HPV self-sampling in organized cervical cancer screening program: A randomized pilot study in Estonia in 2021

Abstract

Background

Cervical cancer incidence in Estonia ranks among the highest in Europe, but screening attendance has remained low. This randomized study aimed to evaluate the impact of opt-in and opt-out human papillomavirus (HPV) self-sampling options on participation in organized screening.

Methods

A random sample of 25,591 women were drawn from the cervical cancer screening target population who were due to receive a reminder in autumn 2021 and thereafter randomly allocated to two equally sized intervention arms (opt-out and opt-in) receiving a choice between HPV self-sampling or clinician sampling. In the opt-out arm, a self-sampler was sent to home address by regular mail; the opt-in arm received an e-mail containing a link to order a self-sampler online. The remaining 30,102 women in the control group received a standard reminder for conventional screening. Participation by intervention arm, age and region of residence was calculated; a questionnaire was used to assess self-sampling user experience.

Results

A significant difference in participation was seen between opt-out (41.7%) (19.8% chose self-sampling and 21.9% clinician sampling), opt-in (34.1%) (7.9% self-sampling, 26.2% clinician sampling) and control group (29.0%, clinician sampling only). All age groups and regions in the intervention arms showed higher participation compared to the control group, but the size of the effect varied. Among self-sampling users, 99% agreed that the device was easy to use and only 3.5% preferred future testing at the clinic.

Conclusion

Providing women with a choice between self-sampling and clinician sampling significantly increased participation in cervical cancer screening. Opt-in and opt-out options had a different effect across age groups, suggesting the need to adapt strategies.

Keywords

Cervical cancer screening self-sampling opt-in and opt-out strategies

Introduction

Cervical cancer is one of the most common cancers in women, although it is a highly preventable disease through vaccination and regular screening. Early detection and timely treatment of cervical precancerous lesions prevent the development of cervical cancer.¹ High participation is needed for a screening program to be effective, as the majority of women who develop cervical cancer have been shown to be never or inadequately screened.²

Estonia is a country in Northern Europe with a population of 1.33 million, of which 95.8% was covered by health insurance in 2021.³ A nation-wide organized cervical cancer screening program in Estonia started in 2006 when all women aged 30–55 with valid health insurance were invited to undergo conventional cytology (Pap smear) every five years. Since 2021, the target group has been extended to all women aged 30–65 regardless of their health insurance status, all being invited to cervical cancer screening every five years. The previously used Pap smear was replaced by screening for human papillomavirus (HPV), the main causative agent of cervical cancer.⁴ Each year, the Estonian Cancer Screening Registry determines the target population and creates an electronic referral in the electronic health record for eligible women, which can be used at any time during the year. Additionally, during the first half of the year, the cancer screening registry sends invitations in randomly selected batches by e-mail and by regular mail to those with a missing or incorrect e-mail address in the Estonian Population Register. Reminders are sent to non-participants in the second half of the year. Women must make their own appointment with a screening service provider. The screening sample is collected by a midwife or gynaecologist. If a person has not been screened by January 31 of the year following the screening invitation, she is considered a non-participant and will be invited to the next screening in five years. In 2021, there were five laboratories in Estonia that analysed HPV samples within the organized screening program.

Despite the existence of a cervical cancer screening program for more than 15 years, the population coverage was constantly below 50% before 2021.⁵ Although the age-standardized cervical cancer incidence has decreased in Estonia since 2013,⁶ the overall incidence and mortality rates remain high with an estimated age-standardized (European Standard Population 2013) incidence rate of 16.8 and a mortality of 7.3 per 100,000 cases in 2022.⁷

In 2015, European guidelines recommended the HPV test as the primary test for cervical cancer screening.⁸ The shift to primary HPV screening in Estonia in 2021 opened the possibility for self-sampling as an additional strategy to bring more women to screening. A vaginal sample self-collection may be an appropriate alternative to HPV testing at a healthcare facility. Self-sampling enables women to take a screening sample privately at a suitable moment, and this may help to overcome various psychological, physical and logistical barriers such as anxiety about the procedure, embarrassment and lack of time that may withhold women from regular screening at a healthcare provider.⁹ Recent meta-analyses and systematic reviews confirm that self-sampling is highly accepted and may significantly increase screening rates.^10,11 The self-collected vaginal samples are comparable with samples collected by a healthcare provider, considered equally accurate,¹⁰ and self-sampling has proved to be an effective way to increase screening uptake in European countries and elsewhere.¹²

In 2020, an HPV self-sampling feasibility study was carried out among long-term non-attenders of cervical cancer screening in Estonia, and the results demonstrated high acceptance and feasibility of vaginal sample self-collection.¹³

The aim of this pilot study was to examine whether different HPV self-sampling options (opt-in or opt-out) increase the participation rate among the cervical cancer screening target population compared to conventional screening and to assess women’s user experience as well as their interest and willingness to use self-sampling for cervical cancer screening in the future.

