The impact of endoscopists' experience and learning curves and interendoscopist variation on colonoscopy completion...

Since its introduction in the late 1960s, the technology of colonoscopy has developed rapidly and the competence of endoscopists has increased. Nevertheless, it is not always possible to perform a complete colonoscopy. The aim of this study was to assess, in a population-based setting, the endoscopist-related factors influencing the completion rate, taking into account patient characteristics and changes in the technology over time.

All colonoscopy records between 1979 and 1995 in one Swedish county (population 258,000) were retrieved. Information was obtained about patient demographics, date of examination, endoscopist, indications, presence of diverticulosis, type of colonoscopy, findings, level of completion, complications, and reasons for incomplete colonoscopy. Completion rates were examined by univariate and multivariate analyses.

Of 5,494 colonoscopies, 4,153 (75%) were complete, and 78% were diagnostic and 22% therapeutic. In 70 %, findings were pathological, and 30 % were normal. The overall 30-day endoscopist-related complication rate was 0.3 %. Over time, the proportion of colonoscopies performed by endoscopists with greater experience increased and so did the completion rates. Completion rates were influenced by endoscopist's experience and to some extent by the intensity (the number of colonoscopies performed by the endoscopist during the previous 90 days). There was a large interendoscopist variation, at each level of experience, in the ability to perform a complete colonoscopy, implying substantial differences between individual learning curves.

The completion rate has increased over time, the major reason being greater competence of endoscopists. The finding of large interendoscopist variation, at each level of experience, in the ability to perform a complete colonoscopy supports the monitoring of endoscopists to maintain and improve performance.