Text reminders vs patient navigation to improve FIT screening uptake

Purpose
Bowel cancer is a leading cause of cancer-related mortality in England. Several large randomised controlled trials have shown that biennial faecal occult blood test screening can improve bowel cancer outcomes by detecting cases early. In light of this, NHS England initiated a national bowel cancer screening programme (BCSP) in 2006. Since then, uptake has been consistently low, falling from 54% in 2010, to 49% in 2015. To reverse this trend, the English BCSP transitioned to a new test in June 2019, called: the ‘faecal immunochemical test’ (FIT), which has been shown to be more acceptable than its predecessor, improving uptake by ~7%.

One perennial issue, which has not yet been addressed, and which is likely to remain an important issue, is the lower uptake of bowel cancer screening in London. Already, several interventions to improve participation are in place (‘pre-invitation letters’, ‘GP-endorsement’, ‘mailed reminders’, etc.). Two interventions, which are not currently implemented, but show promise, are text-message reminders and patient navigation (PN). Evidence from a recent service evaluation, conducted in Southeast London, demonstrated that a combination of text-message reminders and PN facilitated uptake in over 13% of breast screening non-responders. As participants were not randomised, however, the extent to which each intervention improves uptake is not known. Further research is needed to determine whether
these interventions can similarly be used to improve the uptake of bowel cancer screening, and the extent to which each contributes toward the overall impact of the intervention strategy.

Aim
The principal research objectives of this study are to:
1.    Test the effectiveness of text-message reminders to improve participation in bowel cancer screening and,
2.    Test whether a combination of text-message reminders and patient navigation is more effective and cost-effective than using text-message reminders alone.

Method
This study will be a randomised controlled trial with three parallel arms and the participants will be men and women, who are registered with a participating general practice, and have not participated in bowel cancer screening within 13 weeks of being sent an invitation from the programme. Participants will be selected through convenience sampling (i.e. those who are non-participants at the time of the trial will be eligible for inclusion). Participants will be anyone who is registered with a participating GP practice, located within the London Boroughs of Brent, Ealing, Lambeth, Lewisham, Redbridge and Barking and Dagenham, and has not returned their bowel cancer screening kit during the study period (certain exclusions will be applied; for example, type II objectors will not be included).

The total size of the study will be 2,703, with 901 participants allocated to each group (see A60 for details).The study is expected to last approximately 5 months, from initial GP recruitment to follow-up and statistical analysis.

During the pre-trial period, at week 0, pre-invitation letters will be sent to the potentially eligible participants. At week 1, the screening kit will be dispatched. At week 5, a reminder letter will be sent to the bowel non-responders. At week 13, it will be the end of screening episode, and non-responders will be identified and randomised at this point. These participants will be identified on a weekly basis (for four consecutive weeks) by NHS Digital, who will remove Type-2 objectors, prior to transferring IDs to iPlato via an encrypted cloud-based server. It is anticipated that recruitment of participants will take approximately one month. iPlato will randomise these participants (1:1:1) to receive either: 1) no intervention ('usual care'), 2) a text-message reminder, which will be sent 13 weeks after invitation, followed by additional text-message reminders at 15, 17 and 19 weeks if there is no response, or 3) a text-message reminder, sent 13 weeks after invitation, followed by PN calls at 15, 17 and 19 weeks if there is no response. The end of data accrual is defined as week 24 of the final randomised participants, when the data are returned to NHS Digital, outcomes and demographics are added, and anonymised data returned to UCL. This is the study end date as far as participant procedures are concerned. Statistical analysis of the data will then be conducted by the UCL research team.

Principal investigator: Dr Robert Kerrison
Contact: Dr Robert Kerrison 
Collaborators: Dr Christian von Wagner