Introduction

Cervical cancer (CC), primarily caused by high-risk human papillomavirus (HR-HPV), remains a major global health issue, particularly in low- and middle-income countries [1,2]. While the majority of HR-HPV infections resolve spontaneously, persistent infection is often linked to socioeconomic and behavioral factors, and it may promote progression to cervical intraepithelial neoplasia and invasive cancer [3,4].

Women who have never been screened or are under-screened face a significantly higher risk of invasive cervical cancer and are often difficult to reach through conventional methods [5,6]. Optimal screening coverage is a challenge that many countries face. In the UK, approximately 30% of women had not undergone screening in the last ten years. Meanwhile, Sweden saw a 10% rise in CC screening coverage after implementing HR-HPV self-sampling in April 2020 amidst the COVID-19 restrictions [7–9].

Latvia remains among the European countries with the highest cervical cancer incidence and mortality rates despite the introduction of an organized cervical cancer screening programme in 2009. Participation in screening has historically been low, and important implementation challenges persist, particularly among women living in rural areas, women with lower health literacy, and socioeconomically disadvantaged groups. In July 2022, Latvia initiated the transition from cytology-based screening to primary HR-HPV testing. Latvia also has a diverse ethnic composition, and previous studies have shown disparities in screening participation and healthcare access among minority populations. These factors highlight the need for acceptable and accessible screening approaches, including HR-HPV self-sampling, particularly in Latvia and the Baltic region as well [10,11–13,14].

HR-HPV self-sampling has become a promising approach to enhance participation in cervical cancer screening, effectively addressing significant practical and emotional obstacles [6]. The World Health Organization (WHO) advises that self-sampling should be made available as an alternative for women between the ages of 30 and 60, aiming for 70% screening coverage by the year 2030. [1].

Comprehensive worldwide evidence indicates that the accuracy of HR-HPV self-sampling is similar to that of samples collected by clinicians [15,16]. HR-HPV self-sampling could enhance screening participation since it is convenient for users, straightforward to integrate into screening initiatives, and efficiently utilizes medical resources, thus reducing the burden on healthcare systems [17].

Preference for clinician-collected cervical sampling has been associated with several factors, including older age, higher educational attainment, concerns about the reliability of self-sampling, cultural beliefs and norms, also, in some studies, higher body mass index (BMI) has been associated with this preference. Evidence also indicates that many patients value the opportunity for in-person interaction with a healthcare professional, which allows them to receive individualized counselling and to discuss additional health-related concerns [18–21].

Moreover, women’s choice of cervical cancer sampling modality is shaped not only by sociodemographic and practical considerations but also by the perceived acceptability of the procedure from the user’s perspective. In this context, acceptability is conceptualized broadly as women’s subjective experience of a sampling method, which may influence both the initiation and maintenance of its use. Although acceptability is a multidimensional construct, the present study concentrates specifically on feelings experienced during HR-HPV self-sampling, as these were directly assessed in the questionnaire and are particularly relevant for understanding preferences regarding future mode of cervical screening sampling and indirectly also participation [22].

Within this framework, negative feelings such as embarrassment and discomfort may function as perceived barriers, increasing the subjective burden associated with self-sampling. In contrast, confidence in one’s ability to carry out the procedure on their own reflects perceived self-efficacy and perceived behavioural control, both of which are central predictors of screening behaviour in established behavioural theories [23].

Findings from empirical studies on HR-HPV self-sampling consistently show that lower levels of embarrassment, enhanced privacy, and a greater sense of autonomy support the use of self-sampling. However, concerns about performing the test correctly and trusting the accuracy of the result continue to pose significant barriers for some women [24–26]. The objective of this study is to identify the determinants of preference for cervical screening sampling modality (self-sampling vs. clinician-collected) to ensure that HR-HPV self-sampling is universally adopted without creating new disparities in cervical cancer prevention.

