To stop the spread of sexually transmitted infections (STIs), early diagnosis and timely treatment are essential. However, many STI cases are diagnosed late or not at all for a variety of reasons: infections are often asymptomatic, many patients are reluctant to get tested, and access to clinics, trained health care workers, and/or laboratory testing is often poor.1 Some of these barriers can be addressed by encouraging at-risk populations to sample and/or test themselves with the use of self-sampling devices.1–5
The Delphi Screener, developed by gynecologists in the Netherlands, received Conformité Européenne marking in 2011 and is marketed by Delphi Bioscience BV (Scherpenzeel, the Netherlands). More than 10,000 European and American women have used the Screener in cervical cancer screening programs to collect samples for subsequent liquid cytology and/or human papillomavirus testing.6–12 Some studies showed that adding the possibility of self-sampling increased participation rates in cervical cancer screening programs,6,10,11,13 and 2 studies found good test result concordance with samples obtained by gynecologists.12,14 Self-sampling has also been successfully used for the diagnosis of other STIs and vaginal infections in a variety of settings including African settings.3,15–18 However, thus far, the Delphi Screener has not yet been used in African settings or to test for STIs or vaginal infections. We therefore conducted a pilot study in Kigali, Rwanda, to determine the feasibility and acceptability of the Delphi Screener among Rwandan women.
The study population consisted of a subsample of 60 Rwandan women participating in a multicountry, observational, prospective cohort study to establish baseline ranges of vaginal microbiome and immunology biomarkers: 30 HIV-negative female sex workers (referred to as the FSW group) and 30 HIV-positive women attending an HIV clinic at Muhima Hospital (referred to as the clinic group). All participants were sexually active, not pregnant, and between 22 and 35 years old. All clinic group participants were asymptomatic, on antiretroviral treatment for at least 6 months, and had a CD4+ count of at least 350 cells/μL. Participants provided written informed consent for the main study and the self-sampling substudy separately. The study was approved by the Rwanda National Ethics Committee and the Ethics Committee of Ghent University Hospital in Ghent, Belgium.
The main study consisted of 7 study visits: visits 1 to 5 at various time points over 2 menstrual cycles and visits 6 and 7 at 3-monthly intervals thereafter. The self-sampling visits were scheduled between visits 5 and 7, with a minimum of 1 week between any 2 main or substudy visits. Participants were shown how to use the Screener, and posters with user instructions in the local language (Kinyarwanda) were available in the clinic room. At the first self-sampling visit, the study clinician was present in the room unless the participant asked her to leave (opt-out), and during the second self-sampling visit, the participant was asked whether she preferred the clinician to be in the room (opt-in). The self-sampling procedures consisted of washing hands, unscrewing a test tube, opening the plastic pouch containing the Screener, inserting the Screener in the vagina in supine or squatting position, pushing a button to release its 3 mL liquid contents (buffered saline) into the vagina, counting to 3, releasing the button to collect the fluid sample back into the screener, removing the Screener from the vagina, transferring the content in the test tube, closing the test tube, discarding the Screener, and washing the hands again. Participants were interviewed at each visit. In addition, the clinician recorded her observations when present in the room. The duration of self-sampling was measured, and the collected samples were weighed.
Data analysis was performed using SPSS.19 Comparisons between groups at the same visit were made by Pearson χ2 test for proportions and Mann-Whitney U test for medians. Comparisons between visits were made by McNemar test for proportions and Wilcoxon paired signed rank test for medians. Multivariable logistic regression models included age and variables that were statistically significantly associated with the outcome of interest in bivariable models; 90% confidence intervals were calculated owing to small sample sizes.
All participants completed both self-sampling visits, except for 1 woman in the clinic group. The median age was 26.5 years (range, 22–33) years in the FSW group and 30.5 (22–35) years in the clinic group. Fifty-seven percent of the women in the clinic group had attended secondary school, but only 13% of the FSWs had. None of the participants had ever used tampons or self-sampling devices, but 13% of the FSW group and 37% of the clinic group had ever inserted vaginal medications. All of these differences reached statistical significance (P < 0.05).