Methods

Data sources

The main data sources were the Estonian Health Insurance Fund database, a national administrative database containing individual data on health insurance status and claims for medical procedures performed,¹⁴ and the Estonian Cancer Screening Registry, which collects data on screening participation, results of screening tests, additional examinations and treatment of nation-wide colorectal, breast and cervical cancer screening programs.¹⁵ Additional inquiries were made to all five medical laboratories involved in organized cervical screening to ensure data quality.

Study population and design

In 2021, women born in 1956, 1961, 1966, 1971, 1976, 1981, 1986 and 1991 were invited to cervical cancer screening. The study population included all women in the target group who had not participated in the screening by 17 August 2021, according to data from the Estonian Health Insurance Fund and Estonian Cancer Screening Registry, and were due to receive their screening reminder (n = 58,008). Two-stage randomization was used to form two intervention groups and one control group (Figure 1). First, a random sample of 26,000 women who had a valid residential address in the cancer screening registry were selected for the intervention group. The remaining 32,008 women formed the control group. At the second stage of randomization, the women in the intervention group were randomly allocated to two equally sized arms (opt-in and opt-out). The calculation of the sample size was based on a detection difference of 4% between the intervention and control groups, with a statistical significance level of 0.05 and 90% power. It also assumed a maximum participation rate of 30% in one study group.

Figure 1.

Design of the randomized pilot study of HPV self-sampling, Estonia 2021.

Before data analysis, additional inquiries were made to all five medical laboratories involved in the organized cervical cancer screening. According to the data obtained, 963 women from the initial sample had completed an HPV test before the cut-off date of 17 August, but information on their participation had been delayed: 205 women in the opt-out arm, 204 in the opt-in arm and 554 in the control group, respectively. These women were excluded from the analyses. In addition, we excluded from the control group 1352 persons with missing residential address to ensure comparability of intervention and control groups.

Women in the intervention arms received a screening reminder, offering a choice between self-sampling and conventional sampling at a healthcare provider. Women in the opt-out arm received the reminder and a self-sampling kit by regular mail to their home address. In the opt-in arm, the e-mail reminder included a link to order a self-sampling kit online. The ordering platforms were provided by partner laboratories and were available in Estonian, English and Russian. Women interested in self-sampling who were unable to order the kit through the website had an opportunity to order it by phone or e-mail. The control group received a standard e-mail reminder for screening at a clinic. Women in all groups with an incorrect or missing e-mail address received a reminder letter by regular mail.

The list of each batch of invitees was double-checked by the cancer screening registry before sending the reminders. The first reminders were sent out from August to October 2021. A second reminder was sent to all women not screened by 16 November and to those who had ordered the self-sampling kit but had not returned the self-collected sample to the laboratory.

Throughout the study period, all women had an opportunity to ask for additional information by calling a client information line of the Estonian Health Insurance Fund, or via e-mail from the study organizer, the National Institute for Health Development.

The self-sampling kit included the sampling device Qvintip^® (Aprovix AB, Uppsala, Sweden), an information sheet, self-sampling instructions, a questionnaire and a prepaid return envelope addressed to the laboratory. Several options for returning the kit were available, as it was possible to send the sample by regular mail or choose between two different parcel locker services. The self-sampling kit also included a short questionnaire (Appendix 1) that contained questions and statements about the self-sampling experience and screening method preference. The questionnaire was designed using examples from previous studies.^13,16,17 Women who did not choose self-sampling were asked to return the unopened sampling device to the lab. All study materials were made available in Estonian, Russian and English.

Laboratory methods

The laboratory service providers responsible for online platforms, posting the self-sampling kits, performing HPV analyses and reporting test results both to women and the Electronic Health Record database were selected via a public tender process. The laboratories had to meet the qualification criteria, which included adherence to the analysis methodology outlined in the Meijer guidelines,¹⁸ and the Estonian code of conduct for cervical cancer screening.¹⁹ Two accredited laboratories were selected. The laboratories used different HPV methodologies to analyse self-collected samples.

In United Laboratories of Tartu University Hospital, the self-sampling device tip was transferred to a vial with 20 mL of ThinPrep medium and stored at room temperature for at least 1 h before starting the hrHPV analysis. The samples were tested for hrHPV using the Alinity m hrHPV Assay (Abbott Molecular, Des Plaines, IL, USA). This test reports individually 3 hrHPV genotype results (HPV 16, HPV 18 and HPV 45) and pooled results of another 11 hrHPV genotypes (35, 39, 51, 56, 59, 66, 68, 31, 33, 52 and 58). The human beta-globin gene was used as an internal control for evaluating sample cell adequacy and amplification efficiency.

