Ordinal responses are common in clinical studies. Although the proportional odds model is a popular option for analyzing ordered-categorical data, it cannot control the type I error rate when the proportional odds assumption fails to hold. The latent Weibull model was recently shown to be a superior candidate for modeling ordinal data, with remarkably better performance than the latent normal model when the data are highly skewed. In clinical trials with ordinal responses, a balanced design is common, with equal sample allocation for each treatment. However, a more ethical approach is to adopt a response-adaptive allocation scheme in which more patients receive the better treatment. In this paper, we propose the use of the doubly adaptive biased coin design to generate treatment allocations that benefit the trial participants. The proposed treatment allocation scheme not only allows more patients to receive the better treatment, it also maintains compatible test power for the comparison of treatment efficiencies. A clinical example is used to illustrate the proposed procedure.
YoudomSWSamsonABascoLK, et al.Multiple treatment comparisons in a series of anti-malarial trials with an ordinal primary outcome and repeated treatment evaluations. Malaria J2012; 11: 147.
2.
World Health Organization. Assessment and monitoring of antimalarial drug efficacy for the treatment of uncomplicated falciparum malaria. Geneva: World Health Organization, 2003.
3.
RossomRCSolbergLIVazquez-BenitezG, et al.Predictors of poor response to depression treatment in primary care. Psychiatr Serv2016; 67: 1362–1367.
4.
WhiteheadJ. Sample size calculation for ordered categorical data. Stat Med1993; 12: 2257–2271.
5.
McCullaghP. Regression models for ordinal data. J R Statist Soc B1980; 42: 109–142.
6.
LuTYPoonWYCheungSH. Multiple comparisons with a control for a latent variable model with ordered categorical responses. Stat Methods Med Res2015; 24: 949–967.
7.
BekeleBNThallPF. Dose-finding based on multiple toxicities in a soft tissue sarcoma trial. J Am Stat Assoc2004; 99: 26–35.
8.
Leon-NoveloLGZhouXBekeleBN, et al.Assessing toxicities in a clinical trial: Bayesian inference for ordinal data nested within categories. Biometrics2010; 66: 966–974.
9.
PoonWYWangHB. Bayesian analysis of multivariate probit models with surrogate outcome data. Psychometrika2010; 75: 498–520.
10.
PoonWY. A latent normal distribution model for analysing ordinal responses with applications in meta-analysis. Stat Med2004; 23: 2155–2172.
11.
LinYCheungSHPoonWY, et al.Pairwise comparisons with ordered categorical data. Stat Med2013; 32: 3192–3205.
12.
LinYKwongKSCheungSH, et al.Step-up testing procedure for multiple comparisons with a control for a latent variable model with ordered categorical responses. Stat Med2014; 33: 3629–3638.
13.
LuTYPoonWYCheungSH. Comparison of two treatments with skewed ordinal responses. Stat Med2016; 35: 189–201.
14.
LuTYPoonWYCheungSH. Multiple comparison of treatments with skewed ordinal responses. Comput Stat Data Anal2016; 104: 223–232.
15.
HonarmandASafaviM. Magnesium sulphate pretreatment to alleviate pain on propofol injection: a comparison with ketamine or lidocaine. Acute Pain2008; 10: 23–29.
16.
L'HermiteJDuboutEBouvetS, et al.Sore throat following three adult supraglottic airway devices: a randomised controlled trial. J Anaesthesiol2017; 7: 417–424.
17.
HuFRosenbergerWF. The theory of response-adaptive randomization in clinical trials, New Jersey: Wiley, 2006.
18.
MontgomeryCM. From standardization to adaptation: clinical trials and the moral economy of anticipation. Sci Cult2017; 2: 232–254.
19.
HuFZhangL. Asymptotic properties of doubly adaptive biased coin designs for multitreatment clinical trials. Ann Stat2004; 32: 268–301.
20.
LemeACHajjarLAVolpeMS. Effect of intensive vs moderate alveolar recruitment strategies added to lung-protective ventilation on postoperative pulmonary complications: a randomized clinical trial. J Am Med Assoc2017; 14: 1422–1432.
21.
AndersonJAPhilipsPR. Regression, discrimination and measurement models for ordered categorical variables. Appl Stat1981; 30: 22–31.
22.
LuTYPoonWYCheungSH. A unified framework for the comparison of treatments with ordinal responses. Psychometrika2014; 79: 605–520.
23.
LawlessJF. Statistical models and methods for lifetime data., 2nd ed. Wiley: New Jersey, 2003.
24.
VeraJOchoaJRomeroM, et al.Intracanal cryotherapy reduces postoperative pain in teeth with symptomatic apical periodontitis: a randomized multicenter clinical trial. J Endod2018; 44: 4–8.
BerrySMPetzoldEADullP, et al.A response adaptive randomization platform trial for efficient evaluation of Ebola virus treatments: a model for pandemic response. Clin Trials2016; 13: 22–30.
27.
DasSLoAW. Re-inventing drug development: a case study of the I-SPY 2 breast cancer clinical trials program. Contemp Clin Trials2017; 62: 168–174.
28.
EiseleJR. The doubly adaptive biased coin design for sequential clinical trials. J Stat Plan Inference1994; 38: 249–261.
29.
HuFRosenbergerWF. Optimality, variability, power: evaluating response-adaptive randomization procedures for treatment comparisons. J Am Stat Assoc2003; 98: 671–678.
30.
HuFZhangLHeX. Efficient randomized-adaptive designs. Ann Stat2009; 37: 2543–2560.
31.
RosenbergerWFLachinJM. Randomization in clinical trials: theory and practice, New York: Wiley, 2002.
32.
ZhangLRosenbergerWF. Response-adaptive randomization for clinical trials with continuous outcomes. Biometrics2006; 62: 562–596.
33.
XuWGaoJHuF, et al.Response-adaptive treatment allocation for non-inferiority trials with heterogeneous variances. Comput Stat Data Anal2018; 124: 168–179.
