Statistical comparison of random allocation methods in cancer clinical trials

Comparison between asymptotic and re-randomisation tests under non-proportional hazards in a randomised controlled trial using the minimisation method

BACKGROUND

Pocock-Simon's minimisation method has been widely used to balance treatment assignments across prognostic factors in randomised controlled trials (RCTs). Previous studies focusing on the survival outcomes have demonstrated that the conservativeness of asymptotic tests without adjusting for stratification factors, as well as the inflated type I error rate of adjusted asymptotic tests conducted in a small sample of patients, can be relaxed using re-randomisation tests. Although several RCTs using minimisation have suggested the presence of non-proportional hazards (non-PH) effects, the application of re-randomisation tests has been limited to the log-rank test and Cox PH models, which may result in diminished statistical power when confronted with non-PH scenarios. To address this issue, we proposed two re-randomisation tests based on a maximum combination of weighted log-rank tests (MaxCombo test) and the difference in restricted mean survival time (dRMST) up to a fixed time point τ , both of which can be extended to adjust for randomisation stratification factors.

METHODS

We compared the performance of asymptotic and re-randomisation tests using the MaxCombo test, dRMST, log-rank test, and Cox PH models, assuming various non-PH situations for RCTs using minimisation, with total sample sizes of 50, 100, and 500 at a 1:1 allocation ratio. We mainly considered null, and alternative scenarios featuring delayed, crossing, and diminishing treatment effects.

RESULTS

Across all examined null scenarios, re-randomisation tests maintained the type I error rates at the nominal level. Conversely, unadjusted asymptotic tests indicated excessive conservatism, while adjusted asymptotic tests in both the Cox PH models and dRMST indicated inflated type I error rates for total sample sizes of 50. The stratified MaxCombo-based re-randomisation test consistently exhibited robust power across all examined scenarios.

CONCLUSIONS

The re-randomisation test is a useful alternative in non-PH situations for RCTs with minimisation using the stratified MaxCombo test, suggesting its robust power in various scenarios.

Minimization in randomized clinical trials

In randomized trials, comparability of the treatment groups is ensured through allocation of treatments using a mechanism that involves some random element, thus controlling for confounding of the treatment effect. Completely random allocation ensures comparability between the treatment groups for all known and unknown prognostic factors. For a specific trial, however, imbalances in prognostic factors among the treatment groups may occur. Although accidental bias can be avoided in the presence of such imbalances by stratifying the analysis, most trialists, regulatory agencies, and other stakeholders prefer a balanced distribution of prognostic factors across the treatment groups. Some procedures attempt to achieve balance in baseline covariates, by stratifying the allocation for these covariates, or by dynamically adapting the allocation using covariate information during the trial (covariate-adaptive procedures). In this Tutorial, the performance of minimization, a popular covariate-adaptive procedure, is compared with two other commonly used procedures, completely random allocation and stratified blocked designs. Using individual patient data of 2 clinical trials (in advanced ovarian cancer and age-related macular degeneration), the procedures are compared in terms of operating characteristics (using asymptotic and randomization tests), predictability of treatment allocation, and achieved balance. Fifty actual trials of various sizes that applied minimization for treatment allocation are used to investigate the achieved balance. Implementation issues of minimization are described. Minimization procedures are useful in all trials but especially when (1) many major prognostic factors are known, (2) many centers of different sizes accrue patients, or (3) the trial sample size is moderate.