None of the women asked the clinician to leave the room during self-sampling at the first visit; during the second visit, 30 women (33% of the FSW group and 69% of the clinic group, P = 0.006) asked for the clinician to be present. One woman in each group chose to collect the sample while squatting; all other women chose the supine position. The median duration of self-sampling at both visits was 4 minutes, but the range decreased from 2 to 14 minutes at visit 1 to 2 to 6 minutes at visit 2 (Table 1). The median amount of fluid collected improved significantly, from 1,716 mg at visit 1 to 2322 mg at visit 2 (P < 0.001). Fluid leakage from the vagina after self-sampling was observed by the clinicians 10 times (10/60) at visit 1 and 5 times (5/30) at visit 2 (P = 1.000). The clinicians also observed that 6 women (6/60) at visit 1 and 4 women (4/30) at visit 2 were trembling during self-sampling (P = 0.625). Four women at visit 1 and none at visit 2 asked for the instructions to be repeated.
At the first and second visits, 63% and 71% of the participants, respectively, deemed the Screener to be both very easy to use and very comfortable (P = 0.267). Only 1 participant throughout the study reported a physical complaint (“pain”) after self-sampling. At the first visit, 42 (70%) of the participants stated that they preferred the Screener, whereas 18 (30%) preferred the pelvic examination performed by a clinician (Table 1). At the second visit, 37 preferred the Screener (63%), whereas 22 preferred the pelvic examination (37%). Of those preferring a pelvic examination, 77% reported as the main reason that pelvic exams allow for internal inspection by a clinician.
Good feasibility (defined as having collected at least 1.5 g of fluid, which is 50% of the volume inserted, combined with no self- or clinician-reported problems), was achieved by 51% of participants at visit 1 and 61% at visit 2 (P = 0.238). In bivariable analyses using data from visit 2, good feasibility and good acceptability (defined as participants preferring the Screener over a pelvic examination performed by clinicians) were statistically significantly associated with a higher educational level and reporting self-sampling to be both very easy and very comfortable. In multivariable models, only the latter was statistically significantly associated with good feasibility and acceptability (Table 2). Study group, physical complaints before and after Screener use, self-reported difficulties, and ever use of vaginal medication were not associated with good feasibility or good acceptability (P > 0.100).
This study showed reasonable feasibility and good acceptability of the Delphi Screener among Rwandan women. Approximately two thirds of the women preferred the Screener over a pelvic examination, which is slightly lower than reports from the Netherlands (75%), the United States (79%), and Italy (78%).7–9 Some study participants were visibly nervous during self-sampling, or reported to feel too inexperienced or too insecure for self-sampling, but this improved with practice. Others said that they prefer speculum examinations because internal inspection by a clinician may identify abnormalities that would otherwise not be detected. This was also reported in other African self-sampling studies, probably because most African women rarely have the opportunity to be examined by a clinician.2,4 Therefore, we think that feasibility and acceptability could be maximized by allowing women to practice with self-sampling in the presence of a clinician and by reassuring women that self-sampling should be considered an addition to, and not a replacement of, pelvic examinations.
The limitations of our study include a limited generalizability (self-sampling was only performed in women of reproductive age in an urban clinic setting), selection bias (all women were already participating in a research study), and social desirability bias. The strengths of our study are that we only included women who had experience with pelvic examinations, that we asked each woman to self-sample twice with a median of 7 weeks in between self-sampling visits, and that we used multiple methods to assess feasibility and acceptability. For example, the agreement between participant-reported and clinician-observed findings was excellent (data not shown).
To conclude, we believe that the Screener could be used by Rwandan women within research studies and in urban clinic settings. Widespread use in cervical cancer or other reproductive health screening programs in rural areas or in home settings would require additional pilot testing and evaluation. The fluid collected by the Screener can be used for molecular diagnostics (as is the case with self-sampled swabs) but could also be used for cytology and to test for soluble molecules, such as antibodies and other immune mediators. However, the performance of the Screener in combination with each of these tests would have to be evaluated.
1. Garland SM, Tabrizi SN. Diagnosis of sexually transmitted infections using self-collected non-invasive specimens. Sex Health 2004; 1: 121–126.
