Quality improvement programme to achieve acceptable colonoscopy completion rates: prospective before and after study
http://www.100md.com
《英国医生杂志》
1 North Tyneside General Hospital, North Shields NE29 8NH
Correspondence to: M R Welfare doctormarkw@aol.com
Abstract
Outline of problem
Colonoscopy is the optimal procedure for examining the colon.1 Performance is operator dependent, and completion rates vary.2 Completion to the caecum is confirmed through use of a combination of signs—indentation or transillumination in the right iliac fossa and view of the appendix, the triradiate fold, or the ileocaecal valve.3 A completion rate of 90% is considered acceptable and since the start of our programme has been accepted by the UK endoscopy community.4 Median colonoscopy completion rates found in an audit in three regions in the United Kingdom were between 57% and 73%, depending on how completion is defined,5 although some institutions report adjusted completion rates of more than 90%.6 In the United States, crude completion rates of 95% have been reported in large series (such as one series of 3465 colonoscopies7), suggesting that a 90% completion rate is achievable in routine practice. The impact of incomplete colonoscopies on the success of a proposed national colorectal screening programme has been highlighted.5 8 We were aware that our colonoscopy completion rate was low, and we wished to attain the suggested standard so that our patients would benefit by avoiding subsequent barium enema or missed lesions.
Outline of context
North Tyneside General Hospital provides secondary care services to approximately 210 000 people in northeast England. Colonoscopies were carried out in a dedicated endoscopy unit staffed by nurses with endoscopy training, led by a senior nurse with more than 10 years' experience. Most endoscopy lists consisted of two colonoscopies and six to eight gastroscopies, and assistance was usually by one trained endoscopy nurse and one healthcare assistant at the beginning of the audit. No lists were dedicated to colonoscopy alone. Doctors at all stages of training were doing colonoscopy, including consultants, academics (a professor, senior lecturers, and research fellows), specialist registrars in gastroenterology and surgery, clinical assistants, and staff grades. Nurses did not do colonoscopy but were training in sigmoidoscopy at the time. Bowel preparation consisted of a low residue diet for 48 hours, with clear fluids only for the last 24 hours, and two sachets of Fleet Phosphosoda (De Witt, Runcorn, Cheshire) to be taken 12 hours apart. Midazolam and pethidine were used for conscious sedation as needed.
Key outcome measure
We used audit against the suggested minimum standards to identify reasons for incomplete colonoscopies and instituted appropriate changes to improve performance. We carried out two full cycles of audit with the following format. We did a detailed examination of each colonoscopy during a defined time period, examining reported reasons for incomplete examination. Results for individual colonoscopists were known only by themselves. We held departmental meetings to review the results, achieving consensus on methods of improving completion rates by using the results of the audit and considering the views of endoscopists and nursing staff. An agreed action plan was then put in place. We examined completion rates at the end of the cycle.
For the first cycle the baseline was all colonoscopies in 1998, when the crude completion rate was 60% (480/600) and no colonoscopist had a completion rate of greater than 90%. We reviewed in detail 124 colonoscopies from 8 April to 17 May 1999 to determine reasons for incomplete colonoscopy, an action plan was instituted in mid-1999, and we assessed the effectiveness of the change on all 1328 colonoscopies done in 2000. For the second cycle detailed analysis of all colonoscopies in 2000 served as the baseline, and an action plan was instituted in mid-2001. We assessed the effectiveness of the change on all 1166 colonoscopies done in 2002.
Strategy for change
The crude completion rate for the colonoscopies done in 2002 improved to 88.1% (984/1166). Thirty one (2.6%) were incomplete because of retained stool, and 24 (2.0%) were incomplete owing to impassable strictures or cancers. Excluding these cases from the analysis produced an adjusted completion rate of 93.8%.
The performance of all endoscopists improved over time. For example, one gastroenterologist (MRW) improved his crude completion rate from 79% to 95% between 1998 and 2002. This suggests that the quality improvement programme had a specific effect on both individual and departmental performance.
The figure summarises the changes in crude completion rates between 1999 and 2002 and gives the adjusted rate for 2002. The audit target of 90% crude completion rate has been approximately achieved. Between December 2000 and June 2002, prospective audit of complications revealed only two colonic perforations in 2077 procedures—a rate of 1 in 1000, which is lower than the rate of 1 in 769 seen in the national audit.3
Crude completion rate over a five year period and adjusted rate in 2002
Lessons learnt and next steps
Lindsay DC, Freeman JG, Cobden I, Record CO. Should colonoscopy be the first investigation for colonic disease? BMJ 1998;296: 167-9.
Balfour TW. Training for colonoscopy. J R Soc Med 1999;94: 160-1.
Bowles CJA, Leicester R, Swarbrick E, Williams CB, Romaya C, Epstein O. Intercollegiate-BSG national colonoscopy (IBNC) audit: methods used to identify the caecum and caecal intubation rate. Gut 2001;48(suppl I): A10, 036.
Royal College of Physicians. Guidelines for the training, appraisal and assessment of trainees on GI endoscopy. London: RCP, 2001.
Bowles CJA, Leicester R, Romaya C, Swarbrick E, Williams CB, Epstein O. A prospective study of colonoscopy practice in the UK today: are we adequately prepared for national colorectal screening tomorrow? Gut 2004;53: 277-83.
Thomas-Gibson S, Thapar C, Shah S, Saunders B. Colonoscopy at a combined district general hospital and specialist endoscopy unit: lessons from 505 consecutive examinations. J R Soc Med 2002;95: 194-7.
Nelson BD, McQuaid KR, Bond JH, Lieberman DA, CCSSG. Population-based colonoscopy screening for colorectal cancer is feasible and safe: preliminary results from the VA colonoscopy screening programme. Gastrointestinal Endosc 1999;49:abstract 65.
Palmer KR, Morris AI. A snapshot of colonoscopy practice in England: stimulus for improvement. Gut 2004;53: 163-5.
Mitchell R, McCallion K, Gardiner K, Watson RG, Collins JS. Successful colonoscopy; completion rates and reasons for incompletion. Ulster Med J 2002;71: 34-7.
Dafnis G, Granath F, Pahlman L. The impact of endoscopists' experience and learning curves and interendoscopist variation on colonoscopy completion rates. Endoscopy 2001;33: 511-7.(Jennifer E Ball, pre-regi)
Correspondence to: M R Welfare doctormarkw@aol.com
Abstract
Outline of problem
Colonoscopy is the optimal procedure for examining the colon.1 Performance is operator dependent, and completion rates vary.2 Completion to the caecum is confirmed through use of a combination of signs—indentation or transillumination in the right iliac fossa and view of the appendix, the triradiate fold, or the ileocaecal valve.3 A completion rate of 90% is considered acceptable and since the start of our programme has been accepted by the UK endoscopy community.4 Median colonoscopy completion rates found in an audit in three regions in the United Kingdom were between 57% and 73%, depending on how completion is defined,5 although some institutions report adjusted completion rates of more than 90%.6 In the United States, crude completion rates of 95% have been reported in large series (such as one series of 3465 colonoscopies7), suggesting that a 90% completion rate is achievable in routine practice. The impact of incomplete colonoscopies on the success of a proposed national colorectal screening programme has been highlighted.5 8 We were aware that our colonoscopy completion rate was low, and we wished to attain the suggested standard so that our patients would benefit by avoiding subsequent barium enema or missed lesions.
Outline of context
North Tyneside General Hospital provides secondary care services to approximately 210 000 people in northeast England. Colonoscopies were carried out in a dedicated endoscopy unit staffed by nurses with endoscopy training, led by a senior nurse with more than 10 years' experience. Most endoscopy lists consisted of two colonoscopies and six to eight gastroscopies, and assistance was usually by one trained endoscopy nurse and one healthcare assistant at the beginning of the audit. No lists were dedicated to colonoscopy alone. Doctors at all stages of training were doing colonoscopy, including consultants, academics (a professor, senior lecturers, and research fellows), specialist registrars in gastroenterology and surgery, clinical assistants, and staff grades. Nurses did not do colonoscopy but were training in sigmoidoscopy at the time. Bowel preparation consisted of a low residue diet for 48 hours, with clear fluids only for the last 24 hours, and two sachets of Fleet Phosphosoda (De Witt, Runcorn, Cheshire) to be taken 12 hours apart. Midazolam and pethidine were used for conscious sedation as needed.
Key outcome measure
We used audit against the suggested minimum standards to identify reasons for incomplete colonoscopies and instituted appropriate changes to improve performance. We carried out two full cycles of audit with the following format. We did a detailed examination of each colonoscopy during a defined time period, examining reported reasons for incomplete examination. Results for individual colonoscopists were known only by themselves. We held departmental meetings to review the results, achieving consensus on methods of improving completion rates by using the results of the audit and considering the views of endoscopists and nursing staff. An agreed action plan was then put in place. We examined completion rates at the end of the cycle.
For the first cycle the baseline was all colonoscopies in 1998, when the crude completion rate was 60% (480/600) and no colonoscopist had a completion rate of greater than 90%. We reviewed in detail 124 colonoscopies from 8 April to 17 May 1999 to determine reasons for incomplete colonoscopy, an action plan was instituted in mid-1999, and we assessed the effectiveness of the change on all 1328 colonoscopies done in 2000. For the second cycle detailed analysis of all colonoscopies in 2000 served as the baseline, and an action plan was instituted in mid-2001. We assessed the effectiveness of the change on all 1166 colonoscopies done in 2002.
Strategy for change
The crude completion rate for the colonoscopies done in 2002 improved to 88.1% (984/1166). Thirty one (2.6%) were incomplete because of retained stool, and 24 (2.0%) were incomplete owing to impassable strictures or cancers. Excluding these cases from the analysis produced an adjusted completion rate of 93.8%.
The performance of all endoscopists improved over time. For example, one gastroenterologist (MRW) improved his crude completion rate from 79% to 95% between 1998 and 2002. This suggests that the quality improvement programme had a specific effect on both individual and departmental performance.
The figure summarises the changes in crude completion rates between 1999 and 2002 and gives the adjusted rate for 2002. The audit target of 90% crude completion rate has been approximately achieved. Between December 2000 and June 2002, prospective audit of complications revealed only two colonic perforations in 2077 procedures—a rate of 1 in 1000, which is lower than the rate of 1 in 769 seen in the national audit.3
Crude completion rate over a five year period and adjusted rate in 2002
Lessons learnt and next steps
Lindsay DC, Freeman JG, Cobden I, Record CO. Should colonoscopy be the first investigation for colonic disease? BMJ 1998;296: 167-9.
Balfour TW. Training for colonoscopy. J R Soc Med 1999;94: 160-1.
Bowles CJA, Leicester R, Swarbrick E, Williams CB, Romaya C, Epstein O. Intercollegiate-BSG national colonoscopy (IBNC) audit: methods used to identify the caecum and caecal intubation rate. Gut 2001;48(suppl I): A10, 036.
Royal College of Physicians. Guidelines for the training, appraisal and assessment of trainees on GI endoscopy. London: RCP, 2001.
Bowles CJA, Leicester R, Romaya C, Swarbrick E, Williams CB, Epstein O. A prospective study of colonoscopy practice in the UK today: are we adequately prepared for national colorectal screening tomorrow? Gut 2004;53: 277-83.
Thomas-Gibson S, Thapar C, Shah S, Saunders B. Colonoscopy at a combined district general hospital and specialist endoscopy unit: lessons from 505 consecutive examinations. J R Soc Med 2002;95: 194-7.
Nelson BD, McQuaid KR, Bond JH, Lieberman DA, CCSSG. Population-based colonoscopy screening for colorectal cancer is feasible and safe: preliminary results from the VA colonoscopy screening programme. Gastrointestinal Endosc 1999;49:abstract 65.
Palmer KR, Morris AI. A snapshot of colonoscopy practice in England: stimulus for improvement. Gut 2004;53: 163-5.
Mitchell R, McCallion K, Gardiner K, Watson RG, Collins JS. Successful colonoscopy; completion rates and reasons for incompletion. Ulster Med J 2002;71: 34-7.
Dafnis G, Granath F, Pahlman L. The impact of endoscopists' experience and learning curves and interendoscopist variation on colonoscopy completion rates. Endoscopy 2001;33: 511-7.(Jennifer E Ball, pre-regi)