Materials and methods

Results

Participants’ emotional/psychological responses regarding HR-HPV self-sampling

As shown in Fig 1, the majority of participants did not express negative emotional/psychological responses, with 63.1% (95% CI: 60.1–66.1) reporting ‘’The feeling was not at all expressed/rather not expressed for embarrassment, and 50.0% (95% CI: 47.0–53.1) for discomfort. Conversely, positive emotional/psychological responses were more frequently reported, though with varying intensity. Specifically, 26.3% (95% CI: 23.7–29.1) of women stated ‘’The feeling was rather/ very strongly expressed” strongly expressed feeling “intrigued/interested,” and 20.2% (95% CI: 17.9–22.8) reported feeling “confident” about performing self-sampling. A considerable proportion of participants remained neutral across all emotional dimensions: 29.4% (95% CI: 26.6–32.2) for discomfort, 38.4% (95% CI: 35.5–41.4) for confidence, 24.7% (95% CI: 22.1-27.5) for embarrassment, and 33.5% (95% CI: 30.7-36.5) for intrigued/interested. Only a small percentage, ranging from 7.2% (95% CI: 5.7–8.9) to 11.7% (95% CI: 9.9–13.8) across different emotional/psychological responses, selected the answer ‘’Hard to say/don’t know,” indicating some uncertainty.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Proportion of women (%) as per their emotional/psychological responses about the HR-HPV self-sampling procedure.

https://doi.org/10.1371/journal.pone.0353180.g001

Table 1 presents a comparison of participants’ emotional/psychological responses to the HR-HPV self-sampling procedure according to their preferred mode of cervical sampling in the next round of screening. Statistically significant differences were observed for embarrassment (p < 0.001), discomfort (p < 0.001), and confidence (p = 0.017). Women preferring clinician-collected cervical sampling more frequently reported both embarrassment and discomfort as ‘rather/very strongly expressed’ compared with women preferring self-sampling (embarrassment 6.4% vs. 1.6%; discomfort 18.2% vs. 9.4%). These differences were significant in pairwise comparisons with both the ‘neutral’ and ‘not expressed’ categories, as indicated by the letter groupings in Table 1. Conversely, participants who favored self-sampling more frequently rated these emotional/psychological responses as “not expressed” or “neutral.” Regarding confidence, women who preferred self-sampling more often reported “the feeling was rather/very strongly expressed” (23.2%) compared with those favoring clinician-collected sampling (17.1%), indicating a greater comfort with the procedure. No statistically significant differences were observed for the feeling of interest or intrigue (p = 0.088), suggesting that curiosity and engagement toward the self-sampling procedure were comparable between the groups.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Number and proportion (%) of responders according to emotional/psychological responses about self-sampling procedure, as per the preference of sampling method for the next round of CC screening.

https://doi.org/10.1371/journal.pone.0353180.t001

The most frequently reported reason for preferring clinician-collected cervical sampling was a lack of confidence in performing self-sampling correctly (81.0%). Other reasons included greater convenience and reassurance associated with the clinician’s involvement (14.6%), and less common reasons like discomfort with self-sampling (2.5%), habit or prior experience with clinician-collected sampling (0.8%), identifying herself as a high-risk patient (0.3%), and fear of causing self-injury during self-sampling (0.3%).

Using multivariate logistic regression analysis, several factors were found to be significantly associated with a preference for clinician-collected cervical sampling (Table 2). Women belonging to the colposcopy study group had higher odds of preferring clinician-collected testing compared with those from the general population (aOR = 1.84, 95% CI: 1.37–2.47). Women who had had a visit with a gynecologist in the last year were more likely to prefer clinician-collected sampling than those whose last visit to a gynecologist was less recent (aOR = 0.34, 95% CI: 0.18–0.64). Similarly, those who reported having quit or only occasionally smoked were less likely to prefer clinician-collected sampling compared with never-smokers (aOR = 0.64, 95% CI: 0.46–0.90). In the unadjusted model, non-Latvian women were more likely to prefer clinician-collected cervical sampling compared with Latvian participants; however, this association did not remain significant after adjustment for other variables. Other sociodemographic and health-related variables, including age, marital status, education, monthly income, alcohol use, number of sexual partners, BMI, and chronic disease status, were not significantly associated with testing preference in the adjusted model.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Factors associated with preferring cervical sampling performed by a clinician in the next screening round (reference: self-sampling).

