Printer Friendly

Self-collected HPV testing improves participation in cervical cancer screening: a systematic review and meta-analysis.

Cervical cancer screening has dramatically reduced the incidence of cervical cancer in those countries that have implemented screening programs; (1) however, the majority of cervical cancers continue to occur in women who do not attend regular screening (underscreened) or who have never been screened. (2) The Papanicolaou (Pap) test is one of the standard screening tests used for the detection of cervical cancer. The Pap test, which is administered during a pelvic examination, involves the examination of cervical cells for abnormal histological changes consistent with cervical cell dysplasia and cervical cancer. Barriers to Pap testing can be grouped into three broad categories: clinic-level, testlevel, and personal-level barriers. (3-6) Clinic-level barriers include such factors as the lack of a family physician, inconvenient clinic hours and lack of available transportation to the clinic. (5,7) Barriers at the test level are reported as physical discomfort or anticipation of discomfort. (6) Personal-level barriers include those that pertain to religious and cultural beliefs, language barriers, and lack of knowledge around cervical cancer screening. (6,7)

The human papillomavirus (HPV) is a necessary cause, but not a sufficient cause, for cervical cancer. (8) More and more, HPV testing is being considered an important addition to cervical cancer screening programs, either by way of co-testing with the Pap test or as a primary screening test used to triage women for subsequent Pap testing. (9-14) HPV testing seeks to identify the presence of HPV in the vaginal canal. The presence of oncogenic HPV types confers an elevated risk for the future development of cervical cancer. In women over 30 years of age, HPV testing has been shown to be more sensitive than Pap testing for the detection of cervical interstitial neoplasia (CIN) grade 2/3+. (15-17) Due to the lower specificity of HPV testing compared to Pap screening, it has been recommended that HPV testing be used only as a screening modality in women older than 30, as a co-test or that those who test HPV-positive be triaged to undergo Pap testing. (9,16) In order for a two-phase double-screening regimen to take place, adequate follow-up procedures and infrastructure are required at the primary care and/or population level.

Trials comparing self-collected HPV test samples to physiciancollected samples found that both methods provided equally viable samples for detecting HPV. (18-20) In addition, women who performed self-collection were found to have similar cancer-related outcomes to women who underwent HPV testing by a physician; (21) as such, self-collected HPV testing has been investigated as an alternative to physician-collected samples.

Beyond the efficiency and effectiveness of HPV testing as a diagnostic test, studies have also considered women's attitudes toward self-collection and found that women have a high acceptance of and positive attitudes toward the use of self-collected HPV testing. (22-25)

Being able to collect one's own sample has the potential to address many of the current barriers around screening and improve the participation of women who are most at risk for cervical cancer, i.e., those who are underscreened and/or hard to reach. (26) Self-collected testing, especially when conducted at home, removes the dependence on clinic hours, transportation, discomfort with the physical exam, and language barriers with a care provider, and could provide a culturally or religiously safe procedure. In short, it appears that the use of self-collected testing could play a major role in the use of HPV testing for cervical cancer screening. (27,28)

While studies have examined and considered many aspects relating to HPV self-testing, there is a need to examine the evidence for whether self-collected HPV testing in practice increases participation and compliance in cervical cancer screening.

The objective of this review is to determine whether the option of self-collected HPV testing increases cervical cancer screening participation (through HPV testing) for women living in developed countries who are never or underscreened for cervical cancer.


Selection of studies

A systematic literature review was conducted to identify relevant articles that examined the use of HPV self-collected testing for cervical cancer screening in never- and underscreened women. The databases Medline and Embase were systematically searched for peer-reviewed articles published between January 1, 1990 and July 15, 2012, with 87 and 155 relevant citations identified within each respective database.

Keywords and Medical Subject Headings (MeSH) were chosen to capture the constructs of HPV infection, cervical cancer, self sampling and HPV DNA testing (Table 1).

The literature search was restricted to peer-reviewed articles that clearly demonstrated the comparison of self-collected HPV testing as a primary screening test to standard Pap testing in women who did not routinely participate in cervical cancer screening programs. This review was restricted to developed countries where Pap testing is the standard for cervical cancer screening. One exception is Mexico. Mexican studies were also considered due to the importance of Mexico in the North American context both geopolitically and from an immigration perspective.

Studies were included if group allocation was clearly described and compliance was available for both the intervention group and the control group. Accepted control groups were those that offered a standard invitation to undergo Pap testing at a local Health Care Clinic or that offered Pap testing via the normal procedures of the jurisdiction within which the study was conducted.

Studies that employed an ecological design were excluded as individual-level crude rates of compliance in testing could not be determined. Conference abstracts, editorials, commentaries and other unpublished manuscripts were excluded, in addition to articles that included duplicate datasets or male participants. Studies involving male participants were excluded as the objective of this review was to examine compliance specifically in cervical cancer screening.

There were no language restrictions on publications included. Of all the publications, only one was published in a foreign language (French). For this paper, the data were extracted and confirmed by a second, independent reviewer comfortable in French.

Articles were reviewed in full if the study abstract met the inclusion criteria or if an article lacked sufficient information in the abstract to make an inclusion/exclusion judgement, so as to minimize errors of omission.

Data extraction and outcome measures

The following information was extracted for each study: first author, publication journal, date of publication, country and city of study, dates of study completion, HPV detection method, sample collection device, population description, age distribution of participants, definition of non-attending or underscreened, study design, total number of participants in each group, compliance in each group by test (HPV and Pap), and percent positivity of high risk HPV.

Two independent abstractors (SR and DW) extracted all data for quality control, and any discordance was resolved by consensus. The most recent article was included for articles that presented on the same dataset.

Methodological quality of studies

Study quality was assessed by looking at factors of appropriate randomization, reporting of allocation and loss to follow-up or dropouts, and representativeness of the sample to the underlying target population. A modified Downs and Black Tool (29) was used to guide the quality assessment of the studies. This allowed for the assessment of both randomized and non-randomized health intervention studies, by examining both internal and external validity. The terms low, medium and high referred to a qualitative judgement of whether the studies met few (low), some (medium) or most (high) of the criteria around randomization, reporting of group allocation and loss to follow-up, and representativeness of sample to the underlying target population.

Statistical analysis

All studies included in this review were used in the analysis to produce a pooled estimate of the relative compliance of HPV self collected testing compared to Pap testing. Additional sensitivity analyses were conducted based on different study designs to ensure that variations in study design did not meaningfully impact the overall pooled estimate. The study-specific relative compliances and the overall pooled relative compliance were calculated using an intention-to-treat analysis. Women randomized to the HPV arm who elected to undergo Pap testing were analyzed as belonging to the HPV arm. However, for each study the compliance was reported by testing modality for each study arm.

The overall pooled relative compliance, with 95% confidence intervals, between the HPV self-testing groups and the control groups was calculated using a random effects model. We had to decide between using a fixed or random effects model for our metaanalysis. A fixed effects model is commonly used when all studies are attempting to estimate one true (fixed) effect size with a narrow distribution. A study with low variance is given more weight in a fixed effect model because it is seen as a better estimation of the true effect. A random effects model is used when individual studies are estimating different true effects, with their own distributions. In our meta-analysis, each country has its own true and valid measure of the effect of HPV self-testing on cervical cancer screening participation, which is impacted by the social norms and values of that country's culture. To allow the true effect to vary between studies, we used a random effects model so that each country could contribute equally (i.e., equal weights) to the overarching mean distribution of effect. The statistical heterogeneity was assessed using the I2 statistic, which measures the variation across studies that is due to heterogeneity rather than chance. (30) All statistical tests were two-sided and analysis was performed using Stata 11.2 (Stata Corporation, College Station, TX, USA).


The systematic literature search identified a total of 242 articles, 87 from Medline and 155 from Embase (Figure 1). After removing duplicate articles, 178 article titles and abstracts were reviewed for inclusion. A total of 21 full-text articles were retrieved for full review, of which 10 met both the inclusion and exclusion criteria and were included in this review (Table 2).

Study characteristics

All studies were conducted between 2003 and 2010. The majority of the studies took place in Europe, specifically, the Netherlands, (31-33) Italy, (34) United Kingdom, (35) Sweden, (36) Finland, (37) and France. (38) All of the European studies used a randomized controlled trial design and took advantage of population registries to be able to identify nonattendees to cervical cancer screening programs. Non-attendees were subsequently randomized to either receive a HPV self-collection kit by mail or receive an invitation to undergo Pap testing.

Two studies were set in North America: one in Mexico (39) and one in the United States. (40) The North American studies differed from the European studies in their design and population definitions. In both North American settings, a door-to-door recruitment approach was used. The US study was non-randomized and was conducted in a neighbourhood that had a low rate of screening. Women who reported not receiving routine Pap testing were offered the option of performing a HPV self-test (delivered to the door) or of receiving a coupon to attend a free Pap-testing clinic. The Mexican study targeted underscreened women, i.e., those with "limited access to health services" and who would only be screened a few times over their lifetime. (39) Mexican participants were randomized from a database of women enrolled in a community-based program for women with limited access to health services. Randomized women received either an invitation to a free Pap-testing clinic, or a self-collected HPV testing kit delivered to her home by a study nurse. Despite the differences in study design, both of these studies employed similar methods for HPV self-testing and obtained similar HPV-positivity results comparable to those found in the European trials.

All of the studies included in the review were similar with regard to targeted age demographic, methods for HPV self-collection and Pap testing, and urban setting. In addition, all studies targeted women who were considered overdue for or who did not attend regular cervical cancer screening in their own jurisdictions.

The studies included did differ in the use of the self-collection device (Table 2). A wide variety of devices were employed across studies: cervovaginal brush, (31,33,39) vaginal swabs, (35,36,38,40) and lavage methods. (32,34,37)

Methodological quality of studies

Study quality was deemed to be high in all studies reviewed. All studies reported on allocation of participants and attrition, and 9 out of 10 used a randomized allocation design. The level of compliance in testing varied between studies but was consistent within a given study, which provided confidence in the study recruitment and internal validity.

