevention information [ 32 ] and have also proven effective for women with limited native language proficiency [ 33 ]. Health professionals travelling to rural areas through mobile units have also proven to be effective, as they can provide adequate information to remote communities, perform screenings, and distribute self-sampling kits [ 34 ]. Thus, a tailored approach that adapts invitation modalities, personalizes reminders according to the screening strategy proposed, i.e., SS versus PBS and enhances awareness is needed to increase uptake to CCS. The key remaining question is: which targeted interventions best align with women's preferences? Few studies have considered the consistency between the design of these programs and women's preferences. Using a discrete choice experiment (DCE), one study [ 35 ] reported a preference for SS among young and never-screened women in the UK. Some DCEs have also been conducted on the appropriateness of SS for CCS in low- and middle-income countries [ 36 , 37 ], surveying women in clinics and mobile units and reporting a preference for PBS over SS in Zambia and South Africa, respectively. More recently, another DCE conducted among rural women in Zimbabwe, highlighted preferences for performing the SS at a health facility (rather than at home), as well as preferences for face-to-face counselling and education over written materials [ 38 ]. Considering these barriers and facilitators to improving CCS, we designed a DCE to better understand the trade-offs made by these women and to identify actionable strategies that could enhance screening uptake. Indeed, this method highlights women’s relative propensity to forego certain level of attributes in order to obtain others. We aim to assess whether these levers, and SS in particular, are in line with women's preferences and whether they are effective in improving overall CCS compliance or, on the contrary, whether their introduction could lead to a disincentive effect and reduce participation. Discrete choice experiment Several preparatory steps informed the design of the DCE. While the main goal of the DCE was to inform strategies to increase CCS for specific women in European countries, it focuses on attributes and levels that could concretely be implemented; for instance, as the collection of biological samples by mail is not feasible in all countries. Results from the CBIG-SCREEN-research project, including studies on screening barriers and qualitative interviews [ 22 , 39 ] and a comprehensive international literature review were combined to identify the most relevant attributes and levels. The final selection of attributes and levels was determined by the project’s scientific committee. Given the central role of the method of screening, the DCE was essentially designed around SS and PBS. The inclusion of the method for returning the kit (e.g. via mail or through a healthcare professional) was discussed but ultimately not included in the DCE, due to country-specific implementation constraints. Instead, an additional question was added to the questionnaire to address this issue. Furthermore, the mode of kit reception, whether directly or following an opt-in process, was considered but not retained in the final implementation. The option of performing SS with the assistance of a healthcare professional, initially proposed as an attribute, was also explored and ultimately included as an additional question. The decision to keep the DCE with only three attributes, was driven by the desire to minimize cognitive burden, particularly given the strong focus in France on reaching vulnerable populations. The cost of screening was not included as an attribute because, within the framework of organised screening, there are no direct costs to participants, and the implementation phase did not involve any costs either. Additionally, to ensure comparability of samples, the DCE needed to be identical in both countries, offering alternatives applicable to both contexts. Attributes and levels Our three attributes are: 1) the sampling method, 2) the invitation and reminder modes and 3) the information campaign (Table 1 ). The choice of levels was guided by an analysis of the literature and feasibility during the implementation of the CBIG-SCREEN project. No cost attribute was added, as screening is assumed to be free of charge within the context of this DCE. Table 1. Attributes and levels selected for the DCE Attribute Levels Sampling Method Physician-based-sampling (PBS) Self-sampling (SS) Invitation and Reminders Postal Mail Email SMS Digital Portal (E-portal) Information Campaign Social Media Leaflets and Posters in your community Door-to-Door Health Mediators Free Prevention Workshops Open in a new tab The DCE design The DCE is composed of 9 different choice sets, in which respondents have to choose between two scenarios that are combinations of the different levels of the three attributes. An example of a choice set is provided in Appendix A. The experimental design was generated using SAS, producing a D-efficient fractional-factorial design. Among the candidate designs evaluated, the final solution achieved a D-efficiency score of 81.9, indicating a statistically efficient design with good level balance and variation across choice tasks. Attribute levels were arranged so that alternatives within each choice set differed meaningfully on at least one dimension to avoid identical profiles within the same choice set, ensuring that respondents were required to make trade-offs. All respondents were presented with the same set of choice tasks, and no blocking was applied. The final questionnaire, designed for this experiment, includes questions on previous experience with CCS, the choice sets, socio-demographic characteristics and the perceived understandability of the questionnaire to assess participants’ comprehension. The English version of the questionnaire is available in the supplementary files. The DCE was pilot-tested among 20 women (10 in each country) from various socioeconomic backgrounds to ensure its accessibility, clarity and to inform the design of the questionnaire and estimate completion time. Based on the feedback, several improvements were made, including the simplification of the vocabulary, the inclusion of graphic aids and the addition of an information sheet detailing the attributes and levels as well as an introductory video providing information on CCS and the different attributes and levels chosen for the DCE, including SS and PBS. In the description of attributes and levels, SS is presented as a procedure that can be completed independently, at home, without the need for assistance. Completion time was estimated between 14 and 17 min. While the DCE has remained consistent across countries, the questionnaire was adapted to each national context and translated into the relevant languages. Data collection and methods Data collection The DCE was programmed using Qualtrics software and distributed in Estonia and France with the objective of specifically targeting women who are remote from the health care