2. van de Wijgert JHHM, Altini L, Jones HE, et al. Two methods of self-sampling compared to clinician sampling to detect reproductive tract infections in Gugulethu, South Africa. Sex Transm Dis 2006; 33: 516–523.
3. Lippman SA, Jones HE, Luppi CG, et al. Home-based self-sampling and self-testing for sexually transmitted infections: Acceptable and feasible alternatives to provider-based screening in low-income women in São Paulo, Brazil. Sex Transm Dis 2007; 34: 421–428.
4. Jones HE, Altini L, de Kock A, et al. Home-based versus clinic-based self-sampling and testing for sexually transmitted infections in Gugulethu, South Africa: Randomised controlled trial. Sex Transm Infect 2007; 83: 552–557.
5. Stewart DE, Gagliardi A, Johnston M, et al. Self-collected samples for testing of oncogenic human papillomavirus: A systematic review. J Obstet Gynaecol Can 2007; 29: 817–828.
6. Gök M, Heideman DAM, Van Kemenade FJ, et al. HPV testing on self collected cervicovaginal lavage specimens as screening method for women who do not attend cervical screening: A cohort study. BMJ 2010; 340: c1040.
7. Jones HE, Wiegerinck MAHM, Nieboer TE, et al. Women in the Netherlands prefer self-sampling with a novel lavaging device to clinician collection of specimens for cervical cancer screening. Sex Transm Dis 2008; 35: 916–917.
8. Jones HE, Brudney K, Sawo DJ, et al. The acceptability of a self-lavaging device compared to pelvic examination for cervical cancer screening among low-income women. J Womens Health 2012; 21: 1275–1281.
9. Igidbashian S, Boveri S, Spolti N, et al. Self-collected human papillomavirus testing acceptability: Comparison of two self-sampling modalities. J Womens Health 2011; 20: 397–402.
10. Virtanen A, Antilla A, Luostarinen T, et al. Self-sampling versus reminder letter: Effects on cervical cancer screening attendance and coverage in Finland. Int J Cancer 2010; 128: 2681–2687.
11. Rossi PG, Marsili LM, Camilloni L, et al. The effect of self-sampled HPV testing on participation to cervical cancer screening in Italy: A randomised controlled trial. Brit J Cancer 2011; 104: 248–254.
12. Deleré Y, Schuster M, Vartazarowa E, et al. Cervicovaginal self-sampling is a reliable method for determination of prevalence of human papillomavirus genotypes in women aged 20 to 30 years. J Clin Microbiol 2011; 49: 3519–3522.
13. Virtanen A, Nieminen P, Luostarinen T, et al. Self-sample HPV tests as an intervention for nonattendees of cervical cancer screening in Finland: A randomized trial. Cancer Epidemiol Biomarkers Prev 2011; 20: 1960–1969.
14. Brink AATP, Meijer CJLM, Wiegerinck MAHM, et al. High concordance of results of testing for human papillomavirus in cervicovaginal samples collected by two methods, with comparison of a novel self-sampling device to a conventional endocervical brush. J Clin Microbiol 2006; 44: 2518–2523.
15. Mitchell S, Ogilvie G, Steinberg M, et al. Assessing women’s willingness to collect their own cervical samples for HPV testing as part of the ASPIRE cervical screening project in Uganda. Int J Gynecol Obstet 2011; 114: 111–115.
16. Safaeian M, Kiddugavu M, Gravitt PE, et al. Prevalence and risk factors for carcinogenic human papillomavirus infections in Rural Rakai, Uganda. Sex Transm Infect 2008; 84: 306–311.
17. Safaeian M, Kiddugavu M, Gravitt PE, et al. Comparability of self-collected vaginal swabs and physician-collected cervical swabs for detection of human papillomavirus infections in Rakai, Uganda. Sex Transm Dis 2007; 34: 429–436.
18. Shih SL, Graseck AS, Secura GM, et al. Screening for sexually transmitted infections at home or in the clinic? Curr Opin Infect Dis 2011; 24: 78–84.
19. SPSS [computer program]. Version 19.00. Armonk, NY: IBM Corp, 2010.