Compliance of HPV self-testing compared to Pap testing

The relative compliance of HPV self-collected testing compared to Pap testing was significantly greater than 1.0 in all reviewed studies; indicating that women were significantly more likely to participate in screening if they were offered HPV self-testing (Table 3). The pooled relative compliance was 2.14 (95% CI 1.30-3.52). These results indicate that women were twice as likely to participate in cervical screening if they were offered a HPV self-testing home kit compared to women who were invited to the clinic to undergo Pap testing. A Forest plot of the studies and the pooled effect estimate (Figure 2) indicated that statistical heterogeneity (I2) was 99.5% (p<0.0001), representing significant heterogeneity between studies.

Overall, the compliance in screening varied widely between studies, with compliance reported in the HPV arms of between 10.2% and 98.2% and in the Pap test arms of between 4.5% and 86.8%. The highest study-specific compliances for both HPV testing and Pap testing were reported in the US and Mexican studies. All compliance and relative compliance estimates were unadjusted for any additional covariates, such as age, educational and marital status. Only the non-randomized study (US) examined age and education level, and found that younger age and higher education were associated with HPV self-testing compliance. (40)

HPV prevalence

All of the studies reported high specimen quality from self-collected samples for HPV testing. The percent positivity of high-risk HPV among those who administered the self-collected test ranged from 6.04% to 21.34%. Of the 10 studies, 8 used the hybrid-two capture assay for HPV detection, (32-37,39,40) and 2 used PCR genotype specific assays. (31,38)


To our knowledge, this is the first systematic literature review and meta-analysis that addresses the specific question of whether offering HPV self-testing, compared to Pap testing, improves participation in cervical cancer screening among women who are never or underscreened for cervical cancer. We found that never/underscreened women offered HPV self-testing were twice as likely to comply with/participate in cervical cancer screening. A large variety of self-collection devices were used so it is still unclear what the best HPV self-collection devices are for collecting reliable samples and maximizing comfort for women.

Many studies have examined the acceptance, reliability, and accuracy of self-collected HPV testing. (22-25) Overwhelmingly, HPV selftesting has been shown to have high acceptance among women, and women are able to collect good samples for testing using this method. (18-20) These factors support the notion that HPV self-collected testing has the potential to significantly improve cervical cancer screening compliance in women. Given that the studies reviewed in this analysis provided significant evidence that offering HPV selftesting did improve participation, HPV self-testing should be pur sued as a complement or alternative to Pap testing in women older than 30 who do not attend regular screening programs.

It is worth noting that HPV testing can potentially produce a relatively larger number of false positives (compared to Pap testing), due to transient infections. Infrastructure to provide timely and effective follow-up is an important component of any health care system that provides wide-scale HPV testing to ensure that women who are at an increased risk of cervical cancer are provided with appropriate follow-up care. (10)

Limitation of the studies reviewed

Large heterogeneity in compliance was observed between studies with reported compliance of HPV self-testing as high as 80.5% and others as low as 10.2%. The magnitude of the compliance was study-specific, however the relative compliance between studies was largely similar. This wide difference in observed compliance between studies was the main rationale for selecting a randomeffects model. It was theorized that these studies and their populations were measuring different true effects and reflected the possible distribution of participation.

The studies included were predominately conducted in urban settings, with the exception of those in the US, Mexico, and one substudy site in rural Italy. The Italian study did find a significantly different relative compliance overall, favouring the use of HPV self-testing; however, within the rural substudy site, the compliance was not significantly different between the HPV self-test and the Pap test. The authors did note that the recruitment time period for this rural site was not optimal and so it is inconclusive whether there are true urban-rural differences. The US and Mexico studies both used door-to-door recruitment methods and had much higher response rates to both HPV self-testing and Pap testing; however, the door-to-door recruitment may have artificially increased their overall participation rates, plus - unfortunately - screening programs using door-to-door delivery are not sustainable in many areas.

The US and Mexico studies were included in the pooled estimate, despite obvious differences from the European trials. The four main differences were: rural locale, high overall response rate, lack of randomization in the US study, and lack of definitive Pap testing history in the Mexican study. A sensitivity analysis was conducted to estimate the pooled relative compliance without the inclusion of those two studies. The sensitivity analysis yielded an estimated pooled relative compliance of 2.34 (95% CI 1.47-3.70), which indicated that the addition of these two community trials did not alter the overall results, and if anything their inclusion provides a more conservative estimate of effect. However, potential differences in compliance between rural and urban women may still exist due to the observed lower relative compliance when both rural studies are taken into account, coupled with the non-significant finding in the rural Italian substudy site. Differences between rural and urban settings should continue to be elucidated, especially given that rural areas have been shown to have lower screening participation and increased burden of cervical cancer. (41-43)

Limitations of the review

One of the limitations of this review is the scope of articles included. The exclusion of middle-/low-income study sites led to the omission of four studies - three from India and one from China that explored HPV self-testing in a developing country context. (44-46) The authors of the three Indian studies concluded that selfcollected testing improved screening participation in women. Despite not including the data in the review, the overarching message of increased participation with the provision of HPV self-testing was echoed throughout these studies. Four additional studies, which examined uptake of HPV self-testing, were omitted due to a lack of comparison group. (24,47-49) Within this group of studies, the uptake of HPV self-test was reported as between 32.0% and 58.0%, a comparable estimation to the HPV self-testing uptake reported by the studies included in this review.