In Synlab Estonia, the vaginal swabs were held for up to 3 days at 4 °C prior to DNA extraction, and HPV was genotyped within 5 days of receiving at the lab. The material from swab specimens was suspended in 1 mL 1× PBS. DNA was extracted using the Roche MagNAPure96 system with Roche Small Volume Kit according to the manufacturer’s instructions (200 uL as input volume and 100 uL of NA extraction volume). Amplification of the human β-globin gene was performed to confirm the integrity of the DNA in the samples. HPV DNA genotyping was done by targeting the E6/E7 region based on multiplex-PCR and Luminex hybridization (Luminex, Austin, TX, USA), which detects 16 individual HPV types: 16, 18, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68 and 82.

Informing of test results

The results of cervical cancer screening tests were reported to the central e-Health Record, a system that integrates data from Estonia’s different healthcare providers to create a common database.²⁰ Patients and healthcare providers can access the system and view their test results, by using a digital identification method. In addition, the laboratory contacted all participants with a hrHPV-positive self-sample, using the woman’s preferred contact method as provided in the questionnaire (by phone, encrypted e-mail and regular mail). Information about a positive test result along with instructions for additional examinations was given. In the absence of a preferred contact method, the information was sent similarly as the initial reminder, either by e-mail or by regular mail. The laboratory also contacted all women with an inadequate result and offered a retest.

Statistical analysis

The participation rate in the control group and two intervention arms was considered as the primary outcome. Study participation was defined as having sent a self-collected HPV sample to the laboratory for analysis or having attended a cervical cancer screening at a healthcare facility. The secondary outcomes were (a) participation rates in intervention arms by screening method (self-sampling vs clinician sampling) and (b) self-sampling user experience assessment based on questionnaires, where at least one of the nine statements had been answered.

We analysed participation rates using the chi-square (χ2) test to examine the differences between the study groups. Four age groups were formed as follows: women aged 30 and 35, 40 and 45, 50 and 55 and 60 and 65. Places of residence were grouped into five regions according to the Nomenclature of Territorial Units for Statistics (NUTS level 3): Northern, North-Eastern, Central, Western and Southern Estonia.²¹ Data on HPV self-sampling user experience were collected with a questionnaire and the analysis based on participants’ agreement with the statements.

Statistical analysis was performed using StataMP17 version (StataCorp, College Station, TX, USA).

Ethics

The study protocol was approved by the Research Ethics Committee of the National Institute for Health Development (decision No. 741, 30 April 2021).

Results

Characteristics of study population

In total, 55,693 women were included in the analysis. Table 1 shows the characteristics of the study population. The proportion of women aged 60 and 65 was slightly higher compared to other age groups (27.5%), and nearly half (47.0%) were from the Northern region of Estonia including the capital city Tallinn.

Table 1.

Characteristics of the study population of the randomized pilot study of HPV self-sampling, Estonia 2021.

			Intervention group
	Total	Control group	Opt-out arm	Opt-in arm
	n (%)	n (%)	n (%)	n (%)
Total	55,693 (100.0)	30,102 (100.0)	12,795 (100.0)	12,796 (100.0)
Age
30 and 35	13,578 (24.3)	7361 (24.4)	3106 (24.3)	3111 (24.3)
40 and 45	13,493 (24.2)	7214 (24.0)	3130 (24.4)	3149 (24.6)
50 and 55	13,311 (23.9)	7219 (24.0)	3018 (23.6)	3074 (24.0)
60 and 65	15,311 (27.5)	8308 (27.6)	3541 (27.7)	3462 (27.1)
Region of residence
Northern	26,151 (47.0)	13,869 (46.1)	6077 (47.5)	6205 (48.5)
North-Eastern	231 (11.2)	3341 (11.1)	1445 (11.3)	1445 (11.3)
Central	4499 (8.1)	2495 (8.3)	1052 (8.2)	952 (7.4)
Western	5590 (10.0)	3042 (10.1)	1286 (10.1)	1262 (9.9)
Southern	13,222 (23.7)	7355 (24.4)	2935 (22.9)	2932 (22.9)

In the opt-out group, 351 unopened self-sampling kits were returned as undelivered, mostly due to the following reasons: an incorrect address and the mailbox was either too small, full, missing or inaccessible. An unopened self-sampling kit was returned by 322 women in the opt-out group and 13 women in the opt-in group for the following main reasons: clinician sampling preference (n = 124), recent screening (n = 43), fear or uncertainty of self-sampling (n = 29), pregnancy (n = 10) or recent childbirth (n = 2), hysterectomy (n = 12) and disability (n = 4).