Impact of minimal sufficient balance, minimization, and stratified permuted blocks on bias and power in the estimation of treatment effect in sequential clinical trials with a binary endpoint

Minimization is among the most common methods for controlling baseline covariate imbalance at the randomization phase of clinical trials. Previous studies have found that minimization does not preserve allocation randomness as well as other methods, such as minimal sufficient balance, making it more vulnerable to allocation predictability and selection bias. Additionally, minimization has been shown in simulation studies to inadequately control serious covariate imbalances when modest biased coin probabilities (≤0.65) are used. This current study extends the investigation of randomization methods to the analysis phase, comparing the impact of treatment allocation methods on power and bias in estimating treatment effects on a binary outcome using logistic regression. Power and bias in the estimation of treatment effect was found to be comparable across complete randomization, minimization, and minimal sufficient balance in unadjusted analyses. Further, minimal sufficient balance was found to have the most modest impact on power and the least bias in covariate-adjusted analyses. The minimal sufficient balance method is recommended for use in clinical trials as an alternative to minimization when covariate-adaptive subject randomization takes place.

Acupuncture for cancer pain: protocol for a pilot pragmatic randomised controlled trial

INTRODUCTION

Although acupuncture has been recommended for alleviating cancer pain by clinical guidelines, the level of the supporting evidence needs to be improved. A pragmatic randomised controlled trial (pRCT) in a hospital setting would provide real-world assessments of the overall clinical effects of acupuncture. This pilot trial aims to explore the feasibility and provide data for sample size calculations for a pRCT evaluating the effectiveness of acupuncture as an adjunctive therapy to routine medical care for cancer pain.

METHODS AND ANALYSIS

Thirty patients with cancer admitted to the oncology department with moderate or severe pain will be recruited. Participants will be randomised at a ratio of 1:1 to the adjunctive acupuncture group or a control group which receives routine pain management without acupuncture. The standardised section of the acupuncture protocol will be developed based on the results of reviews of the literature, recommendations in clinical guidelines and interviews with clinical experts. The acupuncturist will be allowed to tailor the protocol according to the individual situation of each participant. Primary outcomes relevant to the feasibility of conducting a fully powered trial include: numbers and proportions of participants recruited, screened, consented and randomised; numbers and reasons for withdrawals and dropouts; numbers and types of adverse events; feasibility of implementing the trial procedures; evaluation of the comprehensiveness and ease-of-use of the case report form. Secondary outcomes are clinical measurements of the effectiveness of the treatment that are intended for use in the full-scale trial. Analysis of feasibility will be descriptive and pain intensity measures will be analysed using mixed-effects regression.

ETHICS AND DISSEMINATION

Ethics approval was obtained from the Institutional Ethics Committee of Guangdong Provincial Hospital of Chinese Medicine (approval no: Z2017-184-01) and RMIT University Human Research Ethics Committee (reference no: 21361). Results will be disseminated in a peer-reviewed journal, and trial participants will be informed via email and/or phone calls.

TRIAL REGISTRATION NUMBER

ChiCTR1800017023.

Testing hypotheses under adaptive randomization with continuous covariates in clinical trials

Covariate-adaptive designs are widely used to balance covariates and maintain randomization in clinical trials. Adaptive designs for discrete covariates and their asymptotic properties have been well studied in the literature. However, important continuous covariates are often involved in clinical studies. Simply discretizing or categorizing continuous covariates can result in loss of information. The current understanding of adaptive designs with continuous covariates lacks a theoretical foundation as the existing works are entirely based on simulations. Consequently, conventional hypothesis testing in clinical trials using continuous covariates is still not well understood. In this paper, we establish a theoretical framework for hypothesis testing on adaptive designs with continuous covariates based on linear models. For testing treatment effects and significance of covariates, we obtain the asymptotic distributions of the test statistic under null and alternative hypotheses. Simulation studies are conducted under a class of covariate-adaptive designs, including the p-value-based method, the Su’s percentile method, the empirical cumulative-distribution method, the Kullback–Leibler divergence method, and the kernel-density method. Key findings about adaptive designs with independent covariates based on linear models are (1) hypothesis testing that compares treatment effects are conservative in terms of smaller type I error, (2) hypothesis testing using adaptive designs outperforms complete randomization method in terms of power, and (3) testing on significance of covariates is still valid.