Another limitation of this review is that studies employing multiple intervention arms were only analyzed based on the comparison between the Pap testing control group and the intervention arm that employed the use of mailed HPV self-collection test kits. The other intervention arms

that were included in some trials consisted of providing HPV testing in a clinic setting or having participants request a self-collection kit by phone. The restriction to only analyze participants who were provided with HPV self-collected testing by mail was to improve comparability across studies. Additionally, if a study employed a multi-phase or crossover design, only the initial phase of the study was included in the analysis, again to ensure that the comparison groups across the trials were as similar as possible, as participation response in subsequent phases may have been impacted due to increased study exposure from a prior invitation or study information relating to cervical cancer screening. This review attempted to simplify the complexity in many of these trials so that they could be reasonably compared, which allowed the authors to identify the overarching trend in women's participation in cervical cancer screening programs when provided with an offer of HPV self-collected testing.

Potential publication bias (for null results) was addressed by searching conference proceedings. All conference abstracts were accounted for by identified publications. Despite the potential for outstanding null effect studies, a null effect would not change the pooled effect from these 10 positive studies. A broad search criterion was used to reduce the potential for missing relevant articles. It is unlikely that the literature is missing a large negative study, and though one negative study might pull the pooled effect towards the null, it would not negate the significantly positive effect.


HPV self-testing could significantly improve cervical cancer screening participation, especially in those who are never or underscreened for cervical cancer. As HPV testing becomes more widely accepted as a primary screening tool or co-testing approach with Pap testing, new approaches to cervical cancer screening delivery should be considered. However, appropriate follow-up and treatment for women who do test high-risk HPV-positive need to develop concurrently with HPV self-testing delivery. Future research efforts in HPV self-testing should focus on exploring effective HPV self-test delivery methods and infrastructure, and examining the uptake in rural areas, where the use of HPV self-testing has the potential to make dramatic improvements in those communities.

Acknowledgement: Funded by Cancer Care Ontario.

Conflict of Interest: None to declare.


(1.) Gustafsson L, Ponten J, Zack M, Adami HO. International incidence rates of invasive cervical cancer after introduction of cytological screening. Cancer Causes Control 1997;8(5):755-63.

(2.) Decker K, Demers A, Chateau D, Musto G, Nuget Z, Lotocki R, Harrison M. Papanicolaou test utilization and frequency of screening opportunities among women diagnosed with cervical cancer. Open Med 2009;3(3):140-47.

(3.) Black AT, McCulloch A, Martin RE, Kan L. Young women and cervical cancer screening: What barriers persist? Can J Nurs Res 2011;43(1):8-21.

(4.) Redwood-Campbell L, Fowler N, Laryea S, Howard M, Kaczorowski J. 'Before you teach me, I cannot know': Immigrant women's barriers and enablers with regard to cervical cancer screening among different ethnolinguistic groups in Canada. Can J Public Health 2011;102(3):230-34.

(5.) Waller J, Jackowska M, Marlow L, Wardle J. Exploring age differences in reasons for nonattendance for cervical screening: A qualitative study. BJOG 2012;119(1):26-32.

(6.) Logan L, McIlfatrick S. Exploring women's knowledge, experiences and perceptions of cervical cancer screening in an area of social deprivation. Eur J Cancer Care (Engl) 2011;20(6):720-27.

(7.) Brown DR, Wilson RM, Boothe MAS, Harris CES. Cervical cancer screening among ethnically diverse black women: Knowledge, attitudes, beliefs, and practices. J Natl Med Assoc 2011;103(8):719-28.

(8.) Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189(1):12-19.

(9.) Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. Am J Clin Pathol 2012;137(4):516-42.

(10.) Petignat P, Vassilakos P. Is it time to introduce HPV self-sampling for primary cervical cancer screening? J Natl Cancer Inst 2012;104(3):166-67.

(11.) Agorastos T, Sotiriadis A, Chatzigeorgiou K. Can HPV testing replace the pap smear? Ann N Y Acad Sci 2010;1205:51-56.

(12.) Bhatla N, Dar L, Patro AR, Kumar P, Kriplani A, Gulati A, et al. Can human papillomavirus DNA testing of self-collected vaginal samples compare with physician-collected cervical samples and cytology for cervical cancer screening in developing countries? Cancer Epidemiol 2009;33(6):446-50.

(13.) Schiffman M, Wentzensen N, Wacholder S, Kinney W, Gage JC, Castle PE. Human papillomavirus testing in the prevention of cervical cancer. J Natl Cancer Inst 2011;103(5):368-83.