https://doi.org/10.1371/journal.pone.0353180.t002

Table 3 presents the factors associated with having no preference for the mode of cervical sampling in the next round of screening. In the adjusted model, women with higher BMI (≥25) were significantly more likely to report no clear preference compared with those with lower BMI (<25) (aOR = 2.13, 95% CI: 1.37–3.32). Non-Latvian nationality was also associated with a higher likelihood of expressing no preference (aOR = 1.64, 95% CI: 1.02–2.64). No significant associations were found for age, education, marital status, income, smoking, alcohol use, number of lifetime sex partners, chronic diseases, time since last gynecologist visit, or study group.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Factors associated with having no preference (“hard to say”) regarding the mode of cervical sampling in the next screening round (reference: self-sampling).

https://doi.org/10.1371/journal.pone.0353180.t003

Discussion

This study focuses on women’s preferences regarding different modes of cervical sampling, specifically comparing HR-HPV self-sampling with clinician-collected cervical sampling. This study assessed the emotional/psychological responses to the self-sampling procedure and preferred mode of sampling in the next round of screening among different groups of women, including those from the general population and those referred for colposcopy. To the best of our knowledge, this is the first study in the Baltic region to examine women’s feelings during the HR-HPV self-sampling procedure. These findings are particularly relevant in Latvia, where HR-HPV self-sampling has not yet been implemented within the organized screening programme.

The present study found that participants generally reported positive or neutral emotional/psychological responses during the HR-HPV self-sampling procedure, with few women expressing embarrassment or discomfort. These findings align with previous research indicating that self-sampling is usually perceived as a comfortable and acceptable procedure among women in different populations and especially among screening non-attenders [24,35].

When asked about their preferred mode of sampling for the next round of cervical cancer screening, participants in this study demonstrated an almost equal distribution between self-sampling (42.6%) and clinician-collected cervical sampling (45,9%), with a smaller proportion reporting no clear preference (11.5%). This distribution aligns with findings from settings where women can choose their screening method, though the balance varies by programme and population. In England, when self-sampling was offered at invitation, 51.4% intended to choose it, 36.5% preferred clinician-based screening, and 10.5% were unsure [[36,37]]. In the Belgian VALHUDES study among screening participants, 57% preferred self-sampling for the next round and 41% preferred clinician collection [38]. Among colposcopy clinic attenders – more comparable to our secondary recruitment setting – preferences were more mixed: 37% preferred self-sampling, 35% had no preference, and smaller proportions favoured clinician-collected options [39]. Evidence from neighbouring Estonia, which has a similar cervical cancer burden to Latvia but more advanced HR-HPV-based screening, also supports these patterns. Estonian studies show high acceptability of HPV self-sampling and increased participation, especially among previous non-attenders, with few women preferring clinician-based sampling after trying self-sampling [12,13,40–43]. Overall, these comparisons indicate that Latvia’s near-equal split between clinician and self-sampling preferences is plausible during transition, and that indecision may be more common in clinically followed populations, underscoring the value of offering choice and clear information to build confidence in self-sampling.

Although this study was carried out in Latvia, its results may be applicable to other countries confronting comparable obstacles in cervical cancer prevention, especially those with a high disease burden, historically low screening uptake, or that are in the process of shifting from cytology-based to HR-HPV-based screening. Evidence from these settings suggests that emotional and psychosocial dimensions – such as perceived self-efficacy, autonomy, and reassurance about test accuracy – are central to shaping the acceptability of self-sampling, often exerting greater influence than practical aspects alone. Against this backdrop, the almost evenly split preferences observed in our study, together with the strong link between confidence and preference for self-sampling, highlight the need for implementation approaches that not only make self-sampling available but also explicitly foster confidence through clear instructions, education, and effective communication. These findings may thus guide HR-HPV self-sampling initiatives outside the Baltic region, particularly in contexts where building trust and ensuring long-term engagement remain major challenges.

In this study, most women reported no negative emotional/psychological responses during HR-HPV self-sampling. These findings suggest that self-sampling is generally perceived as emotionally acceptable and non-threatening. Emotional responses are important to consider, as they influence willingness to repeat the test and recommend it to others. Women’s feelings during the procedure, particularly comfort, confidence, and a sense of control, play a central role in building trust in self-sampling and promoting sustained participation in screening [44–46]. These results highlight that emotional acceptance, alongside technical reliability, should be prioritized when integrating HPV self-sampling into national screening programs.

