ASCEND research summary

Overview

The ASCEND study was led by Professor Rosalind Raine at University College London (UCL) and involved a collaboration between the NHS Bowel Cancer Screening Programme (BCSP), the Health Behaviour Research Centre (UCL), the Centre for Cancer Screening (Queen Mary University of London) and the Cancer Screening & Prevention Research Group (CSPRG) (Imperial College London).

The trial is registered with the ISRCTN registry. Its unique number is 74121020.

This study was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research (PGfAR) scheme.

Background

Colorectal cancer screening reduces cancer-specific mortality. Uptake in the BCSP in England varies by socioeconomic status, which may lead to inequalities in cancer outcomes.

ASCEND assessed four interventions aimed at reducing this inequality, with the intention of improving the health benefits of screening.

Method

ASCEND was an intervention study comprising three workstreams aiming to reduce socioeconomic inequalities in uptake of the BCSP’s guaiac faecal occult blood test (gFOBt) without compromising uptake in any socioeconomic group.

• Workstream 1 explored psychosocial determinants of non-uptake of gFOBt in focus groups and interviews.
• Workstream 2 developed and tested four theoretically based interventions: (1) ‘gist’ information, (2) a ‘narrative’ leaflet, (3) ‘general practice endorsement’, and (4) an ‘enhanced reminder’.
• Workstream 3 comprised four national cluster randomised controlled trials (RCTs) of the cost-effectiveness of each intervention.

Interventions were co-designed with user panels, user tested using interviews and focus groups, and piloted with postal questionnaires. All people eligible for screening in 2012 (men and women aged 60–74 years) across England were included in four cluster-RCTs. These RCTs compared ‘usual care’ (existing BCSP invitations) with usual care plus each intervention. The four trials tested: (1) ‘gist’ leaflet (n = 163,525), (2) ‘narrative’ leaflet (n = 150,417), (3) ‘general practice endorsement’ on the invitation letter (n = 265,434) and (4) enhanced reminder’ (n = 168,480). Randomisation was based on day of mailing of the screening invitation. The Index of Multiple Deprivation (IMD) score associated with each individual’s home address was used as the marker of socioeconomic circumstances (SECs). Change in the socioeconomic gradient in uptake (interaction between treatment group and IMD quintile) was the primary outcome. Screening uptake was defined as the return of a gFOBt kit within 18 weeks of the invitation that led to a ‘definitive’ test result of either ‘normal’ (i.e. no further investigation required) or ‘abnormal’ (i.e. requiring referral for further testing). Difference in overall uptake was the secondary outcome.

Results

The gist and narrative trials showed no effect on the SECs gradient or overall uptake (57.6% and 56.7%, respectively, compared with 57.3% and 58.5%, respectively, for usual care; all p-values > 0.05). The general practice endorsement showed no effect on the gradient (p = 0.5) but increased overall uptake [58.2% vs. 57.5% for usual care, odds ratio (OR) = 1.07, 95% confidence interval (CI) 1.04 to 1.10; p < 0.0001]. The enhanced reminder showed a significant interaction with SECs (p = 0.005), with a stronger effect in the most deprived IMD quintile (14.1% vs. 13.3% for usual care, OR = 1.11, 95% CI 1.04 to 1.20; p = 0.003) than the least deprived (34.7% vs. 34.9% for usual care OR = 1.00, 95% CI 0.94 to 1.06; p = 0.98), and higher overall uptake (25.8% vs. 25.1% for usual care, OR = 1.07, 95% CI 1.03 to 1.11; p = 0.001).

All interventions were inexpensive to provide. As all four trials were embedded in the screening programme, loss to follow-up was minimal (less than 0·5%).

In line with NHS policy, the gist and narrative leaflets supplemented rather than replaced existing BCSP information. This may have undermined their effect.

The enhanced reminder reduced the socioeconomic gradient in uptake and modestly increased overall uptake, whereas GPE increased overall uptake but did not reduce the gradient. Therefore, given their effectiveness and very low cost, the findings suggest that implementation of both by the BCSP would be beneficial. The gist and narrative trial results highlight the challenge of achieving equitable delivery of the screening offer when all communication is written; the format is universal and informed decision-making mandates extensive medical information.

Of four evidence-based interventions, the enhanced reminder letter reduced the socioeconomic gradient in screening uptake, but further reducing inequalities in screening uptake through written materials alone will be challenging.

Results were published in The Lancet and are also available on the UCL Discovery open access deposit.

We hope that the results will be used to inform the NHS BCSP on how best to invite people for screening in the future and to improve the effectiveness of the programme.