(14.) Wright TC, Jr. Cervical cancer screening in the 21st Century: Is it time to retire the PAP smear? Clin Obstet Gynecol 2007;50(2):313-23.

(15.) Ronco G, Giorgi-Rossi P, Carozzi F, Confortini M, Dalla Palma P, Del Mistro A, et al. Efficacy of human papillomavirus testing for the detection of invasive cervical cancers and cervical intraepithelial neoplasia: A randomized controlled trial. Lancet Oncology 2010;11:249-57.

(16.) Leinonen M, Nieminen P, Kotaniemi-Talonen L, Malila N, Tarkkanen J, Laurila P, et al. Age-specific evaluation of primary human papillomavirus screening vs conventional cytology in a randomized setting. J Natl Cancer Inst 2009;101(23):1612-23.

(17.) Mayrand M-H, Duarte-Franco E, Rodrigues I, Walter SD, Hanley J, Ferenczy A, et al. Human papillomavirus DNA versus Papanicolaou screening tests for cervical cancer. N Engl J Med 2007;357(16):1579-88.

(18.) Karwalajtys T, Howard M, Sellors JW, Kaczorowski J. Vaginal self sampling versus physician cervical sampling for HPV among younger and older women. Sex Transm Infect 2006;82(4):337-39.

(19.) Ogilvie GS, Patrick DM, Schulzer M, Sellors JW, Petric M, Chambers K, et al. Diagnostic accuracy of self collected vaginal specimens for human papillomavirus compared to clinician collected human papillomavirus specimens: A meta-analysis. Sex Transm Infect 2005;81(3):207-12.

(20.) Petignat P, Faltin DL, Bruchim I, Tramer MR, Franco EL, Coutlee F. Are selfcollected samples comparable to physician-collected cervical specimens for human papillomavirus DNA testing? A systematic review and meta-analysis. Gynecol Oncol 2007;105(2):530-35.

(21.) Moscicki AB, Widdice L, Ma Y, Farhat S, Miller-Benningfield S, Jonte J, et al. Comparison of natural histories of human papillomavirus detected by clinician- and self-sampling. Int J Cancer 2010;127(8):1882-92.

(22.) Huynh J, Howard M, Lytwyn A. Self-collection for vaginal human papillomavirus testing: Systematic review of studies asking women their perceptions. J Low Genit Tract Dis 2010;14(4):356-62.

(23.) Igidbashian S, Boveri S, Spolti N, Radice D, Sandri MT, Sideri M. Self-collected human papillomavirus testing acceptability: Comparison of two selfsampling modalities. J Womens Health (Larchmt) 2011;20(3):397-402.

(24.) Wikstrom I, Stenvall H, Wilander E. Attitudes to self-sampling of vaginal smear for human papilloma virus analysis among women not attending organized cytological screening. Acta Obstet Gynecol Scand 2007;86(6):720-25.

(25.) Zehbe I, Moeller H, Severini A, Weaver B, Escott N, Bell C, 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(1):e000030-e.

(26.) Ogilvie G, Krajden M, Maginley J, Isaac-Renton J, Hislop G, Elwood-Martin R, et al. Feasibility of self-collection of specimens for human papillomavirus testing in hard-to-reach women. CMAJ2007;177(5):480-83.

(27.) Schmeink CE, Bekkers RL, Massuger LF, Melchers WJ. The potential role of self-sampling for high-risk human papillomavirus detection in cervical cancer screening. Rev Med Virol 2011;21(3):139-53.

(28.) Zhao FH, Lewkowitz AK, Chen F, Lin MJ, Hu SY, Zhang X, et al. Pooled analysis of a self-sampling HPV DNA test as a cervical cancer primary screening method. J Natl Cancer Inst 2012;104(3):178-88.

(29.) Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 1998;52:377-84.

(30.) Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ2003;327(7414):557-60.

(31.) Bais AG, van Kemenade FJ, Berkhof J, Verheijen RH, Snijders PJ, Voorhorst F, et al. Human papillomavirus testing on self-sampled cervicovaginal brushes: An effective alternative to protect nonresponders in cervical screening programs. Int J Cancer 2007;120(7):1505-10.

(32.) Gok M, Heideman DAM, van Kemenade FJ, Berkhof J, Rozendaal L, Spruyt JWM, et al. HPV testing on self collected cervicovaginal lavage specimens as screening method for women who do not attend cervical screening: Cohort study. BMJ 2010;340:c1040. doi: 10.1136/bmj.c1040.

(33.) Gok M, van Kemenade FJ, Heideman DA, Berkhof J, Rozendaal L, Spruyt JW, et al. Experience with high-risk human papillomavirus testing on vaginal brush-based self-samples of non-attendees of the cervical screening program. Int J Cancer 2012;130(5):1128-35.

(34.) Giorgi Rossi P, Marsili LM, Camilloni L, Iossa A, Lattanzi A, Sani C, et al. The effect of self-sampled HPV testing on participation to cervical cancer screening in Italy: A randomised controlled trial (ISRCTN96071600). Br J Cancer 2011;104(2):248-54.