system (Estonia), and financially vulnerable women (France), in line with the CBIG-SCREEN project’s objectives. While the choice sets were identical across countries, sociodemographic questions were adapted to each country’s specific context. Participant recruitment relied on different strategies: in Estonia, the National Health Registry was used to distribute the DCE to a random sample of 26,000 women from a general population via email; the initial invitation was followed by two reminders, each sent at three-weeks intervals. In France, the distribution strategy involved both social media and face-to-face interviews, conducted with the support of various NGOs, while leaflets and posters were also distributed in pharmacies, medical facility waiting rooms. The interviews were primarily conducted in two departments in metropolitan France, Paris and Bouches du Rhône, by NGO staff or by research assistants recruited specifically for this purpose. Participants completed the questionnaire on tablets and were able to request assistance from research assistants if they encountered any difficulties. Data collection in France primarily aimed to target vulnerable women, whereas in Estonia, distribution within the general population was intended to enable heterogeneity analysis, thereby enriching the overall analysis. Data were collected from July to December 2023. Informed consent to participate was obtained from all participants. An information sheet outlining their rights and the purpose of the study was made available to all participants, who agreed to them by ticking a box before completing the questionnaire. Only women eligible for CCS in their respective countries were included, specifically those aged 30 to 65 in Estonia and 25 to 65 in France. Our final samples are composed of 201 and 4,573 completed answers for France and Estonia, respectively. This resulted in a response rate of 17.6% for Estonia, while the response rate for France could not be determined due to face-to-face recruitment and social media dissemination. The significant difference in sample sizes is largely due to the different data collection methods employed in each country. We pooled our samples to reflect broader patterns in screening preferences across the different healthcare and cultural contexts. This approach allows to identify commonalities regarding CCS, while a subsequent heterogeneity analysis assesses the extent to which country-specific factors influence these preferences. Estimation method We use a two-stage pairwise choice experiment. The first stage forces participants to choose between two screening programs to elicit preferences for the intervention's attributes and levels. The second stage, the 'adoption preference task', allows participants to decide if they are willing to be screened under the program they chose in the first stage, identifying the determinants of intervention adoption. In the first stage , choice data are analysed within a random utility maximization framework, i.e., a random utility function is assumed for each hypothetical scenario (including a systematic and unobserved portion of utility), and a utility maximization decision rule [ 40 ] is assumed. A mixed multinomial logit (MIXL) model was used, which allows the preference parameters to be randomly distributed across the sample and accounts for (i) unobserved preference heterogeneity and (ii) the correlation of choices between respondents [ 41 ]. MIXL (using 500 Halton draws) was estimated with correlated random coefficients between all three attribute levels, assuming a normal distribution for each parameter . This model also allows for scale heterogeneity, that is, various degrees of decision consistency across women [ 42 ]. For women n , scenario j (j = 1,2) , and choice set t (t = 1,…,9) , the utility U is given by: where represents the deterministic component of utility of woman n, scenario j and choice set t , such that the specification captures the combination of attributes and levels in a linear form, without interactions: are the estimated individual-specific utility weights for attribute k and where the second component, , is given by a random component that is assumed to be an independently, identically distributed extreme Value Type I function [ 43 ], thus leading to the multinomial logit choice specification [ 40 ]. This first stage provides conditional preferences by estimating utilities based on the attributes and levels. It allows us to identify the factors that influence decision-making, providing valuable insights into the trade-off’s respondents are willing to make. The second stage analyses the decision-making process by employing a panel data approach. Each individual evaluates several screening campaigns across multiple choice sets, generating repeated observations for the same respondent. The fixed effect α n is included to control for unobserved, time invariant characteristics specific to each respondent, ensuring that the estimated coefficients capture the influence of the program attributes on choices within individuals across the different scenarios. The dependent variable, denoted as the willingness to participate in the screening campaign, indicates whether the respondent will adopt a specific campaign. This model is estimated using a logistic regression: where α n is an individual fixed effect, capturing any unobserved heterogeneity at the individual level. A heterogeneity analysis was conducted via latent class analysis to identify unobserved subgroups within the population on the basis of response patterns in the DCE. This analysis enables probabilistic clustering, wherein individuals are classified into latent classes representing distinct profiles on the basis of their likelihood of choosing specific answers. We estimated models with an increasing number of classes (from 2 to 6). The optimal number of classes was determined using fit criteria such as the Bayesian information criterion (BIC), log likelihood and entropy [ 44 , 45 ]. To identify the factors influencing the likelihood of belonging to different latent classes, we conducted a multinomial logistic regression. This approach allows us to model the probability that a patient belongs to a specific class based on a set of sociodemographic and clinical characteristics. The selected characteristics included the following: Results Descriptive statistics Descriptive statistics are presented in Tables 2 and 3 . Women in the Estonian sample are relatively older, with 42% aged 50–65 years compared with 8% in France (Table 2 ). A total of 77% of Estonian women have a higher education degree, compared with 67% in France. Employment rates are also high, with Estonian women showing a higher proportion of employed individuals (87% vs. 73% in France). Financial struggle is more common in the French study population, where 11% of respondents “always” face difficulties, compared to Estonia (61% “never” struggle). Urban residence was reported by 83% of the sample in Estonia which is above national average of 70%, as in France where 87% of respondents reported