The present study also explored participants’ emotional/psychological responses to HR-HPV self-sampling and how these feelings related to their preferred sampling method for the next round. Women who preferred clinician-collected cervical sampling in the next screening round more frequently reported embarrassment (6.4%) and discomfort (18.2%) during the procedure, whereas those who favored self-sampling in the next screening round predominantly described neutral or positive emotions (embarrassment 1.6%; discomfort 9.4%). These findings are consistent with international studies demonstrating that self-sampling reduces embarrassment and discomfort compared with clinician-based methods, likely because it enhances privacy, autonomy, and perceived control [24,39,44,45]. In our study, women who preferred self-sampling in the next screening round also expressed stronger confidence in performing the procedure (23.2% vs. 17.1%), suggesting increasing trust in their ability to collect samples correctly.

Importantly, our results indicate that positive emotional factors, especially confidence, may exert a stronger influence on preferences for screening modality than simply minimizing negative emotions such as embarrassment or discomfort. Although mitigating negative emotional reactions is likely a necessary step toward enhancing acceptability, it may not be enough to ensure ongoing use if women do not feel confident in their ability to carry out self-sampling correctly or to trust the test result. This view is consistent with prior research showing that worries about collecting the sample properly and doubts about result accuracy remain among the most frequently cited reasons for favoring clinician-collected sampling, even when self-sampling is regarded as less embarrassing [23–26].

While acceptability is multifaceted, the emotional dimension examined here provides valuable insight into mechanisms influencing screening preferences. From a behavioural standpoint, confidence can be viewed as an indicator of perceived self-efficacy and perceived behavioural control, both of which are central determinants of health-related actions [23–26]. In this framework, the self-sampling experience may shape emotional reactions, with confidence potentially mediating the relationship between that experience and subsequent screening preferences. While the cross-sectional nature of this study precludes causal conclusions, this proposed pathway offers a credible account of the observed relationships and highlights the need for confidence-enhancing strategies when implementing HR-HPV self-sampling within screening programmes.

Analysis involving multiple variables identified several important factors associated with women’s preference for clinician-collected cervical sampling. In our study, non-Latvian women were more likely either to prefer clinician-collected sampling or to report uncertainty regarding future screening modality. These findings may reflect cultural, linguistic, or informational differences influencing trust in self-sampling and confidence in performing the procedure independently. Previous studies have similarly shown that some ethnic minority groups may be less likely to engage with HR-HPV self-sampling because of concerns regarding test accuracy, unfamiliarity with the method, or preferences for direct interaction with healthcare professionals [30,47]. At the same time, recent evidence suggests that attitudes toward self-sampling among ethnically diverse populations are complex and context-dependent. A qualitative study from Northwest England reported that many women from ethnically diverse backgrounds considered self-sampling acceptable and empowering because of the increased privacy, convenience, and autonomy associated with the method, although concerns regarding correct sample collection and trust in test accuracy persisted among some participants [48]. Similarly, findings from the YouScreen study demonstrated that women who had previously not attended screening often reported positive experiences with self-sampling and expressed willingness to use it again in future screening rounds [35]. These findings suggest that cultural background alone does not determine acceptability; rather, access to understandable information, confidence in the procedure, health literacy, and trust in healthcare systems may play more important roles in shaping screening preferences.

An interesting finding was that educational level was not independently associated with screening modality preference in the adjusted analyses, suggesting that formal education alone may not adequately explain women’s screening preferences in this population. Instead, factors such as health literacy, trust in healthcare systems, perceived accuracy of self-sampling, and familiarity with screening procedures may play more important roles. Religious affiliation was not assessed in the present study; therefore, no direct conclusions regarding religious influences can be drawn. Nevertheless, cultural norms and attitudes toward intimate health procedures may influence the acceptability of HPV self-sampling in some populations. Together, these findings highlight the importance of culturally sensitive communication strategies, multilingual educational materials, and clear explanations regarding the accuracy and safety of HR-HPV self-sampling when implementing population-based screening programmes [48,35].