(35.) Szarewski A, Cadman L, Mesher D, Austin J, Ashdown-Barr L, Edwards R, et al. HPV self-sampling as an alternative strategy in non-attenders for cervical screening - a randomised controlled trial. Br J Cancer 2011;104(6):915-20.

(36.) Wikstrom I, Lindell M, Sanner K, Wilander E. Self-sampling and HPV testing or ordinary Pap-smear in women not regularly attending screening: A randomised study. Br J Cancer 2011;105(3):337-39.

(37.) 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):1960-69.

(38.) Piana L, Leandri FX, Le Retraite L, Heid P, Tamalet C, Sancho-Garnier H. L'auto-prelevement vaginal a domicile pour recherche de papilloma virus a haut risque. Campagne experimentale du departement des Bouches-duRhone. Bulletin du Cancer 2011;98(7):723-31.

(39.) Lazcano-Ponce E, Lorincz AT, Cruz-Valdez A, Salmeron J, Uribe P, Velasco-Mondragon E, et al. Self-collection of vaginal specimens for human papillomavirus testing in cervical cancer prevention (MARCH): A community-based randomised controlled trial. Lancet 2011;378(9806):1868-73.

(40.) Castle PE, Rausa A, Walls T, Gravitt PE, Partridge EE, Olivo V, et al. Comparative community outreach to increase cervical cancer screening in the Mississippi Delta. Prev Med 2011;52(6):452-55.

(41.) Edwards JB, Tudiver F. Women's preventive screening in rural health clinics. Womens Health Issues 2008;18(3):155-66.

(42.) Fehringer G, Howlett R, Cotterchio M, Klar N, Majpruz-Moat V, Mai V. Comparison of papanicolaou (Pap) test rates across Ontario and factors associated with cervical screening. Can J Public Health 2005;96(2):140-44.

(43.) Yabroff KR, Lawrence WF, King JC, Mangan P, Washington KS, Yi B, et al. Geographical disparities in cervical cancer mortality: What are the roles of risk factor prevalence, screening, and use of recommended treatment. J Rural Health 2005;21(2):149-57.

(44.) Nene B, Jayant K, Arrossi S, Shastri S, Budukh A, Hingmire S, et al. Determinants of women's participation in cervical cancer screening trial, Maharashtra, India. Bull WHO 2007;85(4):264-72.

(45.) Sankaranarayanan R, Nene BM, Dinshaw KA, Mahe C, Jayant K, Shastri SS, et al. A cluster randomized controlled trial of visual, cytology and human papillomavirus screening for cancer of the cervix in rural India. Int J Cancer 2005;116(4):617-23.

(46.) Sowjanya AP, Paul P, Vedantham H, Ramakrishna G, Vidyadhari D, Vijayaraghavan K, et al. Suitability of self-collected vaginal samples for cervical cancer screening in Periurban Villages in Andhra Pradesh, India. Cancer Epidemiol Biomarkers Prev 2009;18(5):1373-78.

(47.) Lindell M, Sanner K, Wikstrom I, Wilander E. Self-sampling of vaginal fluid and high-risk human papillomavirus testing in women aged 50 years or older not attending Papanicolaou smear screening. BJOG 2012;119(2):245-48.

(48.) Sanner K, Wikstrom I, Strand A, Lindell M, Wilander E. Self-sampling of the vaginal fluid at home combined with high-risk HPV testing. Br J Cancer 2009;101(5):871-74.

(49.) Stenvall H, Wikstrom I, Wilander E. High prevalence of oncogenic human papilloma virus in women not attending organized cytological screening. Acta Derm Venereol 2007; 87(3): 243-45.

Received: November 24, 2012

Accepted: January 25, 2013

C. Sarai Racey, MPH, Diana R. Withrow, MPH, Dionne Gesink, PhD

Author Affiliations

Dalla Lana School of Public Health, University of Toronto, Toronto, ON Correspondence: C. Sarai Racey, Dalla Lana School of Public Health, 155 College Street Health Science Building, 6th Floor, Toronto, ON M5T 3M7, Tel: 647-993-6093, E-mail:

Table 1. The Key Words and Medical Subject Headings (MeSH)
Used to Identify Peer-reviewed Articles in the Databases
Medline and Embase From January 1, 1990-July 15, 2012

             Medline                               Embase
  [Papillomavirus infections OR          [Vagina smear OR papilloma
  cervical intraepithelial neo-        virus OR papillomavirus infec-
    plasia OR uterine cervical          tion OR wart virus OR papil-
   neoplasmas OR vaginal smears        loma OR uterine cervix carci-
       OR papillomaviridae]            noma OR uterine cervix carci-
                                               noma in situ]

               AND                                  AND

  [Self care OR patient accepta-        [Self evaluation OR patient
   nce of health care OR Self-         participation OR patient com-
     sampl * or Self test *]            pliance OR self care OR Self
                                       test * OR Self sampling OR Self
                                       sampl * OR Self sampling Human
                                            papillomavirus test]

