Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit

P.E. Teloken, Katrina Spilsbury, Cameron Platell, A. Heriot, C. Byrne, P. Chapuis, M. Doudle, P. McMurrick, E. Murphy, M. Thompson-Fawcett, A. Brennan, C. Reid, M. O'Regan, A. Hunter

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

© 2016 Royal Australasian College of Surgeons.Background: In the last decade, there has been a significant increase in interest for public reporting of outcome data and performance comparison across institutions and surgeons. This study aims at comparing postoperative mortality after colorectal cancer surgery across units and individual consultants in Australia and New Zealand using funnel plots. Methods: The Bi-National Colorectal Cancer Audit database was used. Unadjusted and adjusted funnel plots of inpatient mortality were constructed. Risk adjustment was based upon multivariable logistic regression models using purposeful covariate selection. Results: A total of 10 008 patients undergoing surgery for colorectal cancer from 56 surgical units and 90 consultants were identified. Overall inpatient mortality was 1.51%, corresponding to 1.1% for elective and 3.9% for urgent cases. Logistic regression identified age, American Society of Anesthesiologists score, urgent surgery and open surgery to be independently associated with inpatient mortality. Unadjusted and adjusted funnel plot analysis identified three (5.3%) units exceeding the inner limit and none exceeding the outer limit. Six (6.6%) consultants had inpatient mortality between the upper inner and outer limits and one (1.1%) between the inferior inner and outer limits. Upon adjustment, seven (7.7%) consultants had inpatient mortality between the inner and outer limit. Potential limitations of this study include: residual confounding being responsible for the association of open surgery and mortality; incomplete case-mix adjustment resulting in outlier identification; and bias towards inclusion of larger institutions. Conclusion: Mortality figures in Australia and New Zealand are comparable to recently reported international data. The vast majority of units and consultants are performing within the expected boundaries.
Original languageEnglish
Pages (from-to)454-458
Number of pages5
JournalANZ Journal of Surgery
Volume86
Issue number6
DOIs
Publication statusPublished - 2016

Fingerprint

Colorectal Surgery
Colorectal Neoplasms
Consultants
Mortality
Inpatients
Risk Adjustment
Logistic Models
New Zealand
Research Design
Databases

Cite this

Teloken, P. E., Spilsbury, K., Platell, C., Heriot, A., Byrne, C., Chapuis, P., ... Hunter, A. (2016). Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit. ANZ Journal of Surgery, 86(6), 454-458. https://doi.org/10.1111/ans.13523
Teloken, P.E. ; Spilsbury, Katrina ; Platell, Cameron ; Heriot, A. ; Byrne, C. ; Chapuis, P. ; Doudle, M. ; McMurrick, P. ; Murphy, E. ; Thompson-Fawcett, M. ; Brennan, A. ; Reid, C. ; O'Regan, M. ; Hunter, A. / Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit. In: ANZ Journal of Surgery. 2016 ; Vol. 86, No. 6. pp. 454-458.
@article{4fe5125ce15d4bd5a3707d93a4b222c3,
title = "Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit",
abstract = "{\circledC} 2016 Royal Australasian College of Surgeons.Background: In the last decade, there has been a significant increase in interest for public reporting of outcome data and performance comparison across institutions and surgeons. This study aims at comparing postoperative mortality after colorectal cancer surgery across units and individual consultants in Australia and New Zealand using funnel plots. Methods: The Bi-National Colorectal Cancer Audit database was used. Unadjusted and adjusted funnel plots of inpatient mortality were constructed. Risk adjustment was based upon multivariable logistic regression models using purposeful covariate selection. Results: A total of 10 008 patients undergoing surgery for colorectal cancer from 56 surgical units and 90 consultants were identified. Overall inpatient mortality was 1.51{\%}, corresponding to 1.1{\%} for elective and 3.9{\%} for urgent cases. Logistic regression identified age, American Society of Anesthesiologists score, urgent surgery and open surgery to be independently associated with inpatient mortality. Unadjusted and adjusted funnel plot analysis identified three (5.3{\%}) units exceeding the inner limit and none exceeding the outer limit. Six (6.6{\%}) consultants had inpatient mortality between the upper inner and outer limits and one (1.1{\%}) between the inferior inner and outer limits. Upon adjustment, seven (7.7{\%}) consultants had inpatient mortality between the inner and outer limit. Potential limitations of this study include: residual confounding being responsible for the association of open surgery and mortality; incomplete case-mix adjustment resulting in outlier identification; and bias towards inclusion of larger institutions. Conclusion: Mortality figures in Australia and New Zealand are comparable to recently reported international data. The vast majority of units and consultants are performing within the expected boundaries.",
author = "P.E. Teloken and Katrina Spilsbury and Cameron Platell and A. Heriot and C. Byrne and P. Chapuis and M. Doudle and P. McMurrick and E. Murphy and M. Thompson-Fawcett and A. Brennan and C. Reid and M. O'Regan and A. Hunter",
year = "2016",
doi = "10.1111/ans.13523",
language = "English",
volume = "86",
pages = "454--458",
journal = "Australian New Zealand Journal of Surgery",
issn = "1445-1433",
publisher = "John Wiley & Sons",
number = "6",

}

Teloken, PE, Spilsbury, K, Platell, C, Heriot, A, Byrne, C, Chapuis, P, Doudle, M, McMurrick, P, Murphy, E, Thompson-Fawcett, M, Brennan, A, Reid, C, O'Regan, M & Hunter, A 2016, 'Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit' ANZ Journal of Surgery, vol. 86, no. 6, pp. 454-458. https://doi.org/10.1111/ans.13523

Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit. / Teloken, P.E.; Spilsbury, Katrina; Platell, Cameron; Heriot, A.; Byrne, C.; Chapuis, P.; Doudle, M.; McMurrick, P.; Murphy, E.; Thompson-Fawcett, M.; Brennan, A.; Reid, C.; O'Regan, M.; Hunter, A.

In: ANZ Journal of Surgery, Vol. 86, No. 6, 2016, p. 454-458.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analysis of mortality in colorectal surgery in the Bi-National Colorectal Cancer Audit

AU - Teloken, P.E.

AU - Spilsbury, Katrina

AU - Platell, Cameron

AU - Heriot, A.

AU - Byrne, C.

AU - Chapuis, P.

AU - Doudle, M.

AU - McMurrick, P.

AU - Murphy, E.

AU - Thompson-Fawcett, M.

AU - Brennan, A.

AU - Reid, C.

AU - O'Regan, M.

AU - Hunter, A.

PY - 2016

Y1 - 2016

N2 - © 2016 Royal Australasian College of Surgeons.Background: In the last decade, there has been a significant increase in interest for public reporting of outcome data and performance comparison across institutions and surgeons. This study aims at comparing postoperative mortality after colorectal cancer surgery across units and individual consultants in Australia and New Zealand using funnel plots. Methods: The Bi-National Colorectal Cancer Audit database was used. Unadjusted and adjusted funnel plots of inpatient mortality were constructed. Risk adjustment was based upon multivariable logistic regression models using purposeful covariate selection. Results: A total of 10 008 patients undergoing surgery for colorectal cancer from 56 surgical units and 90 consultants were identified. Overall inpatient mortality was 1.51%, corresponding to 1.1% for elective and 3.9% for urgent cases. Logistic regression identified age, American Society of Anesthesiologists score, urgent surgery and open surgery to be independently associated with inpatient mortality. Unadjusted and adjusted funnel plot analysis identified three (5.3%) units exceeding the inner limit and none exceeding the outer limit. Six (6.6%) consultants had inpatient mortality between the upper inner and outer limits and one (1.1%) between the inferior inner and outer limits. Upon adjustment, seven (7.7%) consultants had inpatient mortality between the inner and outer limit. Potential limitations of this study include: residual confounding being responsible for the association of open surgery and mortality; incomplete case-mix adjustment resulting in outlier identification; and bias towards inclusion of larger institutions. Conclusion: Mortality figures in Australia and New Zealand are comparable to recently reported international data. The vast majority of units and consultants are performing within the expected boundaries.

AB - © 2016 Royal Australasian College of Surgeons.Background: In the last decade, there has been a significant increase in interest for public reporting of outcome data and performance comparison across institutions and surgeons. This study aims at comparing postoperative mortality after colorectal cancer surgery across units and individual consultants in Australia and New Zealand using funnel plots. Methods: The Bi-National Colorectal Cancer Audit database was used. Unadjusted and adjusted funnel plots of inpatient mortality were constructed. Risk adjustment was based upon multivariable logistic regression models using purposeful covariate selection. Results: A total of 10 008 patients undergoing surgery for colorectal cancer from 56 surgical units and 90 consultants were identified. Overall inpatient mortality was 1.51%, corresponding to 1.1% for elective and 3.9% for urgent cases. Logistic regression identified age, American Society of Anesthesiologists score, urgent surgery and open surgery to be independently associated with inpatient mortality. Unadjusted and adjusted funnel plot analysis identified three (5.3%) units exceeding the inner limit and none exceeding the outer limit. Six (6.6%) consultants had inpatient mortality between the upper inner and outer limits and one (1.1%) between the inferior inner and outer limits. Upon adjustment, seven (7.7%) consultants had inpatient mortality between the inner and outer limit. Potential limitations of this study include: residual confounding being responsible for the association of open surgery and mortality; incomplete case-mix adjustment resulting in outlier identification; and bias towards inclusion of larger institutions. Conclusion: Mortality figures in Australia and New Zealand are comparable to recently reported international data. The vast majority of units and consultants are performing within the expected boundaries.

U2 - 10.1111/ans.13523

DO - 10.1111/ans.13523

M3 - Article

VL - 86

SP - 454

EP - 458

JO - Australian New Zealand Journal of Surgery

JF - Australian New Zealand Journal of Surgery

SN - 1445-1433

IS - 6

ER -