Internet-based treatment for panic disorder: A three-arm randomized controlled trial comparing guided (via real-time video sessions) with unguided self-help treatment and a waitlist control. PAXPD study results

A growing body of evidence suggests that Internet-based cognitive behavioral treatments (ICBT) are effective to treat anxiety disorders. However, the effect of therapist guidance in ICBT is still under debate and guided ICBT offered in a real-time audio-video communication format has not yet been systematically investigated. This three-arm RCT compared the efficacy of guided with unguided ICBT (12 weeks intervention) and a waitlist (WL). A total of 111 individuals meeting the diagnostic criteria for panic disorder (PD) were randomly assigned to one of three conditions. Primary outcomes were the severity of self-report panic symptoms and diagnostic status. Secondary outcomes were symptoms of depression, functional impairment, catastrophic cognitions, fear of sensations and body vigilance. At post-treatment, both active conditions showed superior outcomes regarding PD and associated symptoms (guided ICBT vs. WL: d = 1.04-1.36; unguided ICBT vs. WL: d = 0.70-1.06). At post-treatment, the two active conditions did not differ significantly in self-reported symptom reduction (d = 0.21-0.54, all ps > 0.05), but the guided treatment was superior to the unguided treatment in terms of diagnostic status (χ² (1) = 13.15, p < 0.01). Treatment gains were maintained at successive follow-ups and the guided treatment became superior to the unguided treatment at 6 months follow-up (d = 0.72-1.05, all ps < 0.05).

Validity and power of minimization algorithm in longitudinal analysis of clinical trials

We studied the validity of longitudinal statistical inferences of clinical trials using minimization, a dynamic randomization algorithm designed to minimize treatment imbalance for prognostic factors. Repeated measures analysis of covariance and the random intercept and slope models, were used to simulate longitudinal clinical trials randomized by minimization or simple randomization. The simulations represented a wide range of analyses in real-world trials, including missing data caused by dropouts, unequal allocation of treatment arms, and efficacy analyses on either the original outcome or its change from baseline. We also analyzed the database from the Dominantly Inherited Alzheimer Network (DIAN), and used the estimated parameters to simulate the ongoing DIAN trial. Our analyses demonstrated minimization had conservative type I errors when the prognostic factor used in the minimization algorithm had a relatively strong correlation with the outcome and was not adjusted for in analyses. In contrast, adjusted tests for the prognostic factor as a covariate resulted in type I errors close to the nominal significance level. In many simulation scenarios, the adjusted tests using minimization had slightly greater statistical power than those using simple randomization, whereas in the other scenarios, the power of adjusted tests using these two randomization methods are almost indistinguishable.

Internet-based treatment for Romanian adults with panic disorder: protocol of a randomized controlled trial comparing a Skype-guided with an unguided self-help intervention (the PAXPD study)

BACKGROUND

Efficacy of self-help internet-based cognitive behavior therapy (ICBT) for anxiety disorders has been confirmed in several randomized controlled trials. However, the amount and type of therapist guidance needed in ICBT are still under debate. Previous studies have shown divergent results regarding the role of therapist guidance and its impact on treatment outcome. This issue is central to the development of ICBT programs and needs to be addressed directly. The present study aims to compare the benefits of regular therapist guidance via online real-time audio-video communication (i.e. Skype) to no therapist guidance during a 12-week Romanian self-help ICBT program for Panic Disorder. Both treatments are compared to a waiting-list control group.

METHODS/DESIGN

A parallel group randomized controlled trial is proposed. The participants, 192 Romanian adults fulfilling diagnostic criteria for panic disorder according to a diagnostic interview, conducted via secured Skype or telephone, are randomly assigned to one of the three conditions: independent use of the internet-based self-help program PAXonline, the same self-help treatment with regular therapist support via secured Skype, and waiting-list control group. The primary outcomes are severity of self-report panic symptoms (PDSS-SR) and diagnostic status (assessors are blind to group assignment), at the end of the intervention (12 weeks) and at follow-up (months 3 and 6). The secondary measures address symptoms of comorbid anxiety disorders, depression, quality of life, adherence and satisfaction with ICBT. Additional measures of socio-demographic characteristics, personality traits, treatment expectancies, catastrophic cognitions, body vigilance and working alliance are considered as potential moderators and/ or mediators of treatment outcome.

