Effect of offering colonoscopy at the faecal immunochemical test (FIT) positivity threshold: a regression discontinuity analysis in older adults.

BACKGROUND

Colorectal cancer (CRC) screening programmes often use the faecal immunochemical test (FIT) to identify individuals for follow-up colonoscopy but evidence on its impact on CRC incidence remains limited. Using a regression discontinuity design (RDD), we investigated whether CRC screening affects CRC incidence among adults aged 71-75 years with FIT values close to the threshold-an age group not consistently included in screening programmes.

METHODS

Using registry data from 2014 to 2024, we compared CRC incidence among screening participants aged 71-75 years with FIT values just above versus just below the 100-ng/ml referral threshold, where only those above were offered colonoscopy. We included 2773 individuals with borderline FIT values (80-120 ng/ml). An RDD was used to estimate the effect of colonoscopy referral based on FIT on CRC incidence at values close to the threshold while mitigating healthy-user bias.

RESULTS

Over a median 8.4-year follow-up, 165 CRC cases were observed. As expected from the lead time, early incidence was higher in FIT-positive individuals offered colonoscopy but cumulative incidence later plateaued in this group while it continued to rise in FIT-negative individuals. After 10 years of follow-up, a reduction was seen between the groups from 12.1 to 6.6 cases per 1000 person-years, equivalent to 5.5 fewer cases (95% confidence interval: 1.44-9.56) per 1000 person-years, corresponding to a 45.5% reduction.

CONCLUSION

CRC screening reduces CRC incidence among adults aged 71-75 years with FIT values close to the FIT threshold. This could guide cancer screening policies for an age group that is only eligible for screening in some countries.