               AND                                  AND

           [HPV test *]                    [Cancer screening OR HPV
                                                   Test *]

Table 2. Summary Study Characteristics of Reviewed Studies Published
Between January 1, 1990 and July 15, 2012 Examining HPV Self-collected
Testing vs. Pap Testing Compliance for Women Who Do Not Normally
Participate in Cervical Cancer Screening Programs

Study (reference),    Number     Age    Estimated   Study Characteris-
Year, Country        of Parti-  Range   National     tics (Randomiza-
                     cipants    (Yrs)   Coverage     tion, HPV Detec-

                                           of        tion Assay, Sam-
                                        Cervical     pling Device and
                                         Cancer          Setting)

Gok et al., (33)      25,822    30-60      77%      Randomized. Hybrid-
2012, Netherlands                                   capture II assay
                                                    with Cervovaginal
                                                    brush sampling
                                                    device. Urban

Szarewski, et al.,     3000     25-64      68%      Randomized. Hybrid-
(35) 2011, United                                   capture II assay
Kingdom                                             with swab sampling
                                                    device. Urban set-

Giorgi et al.,         1235     35-65    48-88%     Randomized. Hybrid-
(34) 2011, Italy                                    capture II assay
                                                    with lavage sam-
                                                    pling device.
                                                    Urban and rural

Wikstrom et al.,       4060     39-60      NR*      Randomized. Hybrid-
(36) 2011, Sweden                                   capture II assay
                                                    with swab sampling
                                                    device. Urban set-

Virtanen et al.,       8699     30-60      70%      Randomized. Hybrid-
(37) 2011, Finland                                  capture II assay
                                                    with lavage sam-
                                                    pling device.
                                                    Urban setting.

Castle et al.,         119      26-65      NR       Not randomized.
(40) 201 1, United                                  Hybrid-capture II
States of America                                   assay with swab
                                                    sampling device.
                                                    Rural setting.

Lazcano-Ponce         22,102    25-65      NR       Randomized. Hybrid-
et al., (39)                                        capture II assay
2011, Mexico                                        with Cervovaginal
                                                    brush sampling
                                                    device. Rural set-

Piana et al.,          7854     35-69      60%      Randomized. PCR
(38) 2011, France                                   genotyping assay
                                                    with swab sampling
                                                    device. Urban set-

Gok et al., (32)      27,163    30-60      77%      Randomized. Hybrid-
2010, Netherlands                                   capture II assay
                                                    with lavage sam-
                                                    pling device.
                                                    Urban setting.

Bias et al., (31)      2624     30-50      63%      Randomized. PCR
2007, Netherlands                                   genotyping with
                                                    Cervovaginal brush
                                                    sampling device.
                                                    Urban setting.

Study (reference),      Description of HPV           Definition of
Year, Country          Self-testing Arm and          Underscreened
                     Control Arm Intervention         Population

Gok et al., (33)     Mailed HPV self-test vs.   Women who had not
2012, Netherlands    Pap invitation letter.     attended cervical
                                                cancer screening in
                                                the last year after a
                                                reminder invitation
                                                for screening.

Szarewski, et al.,   Mailed HPV self-test vs.   Women who did not
(35) 2011, United    Pap invitation letter.     respond to 2 invita-
Kingdom                                         tions for screening.

Giorgi et al.,       Mailed HPV self-test vs.   Women who did not
(34) 2011, Italy     Pap invitation letter.     respond to 1 regular
                                                invitation for scree-

Wikstrom et al.,     Mailed HPV self-test vs.   Women who had not
(36) 2011, Sweden    Pap invitation letter.     participated in scree-
                                                ning for >6 years.

Virtanen et al.,     Mailed HPV self-test vs.   Women who did not
(37) 2011, Finland   Pap invitation letter.     respond to 2 invita-
                                                tions for screening.

Castle et al.,       Direct door-to-door        Women who had not had
(40) 201 1, United   offering of HPV self-      a Pap test in the last
States of America    test vs. coupon for free   3 years.
                     Pap test at local clinic.

Lazcano-Ponce        Direct door-to-door        Women in a poverty-
et al., (39)         offering of HPV self-      reduction program, with
2011, Mexico         test vs. door-to-door      limited access to
                     invitation to Pap tes-     health services.
                     ting at nearest clinic.

Piana et al.,        Solicitation for HPV       Did not participate in
(38) 2011, France    self-testing at home vs.   cervical cancer scree-
                     Pap invitation letter.     ning after invitation.

Gok et al., (32)     Mailed HPV self-test vs.   Women who had not had
2010, Netherlands    Pap invitation letter.     a Pap test in 5 years
                                                and did not respond to
                                                1 invitation for scree-

Bias et al., (31)    Mailed HPV self-test vs.   Women who did not
2007, Netherlands    Pap invitation letter.     respond to 2 invita-
                                                tions for screening.

* NR = Not reported.

