Optimal Design for Heart Defibrillators

During heart defibrillator implantation, a physician fibrillates the patient’s heart several times at different test strengths to estimate the effective strength necessary for defibrillation. One strategy is to implant at the strength that de-fibrillates 95% of the time (ED95). Efficient choice and use of the test strengths in this estimation problem is crucial, as each additional observation increases the risk of serious injury or death. Such choice can be formalized as finding an optimal design in, say, a logistic regression problem with interest in estimating the ED95. In practice, important features specific to this problem are the very small sample sizes; the asymmetry of the loss function; and the fact that the prior distribution arises as the distribution for the next draw of patient-specific parameters in a hierarchical model.

[1]  Nancy Flournoy,et al.  A Clinical Experiment in Bone Marrow Transplantation: Estimating a Percentage Point of a Quantal Response Curve , 1993 .

[2]  K. Chaloner,et al.  Optimal Bayesian design applied to logistic regression experiments , 1989 .

[3]  S. Zacks,et al.  Problems and Approaches in Design of Experiments for Estimation and Testing in Non-Linear Models. , 1975 .

[4]  J. Q. Smith,et al.  1. Bayesian Statistics 4 , 1993 .

[5]  E. Fain,et al.  The relationship between successful defibrillation and delivered energy in open-chest dogs: reappraisal of the "defibrillation threshold" concept. , 1987, American heart journal.

[6]  I. Verdinelli,et al.  Bayesian designs for maximizing information and outcome , 1992 .

[7]  J B Greenhouse,et al.  Bayesian methods for phase I clinical trials. , 1992, Statistics in medicine.

[8]  L A Geddes,et al.  Strength-duration curves for trapezoidal waveforms of various tilts for transchest defibrillation in animals. , 1978, Medical instrumentation.

[9]  Nancy Flournoy,et al.  Random Walks for Quantile Estimation , 1994 .

[10]  S. Gupta,et al.  Statistical decision theory and related topics IV , 1988 .

[11]  J O'Quigley,et al.  Continual reassessment method: a practical design for phase 1 clinical trials in cancer. , 1990, Biometrics.

[12]  McDaniel Wc,et al.  The cardiac ventricular defibrillation threshold: inherent limitations in its application and interpretation. , 1987 .

[13]  R. Ideker,et al.  Estimating the 95% effective defibrillation dose , 1993, IEEE Transactions on Biomedical Engineering.

[14]  P. S. Chen,et al.  Effects of pacing rate and timing of defibrillation shock on the relation between the defibrillation threshold and the upper limit of vulnerability in open chest dogs. , 1991, Journal of the American College of Cardiology.

[15]  D. McLeish,et al.  The estimation of extreme quantiles in logit bioassay , 1983 .

[16]  R. Tsutakawa.  Selection of Dose Levels for Estimating a Percentage Point of a Logistic Quantal Response Curve , 1980 .

[17]  Robert K. Tsutakawa,et al.  Design of Experiment for Bioassay , 1972 .

[18]  K. Chaloner,et al.  Bayesian Experimental Design: A Review , 1995 .

[19]  Kathryn Chaloner,et al.  A note on optimal Bayesian design for nonlinear problems , 1993 .

[20]  C. Haffajee,et al.  Clinical experience in seventy-seven patients with the automatic implantable cardioverter defibrillator. , 1989, American heart journal.

[21]  P. Freeman.  Optimal Bayesian sequential estimation of the median effective dose , 1970 .

[22]  B E Storer,et al.  Design and analysis of phase I clinical trials. , 1989, Biometrics.

[23]  Peter Müller,et al.  Optimal Design via Curve Fitting of Monte Carlo Experiments , 1995 .

[24]  Trevor Hastie,et al.  Statistical Models in S , 1991 .

[25]  J. Schuder,et al.  The cardiac ventricular defibrillation threshold: inherent limitations in its application and interpretation. , 1987, Medical instrumentation.

[26]  N V Thakor,et al.  The Defibrillation Success Rate Versus Energy Relationship: Part I‐Curve Fitting and the Most Efficient Defibrillation Energy , 1990, Pacing and clinical electrophysiology : PACE.