living in urban areas, slightly higher than the national benchmark of 80% [ 46 ]. Owing to the distribution strategy, the study populations in Estonia and France differ significantly, with the latter including more vulnerable women. Table 2. Sociodemographic data for Estonian and French study populations Estonia N = 4,573 France N = 201 Total N = 4,774 Age 25–29 - 34% - 30–39 27% 36% 27% 40–49 31% 22% 31% 50–65 42% 8% 42% Nationality French || Estonian 74% 81% - Foreigner a 26% 12% - Dual Nationality - 7% - Residence In a major city 56% 73% 56% In other urban areas 27% 14% 27% In a rural area 17% 13% 17% Housing status Owner 98% 35% 97% Tenant 1% 62% 3% Free user < 1% 3% < 1% Education Below primary school 2% 3% 3% Secondary school/High school 20% 30% 20% University or technical degree b 78% 67% 77% Employment status Yes 87% 73% 86% Financial struggle c Always 2% 11% 3% Often 4% 24% 5% Sometimes 32% 54% 33% Never 61% 10% 59% Open in a new tab a In Estonia, 88% of the foreigners are Russian b of which 29% have applied or vocational training in Estonia c “Do you have difficulty meeting your basic needs (rent, food, heating, etc.)?” Table 3. CCS practices of women in France and Estonia Estonia N = 4,573 France N = 201 Total N = 4,774 HPV Vaccination Yes 6% 29% 7% No 82% 63% 81% I don’t know 13% 8% 12% Awareness of CCS a Yes 97% 93% 97% Invitation to CCS Yes 83% 44% 82% No 12% 46% 13% I don’t know 5% 10% 5% Information Campaign Awareness Yes 88% 72% 88% Knowledge of cervical smear test Yes 94% 97% 94% Cervical smear Test Yes 89% 72% 88% No 6% 18% 6% I don’t know 5% 10% 6% Up-to-Date Yes 79% 64% 78% No 17% 35% 18% I don’t know 4% 1% 4% Regular gynaecological follow-up Once a year 42% 42% 42% Every 2/3 years 32% 21% 31% When necessary 21% 22% 21% Never 5% 14% 6% Open in a new tab a Awareness was measured after the introductory video, which may lead to overestimated results Reported rates of HPV vaccination remain low (29% in France and 5.7% in Estonia), which is consistent with the relatively recent introduction of HPV vaccination programmes -especially in Estonia (2018). As a result, many women in the study cohorts were already older than the target age when vaccination was introduced and are therefore less likely to have been vaccinated (Table 3 ). Receiving an invitation to CCS is more frequent in Estonia (83%), where the organized CCS is universal, than in France (44%), where only women with late for screening receive a reminder. Almost all the women are aware of the existence of CCS information campaigns and what a cervical smear test is (97% in France, 94% in Estonia). A total of 72% and 89% of French and Estonian women, respectively, have had a cervical smear test performed at least once in their life. However, when asked if they are up to date with their screening, 64% for France and 79% in Estonia reported being current with their screening. This self-reported screening rates are above the national participation estimates, which are around 50% in Estonia and 56% in France. This higher rate in Estonia is likely due to the self-selection of respondents, as the questionnaire was distributed via email. Compared with Estonia (5%), more French women (14%) declare never going to gynaecologists. Note that the respondents' assessment of questionnaire understandability revealed that more than 90% of the respondents found the questionnaire "easy" or "relatively easy" (Appendix B), with no statistically significant differences observed across survey administration modes. Women’s preferences analysis We first examine the results of the DCE in each country and then explore heterogeneity in the responses. Table 4 compiles the results of the two-stage models for the pooled sample. The results for the individual country datasets, consistent with the pooled results, can be found in Appendix C. Table 4. Women’s preferences for CCS attributes and determinants of their willingness to participate Pooled Sample N = 4,774 (A) (B) Preferences for attributes (stage 1) Adoption preference (stage 2) Estimate CI Estimate CI Sampling method (Ref = Physician-based-sampling) Self-Sampling −0.77 *** (−0.80, −0.74) −0.30 *** (−0.41, −0.19) Invitation and Reminders (Ref = Postal Mail) SMS 0.34 *** (0.32, 0.36) 0.05 ** (0.00, 0.10) E-portal −0.32 *** (−0.35, −0.30) 0.02 (−0.05, 0.09) Email 0.61 *** (0.58, 0.62) 0.02 (−0.02, 0.07) Information Campaign (Ref = Social Media) Leaflets −0.35 *** (−0.38, −0.33) 0.11 *** (0.06, 0.16) Workshops −0.43 *** (−0.51, −0.48) −0.01 (−0.07, 0.04) Door-to-Door −1.44 *** (−1.55, −1.33) −0.03 (−0.10, 0.03) Open in a new tab Reading grid: Adopting SS is associated with a lower utility compared to PBS (β = −0.77). This corresponds to a reduction in the odds of choosing SS relative to PBS (OR = 0.46) *** Corresponds to p < 0.001, ** to p < 0.01, and * to p < 0.05, Estimates in log-odds Table 4 (Column A) provides the results of women's relative preferences for different levels and attributes. First, the results show that screening campaigns using SS kits decrease women’s utility by 0.77 compared with those offering PBS (OR = 0.46, p < 0.001). Women prefer to receive invitations by email or SMS rather than by postal mail but prefer postal mail compared to an invitation through e-portal. These results are highly significant ( p < 0.001). Regarding awareness campaigns, social media appears as the most preferred source of information for CCS. In contrast, door-to-door communication is the least favoured approach, with substantially lower odds of being chosen compared with social media (OR = 0.24, p < 0.001). However, preferences exhibit substantial heterogeneity, as reflected by the large and highly significant standard deviations. Table 4 , Column B, presents estimations of the women’s willingness to be screened according to the different features of the program. The willingness to get screened remains high across all choice tasks, although it decreases when participants must choose between two options with SS (Appendix D). Interestingly, we show that CCS programs with SS are associated with a reduced participation compared to programs with PBS (OR = 0.74, p < 0.001). Yet, in two choice tasks, when respondent selected an intervention with PBS over one with SS, they were asked whether the option of SS with guidance from a professional would change their choice. Approximately 20% of respondents who initially didn’t choose the SS intervention indicated they might reconsider their decision if offered professional support for SS. Among women who chose the SS kit, preferences regarding sample return were assessed. Overall, 40% reported a preference for postal return, 26% for return via a clinician, and 35% for return at a nearby facility (e.g. pharmacy or laboratory). Furthermore, our results highlight that inviting women through SMS would increase the probability of screening uptake compared to postal mail (0R = 1.05, p < 0.01). Nothing can be concluded for E-portal and Email. Traditional methods of disseminating information, such as leaflets, have a positive effect in willingness to get screened. Heterogeneity