Overall screening participation

The participation rate was significantly higher in the opt-out arm (41.7%) and the opt-in arm (34.1%) compared to the control group (29.0%) (Table 2). The mean participation rate in the intervention arms was 37.9%, 8.9% higher than in the control group.

Table 2.

Screening participation (%) in control group and intervention arms of the randomized pilot study of HPV self-sampling, Estonia 2021.

	Control group, %	Opt-out, %	Difference opt-out vs control	p-value	Opt-in, %	Difference opt-in vs control	p-value	Difference opt-out vs opt-in	p-value
Total	29.0	41.7	12.7	<0.001	34.1	5.1	<0.001	7.6	<0.001
Age
30 and 35	29.7	39.7	10.0	<0.001	36.4	6.7	<0.001	3.3	0.007
40 and 45	30.6	41.6	11.0	<0.001	37.5	6.9	<0.001	4.1	0.001
50 and 55	30.9	43.4	12.5	<0.001	34.4	3.5	0.001	9.0	<0.001
60 and 65	25.5	42.1	16.6	<0.001	28.7	3.2	<0.001	13.4	<0.001
Region of residence
Northern	31.7	43.1	11.4	<0.001	35.2	3.5	<0.001	7.9	<0.001
North-Eastern	18.6	32.1	13.5	<0.001	22.4	3.8	0.002	9.7	<0.001
Central	28.9	44.0	15.1	<0.001	37.4	8.5	<0.001	6.6	0.003
Western	33.1	45.1	12.0	<0.001	41.1	8.0	<0.001	4.0	0.039
Southern	27.1	41.2	14.1	<0.001	33.4	6.3	<0.001	7.8	<0.001

While the overall difference in participation between opt-out arm and control group was 12.7% (41.7% compared to 29.0%), the largest effect was seen in age group 60 and 65 (difference of 16.6% compared to the control group) and then in the Central (15.1%) and Southern (14.1%) regions (Table 2). In the opt-in arm, the overall screening uptake was 5.1% higher than in the control group with the largest difference seen among younger women at age of 30 and 35 (6.7%) and 40 and 45 (6.9%) and in the Central (8.5%) and Western (8.0%) regions. The overall difference between the opt-out and opt-in arms was 7.6% (41.7% compared to 34.1%) with the largest effect observed in age group 60 and 65 (13.4%) and in the North-Eastern region (9.7%) (Table 2).

Participation in intervention arms by screening method

The proportion of women who chose clinician sampling was lower in both intervention arms (26.2% in opt-in arm and 21.9% in opt-out arm) compared to the control group (29.0%) (Table 3). In the opt-out arm, the difference was statistically significant in all age groups and regions with the biggest differences seen among women aged 60 and 65 (−7.6%) and 40 and 45 (−7.8%) and in the Western (−8.9%) and Central (−8.0%) regions. In the opt-in arm, statistically significant differences were observed in all age groups with the biggest differences among women aged 30 and 35 (−3.2%) and in the Northern (−4.4%) and Southern (−2.7%) regions compared to the control group.

Table 3.

Method of cervical cancer screening by age and region of residence in the randomized pilot study of HPV self-sampling, Estonia 2021.

	Control group (clinician sampling only), %	Women who chose clinician sampling						Women who chose self-sampling
				Difference						Difference
		Opt-out, %	Opt-in, %	Opt-out vs control		Opt-in vs control		Opt-out, %	Opt-in, %	Opt-out vs opt-in
	n = 8737	n = 2797	n = 3357		p-value		p-value	n = 2538	n = 1005		p-value
Total	29.0	21.9	26.2	−7.1	<0.001	−2.8	<0.001	19.8	7.9	11.9	<0.001
Age
30 and 35	29.7	22.3	26.5	−7.4	<0.001	−3.2	0.001	17.4	10.0	7.4	<0.001
40 and 45	30.6	22.8	27.8	−7.8	<0.001	−2.8	0.003	18.8	9.7	9.1	<0.001
50 and 55	30.9	25.0	28.3	−5.9	<0.001	−2.6	0.009	18.5	6.1	12.4	<0.001
60 and 65	25.5	17.9	22.8	−7.6	<0.001	−2.7	0.002	24.1	5.9	18.2	<0.001
Region of residence
Northern	31.7	24.0	27.3	−7.7	<0.001	−4.4	<0.001	19.1	7.9	11.2	<0.001
North-Eastern	18.6	15.7	19.2	−2.9	0.017	0.6	0.596	16.4	3.2	13.2	<0.001
Central	28.9	20.9	28.1	−8.0	<0.001	−0.8	0.605	23.1	9.4	13.7	<0.001
Western	33.1	24.2	32.0	−8.9	<0.001	−1.1	0.488	20.9	9.0	11.9	<0.001
Southern	27.1	19.7	24.4	−7.4	<0.001	−2.7	0.004	21.5	9.1	12.4	<0.001

The proportion of women who chose self-sampling was significantly higher in the opt-out than the opt-in group across all age groups and regions (Table 3). The largest differences were observed among the eldest age group 60 and 65 (18.2%) and in the Central (13.7%) and North-Eastern (13.2%) regions.