In our study, women who had not visited a gynecologist for more than three years were significantly less likely to prefer clinician-collected cervical sampling, and even those whose last visit occurred one to three years earlier showed a reduced preference compared to those women who had visited a gynecologist in a year. This suggests that self-sampling may appeal to women who are less regularly engaged with gynecological care. While this variable does not directly measure participation in cervical cancer screening, it may indirectly reflect patterns of healthcare utilization. Similar findings have been reported in other studies, where self-sampling was found to be particularly attractive to women who attend healthcare services less frequently [37,49–53].

Lifestyle-related factors also appeared to play a role. Although the mechanism is not entirely clear, smoking behavior may act as a proxy for broader lifestyle and health-related attitudes. Previous research has shown that smokers are generally less likely to participate in cervical cancer screening and other preventive health programs, possibly due to lower health awareness, reduced engagement with healthcare, and higher perceived barriers to screening participation [54,55]. Therefore, the observed association in our study may reflect differences in preventive health-seeking behavior and perceived convenience rather than smoking itself.

An interesting finding was a lower preference for self-sampling among women recruited from colposcopy settings. This finding is consistent with previous studies showing that women with recent or ongoing clinical follow-up because of changes in their cervix tend to prefer clinician-collected cervical sampling [37,56]. The reasons why these women opted for clinician-collected cervical sampling remain speculative, but there is a theory that respondents attending colposcopy are already engaged with screening and have a divergent perception of cervical cancer risk. This finding aligns with the review by Nelson et al., which reported that women who preferred clinician-collected sampling often expressed doubts about their ability to perform self-sampling correctly and about the reliability of the obtained specimens [24]. Nishimura et al. identified concerns among some women that self-sampling might reduce opportunities for direct consultation with healthcare professionals, leading them to favor clinician-collected sampling [25]. Therefore, lack of confidence in performing the procedure and the perceived importance of clinical interaction appear to be key factors influencing preference for clinician-collected cervical sampling [24,25,39].

Among participants who reported no clear preference regarding the screening modality (“hard to say”), a higher body mass index (BMI ≥ 25) was one independent predictor in the adjusted model. Women with higher BMI were approximately twice as likely to express indecision, which may reflect body image concerns, perceived procedural difficulty, or the trustworthiness of each modality. Also, other study data suggest that women with obesity are historically under-screened for cervical cancer, potentially due to physical discomfort with speculum exams, embarrassment, or stigma [57].

In addition, non-Latvian nationality was associated with a higher likelihood of expressing no preference, suggesting that cultural, linguistic, or informational barriers may influence screening-related decision-making [37,47,58,59]. Overall, emotional and experiential factors appeared more influential than standard sociodemographic characteristics in shaping preferences.

The study has several limitations. Behavioural data were self-reported and may be subject to reporting or recall bias; however, this approach is widely accepted in epidemiological research [60]. The general population sample was not randomly selected, which may limit representativeness, although geographical diversity was ensured. Additionally, self-sampling was performed in a healthcare setting rather than fully at home, which may have influenced emotional responses and limit generalisability to population-based programmes. Despite these limitations, the study provides valuable insight into women’s emotional responses and screening preferences within a real-world national context.

This study provides the first evidence from Latvia on women’s feelings during HR-HPV self-sampling and their preferences for future cervical cancer sampling modalities. The findings indicate that self-sampling is well tolerated and that reluctance is driven primarily by informational, emotional, and cultural factors rather than socioeconomic ones. Targeted communication strategies addressing confidence, accuracy, and follow-up—particularly for women of non-Latvian nationality and those with higher BMI—will be critical to ensure equitable implementation of HR-HPV self-sampling within national screening programmes.

Conclusion

Most participants reported positive or neutral emotional/psychological responses during HR-HPV self-sampling, indicating that the procedure is emotionally well-tolerated. Embarrassment and discomfort were uncommon, while feelings of confidence and interest predominated. Self-sampling appeared particularly appealing to women less regularly engaged in gynecological care, suggesting its potential to reach under-screened populations. In contrast, preference for clinician-collected cervical sampling was more strongly influenced by medical history and cultural background. Importantly, confidence emerged as a key emotional factor associated with preference for self-sampling, suggesting that fostering trust in women’s ability to perform the procedure and in the reliability of test results is central to acceptance. These aspects should be taken into account when developing communication and education strategies for the introduction of HR-HPV self-sampling into the national screening program.

Supporting information