Efficacy of Offering a Self-sampling Device by the GP to Reach Underscreened Women (ESSAG)

Efficacy of Offering a Self-sampling Device by the General Practitioner to Reach Women Underscreened in the Routine Cervical Cancer Screening Program Compared to Sending Reminder Letters by the Screening Organization

Kom Op Tegen Kanker, Sciensano, Centrum voor Kankeropsporing, Belgian Cancer Registry, KU Leuven, Vrije Universiteit Brussel, Universiteit Antwerpen

The ESSAG trial invests the impact of offering a free self-sampling device (SSD) on the cervical cancer screening rate of underscreened women. This study is aimed at women between the age of 31 and 65 who did not have a smear taken during the last 6 years. In order to assess the effect of a) providing the SSD, and b) the intervention of the general practitioner (GP) (either face-to-face, either by sending the SSD by letter), a randomized control trial is set up with three arms. The ESSAG trial evolves from a collaboration between Universiteit Gent and Vrije Universiteit Brussel, Katholieke Universiteit Leuven, Universiteit Antwerpen, Sciensano, het Centrum voor Kankeropsporing en het Belgisch Kankerregister, and is funded by "Kom Op Tegen Kanker".

Following the usual care procedure, a random sample of 1125 Flemish women who meet the same inclusion criteria as in arm A and B and are not included in one of the latter arms is drawn by "Centrum voor Kankeropsporing" (CvKO) from Heracles, the database of the Flemish screening program.

In a group of 45 GP practices, over a course of 9 months, all long-term unscreened women with a GMD will be addressed by their GP when they consult for any reason. The GP will discuss the pros and cons of screening for cervical cancer, the various options for screening for cervical cancer including the possibility of using a SSD. For this, the GP can use on accessible brochure and video materials. After the woman agrees, she is given a self-taking kit that she can use when and where it suits her and send it to the lab with the prepaid envelope. The woman is informed of the result of the Human Papilloma Virus test result by her GP according to the usual way in the practice (e.g. via letter, by phone or during the next consultation).

In the second arm, a second group of 45 GP practices in Flanders will be recruited. With the intervention of the Centre for Cancer Prevention (Flanders), 25 at randomly selected long-term unscreened women with a Global Medical Form (in dutch: 'Globaal Medisch Dossier, GMD) in one of these practices will receive an envelope containing a letter of invitation from their GP for cervical cancer screening, a brochure with the advantages and disadvantages of cervical cancer screening and more specifically the use of a self-taking kit, and a self-taking kit. If the woman wishes, she can use the kit when and where it suits her and send it to the lab with the pre-paid envelope. As in arm A, the woman will be informed of the Human Papilloma Virus test result by her GP according to the usual way in the practice (e.g., by letter, by phone or during the next consultation). All women will receive a study-related reminder letter after 4 months.

Age-standardised response rates in each arm and the differences between arms will be calculated, in a per protocol (PP) and intention-to-treat (ITT) analysis. As the goal is to reveal to what extent the respective approach (providing a self sampling device (SSD) with different levels of general practitioner (GP) involvement) succeeds in reaching vulnerable women, analyses will take into account a comprehensive list of covariates by logistic regression accounting for cluster effects. These covariates include trial arm, GP-office characteristics, as well as demographic information of the patient on origin and socio-economical background provided by the "Kruispuntbank voor Sociale Zekerheid (KSZ)".

The participation rate will be calculated from arm A to answer questions at public health level, i.e. the population effect of the intervention by GP's on increasing cervical cancer screening coverage in underscreened women. For participation rate, the denominator are all women who belong to the target group for that specific GP-office (women between 31-64y, not screened within 6y, etc).

A cost-effectiveness analysis will also be performed, addressing the additional number of women screened per 1,000€ in interventions A and B versus control intervention C.

In order to explore the potential roll-out of the study within the Flemish screening programme, the investigators will also perform a feasibility study. Through interviews with 20-30 GPs who participated in arm A, a feasibility analysis will reveal the strengths and pitfalls of the intervention.

Although this study is not powered for inter-arm differences in follow-up after an abnormal screening test result, compliance analysis will be performed with the following indicators: screen-test positivity rates, proportion with invalid screen-test results, proportion of screen-test positives that have the foreseen follow-up tests.

Inclusion Criteria: women between 31-64 years old living in Flanders eligible for the Flemish actions with regard to population screening without a smear registered in the Belgian Cancer Registry in the last 6 years registered as GMD patient in one of the participating GP practices Exclusion Criteria: hysterectomy pregnancy (past) diagnosis of cervical cancer