In the opt-out group, the proportion of women choosing self-sampling was higher than the proportion choosing clinician sampling among women aged 60 and 65 (24.1% vs 17.9%, p ˂ 0.001) and those living in the Central (23.1% vs 20.9%, p = 0.226), North-Eastern (16.4% vs 15.7%, p = 0.612), and Southern regions (21.5% vs 19.7%, p = 0.076).

Self-sampling experience and screening method preference

Out of 3543 women who chose self-sampling, 3306 (93.3%) also returned a completed questionnaire. The assessment of self-sampling user experience was based on women’s responses to eight statements, as shown in Figure 2. In addition, 3113 women responded to the question about screening method preference, of which 71.9% preferred self-sampling, 24.6% had no preference and 3.5% of respondents preferred clinician sampling for cervical cancer screening in the future. There were no differences in user experience between opt-in and opt-out groups (data not shown).

Figure 2.

Self-sampling user experience in randomized pilot study of HPV self-sampling in organized screening in Estonia, 2021. Total number of responders given in brackets. The percentage indicates the number of responders who agreed with the statement.

Proportion of positive and inadequate hrHPV self-sampling results

In total, there were 488 hrHPV-positive self-samples in this study (n = 13.8%), comprising 302 of the 1805 (16.7%) samples analysed in Synlab Estonia and 186 of the 1742 (10.7%) analysed in United Laboratories of Tartu University Hospital. The proportion of inadequate samples was 0.2% at United Laboratories of Tartu University Hospital. No inadequate samples were identified among the tests analysed in Synlab Estonia.

Discussion

The results of this randomized pilot study in Estonia demonstrated that offering women a choice between HPV self-sampling and conventional clinician sampling significantly increased participation in the organized cervical cancer screening program. The opt-out strategy proved to be the most effective approach, but opt-in also had a significant positive effect – opt-out was preferred among older and opt-in among younger age groups. The impact of opt-in and opt-out on screening participation differed across age groups. The majority of women who chose self-sampling showed a high level of acceptance.

The strengths of this study were the randomized design and a large study sample, allowing to generalize the findings to the cervical screening target population in Estonia. The study also had some limitations. We were unable to determine how many reminders (including self-sampling kits mailed to opt-out group) did not reach the intended recipients due to inaccuracies in e-mail or residential address, potentially leading to underestimation of the effect. Some women who had been randomized into study groups had to be excluded from the analyses as information on their screening participation before the sample cut-off date had been delayed. However, the proportion of excluded participants did not differ between study groups.

Cervical screening participation was higher in both intervention arms compared to control group, while the highest participation was achieved in the opt-out arm. In previous studies, opt-out strategies have resulted in higher response rates than opt-in methods, although the size of effect has varied.^{11,13,22–25} A recent study among women who were due or overdue for screening concluded that the opt-in method resulted in only a minimal increase, and therefore the opt-out strategy should be prioritized.²⁶ The results of the HPV self-sampling feasibility study conducted in Estonia in 2020 among long-term screening non-attenders showed that the highest participation rate was achieved using the opt-out method.¹³ But in terms of negative environmental impact, the direct mailing results in a higher number of wasted sampling kits that are not returned to the laboratory, and this gives some advantage to opt-in strategies. Although we tried to minimize the waste in our study, offering a free-of-charge returning option for unused sampling devices, only 322 kits were returned in the opt-out arm.

There was a difference in the screening behaviour of women in the opt-out and opt-in arms. In the opt-in arm, the proportion of women who attended the clinic was only slightly smaller compared to the control group, and therefore it is likely that the increase in participation occurred on account of women who would not have participated in screening without the option of self-sampling. In addition, the opt-in option may have encouraged screening by both methods. At the same time, more women switched from clinician sampling to self-sampling in the opt-out arm with an additional surge in participation on the account of new attenders.

Self-sampling, regardless of opt-in or opt-out approach, offers the possibility to reduce healthcare costs by reducing the number of healthcare visits, as only women with an hrHPV-positive result (13.8% of all participants) needed a visit to a healthcare facility. This is particularly important in the context of severe shortage of healthcare personnel. This effect was bigger in the opt-out arm.