of preferences To explore preference heterogeneity, we use latent class analysis to identify different preference profiles. A 5-class solution was selected as the optimal model (see the selection criteria model in Appendix E). Interestingly, there is a notable gradient in the preferences for SS between the groups (Table 5 ). Table 5. Five women’s preferences profiles Class 1 N = 1,504 (31.5%) a Class 2 N = 905 (19.0%° Class 3 N = 1,022 (21.4%) Class 4 N = 121 (2.5%) Class 5 N = 1,222 (25.6%) Est SE Est SE Est SE Est SE Est SE Sampling Method (Ref = Physician-based-sampling) Self-sampling −3.04 *** 0.11 −1.75 *** 0.31 −0.59 *** 0.13 0.00 0.32 2.61 *** 0.93 Invitation and Reminders (Ref = Postal mail) SMS 0.16 * 0.08 1.90 *** 0.19 0.49 *** 0.08 2.36 *** 0.43 0.13 0.36 Eportal −0.30 *** 0.06 −0.59 *** 0.16 0.78 *** 0.08 1.71 *** 0.38 −0.02 0.13 Email 0.54 *** 0.06 2.08 *** 0.14 0.61 *** 0.12 5.93 *** 1.15 0.98 *** 0.23 Information Campaign (Ref = Social Media) Leaflets −0.34 *** 0.05 −0.81 *** 0.11 −0.48 *** 0.08 3.22 *** 0.71 −0.67 *** 0.13 Workshops −0.52 *** 0.08 −0.60 *** 0.10 −0.11 0.07 1.43 *** 0.33 −2.15 *** 0.37 Door to Door −0.52 *** 0.06 −2.23 *** 0.14 −1.23 *** 0.09 5.04 *** 1.06 −1.83 *** 0.22 Open in a new tab Reading grid: Adopting SS is associated with a substantially lower odds of being chosen by women in C1 compared with PBS (OR = 0.05, p < 0.001) a Percentage of the total sample *** Corresponds to p < 0.001, ** to p < 0.01, and * to p < 0.05 Estimates in log-odds While women in C1 strongly reject SS (OR = 0.05, p < 0.001), they are more likely to be up-to-date in their screening and visit their gynaecologist every year, thus being accustomed to having their CCS performed by a health care professional (Table 6 ). These women, who are older, thus demonstrate high screening compliance and regular gynaecological care, despite experiencing some barriers (less educated, migration history). Table 6. Determinants of class membership probability Class 1 N = 1,504 (31.5%) a Class 2 N = 905 (19.0%) Class 3 N = 1,022 (21.4%) Class 4 N = 121 (2.5%) Class 5 N = 1,222 (25.6%) Est SE Est SE Est SE Est SE Est SE Intercept 0.31 *** 0.00 0.23 *** 0.02 0.25 *** 0.02 0.01 0.01 0.18 *** 0.02 Up To Date (Ref = Yes) - - - - - - - - - - No −0.06 *** 0.02 0.00 0.01 −0.01 0.01 0.02 ** 0.01 0.05 ** 0.02 I don’t know 0.03 0.03 −0.01 0.02 0.01 0.02 0.01 0.01 −0.04 0.03 Age (Ref = 25–39 y.o.) - - - - - - - - - - 40–49 y.o 0.01 0.02 −0.01 0.01 −0.02 0.01 0.00 0.01 0.01 0.02 50–65 y.o 0.04 ** 0.01 −0.04 *** 0.03 0.03 * 0.01 0.01* 0.0 −0.04 ** 0.01 Country (Ref = Estonia) - - - - - - - - - - France 0.11 ** 0.03 −0.10 *** 0.03 0.04 0.03 0.00 0.01 −0.01 0.03 Nationality (Ref = Native) - - - - - - - - - - Foreigner 0.16 *** 0.01 −0.09 *** 0.01 −0.04 *** 0.01 0.03 *** 0.00 −0.06 *** 0.01 Residence location (Ref = in a big city) - - - - - - - - - - In other urban areas −0.01 0.02 −0.01 0.01 −0.01 0.01 0.00 0.01 0.02 0.02 In rural areas −0.01 0.02 −0.04 ** 0.01 −0.01 0.01 0.01 0.01 0.05 * 0.02 Education (Ref = Highschool or below) - - - - - - - - - - Upper education −0.05 *** 0.01 0.04 *** 0.01 −0.03 *** 0.01 0.00 0.00 0.05 *** 0.01 Gynecologist Follow Up (Ref = Every year ) - - - - - - - - - - Once every two years −0.05 *** 0.01 0.00 0.01 −0.02 * 0.01 0.00 0.00 0.08 *** 0.01 When necessary −0.09 *** 0.02 0.00 0.01 −0.04 *** 0.01 −0.01 0.01 0.14 *** 0.02 No follow up −0.19 *** 0.03 −0.03 0.02 −0.06 ** 0.02 −0.01 0.01 0.28 *** 0.03 Self Assessed Health (Ref = Good) Correct 0.01 0.01 −0.02 0.01 0.01 0.01 0.01 * 0.00 −0.01 0.01 Bad −0.01 0.03 −0.02 0.02 0.05 * 0.02 0.03 *** 0.01 −0.07 * 0.03 Experience with healthcare system ( Ref = Positive) Negative −0.01 0.02 0.02 0.01 0.03 * 0.01 0.00 0.01 0.03 0.02 Financial difficulties (Ref = Rarely ) - - - - - - - - - - Often 0.02 0.01 −0.02 * 0.04 0.01 0.01 0.00 0.00 −0.02 0.01 Open in a new tab a Percentage of the total sample *** Corresponds to p < 0.001, ** to p < 0.01, and * to p < 0.05 Women in C2, who also have an aversion to SS are predominantly younger, more likely to be native Estonians, well-educated, more likely to live in urban areas and report few financial difficulties. Although coefficients are non-significant, our findings suggest, that similar to women in C1, they are generally up-to-date in screening and maintain gynaecological follow-up. Similarly, women in C3 exhibit an aversion to SS. They are more likely to be older, native and less educated. They report poorer self-assessed health ( p < 0.05) and more frequent negative experiences with the healthcare system ( p < 0.05) yet maintain regular gynaecological follow-up. Although the coefficient is not statistically significant, women in C4 seem rather indifferent between SS and PBS. Their preferences differ nonetheless from the ones of women in other classes regarding their preference for prevention campaign. Indeed, they don’t prefer social media but rather all other possibilities, with substantially higher odds of choosing door-to-door prevention compared with social media( p < 0.001). These women are less likely to be up-do-date in their screening, report poorer self-assessed health and are more likely to be older and non-native. Finally, women in C5 are the only group that preferred SS to PBS (OR = 1.05, p < 0.001). They are less likely to be up-to-date in screening as well as to have regular gynaecological follow-up. These women are younger, native, report better self-assessed health and more educated as well as more likely to reside in rural areas. With respect to the invitation delivery method, few women still prefer postal mail, except those in C1 and C2, only when compared with e-portals. Email is the preferred options for all women, except women in C3, who are expressing a preference for E-Portal. Except for women in C4, who express their preference for all information campaign but social media, women in all the other classes would rather be informed through social media. Implementation of SS and its impact on CCS uptake On the basis of the results of the latent class analysis, we evaluated the potential impact of implementing SS on CCS uptake (Appendix F). We estimate the average marginal effects of each level of the DCE attributes on the likelihood of participating in a screening program based on its characteristics. For this simulation, we focus on women with the strongest views on SS, i.e. those ranged in class C1 and C5, respectively. Our findings highlight the variability in responses across different population profiles if SS were to become the standard of care in organized screening programs. The group most averse to SS (C1) would decrease its participation by 76 pp. Conversely, implementing SS would increase the likelihood of participation by 37 pp for women in C5. Based on the profiles, C5 is the only group for which an increase in participation can be expected with the introduction of SS compared to PBS, whereas a decline is likely among at least the first three classes. The trajectory for class 4 remains uncertain. Robustness checks and further analysis Four-class solutions The 4-class