The main purpose is to develop and test an IT-infrastructure to provide more individualized ways of inviting and reaching women who have not participated, despite having been invited with annual reminders, in cervix cancer screening program during a ten year period.

BACKGROUND Non-participation in the screening program is the main risk factor for cervical cancer (Andrae et al JNCI, 2008). Therefore it is important to reduce barriers to screening and facilitate participation in the screening program for cervical cancer. Offering self-sampling kits for human papillomavirus (HPV) testing has been examined in several research studies as a way to reach women who have not responded to screening invitations.

Results from a population-based, randomized study in western Sweden concluded that by offering HPV self-sampling kits to screening non-attendees, 17% of cervical cancer cases could be avoided. Offering self-sampling kits to women who have not attended screening in the organized program, can lead to greater participation without substantial increases in the overall health care costs. While there is a growing body of evidence supporting the use of self-sampling, it is unclear how this strategy can be implemented effectively in the context of an organized screening program. An alternative is to offer women the chance to speak with a coordinating midwife who can answer questions and address concerns regarding screening. It is unclear how the results of different research studies should be put into practical use within an organized screening program. Currently, women who have not attended screening are sent an annual reminder invitation. We want to examine whether these alternative strategies (self-sampling and contact with a midwife) may work as well, or better, than current strategy of reminder invitations to non-attenders.

AIM The aim of this randomized health services study is therefore both to test the feasibility of providing alternative strategies instead of the customary reminder invitations to women in Stockholm (so-called feasibility study, primary aim), and whether such strategies will increase overall participation in screening (secondary aim).

STUDY DESIGN AND METHODOLOGY A new IT infrastructure for ordering a self-sampling kit has been developed by LabMed, KI. This in conjunction with the screening database, will be used to identify and invite women who are long-term non-participants in cervical screening. Women who have not participated for ten years, despite annual reminder invitations, will be identified and randomized to one of four strategies.

1. A validated HPV self-sampling kit will be sent directly to women with a study invitation letter and instructions.
2. An invitation to order an HPV self-sampling kit through an online application will be sent.
3. An invitation to call the coordinating midwife with questions and concerns regarding screening will be sent.
4. The standard annual reminder invitation will be sent (control, routine practice)

Women who choose to participate will be sampled and monitored according to the standards of care in the regional cancer screening program. HPV-positive women will be followed by colposcopy and biopsy according to standard clinical guidelines. Women will be invited over the course of 3 months. All letters will be printed in advance with pre-marked dates. Batches will be sent out weekly and checked against changes in the population registry data (for individuals that have moved, taken a smear, died, or unregistered, invitation letters will be sorted out and those individuals marked in the database).

2 Notation and abbreviations

• PNR = person nummer, personal ID number 3 Study population 3.1 Inclusion criteria
• Women who have lived in Stockholm for the past 10 years and have been sent annual reminder at least once per year for 10 years without attending screening.

Specific selection process:

Include women without an "H-prov" on record for the past 10 years Include women have actively been sent a reminder invitation (invitation type=1) during 10 years time. In other words, resident in Stockholm for the past 10 years.

Include women over the age of 33 3.2 Exclusion criteria Women who have opted out of the screening program Hysterectomized Return to sender Changed PNR Women who are marked as Expired=1. Women who have reason marked for being unregistered but are not marked as is unregistered=1.

Women who are due to receive their reminder invitation in the next 3 months 4 Statistical analyses Selection of the study population Randomization A random selection of 8000 women will be chosen from all women meeting the study inclusion criteria.

These 8000 women will be randomized to the 4 study arms (age-stratified 30-50 and 51-60).

The selection of the study population and the randomization to study arms will be carried out by Anna Stoltenberg, Statistician at RCC and outside of the study group 1.

Follow-up - study end date 2016-12-31

Inclusion Criteria:

• Non-participants in cervical cancer screening during a 10 year period

Exclusion Criteria:

• Active refusers

Our intention is to share data, but a plan for how this will be done in practise has not been written.