DISCUSSION

To the best of our knowledge, the present study is the first effort to investigate the efficacy of a self-help internet-based intervention with therapist guidance via real-time video communication. A direct comparison between therapist guided versus unguided self-directed intervention for panic disorder will also be addressed for the first time. Findings from this study will inform researchers and practitioners about the added value of online video-therapy guidance sessions and the type of patients who may benefit the most from guided and unguided ICBT for Panic disorder.

TRIAL REGISTRATION

ACTRN12614000547640 (Australian New Zealand Clinical Trials Registry). Registered 22/05/2014.

Creating groups with similar expected behavioural response in randomized controlled trials: a fuzzy cognitive map approach

BACKGROUND

Controlling bias is key to successful randomized controlled trials for behaviour change. Bias can be generated at multiple points during a study, for example, when participants are allocated to different groups. Several methods of allocations exist to randomly distribute participants over the groups such that their prognostic factors (e.g., socio-demographic variables) are similar, in an effort to keep participants' outcomes comparable at baseline. Since it is challenging to create such groups when all prognostic factors are taken together, these factors are often balanced in isolation or only the ones deemed most relevant are balanced. However, the complex interactions among prognostic factors may lead to a poor estimate of behaviour, causing unbalanced groups at baseline, which may introduce accidental bias.

METHODS

We present a novel computational approach for allocating participants to different groups. Our approach automatically uses participants' experiences to model (the interactions among) their prognostic factors and infer how their behaviour is expected to change under a given intervention. Participants are then allocated based on their inferred behaviour rather than on selected prognostic factors.

RESULTS

In order to assess the potential of our approach, we collected two datasets regarding the behaviour of participants (n = 430 and n = 187). The potential of the approach on larger sample sizes was examined using synthetic data. All three datasets highlighted that our approach could lead to groups with similar expected behavioural changes.

CONCLUSIONS

The computational approach proposed here can complement existing statistical approaches when behaviours involve numerous complex relationships, and quantitative data is not readily available to model these relationships. The software implementing our approach and commonly used alternatives is provided at no charge to assist practitioners in the design of their own studies and to compare participants' allocations.

Bias, precision and statistical power of analysis of covariance in the analysis of randomized trials with baseline imbalance: a simulation study

Background

Analysis of variance (ANOVA), change-score analysis (CSA) and analysis of covariance (ANCOVA) respond differently to baseline imbalance in randomized controlled trials. However, no empirical studies appear to have quantified the differential bias and precision of estimates derived from these methods of analysis, and their relative statistical power, in relation to combinations of levels of key trial characteristics. This simulation study therefore examined the relative bias, precision and statistical power of these three analyses using simulated trial data.

Methods

126 hypothetical trial scenarios were evaluated (126 000 datasets), each with continuous data simulated by using a combination of levels of: treatment effect; pretest-posttest correlation; direction and magnitude of baseline imbalance. The bias, precision and power of each method of analysis were calculated for each scenario.

Results

Compared to the unbiased estimates produced by ANCOVA, both ANOVA and CSA are subject to bias, in relation to pretest-posttest correlation and the direction of baseline imbalance. Additionally, ANOVA and CSA are less precise than ANCOVA, especially when pretest-posttest correlation ≥ 0.3. When groups are balanced at baseline, ANCOVA is at least as powerful as the other analyses. Apparently greater power of ANOVA and CSA at certain imbalances is achieved in respect of a biased treatment effect.