solution (Appendix G) is consistent with the 5-class analysis, with a gradient across classes regarding their preference for SS. Consistent with the 5 women profiles model, we observe that women in the class that is most averse to SS have high screening rates, are up-to-date and have regular gynaecological appointments, whereas women who are inclined to SS are less screened, less up-to-date and more rural. The 5-class solution was selected on the basis of the best statistical fit (Appendix E). Analysis of ‘never-screened women’ Two hundred ninety-one women declare never having been screened. These women are significantly younger than the study population is. They show less favourable socioeconomic characteristics, with lower education levels and/or employment rates (Appendix H). Interestingly, women who have never been screened reveal different preferences for screening methods, with a significantly stronger preference for SS compared with PBS ( p < 0.05, Table 7 ). This result indicates the potential for SS to engage first-time screeners, consistent with evidence from Finland [ 13 ], where it was shown that a combination of an invitation letter and a self-sampling kit increased the screening rate from 63 to 78% among women who had never been screened. Table 7. DCE Results for the “never screened women” Never-screened women N = 291 Estimate CI Sampling method (Ref = Physician-based-sampling) Self-Sampling 0.43 * (0.04, 0.82) Invitation and Reminders (Ref = Postal Mail) SMS 0.78 *** (0.49, 1.07) E-portal 0.25 * (0.03, 0.48) Email 1.18 *** (0.87, 1.49) Information Campaign (Ref = Social Media) Leaflets −0.39 ** (−0.65, −0.14) Workshops −0.54 *** (−0.79, −0.30) Door-to-Door −1.42 *** (−1.77, −1.07) Open in a new tab Reading grid: The odds of “never-screened” women participating in a screening campaign using SS are 1.54 times higher than those with PBS *** Corresponds to p < 0.001, ** to p < 0.01, and * to p < 0.05 Estimates in log-odds Discussion Using a DCE approach, we studied women’s preferences for cervical cancer screening with a focus on self-sampling (SS), recognized for its potential to increase screening uptake. We also studied preferences for invitation and information modalities. Preferences regarding SS are characterized by strong heterogeneity. Our latent class analysis reveals a strong gradient towards SS, with great variability in its effect on women’s utility compared with that of physician-based sampling (PBS). Women in favour of SS (C5) are younger and exhibit higher levels of education. However, these women are significantly more likely to live in rural areas than in other classes and may therefore face geographic barriers to accessing screening [ 47 ], as well as higher opportunity costs, which could explain their lower CCS uptake and visit to the gynaecologist. These results are consistent with previous findings [ 48 ], which suggest that SS could improve CCS uptake among women in underserved or rural areas of Minnesota. On the other hand, we find that the women who are the most hostile to SS are those who are already the most involved in screening trajectories, and those from whom we therefore expect the most regularity in use [ 49 ]. Consistently, the introduction of SS in the Netherlands in 2017 did not translate into widespread adoption, with SS only accounting for 8% of total screening uptake [ 50 ]. Our sample likely over-represents women already engaged in screening, which may underestimate the impact of SS among the most had-to-reach populations. Moreover, the DCE presents SS as a procedure to be performed without assistance, but our results indicate that a significant proportion of women respondents might be willing to adopt SS if guided by a healthcare professional. This finding relates to the fear performing the test incorrectly, which has been shown in qualitative studies [ 51 ]. Our analysis identified a small class of women ( N = 121) with distinct preferences regarding prevention campaign; notably a strong preference for door-to-door outreach. These women are less up-to-date in screening and more often non-native to their country of residence to their country. This suggest potential challenges in navigating healthcare recommendations and systems, thus the preference for personalized information and health mediation, whose effectiveness with vulnerable public has been demonstrated [ 52 ]. Contrary to expectations, we observe that introducing SS as the sole screening method could significantly reduce CCS participation, as women already engaged in screening are particularly reluctant to do so, whereas those facing health care access difficulties might increase their participation. As shown in previous studies [ 24 , 36 ], SS can act as a catalyst for initiating screening among women who have never been screened. Our findings are consistent with this, particularly for younger and less educated women, suggesting that SS could facilitate their entry into the screening pathway. This is a critical point, given the long-term persistence of screening behaviours [ 49 ]. Our heterogeneity analysis does not bring out any class predominantly composed of respondents from one country over the other, contrary to what could be expected from cultural and healthcare systems’ differences between these two countries. What emerges as a more significant determinant of latent class membership is the status as native or foreigner within the country, which influences the ability to navigate health care systems. Interactions between migration status and country of residence were tested but did not yield statistically significant results in our analysis. The results of this heterogeneity analysis provide updated preference profiles that can be used to tailor programs to women’s needs to increase their engagement. Such tailoring not only enhances the likelihood of uptake but also allows a more efficient allocation of resources within organised programs. For example, our findings highlight that the majority of respondents prefer communication via social media, which is relatively inexpensive, whereas a subgroup of women, who are less connected to the healthcare system, would benefit more from resource-intensive solutions (e.g. door-to-door campaigns, prevention workshops). Targeting these women more precisely could therefore lead to a more optimal use of resources. Limitations Our study is not without limitations. First, the analysis assumes screening to be free of charge as part of an organized program, although in practice, women may incur certain direct costs, namely out-of-pocket costs depending on their insurance status and the type of provider seen. We also did not consider indirect costs, such as travel expenses or opportunity costs, which can further affect women’s willingness to participate in screening programs. More importantly, there is heterogeneity between the Estonian and French samples, both in terms of sample size and the characteristics of women. Women aged over 50 are significantly underrepresented in the French sample. In addition, the French population includes more women with financial difficulties. In Estonia, the questionnaire was distributed to the general population and presented higher screening rates than average (79% in the study sample compared with 55%, as estimated by the HPV information centre, 2023). These difficulties were partially overcome by pooling our samples, in order to identify commonalities and country-specific features. Nevertheless, differences in recruitment modes across countries may have influenced preferences, particularly regarding the invitation and reminder attribute. Preferences for digital communication channels observed in Estonia may partly reflect the higher level of digitalisation in the country and the characteristics of the recruited sample, as well as the online mode of survey administration. Online data collection may have introduce a digital literacy selection bias, whereby women comfortable with digital tools were more likely to participate and to reveal favourable preferences towards digital communication and information. Consequently, preferences for digital communication strategies in our sample may be upwardly biased, as women with lower digital literacy are probably underrepresented. Conversely, the aversion for Eportal invitation observed mostly in French respondents may be related to the more vulnerable profile of respondents and the predominantly face-to-face data collection. Therefore, cross-country comparison should be interpreted with caution, as some of the observed heterogeneity may reflect differences in sampling and administration rather than contextual preferences. Finally, no consistency checks were added to provide insight into the understanding or engagement of participants. Although the self-reported difficulty question provides some insight, its result should be interpreted with caution due to potential desirability bias. Nevertheless, we believe that both the pilot study and the explanatory video helped mitigate these risks by testing the questionnaire for clarity and comprehensibility. In addition, our assessment of participants' knowledge regarding CCS may have been influenced by the introductory video. All participants showed a high level of CCS awareness, which may not reflect their knowledge prior to participating in the DCE. The decision to use an introductory video to explain screening and SS was based on the results of the pilot phase, in order to ensure that participants could make an informed choice of intervention. Conclusion and policy implications Our study contributes to the literature on women's preferences for CCS in countries with organized CCS programs. Similar to previous findings [ 35 ], we find that offering SS could help engage young, never-screened women in a screening trajectory, as well as women living in rural areas with limited access to health care. While our results confirm that SS may encourage some women to enter a screening program, we also show that its generalization as the sole screening modality would reduce uptake for women already engaged in screening pathways. Our preference analysis provides a more nuanced perspective on the validity of offering SS to the most vulnerable women, as they tend to prioritize contact with a health professional and more personalized information about the CCS. Thus, offering SS as an option in organized CCS programs may indeed increase screening rates among women with limited health care access but not among the most vulnerable ones. This highlights the limitations of universal programs that may not be well suited for certain subgroups of women, resulting in costly and inefficient strategies. Screening strategies could be tailored as to better meet women’s diverse needs. Our results, for instance, highlight the necessity to offer screening through healthcare professionals for women who are already engaged in screening and averse to SS. However, for rural populations, providing the option to receive a SS kit could be an effective alternative. SS could also be made available to young women who have not yet entered their screening pathway. This tailoring could significantly improve the effectiveness of screening programs, ensuring they are more accessible and relevant to all subgroups. Supplementary Information Supplementary Material 1. (468.9KB, docx) Supplementary Material 2. (5.5MB, docx) Acknowledgements The authors would like to thank the [XXX] Consortium members as well as the participants of the 2024 JESF and the 2024 EUHEA PhD Workshop for their feedback. Abbreviations BIC Bayesian information criteria CC Cervical Cancer CCS Cervical Cancer Screening CSS Complémentaire Santé Solidaire -France’s means-tested complementary health insurance DCE Discrete Choice Experiment EU European Union HPV Human Papillomavirus PBS Physician-based sampling SS Self-sampling Authors’ contributions - **VD:** conceptualization, methodology, formal analysis, investigation, validation, visualisation, writing – original draft, writing- review & editing—**LR:** conceptualization, methodology, formal analysis, investigation writing -review & editing—**MC:** formal analysis, visualisation, funding acquisition, writing– review & editing—**AU:** conceptualization, investigation, writing – review & editing—**AT:** conceptualization, investigation, writing – review & editing—**CF:** formal analysis, visualisation, investigation, writing -review & editing All authors read and approved the final manuscript. Funding CBIG-SCREEN has received funding from the EU’s Horizon 2020 research and innovation program under grant agreement No 964049. The work was supported by European Commission’s Horizon 2020 program as part of the CBIG-SCREEN research project; however, the views expressed do not necessarily reflect the European Commission’s official policies. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Data availability The datasets generated and/or analysed during the current study are not publicly available due to the ethical agreements. Declarations Ethics approval and consent to participate This study was conducted in accordance with the ethical standards of the 1964 Helsinki Declaration and its later amendments and was approved by the Institutional Review Board (IRB00003888, IORG0003254, WA00005831) of the French Institute of medical research and Health in June 2023 (Opinion Number 23–1014). Informed consent to participate was obtained from all participants in the study. Data were anonymized and handled in compliance with the General Data Protection Regulation (GDPR). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Singh GK, Williams SD, Siahpush M, Mulhollen A. Socioeconomic, Rural-Urban, and Racial Inequalities in US Cancer Mortality: Part I—All Cancers and Lung Cancer and Part II—Colorectal, Prostate, Breast, and Cervical Cancers. J Cancer Epidemiol. 2011;2011:1–27. 10.1155/2011/107497. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Ritchie D, Arbyn M, Basu P, Corbex M, Fidarova E, Ivanuš U, et al. Europe’s path to eliminating cervical cancer as a public health problem. Lancet Reg Health Eur. 2022;12:100276. 10.1016/j.lanepe.2021.100276. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Franceschi S, Plummer M, Clifford G, De Sanjose S, Bosch X, Herrero R, et al. Differences in the risk of cervical cancer and human papillomavirus infection by education level. Br J Cancer. 2009;101(5):865–70. 10.1038/sj.bjc.6605224. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Brønnum-Hansen H. Socially disparate trends in lifespan variation: a trend study on income and mortality based on nationwide Danish register data. BMJ Open. 2017;7(5):e014489. 10.1136/bmjopen-2016-014489. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Kinge JM, Modalsli JH, Øverland S, Gjessing HK, Tollånes MC, Knudsen AK, et al. Association of household income with life expectancy and cause-specific mortality in Norway, 2005-2015. JAMA. 2019;321(19):1916. 10.1001/jama.2019.4329. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Greenley R, Bell S, Rigby S, Legood R, Kirkby V, McKee M, et al. Factors influencing the participation of groups identified as underserved in cervical cancer screening in Europe: a scoping review of the literature. Front Public Health. 2023;11:1144674. 10.3389/fpubh.2023.1144674. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. International Agency for Research on Cancer. Handbook 10: Cancer Screening. Lyon, France: IARC; 2020. Retrieved from https://screening.iarc.fr/doc/HANDBOOK10.pdf . 8. Bruni L, Albero G, Serrano B, Mena M, Gómez D, Muñoz J, et al. Human Papillomavirus and related diseases report – Europe. ICO/IARC HPV Information Centre; 2024. Retrieved from https://hpvcentre.net/statistics/reports/EST_FS.pdf . 9. Rigby S, Greenley R, Uuskula A, Tisler A, Suurna M, Jesse M, et al. Expanding eligibility and improving quality of cervical cancer screening in Estonia: the 2021 reforms. Health Policy. 2024;144:105077. 10.1016/j.healthpol.2024.105077. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Kivistik A, Lang K, Baili P, Anttila A, Veerus P. Women’s knowledge about cervical cancer risk factors, screening, and reasons for non participation in cervical cancer screening programme in Estonia. BMC Women’s Health. 2011 ;11(43). 10.1186/147268741143. [ DOI ] [ PMC free article ] [ PubMed ] 11. International Agency for Research on Cancer & ICO/IARC HPV Information Centre. France: Human Papillomavirus and related cancers, fact sheet. ICO/IARC HPV Information Centre; 2023. Retrieved from https://hpvcentre.net/statistics/reports/FRA_FS.pdf . 12. Santé publique France. Participation au programme de dépistage organisé du cancer du col de l’utérus : Période 2014 2023 (Bulletin national). 2024. 13. World Health Organization. Global strategy to accelerate the elimination of cervical cancer as a public health problem. World Health Organization; 2020. 10.1016/j.healthpol.2024.105077. 14. Rosenstock IM. Historical origins of the health belief model. Health Educ Monogr. 1974;2(4):328–35. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36(1):1–10. 10.2307/2137284. [ PubMed ] [ Google Scholar ] 16. Suurna M, Orumaa M, Ringmets I, Pärna K. Inequalities in reported use of cervical screening in Estonia: results from cross-sectional studies in 2004–2020. BMC Women’s Health. 2022;22(545). 10.1186/s12905-022-02123-z [ DOI ] [ PMC free article ] [ PubMed ] 17. Laranjeira CA. Portuguese women’s knowledge and health beliefs about cervical cancer and its screening. Soc Work Public Health. 2013;28(2):150–7. 10.1080/19371918.2011.592042. [ DOI ] [ PubMed ] [ Google Scholar ] 18. Ouanhnon L, Bugat M-ER, Lamy S, Druel V, Delpierre C, Grosclaude P. Social and territorial inequalities in breast and cervical cancers screening uptake: a cross-sectional study in France. BMJ Open. 2022;12(2):e055363. 10.1136/bmjopen-2021-055363. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Mansfield LN, Silva SG, Merwin EI, Chung RJ, Gonzalez-Guarda RM. Factors associated with human papillomavirus vaccine series completion among adolescents. Am J Prev Med. 2021;61(5):701–8. 10.1016/j.amepre.2021.04.031. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Hawkes D, Keung MHT, Huang Y, McDermott TL, Romano J, Saville M, et al. Self-collection for cervical screening programs: from research to reality. Cancers. 2020;12(4):1053. 10.3390/cancers12041053. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Nakalembe M, Makanga P, Kambugu A, Laker-Oketta M, Huchko MJ, Martin J. A public health approach to cervical cancer screening in Africa through community-based self-administered HPV testing and mobile treatment provision. Cancer Med. 2020;9(22):8701–12. 10.1002/cam4.3468. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Serrano B, Ibáñez R, Robles C, Peremiquel-Trillas P, De Sanjosé S, Bruni L. Worldwide use of HPV self-sampling for cervical cancer screening. Prev Med. 2022;154:106900. 10.1016/j.ypmed.2021.106900. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Kirkegaard P, Larsen MB, Andersen B. It’s cancer screening after all”. Barriers to cervical and colorectal cancer screening and attitudes to promotion of self-sampling kits upon attendance for breast cancer screening. J Med Screen. 2023;30(2):74–80. 10.1177/09691413221137852. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Parker S, Deshmukh AA, Chen B, Lairson DR, Daheri M, Vernon SW, et al. Perceived barriers to cervical cancer screening and motivators for at-home human papillomavirus self-sampling during the COVID-19 pandemic: results from a telephone survey. Elife. 2023;12:e84664. 10.7554/eLife.84664. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Arbyn M, Rebolj M, De Kok IMCM, Fender M, Becker N, O’Reilly M, et al. The challenges of organising cervical screening programmes in the 15 old member states of the European Union. Eur J Cancer. 2009;45(15):2671–8. 10.1016/j.ejca.2009.07.016. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Virtanen A, Nieminen P, Luostarinen T, Anttila A. Self-sample HPV tests as an intervention for nonattendees of cervical cancer screening in Finland: a randomized trial. Cancer Epidemiol Biomarkers Prev. 2011;20(9):1960–9. 10.1158/1055-9965.EPI-11-0307. [ DOI ] [ PubMed ] [ Google Scholar ] 27. Tranberg M, Bech BH, Blaakær J, Jensen JS, Svanholm H, Andersen B. HPV self-sampling in cervical cancer screening: the effect of different invitation strategies in various socioeconomic groups - a randomized controlled trial. Clin Epidemiol. 2018;10:1027–36. 10.2147/CLEP.S164826. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Lozar T, Nagvekar R, Rohrer C, Dube Mandishora RS, Ivanus U, Fitzpatrick MB. Cervical cancer screening postpandemic: self-sampling opportunities to accelerate the elimination of cervical cancer. Int J Womens Health. 2021;13:841–59. 10.2147/IJWH.S288376. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Paulauskiene J, Stelemekas M, Ivanauskiene R, Petkeviciene J. The cost-effectiveness analysis of cervical cancer screening using a systematic invitation system in Lithuania. Int J Environ Res Public Health. 2019;16(24):5035. 10.3390/ijerph16245035. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Firmino-Machado J, Varela S, Mendes R, Moreira A, Lunet N, Carmo A, et al. A 3-step intervention to improve adherence to cervical cancer screening: the SCAN randomized controlled trial. Prev Med. 2019;123:250–61. 10.1016/j.ypmed.2019.03.025. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Agide FD, Garmaroudi G, Sadeghi R, Shakibazadeh E, Yaseri M, Koricha ZB, et al. A systematic review of the effectiveness of health education interventions to increase cervical cancer screening uptake. Eur J Public Health. 2018;28(6):1156–62. 10.1093/eurpub/cky197. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Alimena S, Manning-Geist B, Pena N, Vitonis AF, Feldman S. Outcomes by race among women referred to an academic colposcopy clinic with a patient navigation program. J Womens Health. 2021;30(6):902–9. 10.1089/jwh.2020.8381. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Genoff MC, Zaballa A, Gany F, Gonzalez J, Ramirez J, Jewell ST, et al. Navigating language barriers: a systematic review of patient navigators’ impact on cancer screening for limited English proficient patients. J Gen Intern Med. 2016;31(4):426–34. 10.1007/s11606-015-3572-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Andreassen T, Weiderpass E, Nicula F, Suteu O, Itu A, Bumbu M, et al. Controversies about cervical cancer screening: a qualitative study of Roma women’s (non)participation in cervical cancer screening in Romania. Soc Sci Med. 2017;183:48–55. 10.1016/j.socscimed.2017.04.040. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Campbell HE, Gray AM, Watson J, Jackson C, Moseley C, Cruickshank ME, et al. Preferences for interventions designed to increase cervical screening uptake in non‐attending young women: how findings from a discrete choice experiment compare with observed behaviours in a trial. Health Expect. 2020;23(1):202–11. 10.1111/hex.12992. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Chamot E, Mulambia C, Kapambwe S, Shrestha S, Parham GP, Macwan’gi M, et al. Preference for human papillomavirus–based cervical cancer screening: results of a choice-based conjoint study in Zambia. J Low Genit Tract Dis. 2015;19(2):119–23. 10.1097/LGT.0000000000000081. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Oberlin AM, Pasipamire T, Chibwesha CJ. Exploring women’s preferences for HPV‐based cervical cancer screening in South Africa. Int J Gynaecol Obstet. 2019;146(2):192–9. 10.1002/ijgo.12869. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Dzobo M, Dzinamarira T, Strauss M, Mashamba-Thompson T. Rural women’s preferences for cervical cancer screening via HPV self-sampling: a discrete choice experiment study in chidamoyo, Hurungwe District, Zimbabwe. AJOG Glob Rep. 2024:100414. 10.1016/j.xagr.2024.100414. [ DOI ] [ PMC free article ] [ PubMed ] 39. Bøje RB, Bardou M, Mensah K, Rico Berrocal R, Giorgi Rossi P, Bonvicini L, et al. What are the barriers towards cervical cancer screening for vulnerable women? A qualitative comparative analysis of stakeholder perspectives in seven European countries. BMJ Open. 2024;14(5):e079921. 10.1136/bmjopen-2023-079921. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. McFadden D. The measurement of urban travel demand. J Public Econ. 1974;3(4):303–28. 10.1016/0047-2727(74)90003-6. [ Google Scholar ] 41. McFadden D, Train K. Mixed MNL models for discrete response. J Appl Econom. 2000;15(5):447–70. 10.1002/1099-1255(200009/10)15:5/3C447::AID-JAE570/3E3.0.CO;2-1. [ Google Scholar ] 42. Hess S, Rose JM. Can scale and coefficient heterogeneity be separated in random coefficients models? Transportation. 2012;39(6):1225–39. 10.1007/s11116-012-9394-9. [ Google Scholar ] 43. Hensher DA, Rose J, Greene WH. The implications on willingness to pay of respondents ignoring specific attributes. Transportation. 2005;32(3):203–22. 10.1007/s11116-004-7613-8. [ Google Scholar ] 44. Weller BE, Bowen NK, Faubert SJ. Latent class analysis: a guide to best practice. J Black Psychol. 2020;46:287–311. 10.1177/0095798420930932. [ Google Scholar ] 45. Sinha P, Calfee CS, Delucchi KL. Practitioner’s guide to latent class analysis: methodological considerations and common pitfalls. Crit Care Med. 2021;49:e63–79. 10.1097/CCM.0000000000004710. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. World Bank. Urban population (% of total population) – France. World Development Indicators. https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?locations=FR 47. Inspection générale des affaires sociales. Le dépistage organisé des cancers en France. IGAS; 2024. https://igas.gouv.fr/sites/igas/files/2024-04/Le%20d%C3%A9pistage%20organis%C3%A9%20des%20cancers%20en%20France.pdf . 48. Zhu X, MacLaughlin KL, Fan C, Jacobson DJ, Jenkins GD, Jacobson RM, et al. Awareness of HPV testing and acceptability of self-sampling for cervical cancer screening among women in Minnesota. J Gen Intern Med. 2022;37(6):1565–8. 10.1007/s11606-021-06854-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Dugord C, Franc C. Trajectories and individual determinants of regular cancer screening use over a long period based on data from the French E3N cohort. Soc Sci Med. 2022;294:114663. 10.1016/j.socscimed.2021.114663. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Aitken CA, Van Agt HME, Siebers AG, Van Kemenade FJ, Niesters HGM, Melchers WJG, et al. Introduction of primary screening using high-risk HPV DNA detection in the Dutch cervical cancer screening programme: a population-based cohort study. BMC Med. 2019;17(1):228. 10.1186/s12916-019-1460-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Pourette D, Cripps A, Guerrien M, Desprès C, Opigez E, Bardou M, et al. Assessing the acceptability of home based HPV self sampling: a qualitative study on cervical cancer screening conducted in Reunion Island prior to the RESISTE trial. Cancers. 2022;14(6):1380. 10.3390/cancers14061380. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Laurence S, Lallemand A, Chappuis M, Rochefort J. Médecins du monde et le principe d’universalisme proportionné appliqué à la prévention du cancer du col de l’utérus. Sante Publique. 2020;S2(HS2):53–7. 10.3917/spub.197.0053. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1. (468.9KB, docx) Supplementary Material 2. (5.5MB, docx) Data Availability Statement The datasets generated and/or analysed during the current study are not publicly available due to the ethical agreements. Articles from BMC Women's Health are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.6 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Determinants of cervical cancer screening among women aged 30 to 49 years in 20 low- and middle-income countries: A multilevel analysis.#135989
• Self-sampling Among Long-term Non-attenders to Cervical Cancer Screening#908577
• Real-world evaluation of a population-based vaginal self-sampling program for cervical cancer prevention in Region Skåne, Sweden.#219088
• Cervical Cancer Screening Non-Attendance and Home-Based Self-Sampling in Northern Sweden: A Spatial Ecological Study#1000535
• Cervical Cancer Screening Cascade in Côte d’Ivoire: A Decade of Real-World Program Evaluation According to HIV Status#218084