Table 3. Overall Results of Compliance and Relative Compliance for All
Studies Reviewed, Including High-risk-HPV Percent Positivity

                                       HPV Self-test Arm

Study           Number     HPV     HPV Self-    Pap Test    Total HPV
(reference),      of      Self-    Test Com-    Compli-     Self-test
Year, Country   Parti-     test     pliance       ance      Compliance
                cipants    Arm       n (%)                    n (%)

Gok et al.,     25,822    25,561     7870         n/a         7870#
(33) 2012,                          (30.8%)                  (30.8%)

Szarewski, et    3000      1500       96       57 (3.8%)       153#
al., (35)                           (6.4%)                   (10.2%)
2011, United

Giorgi et        1235      616        103      18 (2.9%)       121#
al., (34)                           (16.7%)                  (19.6%)
2011, Italy

Wikstrom et      4060      2000       679      100 (5.0%)      779#
al., (36)                           (34.0%)                  (39.0%)
2011, Sweden

Virtanen et      8699      2397       663      93 (3.9%)       756#
al., (37)                           (27.7%)                  (31.5%)
2011, Finland

Castle et         119       77        62          n/a          62#
al., (40)                           (80.5%)                  (80.5%)
2011, United
States of

Lazcano-Ponce   22,102     9371      9202         n/a         9202#
et al., (39)                        (98.2%)                  (98.2%)
2011, Mexico

Piana et al.,    7854      3552       939         n/a          939#
(38) 2011,                          (26.4%)                  (26.4%)

Gok et al.,     27,163    26,886     7404      51 (0.19%)     7455#
(32) 2010,                          (27.5%)                  (27.7%)

Bias et al.,     2624      2352       736      70 (3.0%)       806#
(31) 2007,                          (31.3%)                  (34.3%)

                  Pap Test Arm

Study            Pap     Pap Test     Relative     p-value     %
(reference),     Test    Compli-     Compliance              High-
Year, Country    Arm       ance       (95% CI)                risk
                          n (%)                               HPV

Gok et al.,      261       17#         4.727#      <0.0001   8.31%
(33) 2012,                (6.5%)    (2.98-7.49)*

Szarewski, et    1500      68#         2.250#      <0.0001   8.42%
al., (35)                 (4.5%)    (1.70-2.97)*
2011, United

Giorgi et        619       86#         1.414#      0.0073    21.36%
al., (34)                (13.9%)    (1.10-1.82)*
2011, Italy

Wikstrom et      2060      188#        4.268#      <0.0001   6.04%
al., (36)                 (9.1%)    (3.69-4.94)*
2011, Sweden

Virtanen et      6302     1631#        1.219#      <0.0001   12.22%
al., (37)                (25.9%)    (1.13-1.31*)
2011, Finland

Castle et         42       17#         1.989#      0.0004    14.52%
al., (40)                (40.5%)    (1.36-2.92)*
2011, United
States of

Lazcano-Ponce   12,731   11,054#       1.131#      <0.0001   9.61%
et al., (39)             (86.8%)    (1.12-1.14)*
2011, Mexico

Piana et al.,    4305      311#        3.659#      <0.0001   6.18%
(38) 2011,                (7.2%)    (3.25-4.13)*

Gok et al.,      277       46#         1.670#      0.0001    10.25%
(32) 2010,               (16.6%)    (1.28-2.18)*

Bias et al.,     272       48#         1.942       <0.0001   7.98%
(31) 2007,               (17.6%)    (1.49-2.53)*

* Statistically significant p<0.01.

Note: Numbers in bold indicated with #

Figure 2. Forest plot for the pooled estimate of the relative
screening compliance of women using HPV self-testing compared to
invitation to Pap testing

Study                             RR (95% CI)        RR (95% CI)

                                  4.73 (2.98-7.49)   9.45
Gok et al. 2012
                                  2.25 (1.71-2.97)   9.96
Gok et al. 2012
                                  1.41 (1.10-1.82)   10.01
Giorgi et al. 2011
                                  4.27 (3.67-4.94)   10.18
Wikstrom et al. 2011
                                  1.22 (1.13-1.31)   10.25
Virtanen et al. 2011
                                  1.99 (1.36-2.92)   9.69
Castle et al. 2011
                                  1.13 (1.12-1.14)   10.27
Lazcano-Ponce et
al. 2011                          3.66 (3.24-4.13)   10.21

Piana et al. 2011
                                  1.67 (1.28-2.18)   9.99
Gok et al. 2010
                                  1.94 (1.49-2.53)   9.99
Bias et al. 2007
                                  2.14 (1.30-3.52)   100.00
Overall (l-squared =
99.5%, p=0.000)

NOTE: Weights are from random effects analysis
COPYRIGHT 2013 Canadian Public Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:SYSTEMATIC REVIEW; Human papillomavirus
Author:Racey, C. Sarai; Withrow, Diana R.; Gesink, Dionne
Publication:Canadian Journal of Public Health
Article Type:Report
Geographic Code:1CANA
Date:Mar 1, 2013
Previous Article:La recherche qualitative sur les inegalites sociales de sante: le parcours de vie.
Next Article:Profit versus public health: the need to improve the food environment in recreational facilities.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |