IDDF2020-ABS-0142 Performance indicators of organised colorectal cancer screening programmes using faecal immunochemical tests and colonoscopy: a systematic review and meta-analysis

Background The success of population-based colorectal cancer (CRC) screening is dependent on the optimal achievement of various targets. We estimated the pooled performance indicators of CRC screening programmes that used faecal immunochemical tests (FIT) as a primary screening modality and colonoscopy as a subsequent confirmatory test in various countries. Methods We searched PubMed, Ovid MEDLINE, Embase, and Cochrane from inception to 1st Jan 2020. We included original articles published in the English language describing population-based CRC screening programme that used FIT and colonoscopy, and relevant national CRC screening reports by hand searching. We extracted data to pool early performance indicators, including participation rate, invalid FIT rate, FIT positive rate, adenoma/CRC detection rate of FIT, colonoscopy compliance rate, rate of adequate bowel preparation, colonoscopy completion rate, colonoscopy complication rate, and positive predictive values (PPV) of FIT for adenoma/CRC. We used Metaprop to conduct a meta-analysis via R software (version 3.6.3). The Freeman-Tukey double arcsine transformation was used to stabilise the variances, and a random-effects model was used to pool the rates with proportions. Results A total of 85 studies (64 articles and 21 reports) were included in this meta-analysis. The pooled participation rate was 52.2%, and the pooled proportion of invalid FIT was 1.0%. The FIT positive rate was 7.0%, the adenoma detection rate was 21.8‰, and the CRC detection rate was 2.5‰. In addition, the colonoscopy compliance rate (79.1%), rate of adequate bowel preparation (86.3%), and the colonoscopy completion rate (93.1%) were high. We found the pooled complication rate of 0.3%, with PPVs of FIT being 39.2% (for adenoma) and 4.7% (for CRC). The range of all indicators varied substantially. In subgroup analysis, a higher FIT cut-off value indicated lower positivity and lower adenoma detection rate. Compared with CRC screening programmes with one FIT, those with two FIT had lower participation rate (48.8% vs 55.5%) but higher positive rate (8.9% vs 5.8%) and adenoma detection rate (21.8‰ vs 17.3‰). Conclusions Our findings reported the pooled performance indicators of different CRC screening programmes. The summary measures could inform the benchmarking of performance indicator targets across different CRC screening programmes.