|
|
||||||||
J Thorac Cardiovasc Surg 2003;125:1175-1176
© 2003 The American Association for Thoracic Surgery
Letters to the Editor |
Office for Medical Technology Assessmenta, Department of Pulmonologyb, University Hospital Groningen, PO Box 30.001, 9700 RB Groningen, The Netherlands
To the Editor:
In the March 2002 issue of this Journal, Anyanwu and associates
1 reported a prospective multicenter study about cost-effectiveness and cost-utility for the different types of lung transplantation. Although the study is an important addition to the existing literature, the method by which survival gain was calculated and the impact on their sensitivity analyses should be discussed further.
First, in the presented study Anyanwu and associates
1 calculated survival for the first 4 years from actual data and from years 4 to 15 by using a parametric Weibull model, in accordance with a previous analysis by our group.
2 At 15 years, the survival curve after transplantation was cut off despite a survival of at least 25%, depending on the type of transplantation. Survival on the waiting list declined to 0% after 11 years. For a valid comparison of costs and effects of both conditions from a lifetime perspective, a survival curve after transplantation should be constructed with further extrapolation until a survival of 0%.
3,4 Therefore the survival after lung transplantation in the presented study was underestimated, because prolonged survival beyond 15 years was neglected. A rough calculation indicates that extrapolation to 0% survival would amount to an estimated additional survival gain after transplantation of 2 years. This is why, in contrast to others,
5 Anyanwu and colleagues
1 did not find survival as a principal determinant of cost-effectiveness in their sensitivity analyses.
Second, the survival curve for the waiting list, to which the survival after lung transplantation was compared, was not constructed in accordance with previously reported methods.
3,4 In previous studies the transplantation date was chosen as the starting point for this comparison, because it would be unrealistic to assume that differences in survival occur before transplantation. For the situation on the waiting list, no real transplantation date exists, and a fictitious transplantation moment should be created (Figure 1). This moment should be based on the average stay of patients on the waiting list before transplantation (about 12 months in the study by Anyanwu and associates
1). However, day 0 was used as the starting point for the waiting list. Applied to our own data, these two methods result in a difference in the cumulative number of life years on the waiting list of 0.5 years (17%).
|
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |