Colonoscopy in elderly patients: diagnostic yield, pre-referral assessment and quality indicators in two Swedish cohorts

OBJECTIVES: Colonoscopy in elderly, frail patients requires weighing diagnostic yield against procedure-related risk, but real-world data are limited. We describe outcomes from two Swedish cohorts undergoing non-screening colonoscopy in two regions. METHODS: We combined aggregated results from two retrospective cohort studies. The Ersta cohort included all patients aged ≥75 years undergoing colonoscopy in 2022 and compared inpatient versus outpatient examinations. The Örebro cohort included all colonoscopies in patients aged ≥80 years during 2016-2018. We extracted data on indications, pre-referral investigations, findings, quality indicators, and short-term mortality. For Örebro, we used previously calculated odds ratios and positive predictive values for significant findings. RESULTS: At Ersta, 1,066 patients aged ≥75 years underwent 1,144 colonoscopies; 40% were performed in inpatients admitted for bowel preparation. Colorectal cancer was diagnosed in 5%, inflammatory bowel disease in 6%, microscopic colitis in 3%, and other colitis in 4%, with similar proportions in inpatients and outpatients. One-year mortality was 6% after inpatient and 1% after outpatient colonoscopy. In Örebro, 565 colonoscopies in patients aged ≥80 years were analyzed. Significant findings occurred in 26.2% (colorectal cancer 18.8%, high-grade dysplasia 4.6%, any colitis 5.0%, with some overlap). Abnormal rectal examination, abnormal abdominal imaging, low ferritin, and positive f-Hb predicted significant disease. Altered bowel habits was the most common indication but had an odds ratio below 1 for significant disease. CONCLUSIONS: In elderly Swedish patients, colonoscopy detects colorectal cancer or colitis, but many examinations show benign or no findings. Selection using simple pre-referral tests and life expectancy may reduce low-yield examinations.