Application of statistical decision theory to treatment choices: implications for the design and analysis of clinical trials.

Statistics in Medicine (Impact Factor: 2.04). 09/1986; 5(5):411-20. DOI: 10.1002/sim.4780050505
Source: PubMed

ABSTRACT This paper explores the application of statistical decision theory to treatment choices in cancer which involve difficult value judgements in weighing beneficial and deleterious outcomes of treatment. Strengths and weaknesses of using decision theory are illustrated by considering the problem of selecting chemotherapy in advanced ovarian cancer. The paper includes an assessment of individual preferences in 27 volunteers and a discussion of some problems in utility assessment. An alternative approach, using threshold analysis, is presented in which the results of the decision analysis are expressed as a function of utility parameters. By knowing what sets of utilities favour each treatment, the assessment of patient preferences can then be focused on important differences of treatment options. The implications of these results for the design and analysis of clinical trials are discussed.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Those who report a clinical trial should acknowledge the right of the ‘consumer’ to make decisions based on his own valuation of the beneficial and adverse effects which rival treatments may have. Suppose a new patient is inclined to trade one unit of benefit for c units of complication. Then he should (should not) be given the treatment if his estimated utility gain, χ1 -cχ2, is positive (negative) and statistically significant according to the data of the trial: here χ1 (χ2) denotes the observed average benefit (complication level). If the estimated gain is not statistically significant, the data do not allow any firm recommendation. This c-dependent recommendation in general cannot be determined from inspection of a joint confidence region for the two means concerned. Therefore investigators should present the outcome of the significance test as a function of c (inverted inference). Typically there are several types of adverse effect or benefit, in which case the quantity c must be generalized into a vector of personal relative utility weights.
    Statistics in Medicine 09/1987; 6(7):745 - 752. DOI:10.1002/sim.4780060705 · 2.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We use a statistical model to examine the relationship between α level, sample size, trial duration, patient accrual rate and therapeutic innovation rate on the increase in treatment efficacy achieved after a series of two-treatment randomized phase III trials. In a setting where the trials include most of the patients in the target population for inference, as in some paediatric cancers, we show that the traditional criteria by which one determines trial size are difficult to justify and apply. In particular, using as a measure of evidence typeI error levels larger than the typical 5 per cent for judging treatment differences, and performing smaller trials than one would usually consider feasible, yields on average, over a 25-year research course, larger gains in cure rate. Judicious choice of type I error rate and trial size keeps the chance of worsening treatment efficacy at a low level, even while increasing the chance of making large improvements in cure rate. We propose that a more appropriate view of trial design in low-incidence cancer settings is in the overall context of the research setting and long-term goals rather than in the narrow context of the current single trial. From this viewpoint, insistence on large trials and stringent evidence for accepting new treatments can be counter productive, in that likely gains in efficacy of treatment will be smaller over the long term. Copyright © 1999 John Wiley & Sons, Ltd.
    Statistics in Medicine 05/1999; 18(10):1183 - 1197. DOI:10.1002/(SICI)1097-0258(19990530)18:10<1183::AID-SIM122>3.0.CO;2-P · 2.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To assess whether the benefits of adding cisplatin (CDDP) concurrent with radiotherapy, followed by adjuvant CDDP and fluorouracil, justifies the toxicity cost for nasopharyngeal cancer (NPC) using the quality-adjusted time without symptoms or toxicity (Q-TWiST) approach. One hundred seven patients treated with radiotherapy (RT) and 111 with concurrent chemotherapy and radiotherapy (CRT) were analyzed. The overall survival was divided into three health states: time on active treatment only, during which any subjective nonhematologic toxic event of grade > = 3 was reported; time without symptoms of disease relapse; or treatment and time following first disease RELapse. The relative advantage of CRT and RT was examined by conducting the analysis cumulatively at restriction times 3, 6, 24, 36, 48 months. At 48 months, the improvement in disease-free survival was 14.4% for CRT, whereas that for overall survival was 18.9%. The differences in Q-TWiST were -0.4, -0.7, 0.1, 1.6, and 3.6 months at 3, 6, 24, 36, and 48 months, respectively, with positive differences favoring CRT. At 24 months, the difference in Q-TWiST began to favor CRT. At 36 months, CRT may be the preferred option from the patient's viewpoint if the time spent in the REL state is valued to be <0.83, with the value of perfect health being 1. Finally, Q-TWiST accumulated within 48 months indicated a significant advantage in quality-adjusted survival time for CRT (p = 0.020). Irrespective of how patients valued periods of toxicity and delayed disease progression, concurrent chemotherapy and radiotherapy offered NPC patients significantly more quality-adjusted survival than radiotherapy alone in the long term.
    International journal of radiation oncology, biology, physics 10/2010; 78(2):454-60. DOI:10.1016/j.ijrobp.2009.07.1702 · 4.59 Impact Factor