Application of models for multivariate mixed outcomes to medical device trials: coronary artery stenting

Statistics in Medicine - Tập 22 Số 2 - Trang 313-336 - 2003
A. James O’Malley1, Sharon‐Lise T. Normand2,1, Richard E. Kuntz3
1Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, MA 02115, U.S.A.
2Department of Biostatistics, Harvard School of Public Health, Boston, U.S.A.
3Divisions of Clinical Biometrics and Cardiovascular Disease, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, U.S.A.

Tóm tắt

Abstract

The approval process for some medical devices involves a single‐arm trial in which the outcomes associated with the new device are compared to the expected outcomes associated with approved devices, the latter denoted the objective performance criterion (OPC). In this paper, models for multivariate mixed outcomes are applied to derive the OPC for a medical device to be used in clinical evaluations of the same type of device. We illustrate the techniques by determining the OPC for coronary artery stents, metal cages used to widen blocked coronary arteries in patients with coronary artery disease, using data from seven randomized trials of stents approved for use in the U.S.A. involving 5806 patients. The OPC is based on two 9‐month endpoints: target lesion revascularization, a binary outcome, and proportion diameter stenosis, a continuous outcome. To account for the correlation between mixed outcomes we consider factorization of the likelihood into marginal and conditional components, or adoption of a latent variable model. Because the models have different structural forms, standard methods for model comparison (such as the AIC and BIC) cannot be used. We discuss how model identifiability and valid inference are achieved, and then adapt the deviance information criterion (DIC) and the pseudo‐Bayes factor (PSBF) to select the best model. Nine months post‐stenting, we find that the average posterior probability (standard deviation) of target lesion revascularization ranges from 0.086 (0.008) for non‐diabetics with one diseased vessel to 0.163 (0.022) for diabetics with three diseased vessels. When considering proportion diameter stenosis, the corresponding posterior means are 0.375 (0.020) and 0.427 (0.030). The correlation coefficient of the components of the OPC lies in the range 0.042 to 0.116. Copyright © 2003 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

FDA Modernization Act.105th Congress of the United States of America amendment to the FFD & C section 201 21 U.S.C. 301 et seq. XXX 1997.

10.1056/NEJM199408253310801

10.1056/NEJM199408253310802

10.1016/S0002-9149(00)01307-2

10.1016/S0002-9149(00)01308-4

10.1016/S0002-9149(99)00581-0

Gersh BJ, 1986, Issues concerning the clinical‐evaluation of new prosthetic heart valves, Journal of Thoracic and Cardiovascular Surgery, 91, 460, 10.1016/S0022-5223(19)36063-5

Grunkemeier GL, 1992, Prosthetic heart valve performance: Long term follow‐up, Current Problems in Cardiology, 17, 331

Grunkemeier GL, 1993, Will randomized trials detect random valve failure? Reflection on a recent FDA workshop, Journal of Heart Valve Disease, 2, 424

10.1056/NEJM199812033392303

10.1016/S0735-1097(99)00271-5

FDA Pre‐Market Approval P970020. ACS Multi‐link TM coronary stent system (Guidant Corporation).http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMA/pma.cfm.

FDA Pre‐Market Approval P970035. AVE Micro Stent II over‐the‐wire coronary stent system and AVE GFX over‐the‐wire coronary stent system (Medtronic AVE Inc.).http://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfPMA/pma.cfm.

10.1214/aoms/1177705052

10.1080/01621459.1995.10476583

10.1093/biomet/73.1.13

10.2307/2531248

10.1080/01621459.1999.10473815

10.2307/2346482

10.1080/01621459.1992.10475264

10.1093/biomet/79.3.441

10.1111/1467-9868.00236

10.1007/978-1-4757-5644-9_4

10.1111/1467-9868.00090

Agresti A, 1990, Categorical Data Analysis, 319

10.1007/978-1-4757-4286-2

Lad F, 1996, Operational Subjective Statistical Methods

10.1109/TPAMI.1984.4767596

10.1080/01621459.1990.10476213

10.1063/1.1699114

10.1093/biomet/57.1.97

SpiegelhalterDJ BestNG CarlinBP.Bayesian deviance the effective number of parameters and the comparison of arbitrary complex models. Technical report 98‐009 Division of Biostatistics University of Minnesota 1998.

Gelfand AE, 1992, Bayesian Statistics 4, 147, 10.1093/oso/9780198522669.003.0009

Gelfand AE, 1994, Bayesian model choice: asymptotics and exact calculations, Journal of the Royal Statistical Society, Series B, 56, 501, 10.1111/j.2517-6161.1994.tb01996.x

Dempster AP, 1974, Proceedings of Conference on Foundational Questions in Statistical Inference, 335

SpiegelhalterDJ ThomasA BestNG.WinBUGS Version 1.2. MRC Biostatistics Unit Cambridge U.K. 1999.

10.1214/ss/1015346317

Best N, 1995, Convergence Diagnosis and Output Analysis Software for Gibbs Sampling Output

Hellmich M, 1998, A Bayesian approach to a general regression model for ROC curves, Academic Radiology, 18, 436

10.1214/ss/1177011136

Raftery AL, 1992, Bayesian Statistics 4, 763, 10.1093/oso/9780198522669.003.0053

Little RJA, 1987, Statistical Analysis with Missing Data

10.1093/ije/25.6.1107

10.2307/2669855