Conclusions

Across a range of correlations between pre- and post-treatment scores and at varying levels and direction of baseline imbalance, ANCOVA remains the optimum statistical method for the analysis of continuous outcomes in RCTs, in terms of bias, precision and statistical power.

MACT: a manageable minimization allocation system

Background. Minimization is a case allocation method for randomized controlled trials (RCT). Evidence suggests that the minimization method achieves balanced groups with respect to numbers and participant characteristics, and can incorporate more prognostic factors compared to other randomization methods. Although several automatic allocation systems exist (e.g., randoWeb, and MagMin), the minimization method is still difficult to implement, and RCTs seldom employ minimization. Therefore, we developed the minimization allocation controlled trials (MACT) system, a generic manageable minimization allocation system. System Outline. The MACT system implements minimization allocation by Web and email. It has a unified interface that manages trials, participants, and allocation. It simultaneously supports multitrials, multicenters, multigrouping, multiple prognostic factors, and multilevels. Methods. Unlike previous systems, MACT utilizes an optimized database that greatly improves manageability. Simulations and Results. MACT was assessed in a series of experiments and evaluations. Relative to simple randomization, minimization produces better balance among groups and similar unpredictability. Applications. MACT has been employed in two RCTs that lasted three years. During this period, MACT steadily and simultaneously satisfied the requirements of the trial. Conclusions. MACT is a manageable, easy-to-use case allocation system. Its outstanding features are attracting more RCTs to use the minimization allocation method.

An easily accessible Web-based minimization random allocation system for clinical trials

Background

Minimization as an adaptive allocation technique has been recommended in the literature for use in randomized clinical trials. However, it remains uncommonly used due in part to a lack of easily accessible implementation tools.

Objective

To provide clinical trialists with a robust, flexible, and readily accessible tool for implementing covariate-adaptive biased-coin randomization.

Methods

We developed a Web-based random allocation system, MinimRan, that applies Pocock–Simon (for trials with 2 or more arms) and 2-way (currently limited to 2-arm trials) minimization methods for trials using only categorical prognostic factors or the symmetric Kullback–Leibler divergence minimization method for trials (currently limited to 2-arm trials) using continuous prognostic factors with or without categorical factors, in covariate-adaptive biased-coin randomization.

Results

In this paper, we describe the system’s essential statistical and computer programming features and provide as an example the randomization results generated by it in a recently completed trial. The system can be used in single- and double-blind trials as well as single-center and multicenter trials.

Conclusions

We expect the system to facilitate the translation of the 3 validated random allocation methods into broad, efficient clinical research practice.

Algorithm for balancing both continuous and categorical covariates in randomized controlled trials

Minimization as proposed by Pocock and Simon for balancing categorical covariates in clinical trials with individual-level, consecutive randomization has been increasingly used. An extension of the method exists that uses the symmetric Kullback-Leibler divergence index to balance both categorical and continuous covariates, albeit for two-arm randomized controlled trials only. To date, the algorithm has not been made widely available to researchers via publicly accessible software. We adapted Endo et al.'s algorithm to randomized trials with two or more arms. In addition, our algorithm incorporates Efron's biased coin method to decrease the predictability of assignment even when a predefined threshold of difference in the numbers of subjects between treatment arms is reached, whereas Endo et al.'s algorithm assigns the next subject to the treatment of smaller size with certainty. We developed code in the free statistical software R to make the algorithm readily available to trialists. While there are no definitive answers regarding the optimal choices for certain statistical parameters that must be defined prior to algorithm application (P(k), D(n), and p_D(n)), we provide guidance on these. We conducted simulations with actual data from a three-arm randomized trial to compare the modified algorithm and R code to another published biased coin minimization method that can accommodate continuous and categorical covariates in multi-arm trials. The three-arm trial used three categorical covariates (sex, race/ethnicity, and online personal health record access) and four continuous covariates (age, fasting blood glucose, body mass index, and waist circumference). All of the continuous and categorical covariates were well balanced, and the results were comparable to the comparison method.

