Stjepan Škudar
We read with great interest the national analysis by Beaton et al. evaluating faecal immunochemical testing (FIT) for triage of symptomatic colonoscopy using the UK National Endoscopy Database [1]. The scale of the dataset is impressive and the findings strongly support the value of FIT for prioritising colonoscopy in patients at greatest risk of colorectal cancer. However, three issues deserve emphasis before such risk-stratified models are translated into service-level thresholds. First, the study population comprised patients who ultimately underwent colonoscopy. This design is highly informative for estimating diagnostic yield among scoped patients, but it cannot fully establish the safety of deferring colonoscopy in FIT-low patients. Patients with negative FIT who were not referred, did not complete colonoscopy or were managed through alternative pathways were necessarily absent. This may introduce selection or collider bias and could make the observed low cancer yield among patients with FIT < 10 μg Hb/g appear more reassuring than it would be in the full symptomatic referral population. Second, FIT recording was incomplete, with FIT results available for fewer than half of the included symptomatic colonoscopies [1]. Missingness is unlikely to be random in routine practice. Clinicians may be more likely to document FIT when it has actively influenced triage, whereas urgent, high-risk, externally referred or incompletely coded cases may follow different documentation pathways. Sensitivity analyses modelling missing FIT values, or stratifying results by site-level FIT completeness, would help readers judge whether the reported risk gradients are robust. Third, the clinical consequences of a pathway designed primarily around colorectal cancer detection should be separated from its consequences for other important diagnoses. The article appropriately reports inflammatory bowel disease and large-polyp outcomes, but a colonoscopy strategy based mainly on cancer-risk thresholds could unintentionally delay diagnosis in younger patients, those with persistent or evolving symptoms or those with iron-deficiency anaemia but low faecal haemoglobin. Current guidance supports FIT as an adjunct to clinical assessment rather than a stand-alone exclusion test [2]. Future analyses linking FIT, referral, colonoscopy and cancer registry data across the entire symptomatic pathway, including patients not undergoing colonoscopy, would strengthen the evidence base. Calibration by age, sex, iron-deficiency anaemia, symptom pattern and site-level FIT completeness would also support safer implementation. Until such data are available, FIT-based triage should be accompanied by explicit safety-netting and audit of interval cancers and non-cancer diagnoses. These comments do not diminish the importance of Beaton et al.'s study; rather, they highlight the next evidential step needed to move from diagnostic yield among colonoscopy attendees to safe population-level triage. Stjepan Škudar: conceptualization, writing – original draft, writing – review and editing. The author has nothing to report. Artificial Intelligence Generated Content Statement: No AI tools were used. The author has nothing to report. Artificial intelligence generated content statement: No AI tools were used. The author declares no conflicts of interest. This article is linked to Beaton et al. paper. To view this article, visit https://doi.org/10.1111/apt.70537. Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.