Assessing the quality of colonoscopy: validation of quality indicators

  • Tinmouth, Jill J. (PI)
  • Angus, Helen H. (CoPI)
  • Baxter, Nancy N. (CoPI)
  • Cohen, Lawrence (CoPI)
  • Dobrow, Mark M. (CoPI)
  • Jeejeebhoy, Khursheed K. (CoPI)
  • Murray, Iain I. (CoPI)
  • Nijhawan, Pardeep P. (CoPI)
  • Paszat, Lawrence L. (CoPI)
  • Rabeneck, Linda L. (CoPI)
  • Sutradhar, Rinku R. (CoPI)

Project Details


Dr Tinmouth will contribute to improved colonoscopy quality by evaluating the accuracy and effectiveness of commonly-used quality assessment measures for this colorectal cancer prevention and detection tool. She will do a large-scale review of administrative information collected from colonoscopies at 30 Ontario health care centres to see if this provides the right kind of data to properly monitor the quality of these procedures. Project summary: Dr Tinmouth's team propose to test the accuracy of a set of quality measures for colonoscopy, a critical tool for the prevention and diagnosis of colorectal cancer (CRC), developed using administrative data compared to data abstracted from patients' charts. Previous research: Colonoscopy is a critical tool in the early detection and prevention of CRC. Use of colonoscopy is increasing because of new screening programs for CRC. Dr Tinmouth's team has shown that the quality of colonoscopy in Ontario may vary depending on who is doing it and where is is performed. Poor-quality colonoscopy is and important problem as it may lead to increased patient morbidity and mortality. Administrative databases containing health information collected for billing purposes, such as provincial health plan databases, may be used to evaluate the quality of health care as they are easily obtainable and inexpensive to use. However, as they are not collected for research purposes, these data may contain important inaccuracies. Project description: The rationale of the study is to test the accuracy of administrative data definitions of quality measures of colonoscopy before using them for quality assurance, quality improvement or research purposes. The team has created a case definition of colonoscopy and defined five quality measures using codes from the administrative databases housed at the Institute for Clinical Evaluative Sciences (ICES), which include structures (where was the colonoscopy performed, whether an anaesthesiologist assists), processes [(was the colonoscopy completed (i.e., was the entire colon examined?), was polypectomy (i.e., removal of precancerous growths in the colon) performed], and outcomes (was a CRC missed?) of colonoscopy care. They test these definitions by doing a large-scale chart review at 30 institutions in Ontario. They will determine the accuracy of (1) their administrative data definition of structures, process measures and their case definition of colonoscopy compared to the charts of patients receiving colonoscopy in 2007 (Specific Aim #1) and (2) their definition of a key-outcome measure compared to data from previously abstracted charts of patients diagnosed with CRC bettheyen 2000 and 2005 (Specific Aim #2). Impact and relevance: Health policy makers in Canada are already actively using administrative data to evaluate and monitor the quality of care. This study will have important health policy implications as it will identify the best potential measures of the quality of colonoscopy care using administrative data, which might be used by policymakers and health administrators. Furthermore, the development of accurate quality measures using administrative data is critical for future research in the quality of colonoscopy in Canada, which will include studies of targeted interventions designed to improve of the quality of colonoscopy care. The overall goal of this research will be to provide all Canadians with uniformly high-quality colonoscopy services, thereby enhancing their ability to detect and prevent CRC.

Effective start/end date1/07/0930/06/11


  • Canadian Cancer Society Research Institute: $145,341.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.