ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grunkemeier, G. L.
Right arrow Articles by Jin, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Grunkemeier, G. L.
Right arrow Articles by Jin, R.
Related Collections
Right arrow Professional affairs

Ann Thorac Surg 2002;74:1749
© 2002 The Society of Thoracic Surgeons

Correspondence

Reply

Gary L. Grunkemeier, PhDa, K.Jeanne Zerr, RN, MBAa, Ruyun Jin, MDa

a St. Vincent Hospital,Providence Health System,9155 SW Barnes Rd, Suite 33,Portland, Oregon, 97225, USA

e-mail: ggrunkemeier{at}providence.org

To the Editor:

We thank Mr Nashef and Dr Roques for responding to our editorial and for agreeing with our conclusion. We respond to their comments in order.

We agree with their comments about risk predictability but do not believe that our criticism of a C-index of 0.80 is unfair. The C-index, the area beneath the receiver operating characteristic curve, is widely used to measure the discrimination value of an operative risk model. A receiver operating characteristic curve is derived by considering various death probabilities as potential cut-points for defining a test to identify deaths. It is a plot of the sensitivity (ability to predict deaths) of these tests versus their specificity (ability to predict survivors). A C-index of .70 to .80 is considered acceptable and .80 to .90 excellent [1].

To support our criticism, we will demonstrate the predictive value of a very good risk model. We recently applied The Society of Thoracic Surgeons’ operative risk model to diabetic patients who had coronary artery operations and found that it gave excellent predictability, as evidenced by a C-index of .83 (Furnary AP, Gao G, Grunkemeier GL, et al. Continuous insulin infusions reduce mortality in diabetic CABG patients. unpublished data). What does this mean for using this model to predict the death of a patient? Agambler would wager on death or survival according to whether the STS probability was greater or less than 50%, respectively. Only four deaths (of 87) would have been guessed correctly, for a sensitivity of 4.6% (Table 1). To increase the sensitivity to 95%, the probability cut-point would need to be lowered to 1.3%. But then only 30% of the surviving patients are predicted correctly. To obtain 95% specificity, we need to make the cut-point 10.8%; but then the sensitivity is only 46%. We think this makes the point that even with an excellent C-index, the ability to predict, though certainly not worthless, is still far from perfect.


View this table:
[in this window]
[in a new window]
  		Table 1. Test Specificity and Sensitivity for Using STS Operative Risk Model to Predict the Death

 
We fully agree with this comment. In the current environment of ever-increasing emphasis on outcomes and the resulting plethora of statistics in quality-improvement reports, it is critical that the ability to correctly interpret those reports be addressed. The danger of inappropriate response to a perceived outlier is real and could result in misguided judgement and subsequent inappropriate action taken to correct the perceived difference in quality of performance.

We tabulated only risk models that were specific for death after coronary artery procedures, and omitted those that included other cardiac procedures, such as heart valves. The EuroSCORE model was not included because it includes all cardiac procedures, coronary and valve. Nevertheless, we are aware of the excellent work done by the EuroSCORE project, from publications [2, 3] and from the website (www.euroscore.org), which has full information and support for patients and physicians. This is an admirable implementation and dissemination model, which should be emulated.

References

  1. Hosmer D.W., Lemeshow S. Applied logistic regression, 2nd ed New York: Wiley, 2000:162.
  2. Roques F., Nashef S.A.M., Michel P., Guaducheau E., de Vincentiis C., Baudet E., et al. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19030 patients. Eur J Cardiothorac Surg 1999;15:816-823.[Abstract/Free Full Text]
  3. Kurki T.S., Jarvinen O., Kataja M.J., Laurikka J., Tarkka M. Performance of three preoperative risk indices; CABDEAL, EuroSCORE and Cleveland models in a prospective coronary bypass database. Eur J Cardiothorac Surg 2002;21:406-410.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
D. M. Shahian, E. H. Blackstone, F. H. Edwards, F. L. Grover, G. L. Grunkemeier, D. C. Naftel, S. A.M. Nashef, W. C. Nugent, and E. D. Peterson
Cardiac Surgery Risk Models: A Position Article
Ann. Thorac. Surg., November 1, 2004; 78(5): 1868 - 1877.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grunkemeier, G. L.
Right arrow Articles by Jin, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Grunkemeier, G. L.
Right arrow Articles by Jin, R.
Related Collections
Right arrow Professional affairs

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
ANN THORAC SURG ASIAN CARDIOVASC THORAC ANN EUR J CARDIOTHORAC SURG
J THORAC CARDIOVASC SURG ICVTS ALL CTSNet JOURNALS