Supplementary Material for: Variable Learning Curve of Basic Rigid Bronchoscopy in Trainees

Background: Despite increased use of rigid bronchoscopy (RB) for therapeutic indications and recommendations from professional societies to use performance-based competency, an assessment tool has not been utilized to measure the competency of trainees to perform RB in clinical settings. Objectives: The aim of the study was to evaluate a previously developed assessment tool – Rigid Bronchoscopy Tool for Assessment of Skills and Competence (RIGID-TASC) – for determining the RB learning curve of interventional pulmonary (IP) trainees in the clinical setting and explore the variability of learning curve of trainees. Methods: IP fellows at 4 institutions were enrolled. After preclinical simulation training, all RBs performed in patients were scored by faculty using RIGID-TASC until competency threshold was achieved. Competency threshold was defined as unassisted RB intubation and navigation through the central airways on 3 consecutive patients at the first attempt with a minimum score of 89. A regression-based model was devised to construct and compare the learning curves. Results: Twelve IP fellows performed 178 RBs. Trainees reached the competency threshold between 5 and 24 RBs, with a median of 15 RBs (95% CI, 6–21). There were differences among trainees in learning curve parameters including starting point, slope, and inflection point, as demonstrated by the curve-fitting model. Subtasks that required the highest number of procedures (median = 10) to gain competency included ability to intubate at the first attempt and intubation time of