Identification of consensus-based quality end points for colorectal surgery

BACKGROUND: Process and outcome measures for quality assessment of colorectal surgical care are poorly defined. OBJECTIVE: The aim of this study was to develop candidate end points for use in surgeon-specific registries designed for case reporting and quality improvement program development. DESIGN: The study design was based on modified Delphibased development of consensus quality end points. SETTING: This study was undertaken by the American Society of Colon and Rectal Surgeons Executive Council, Quality Committee, and by the ColoRectal Education System Template Committee, American Board of Colon and Rectal Surgery. PATIENTS: No patients were included in this study. INTERVENTIONS: Six areas of colorectal surgery were defined by members of the American Society of Colon and Rectal Surgeons' Executive Council and the American Board of Colon and Rectal Surgery to cover areas of importance for colorectal surgeons. These included colectomy, rectal cancer, hemorrhoidectomy, anal fistula and abscess, colonoscopy, and rectal prolapse. Relevant American Society of Colon and Rectal Surgeons' committee members through a series of 4 panel discussions identified important demographic, process, and outcome measures in each of these 6 areas that might be suitable for the American College of Surgeons case log. Panel size was sequentially expanded from 8 members to 28 members to include all active committee members. Panelists contributed additional process and outcome measures for inclusion during each discussion. Modified Delphi methodology was used to generate consensus, and, after each panel discussion, members rated the relative importance of each end point from 1 (least important) to 4 (most important). MAIN OUTCOME MEASURES: The mean rating for each process and outcome measure after each round was recorded with the use of standardized definitions for relevant variables. RESULTS: Eighty-nine process and outcome measures were compiled and rated. Mean scores following the final round ranged from a low of 1.3 (anal fistula/abscess, preoperative imaging) to a high of 4.0 (colectomyanastomotic leak). LIMITATIONS: The limitations of this study involved the use of consensus, small study size, and the fact that no end points were excluded. CONCLUSIONS: With the use of modified Delphi methodology, a consensus-derived ranked list of 89 process and outcome measures was developed in 6 key areas of colorectal surgery. These data provide a framework for development of guideline standards for case-reporting program development initiatives for colon and rectal surgery.