Using behavioural insights to increase cancer-screening rates

Screening has a significant role to play in delivering earlier diagnosis of most common cancers. We applied to be part of the LGA Behavioural Insights programme to work alongside national experts and develop an innovation to increase cancer-screening rates in targeted communities.

View all Efficiency and income generation articles

We worked with the Behavioural Insights Team, the LGA, the local CCG and primary care practices to increase cancer-screening rates. We sent a sample of 1,316 previous non-attenders of routine bowel cancer screening a reminder letter from the their GP using anticipated regret messaging, i.e. asking people to reflect on how they might feel if they were diagnosed too late. We compared the proportion of people returning their kits ten weeks before and ten weeks after the letter was sent. Two patients (0.2%) returned their kits in the pre-intervention period, compared with 54 patients (4.4%) after the letter was sent.

The challenge

Late presentation and early diagnosis of cancers are important issues in North East Lincolnshire and in many other places in the UK. In Grimsby, there is a wide variation in the under-75 mortality rate for all cancers between wards and GP practices. These inequalities are particularly stark in our most deprived wards where the under-75 mortality rate is up to seven times higher. There is also a wide variation in the uptake of cancer screening programmes, which broadly reflect patterns of mortality. Areas and practices with low uptake of cancer screening are frequently the same as those with high rates of premature cancer mortality.

Screening has a significant role to play in delivering earlier diagnosis of most common cancers. We applied to be part of the LGA Behavioural Insights programme to work alongside national experts and develop an innovation to increase cancer-screening rates in targeted communities. Success with this project and a subsequent wider rollout would meet our aims of increasing the number of people attending screening, reduce the numbers of preventable mortalities and hopefully reduce cancer treatment costs, which are much lower if cancers are diagnosed at an earlier stage.

The solution

We identified a number of practices that serve populations in the two North East Lincolnshire wards with the lowest screening rates. We conducted research with the aim of identifying key barriers to screening uptake, evidence-based solutions and touchpoints for targeting an intervention at patients. A review of academic literature identified a range of barriers such as embarrassment, denial and fatalism, lack of knowledge and awareness, and lack of trust and confidence in being able to navigate the medical system. Many of these barriers were echoed in interviews with professionals working in the borough.

Literature shows that sending reminders can be an effective way of increasing uptake, but most GPs in North East Lincolnshire did not do so. We decided to test a reminder letter from the GPs themselves together with an additional behavioural approach – anticipated regret messaging - asking people to reflect on how they might feel if they were diagnosed too late. We worked with five GP practices to send a letter to all patients who had been invited to, but not completed, bowel cancer screening over a twelve-month period. Our sample consisted of 1,316 patients. We compared the proportion of people returning their kits ten weeks before and ten weeks after the date the letter was sent.

The impact

In our study, two patients (0.2%) returned their kits in the pre-intervention period, compared with 54 patients (4.4%) after the letter was sent. This difference was statistically significant, which, combined with the timing of the increase, suggests the change was due to our intervention rather than chance or other factors.

Adopted at scale by practices in North East Lincolnshire and beyond, this simple nudge could illustrate a real impact on lives and treatment costs with very little outlay. On average 2-3% of people who complete bowel cancer screening have a definitive positive (abnormal) result.* When bowel cancer is detected at the earliest stage, there is an excellent chance of survival - more than 90% of people survive at least five years. When found late, the chances of survival are dramatically reduced.  In terms of budgetary savings - cost for colon cancer, stage 1 treatment costs £3,373, whereas stage 4 treatment costs £12,519.  For rectal cancer, stage 1 treatment costs £4,449, whereas stage 4 treatment costs £11,815.  

How is the new approach being sustained?

We have recommended to local Public Health England and Clinical Commissioning Group colleagues that we work to scale this intervention across primary care settings in North East Lincolnshire, exploring how we can use it for other cancers besides bowel. We are aware that there may be some concerns about the resources required to deliver this approach at scale, but in the most part this would mean staff identifying non-attenders as data is received from the screening hubs and logging changes, and postage time and costs.

Lessons learned

This was a much more difficult project than we had expected and we faced challenges along the way. These were mainly due to a lack of clarity and understanding about the use of behavioural insights in the health field, PHE approval processes, the interaction with the national cancer screening programmes and the source of the data we needed.

Our initial design was a randomised control trial which included those patients overdue for cervical or bowel cancer screening. We ruled out a first proposal to use text messaging due to both GDPR issues and the low number of mobile phone numbers from the target cohort on local GP databases. This meant an early delay.

We moved onto letters and wanted to test two behavioural approaches: anticipated regret and a lottery incentive. After applying to NHS ethics/Health Research Authority, we received approval after a round of clarification questions, and some minor changes to our intervention letters. We then discovered that we also needed to get approval from Public Health England’s cancer screening programme boards. This was a surprise, as we were not using Screening Hub data (instead getting data direct from GPs). After waiting some months for the boards to meet, which again affected project timescales, both the Cervical Screening Programme Research Advisory Committee (CSP RAC) and the Bowel Screening Programme Research Advisory Committee (BSP RAC) rejected our project.

We paused the project and contacted the screening team at PHE, who agreed that the project could restart if we changed the study design and the letter. We moved the focus to bowel cancer screening, agreed to remove the lottery incentive and changed to a pre-post study.  In this sort of study, data is collected on a baseline date and one or more times before or after the intervention, to measure the impact of the intervention itself.

Restarting the project meant reinitiating the project management, which included getting back in touch and meeting with GP practices to check they were still willing to take part after such a delay. Thankfully, their enthusiasm about the potential of the study meant that we were able to move forward soon after.

The key lessons learned for us are relationship based. We would recommend to anyone delivering a project of this type that they contact the relevant national Public Health England department at the outset. These initial conversations ensure that projects are framed in a way to prevent unnecessary delays and offer the opportunity to deliver something of benefit to all parties. It would also be useful to site our project as an example of good practice where the project has gone ahead successfully.

We would also encourage council behavioural insight teams to build and maintain relationships with local GP practices directly – the opportunities for more of this work will increasingly be of value to the new Primary Care Networks. We should all be working together on these issues anyway. We would also add that, though behavioural insight approaches are slowly moving into the mainstream, other councils may face some resistance to trialling something like this locally, we certainly did. We would encourage you to stay determined, but to be prepared to deal with some difficult conversations. We have worked hard over the last couple of years to listen, stay calm, build partnerships and to develop understanding. This is a very different way of trying to change behaviour than some people are used to, don’t give up, it really is worth it.

*According to Cancer Research UK

Contact

Claire Thompson, [email protected]