Background Internationally, self-sampling for human papillomavirus (HPV) has been shown to increase participation in cervical-cancer screening. In Aotearoa New Zealand, there are long-standing ethnic inequalities in cervical-cancer screening, incidence, and mortality, particularly for indigenous Māori women, as well as Pacific and Asian women. Methods We invited never- and markedly under-screened (≥5 years overdue) 30-69-year-old Māori, Pacific, and Asian women to participate in an open-label, three-arm, community-based, randomised controlled trial, with a nested sub-study. We aimed to assess whether two specific invitation methods for self-sampling improved screening participation over usual care among the least medically served populations . Women were individually randomised 3:3:1 to: clinic-based self-sampling (CLINIC – invited to take a self-sample at their usual general practice); home-based self-sampling (HOME – mailed a kit and invited to take a self-sample at home); and usual care (USUAL – invited to attend a clinic for collection of a standard cytology sample). Neither participants nor research staff could be blinded to the intervention. In a subset of general practices, women who did not participate within three months of invitation were opportunistically invited to take a self-sample, either next time they attended a clinic or by mail. Findings We randomised 3,553 women: 1,574 to CLINIC, 1,467 to HOME, and 512 to USUAL. Participation was highest in HOME (14.6% among Māori, 8.8% among Pacific, and 18.5% among Asian) with CLINIC (7.0%, 5.3% and 6.9%, respectively) and USUAL (2.0%, 1.7% and 4.5%, respectively) being lower. In fully adjusted models, participation was statistically significantly more likely in HOME than USUAL: Māori OR=9.7, (95%CI 3.0-31.5); Pacific OR=6.0 (1.8-19.5); and Asian OR=5.1 (2.4-10.9). There were no adverse outcomes reported. After three months, 2,780 non-responding women were invited to participate in a non-randomised, opportunistic, follow-on substudy. After 6 months,192 (6.9%) additional women had taken a self-sample. Interpretation Using recruitment methods that mimic usual practice, we provide critical evidence that self-sampling increases screening among the groups of women (never and under-screened) who experience the most barriers in Aotearoa New Zealand, although the absolute level of participation through this population approach was modest. Follow-up for most women was routine but a small proportion required intensive support. Trial registration ANZCTR Identifier: ACTRN12618000367246 (date registered 12/3/2018) https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=371741&isReview=true ; UTN: U1111-1189-0531 Funding Health Research Council of New Zealand (HRC 16/405) Protocol http://publichealth.massey.ac.nz/assets/Up...

In Aotearoa, New Zealand, the majority of cervical cancer cases occur in women who have never been screened or are under-screened. Wāhine Māori, Pacific and Asian women have the lowest rate of cervical screening. Self-sampling for human papillomavirus (HPV-SS) has been shown to increase participation in cervical cancer screening. A whole-of-system approach, driven by evidence in the most effective delivery of HPV-SS, is required to mitigate further widening of the avoidable gap in cervical screening access and outcomes between groups of women in Aotearoa. This single-arm feasibility and acceptability study of HPV self-sampling invited never- and under-screened (≥5 years overdue) 30–69-year-old women from general practices in Auckland, Aotearoa. Eligible women were identified by data matching between the National Cervical Programme (NCSP) Register and practice data. Focus groups were additionally held with eligible wāhine Māori, Asian and Pacific women to co-design new patient information materials. Questionnaires on HPV knowledge and post-test experience were offered to women. Our follow-up protocols included shared decision-making principles, and we committed to follow-up ≥90% of women who tested positive for HPV. Data matching identified 366 eligible never- and under-screened wāhine Māori, Pacific and Asian women in participating practices. We were only able to contact 114 women, and 17, during the discussion, were found to be ineligible. Identifying and contacting women overdue for a cervical screen was resource-intensive, with a high rate of un-contactability despite multiple attempts. We found the best uptake of self-sampling was at focus groups. Of the total 84 HPV-SS tests, there were five positive results (6%), including one participant with HPV18 who was found to have a cervical Adenocarcinoma at colposcopy. In our feasibility study, self-sampling was acceptable and effective at detecting HPV and preventing cervical cancer in under-screened urban wāhine Māori, Pacific and Asian women in Aotearoa. This is the first report of cervical Adenocarcinoma (Grade 1B) as a result of an HPV-18 positive self-sample in Aotearoa. We co-designed new patient information materials taking a health literacy and ethnicity-specific approach. This work provides policy-relevant information to the NCSP on the resources required to implement an effective HPV self-sampling programme to improve equity in national cervical cancer screening.

BackgroundMāori, Pacific and Asian women in New Zealand have lower cervical-cancer screening rates than European women, and there are persistent inequities in cervical cancer outcomes for Māori and Pacific women. Innovative ways to address access barriers are required. New Zealand is transitioning to screening with human papillomavirus (HPV) DNA testing, which could allow women themselves, rather than a clinician, to take the sample. Internationally, self-sampling has been found to increase screening participation rates. The aim of this open-label community-based randomised controlled trial is to investigate whether self-sampling increases screening participation among un- and under-screened Māori, Pacific and Asian women in New Zealand.Methods/designWe aim to invite at least 3550 un- or under-screened (≥5 years overdue) Māori, Pacific and Asian women (1050, 1250, 1250 respectively), aged 30–69 years, for screening. The three study arms are: usual care in which women are invited to attend a clinic for a standard clinician-collected cytology test; clinic-based self-sampling in which women are invited to take a self-sample at their usual general practice; and mail-out self-sampling in which women are mailed a kit and invited to take a self-sample at home. Women will be randomised 3:3:1 to the clinic and mail-out self-sampling groups, and usual care. There is also a nested sub-study in which non-responding women in all allocation groups, when they subsequently present to the clinic for other reasons, are offered clinic or home-kit self-sampling. The primary outcome will be the proportion of women who participate (by taking a self-sample or cytology test).DiscussionThis trial is the first to evaluate the effectiveness of mailed self-sampling in New Zealand and will be one of the first internationally to evaluate the effectiveness of opportunistic in-clinic invitations for self-sampling. The trial will provide robust evidence on the impact on participation proportions from different invitation approaches for HPV self-sampling in New Zealand un- and under-screened Māori, Pacific and Asian women.Trial registrationANZCTR Identifier: ACTRN12618000367246 (date registered 12/3/2018) https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=371741&isReview=true; UTN: U1111–1189-0531.

No abstract

Following publication of the original article [1], the authors reported an error in the authorship list associated with the paper. Georgina McPherson has therefore been added.