The opt-out arm showed the most significant participation increase, of 16.6% among women aged 60 and 65, compared to the control group. In this sub-group, the women who chose self-sampling outnumbered those who attended screening at a clinic. Before 2021, cervical cancer screening in Estonia stopped at the age of 55. It is possible that many women over 60 had not been screened for a long time, and now the screening was brought to their homes, and therefore the response rate may have been higher. A Swedish study (2020) has found that self-collection was highly accepted by women aged over 60.²⁷ In contrast, in the opt-in arm, younger women were more active self-sampling users.

In North-Eastern, Central and Southern Estonia, the number of women who chose self-sampling was slightly higher than those who attended conventional screening. Some of these regions are remote areas and therefore all health services may not be very accessible. Self-sampling enhances screening accessibility and helps to overcome several barriers related to clinician screening. Earlier studies have described logistical barriers as a factor preventing women from screening attendance.²⁸

Self-sampling was well accepted by women in our study. For the majority, the instructions were clear and the procedure was considered easy. These findings are consistent with earlier studies indicating that self-sampling generally is well accepted, regardless of age and place of residence.²⁸ The majority of women who chose self-sampling preferred it to clinician sampling as a future option, and this corresponds to earlier findings.^29,30 Surprisingly, only 3.5% of women who chose self-sampling preferred future sample collection by a healthcare provider. As expected, we also saw a minor level of hesitancy related to self-sampling, as some women expressed fear of not doing the test correctly or questioned the reliability of the result, as described also in other studies.³¹ Thus, more efforts should be taken to improve the reliability of self-sampling by increasing women’s awareness and confidence in vaginal sample self-collection. Nevertheless, the general feedback shows that self-sampling is a convenient method and is likely to be adopted by many. However, as we also saw a considerable number of women in the intervention arms who chose clinician sampling, it is likely that self-sampling will not yet replace conventional screenings in Estonia; rather it could be an appropriate alternative for those who, for some reason, cannot or do not prefer the screening at a healthcare facility.

In 2021, the cervical cancer screening participation rate in Estonia increased significantly by 6% compared to the previous year.⁵ As the screening program was recently reformed, there were other factors affecting participation in addition to self-sampling, such as the availability of screening for uninsured women and expansion of the target group, which may have helped to improve screening availability. Furthermore, other reasons such as a change in peoples’ attitude in conditions of pandemic, additional screening reminders and more media coverage may have been involved.

Our study was conducted during the Covid-19 pandemic and the self-sampling opportunity enabled a possibility of screening with no need to visit a clinic. This had advantages during the time when social distancing was recommended. The feedback of several participants indicated that self-sampling had been for them the only way of screening at that time. The Stockholm region in Sweden switched to primary HPV self-sampling in 2021, during the Covid-19 pandemic, resulting in a significant screening coverage increase.³² The wider use of self-sampling is growing, as the Covid-19 pandemic has accelerated efforts to introduce HPV self-sampling in more countries, and is considered vital for increasing the cervical screening rate.³³

The results of this study confirm that the availability of self-sampling and opportunity to choose the screening method had a positive impact and increased screening attendance rate. The use of self-sampling also supports the goal set by the WHO to move towards eliminating the burden of cervical cancer.³⁴ Although self-sampling was offered to only a limited sample in our study, 9.4% of all screened women that year had self-collected the sample. The opt-out strategy is the most effective in terms of bringing more women to screening. While previous studies have shown low efficacy of opt-in strategies as having minimal or no significant effect on participation rate, especially among hard-to-reach women,^25,30 in our study, both opt-out and opt-in approaches showed significant results.

The cost-effectiveness of different self-sampling dissemination methods needs further assessment, and the distribution of sampling kits in pharmacies and at primary healthcare providers should also be evaluated. Furthermore, it is crucial to analyse how many women with a positive HPV self-sampling result will have the additional examinations done. Proper follow-up is critical to gain the advantage of increased screening rates achieved with primary HPV self-testing. Analysing the socio-demographic determinants is essential for getting information: in which groups is the self-sampling most efficient and in which groups are the follow-up rates the lowest?

Conclusion

Offering women a choice between HPV self-sampling and conventional clinician sampling for HPV significantly increased attendance in organized screening, and the opt-out strategy proved to be the most effective. High acceptance of self-sampling indicates that it is an appropriate alternative to conventional screening, and it should be made available within the cervical cancer screening program in Estonia. Self-sampling should be available for all screening-eligible women as it enhances the accessibility of screening and helps to minimize the number of unscreened individuals with an additional benefit of cutting healthcare costs. According to the study results, the use of both opt-out and opt-in strategies should be considered in Estonia and adapted to different age groups. The opt-in strategy requires a comprehensive design and good communication, and it could be offered as an option to the whole screening target population. Additional opt-out strategies should be tested for targeting specific sub-groups as well as long-time non-attenders.

Supplemental Material

sj-jpg-1-msc-10.1177_09691413241268819 - Supplemental material for HPV self-sampling in organized cervical cancer screening program: A randomized pilot study in Estonia in 2021

Supplemental material, sj-jpg-1-msc-10.1177_09691413241268819 for HPV self-sampling in organized cervical cancer screening program: A randomized pilot study in Estonia in 2021 by Reeli Hallik, Kaire Innos, Jaak Jänes, Kai Jõers, Kaspar Ratnik and Piret Veerus in Journal of Medical Screening

Footnotes

Acknowledgements

We thank colleagues from the Estonian Health Insurance Fund, Estonian Cancer Screening Registry, the medical laboratories Synlab Estonia and United Laboratories of Tartu University Hospital, the postal companies Omniva and Itella Smartpost and all participants in cervical cancer screening program.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Estonian Research Council (grant number PRG722) and the Estonian Health Insurance Fund.

ORCID iDs

Reeli Hallik

Kaire Innos

Piret Veerus

Supplemental material

Supplemental material for this article is available online.

References

Miller

Nazeer

Fonn

, et al. Report on consensus conference on cervical cancer screening and management. Int J Cancer 2000; 89: 440–447. PMID: 10760836.

Spence

Goggin

Franco

. Process of care failures in invasive cervical cancer: systematic review and meta-analysis. Prev Med (Baltim) 2007; 45: 93–106. Epub 2007 Jun 22. PMID: 17651792.

Statistic Estonia. Health Insurance. https://andmed.stat.ee/en/stat/ sotsiaalelu__sotsiaalne-kaitse__sotsiaalkindlustus__ravikindlustus (accessed 31 May 2024).

Massad

. Preinvasive disease of the cervix. Clinical gynecologic oncology (ninth edition). Philadelphia, PA: Elsevier Inc, 2018, pp. 1–19.e3. https://doi.org/10.1016/B978-0-323-40067-1.00001-2 (accessed 7 April 2023).

Health Statistics and Health Research Database. VSR12: Cervical cancer screening coverage by PAP-test examination in target population by age. PxWeb [Internet]. https://statistika.tai.ee/pxweb/en/Andmebaas/Andmebaas__ 02Haigestumus__07Soeluuringud/VSR12.px/ (accessed 20 December 2023).

Innos

Baburin

Hallik

, et al. Results of screening programmes on breast, cervical and colorectal cancer in Estonia. Eesti Arst 2022; 101: 281–290.

ECIS - European Cancer Information System. Estimates of cancer incidence and mortality in 2022. https://ecis.jrc.ec.europa.eu/ (accessed 27 February 2024).

Karsa

Dillner

Suonio

, et al. European guidelines for quality assurance in cervical cancer screening: second edition: supplements. Publications Office; 2015. https://doi.org/10.2875/85950 (accessed 12 April 2023).

Akinlotan

Bolin

Helduser

, et al. Cervical cancer screening barriers and risk factor knowledge among uninsured women. J Community Health 2017; 42: 770–778. PMID: 28155005; PMCID: PMC5494033.

10.

Arbyn

Smith

Temin

, et al. Detecting cervical precancer and reaching underscreened women by using HPV testing on self-samples: updated meta-analyses on behalf of the collaboration on self-sampling and HPV testing. Br Med J 2018; 363: 4823. PMID: 30518635; PMCID: PMC6278587.

11.

Verdoodt

Jentschke

Hillemanns

, et al. Reaching women who do not participate in the regular cervical cancer screening programme by offering self-sampling kits: a systematic review and meta-analysis of randomised trials. Eur J Cancer 2015; 51: 2375–2385. Epub 2015 Aug 18. PMID: 26296294.

12.

Gupta

Palmer

Bik

, et al. Self-sampling for human papillomavirus testing: increased cervical cancer screening participation and incorporation in international screening programs. Front Public Health 2018; 6: 1. PMCID: PMC5900042; PMID: 29686981.

13.

Veerus

Hallik

Jänes

, et al. Human papillomavirus self-sampling for long-term non-attenders in cervical cancer screening: a randomised feasibility study in Estonia. J Med Screen 2022; 29: 53–60. Epub 2021 Oct 25. PMID: 34694179.

14.

Estonian Health Insurance Fund. https://tervisekassa.ee/en (accessed 20 December 2023).

15.

Estonian Cancer Screening Registry. https://en.tai.ee/en/r-and-d/registers/estonian-cancer-screening-registry (accessed 7 April 2023).

16.

Sultana

Mullins

English

, et al. Women’s experience with home-based self-sampling for human papillomavirus testing. BMC Cancer 2015; 15: 1–10. PMID: 26536865; PMCID: PMC4634718.

17.

Enerly

Bonde

Schee

, et al. Self-sampling for human papillomavirus testing among non-attenders increases attendance to the Norwegian cervical cancer screening programme. PLoS One 2016; 11: e0151978. PMID: 27073929; PMCID: PMC4830596.

18.

Meijer

CJLM

Berkhof

Castle

, et al. Guidelines for human papillomavirus DNA test requirements for primary cervical cancer screening in women 30 years and older. Int J Cancer 2009; 124: 516–520. PMID: 18973271; PMCID: PMC2789446.

19.

Emakakaelavähi sõeluuring, tegevusjuhend. Eesti Haigekassa 2020. https://www.tervisekassa.ee/partnerile/raviasutusele/haiguste-ennetus/emakakaelavahi-soeluuring (accessed 5 March 2023).

20.

E-Health Record. E-Estonia. https://e-estonia.com/solutions/healthcare/e-health-records/ (accessed 13 April 2023).

21.

Statistical regions in the European Union and partner countries. NUTS And statistical regions. Eurostat 2021. https://ec.europa.eu/eurostat/documents/3859598/10967554/KS-GQ-20-092-EN-N.pdf/9d57ae79-3ee7-3c14-da3e-34726da385cf (accessed 3 March 2023).

22.

Tranberg

Bech

Blaakær

, et al. Preventing cervical cancer using HPV self-sampling: direct mailing of test-kits increases screening participation more than timely opt-in procedures – a randomized controlled trial. BMC Cancer 2018; 18: 273. PMID: 29523108; PMCID: PMC5845195.

23.

Ivanus

Jerman

Fokter

, et al. Randomised trial of HPV self-sampling among non-attenders in the Slovenian cervical screening programme ZORA: comparing three different screening approaches. Radiol Oncol 2018; 52: 399–412. PMID: 30216191; PMCID: PMC6287183.

24.

Aasbø

Tropè

Nygård

, et al. HPV self-sampling among long-term non-attenders to cervical cancer screening in Norway: a pragmatic randomised controlled trial. Br J Cancer 2022; 127: 1816–1826. Epub 2022 Aug 23. PMID: 35995936; PMCID: PMC9643532.

25.

Yeh

Kennedy

De Vuyst

, et al. Self-sampling for human papillomavirus (HPV) testing: a systematic review and meta-analysis. BMJ Glob Heal 2019; 4: e001351. PMID: 31179035; PMCID: PMC6529022.

26.

Winer

Lin

Anderson

, et al. Strategies to increase cervical cancer screening with mailed human papillomavirus self-sampling kits: a randomized clinical trial. JAMA 2023; 330: 1971–1981. PMID: 38015219; PMCID: PMC10685881.

27.

Hermansson

Olovsson

Gustavsson

, et al. Elderly women’s experiences of self-sampling for HPV testing. BMC Cancer 2020; 20: 473. PMID: 32456679; PMCID: PMC7249375.

28.

Nishimura

Yeh

Oguntade

, et al. HPV self-sampling for cervical cancer screening: a systematic review of values and preferences. BMJ Glob Heal 2021; 6: e003743. PMID: 34011537; PMCID: PMC8137189.

29.

Zehbe

Moeller

Severini

, et al. Feasibility of self-sampling and human papillomavirus testing for cervical cancer screening in First Nation women from Northwest Ontario, Canada: a pilot study. BMJ Open 2011; 1: e000030. PMID: 22021733; PMCID: PMC3191400.

30.

Tatara

Wnuk

Miazga

, et al. The influence of vaginal HPV self-sampling on the efficacy of populational screening for cervical cancer—an umbrella review. Cancers (Basel) 2022; 14: 5913. PMID: 36497396; PMCID: PMC9740498.

31.

Camara

Zhang

Lafferty

, et al. Self-collection for HPV-based cervical screening: a qualitative evidence meta-synthesis. BMC Public Health 2021; 21: 1503. PMID: 34348689; PMCID: PMC8336264.

32.

Elfström

Gray

Dillner

. Cervical cancer screening improvements with self-sampling during the COVID-19 pandemic. Elife 2023; 12: e80905. PMID: 38085566; PMCID: PMC10715724.

33.

Serrano

Ibáñez

Robles

, et al. Worldwide use of HPV self-sampling for cervical cancer screening. Prev Med 2022; 154: 106900. Epub 2021 Nov 30. PMID: 34861338.

34.

Global strategy to accelerate the elimination of cervical cancer as a public health problem. World Health Organization. https://www.who.int/publications/i/item/9789240014107 (accessed 23 December 2023)

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB