| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:100-101
© 1995 The Society of Thoracic Surgeons
 |
Introduction |
|---|
 Top Introduction |
|---|
DR CHARLES A. S. MARRIN (Lebanon, NH): I think this is a commendable study, and it sheds light on some questions to which many of us, and indeed many of our adversaries, are anxious to know the answers. However, I am somewhat uncomfortable with your conclusions about the model. It is in the nature of prediction models that they do not perform well at the extremes. The fact is that this is an unusual population, and as the model is designed to apply to the majority of patients, it is unlikely that it will predict accurately in high-risk subsets.
I also would take issue with your comment about adjusting for age. Clearly the age weighting assigned by Dr Parsonnet in his original model overpredicts the risk; however, when you eliminated age as a consideration, there was not a dramatic improvement in the correlation between the estimated and observed risks in your patient population. Adjusting for age really made little difference, so there is clearly something else. Maybe your 70-year-old patients are younger than Dr Parsonnet's 70-year-old patients, or there are other factors that we do not know. But I think there is danger in attributing this effect to age alone. I think that there may be other factors that deserve careful analysis.
DR KATZ: Thank you for your comments. First of all, in terms of the results in this era as compared with those of the 1980s, we believe that the new techniques of myocardial preservation are very important in reducing risk. Low cardiac output is very poorly tolerated, especially in elderly patients, and is associated with increased morbidity and mortality.
Another interesting aspect is the use of the internal mammary artery. The Society of Thoracic Surgeons National Database indicates that coronary artery bypass grafting results for all patients, except for elderly patients undergoing reoperation, may be better when the internal mammary artery is used. In this study 78% of the patients in the elderly group received internal mammary artery grafts, and I believe the internal mammary artery has been used more often in elderly patients in other series.
In regard to the factors you mentioned, it is interesting that, as we looked at the individual factors in the Parsonnet model to determine whether they were determinants of operative mortality, we found that only a minority of them were. So I agree with you, I think it is more complex than this system would indicate and we will need to look at more complex predictive models that can be adjusted for new data.
DR JOHN C. ALEXANDER (Evanston, IL): We have been very interested in trying to predict the charges or the costs in patients who undergo coronary artery bypass grafting because we are involved in a variety of contractual relationships and we want our hospital to be financially stable. We have used the Parsonnet score to construct a regression relationship between hospital charges and the Parsonnet number, and as it turns out, it is a pretty reasonable predictor of charges. In diagnosis-related group 107 our charges during the same time frame that you dealt with were approximately $25,000 plus $500 for every Parsonnet score count. That let us know whether we were going to be able to financially provide care in a contracted group, given some assumptions about the cohort. And, as you pointed out, age is an important part of Parsonnet's initial data. It is a tool that we have used to try to help us know when we are going to get into a contractual trap in terms of taking on a group that has an older average age.
This research is interesting as it goes forward, and I think we will see more relationships between our classic predictors of operative mortality and charges or costs. That is certainly what we found during the late 1980s and early 1990s. The standard methods that we have used in the past to predict operative mortality also can be used to predict the costs of providing care.
DR KATZ: We certainly agree, and I might point out that our use of the hospital charge ratio was to some extent determined by the fact that we had a great deal of difficulty determining the actual cost versus the hospital charge. Therefore we chose to normalize the data with the ratio to compare the data between elderly and younger patients.
| ||||||||||||||||||||||||||||||||||||||||||
| 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 |