Generalized multidimensional dynamic allocation method

Dynamic allocation has received considerable attention since it was first proposed in the 1970s as an alternative means of allocating treatments in clinical trials which helps to secure the balance of prognostic factors across treatment groups. The purpose of this paper is to present a generalized multidimensional dynamic allocation method that simultaneously balances treatment assignments at three key levels: within the overall study, within each level of each prognostic factor, and within each stratum, that is, combination of levels of different factors Further it offers capabilities for unbalanced and adaptive designs for trials. The treatment balancing performance of the proposed method is investigated through simulations which compare multidimensional dynamic allocation with traditional stratified block randomization and the Pocock-Simon method. On the basis of these results, we conclude that this generalized multidimensional dynamic allocation method is an improvement over conventional dynamic allocation methods and is flexible enough to be applied for most trial settings including Phases I, II and III trials.

Balancing Treatment Allocation over Continuous Covariates: A New Imbalance Measure for Minimization

In many clinical trials, it is important to balance treatment allocation over covariates. Although a great many papers have been published on balancing over discrete covariates, the procedures for continuous covariates have been less well studied. Traditionally, a continuous covariate usually needs to be transformed to a discrete one by splitting its range into several categories. Such practice may lead to loss of information and is susceptible to misspecification of covariate distribution. The more recent papers seek to define an imbalance measure that preserves the nature of continuous covariates and set the allocation rule in order to minimize that measure. We propose a new design, which defines the imbalance measure by the maximum assignment difference when all possible divisions of the covariate range are considered. This measure depends only on ranks of the covariate values and is therefore free of covariate distribution. In addition, we developed an efficient algorithm to implement the new procedure. By simulation studies we show that the new procedure is able to keep good balance properties in comparison with other popular designs.

Intrapleural use of tissue plasminogen activator and DNase in pleural infection

BACKGROUND

More than 30% of patients with pleural infection either die or require surgery. Drainage of infected fluid is key to successful treatment, but intrapleural fibrinolytic therapy did not improve outcomes in an earlier, large, randomized trial.

METHODS

We conducted a blinded, 2-by-2 factorial trial in which 210 patients with pleural infection were randomly assigned to receive one of four study treatments for 3 days: double placebo, intrapleural tissue plasminogen activator (t-PA) and DNase, t-PA and placebo, or DNase and placebo. The primary outcome was the change in pleural opacity, measured as the percentage of the hemithorax occupied by effusion, on chest radiography on day 7 as compared with day 1. Secondary outcomes included referral for surgery, duration of hospital stay, and adverse events.

RESULTS

The mean (±SD) change in pleural opacity was greater in the t-PA-DNase group than in the placebo group (-29.5±23.3% vs. -17.2±19.6%; difference, -7.9%; 95% confidence interval [CI], -13.4 to -2.4; P=0.005); the change observed with t-PA alone and with DNase alone (-17.2±24.3 and -14.7±16.4%, respectively) was not significantly different from that observed with placebo. The frequency of surgical referral at 3 months was lower in the t-PA-DNase group than in the placebo group (2 of 48 patients [4%] vs. 8 of 51 patients [16%]; odds ratio for surgical referral, 0.17; 95% CI, 0.03 to 0.87; P=0.03) but was greater in the DNase group (18 of 46 patients [39%]) than in the placebo group (odds ratio, 3.56; 95% CI, 1.30 to 9.75; P=0.01). Combined t-PA-DNase therapy was associated with a reduction in the hospital stay, as compared with placebo (difference, -6.7 days; 95% CI, -12.0 to -1.9; P=0.006); the hospital stay with either agent alone was not significantly different from that with placebo. The frequency of adverse events did not differ significantly among the groups.

CONCLUSIONS

Intrapleural t-PA-DNase therapy improved fluid drainage in patients with pleural infection and reduced the frequency of surgical referral and the duration of the hospital stay. Treatment with DNase alone or t-PA alone was ineffective. (Funded by an unrestricted educational grant to the University of Oxford from Roche UK and by others; Current Controlled Trials number, ISRCTN57454527.).

Comparison of dynamic block randomization and minimization in randomized trials: a simulation study

BACKGROUND

Minimizing the imbalance of key baseline covariates between treatments is known to be very important to the precision of the estimate of treatment effect in clinical research. Dynamic randomization allocation techniques have been used to achieve balance across multiple baseline characteristics. However, empirical data are limited on how these techniques compare in terms of balance and efficiency. We are motivated by a newly funded randomized controlled trial, in which we have the option of choosing between two methods of randomization at the subject level: (1) randomizing individual subjects consecutively as they are enrolled, using Pocock and Simon's minimization method, and (2) simultaneously randomizing blocks of subjects once all subjects in a block have been enrolled, using a balance algorithm originally developed for cluster randomized trials.

PURPOSE

To compare dynamic block randomization and minimization in terms of balance on baseline covariates and statistical efficiency. Simple randomization was included as a reference.

METHODS

A simulation study using data from a previous randomized controlled trial was conducted to compare balance statistics and the accuracy and power of hypothesis testing among the randomization methods.

RESULTS

Dynamic block randomization consistently produced the best balance and highest power for various sample and treatment effect sizes, even after post-adjustment of the pre-specified baseline covariates in all three methods. Consistent with previous reports, minimization performed better in balance and power than simple randomization; however, the differences were noticeably smaller compared to those between dynamic block randomization and simple randomization.

LIMITATIONS

In this simulation study, we considered three sample sizes and two block sizes for a two-arm randomized trial. We assumed no interactions among the multiple baseline covariates. It is necessary to evaluate how the results may vary when the simulation conditions are changed before drawing broader conclusions regarding comparisons between the randomization methods.

CONCLUSIONS

This study demonstrates that dynamic block randomization outperforms minimization with regard to achieving balance and maximizing efficiency. Nevertheless, the differences across the three randomization strategies are modest. The statistical advantages associated with dynamic block randomization need to be considered in relation to the planned sample size and the practical issues for its implementation in deciding the preferred method of randomization for a given trial (e.g., the time required to accrue blocks of subjects of adequate size as balanced against the need to commence intervention/treatment immediately in those randomized to that experimental condition).

Randomization by minimization for unbalanced treatment allocation

Minimization is a dynamic randomization technique that has been widely used in clinical trials for achieving a balance of prognostic factors across treatment groups, but most often it has been used in the setting of equal treatment allocations. Although unequal treatment allocation is frequently encountered in clinical trials, an appropriate minimization procedure for such trials has not been published. The purpose of this paper is to present novel strategies for applying minimization methodology to such clinical trials. Two minimization techniques are proposed and compared by probability calculation and simulation studies. In the first method, called naïve minimization, probability assignment is based on a simple modification of the original minimization algorithm, which does not account for unequal allocation ratios. In the second method, called biased-coin minimization (BCM), probability assignment is based on allocation ratios and optimized to achieve an 'unbiased' target allocation ratio. The performance of the two methods is investigated in various trial settings including different number of treatments, prognostic factors and sample sizes. The relative merits of the different distance metrics are also explored. On the basis of the results, we conclude that BCM is the preferable method for randomization in clinical trials involving unequal treatment allocations. The choice of different distance metrics slightly affects the performance of the minimization and may be optimized according to the specific feature of trials.

Development of a minimization instrument for allocation of a hospital-level performance improvement intervention to reduce waiting times in Ontario emergency departments

BACKGROUND

Rigorous evaluation of an intervention requires that its allocation be unbiased with respect to confounders; this is especially difficult in complex, system-wide healthcare interventions. We developed a short survey instrument to identify factors for a minimization algorithm for the allocation of a hospital-level intervention to reduce emergency department (ED) waiting times in Ontario, Canada.

METHODS

Potential confounders influencing the intervention's success were identified by literature review, and grouped by healthcare setting specific change stages. An international multi-disciplinary (clinical, administrative, decision maker, management) panel evaluated these factors in a two-stage modified-delphi and nominal group process based on four domains: change readiness, evidence base, face validity, and clarity of definition.

RESULTS

An original set of 33 factors were identified from the literature. The panel reduced the list to 12 in the first round survey. In the second survey, experts scored each factor according to the four domains; summary scores and consensus discussion resulted in the final selection and measurement of four hospital-level factors to be used in the minimization algorithm: improved patient flow as a hospital's leadership priority; physicians' receptiveness to organizational change; efficiency of bed management; and physician incentives supporting the change goal.

CONCLUSION

We developed a simple tool designed to gather data from senior hospital administrators on factors likely to affect the success of a hospital patient flow improvement intervention. A minimization algorithm will ensure balanced allocation of the intervention with respect to these factors in study hospitals.

Optimal design of clinical trials with computer simulation based on results of earlier trials, illustrated with a lipodystrophy trial in HIV patients

The designer of a clinical trial needs to make many assumptions about real-life practice based on prior knowledge. Simulation allows us to learn from experience by using the information obtained from a trial to improve the original estimators of population parameters. We propose using data from a previous trial to formulate assumptions that can be used to simulate trials and thus improve the design of new trials. To demonstrate our method, we used data from a real clinical trial which had been designed to evaluate cholesterol level changes as a surrogate marker for lipodystrophy in HIV patients. We were able to identify the optimal design that would have minimised the cost of a trial subject to a statistical power constraint which could then be used to design a new trial. In particular, we focused on three factors: the distribution of cholesterol levels in HIV patients, trial recruitment rates and trial dropout rates. We were able to verify our hypothesis that the total cost resulting from carrying out a clinical trial can be minimised by applying simulation models as an alternative to conventional approaches. In our findings the simulation model proved to be very intuitive and a useful method for testing the performance of investigators' assumptions and generating an optimal clinical trial design before being put into practice in the real world. In addition, we concluded that simulation models provide a more accurate determination of power than conventional approaches, thus minimising the total cost of clinical trials.

Randomization with a posteriori constraints: Description and properties

The use of randomization for assigning patients to treatment groups in clinical trials is firmly acknowledged as providing the best quality results. Two standard methods are used in order to achieve well-balanced groups with respect to prognostic factors (i.e. factors influencing the disease outcome): stratification and minimization. Stratification is recommended when the number of strata is not too high--otherwise, minimization is preferred. However, minimization may compromise blinding (since the search for balance is performed a priori) and, furthermore, use of the technique has been questioned by the European Agency for the Evaluation of Medicinal Products. We have developed a new procedure for adaptive randomization, which we have named 'randomization with a posteriori constraints'. By using a search for balance a posteriori, this procedure ensures that patient groups are similar with respect to prognostic factors while being less vulnerable to selection bias. The aim of this work was to describe the new method and to compare it (using simulations) with stratification and minimization. In the case of trials with few prognostic factors, the recourse to minimization or 'randomization with a posteriori constraints' does not appear to be useful. In such a context, stratification has suitable properties and its simplicity of implementation encourages its use. However, when the number of prognostic factors is higher, 'randomization with a posteriori constraints' is less predictable than minimization and the chance of imbalance is lower than for stratification. In conclusion, 'randomization with a posteriori constraints' with an adequate threshold seems to be a good compromise between minimization and stratification.

A generic minimization random allocation and blinding system on web

BACKGROUND

Minimization is a dynamic randomization method for clinical trials. Although recommended by many researchers, the utilization of minimization has been seldom reported in randomized trials mainly because of the controversy surrounding the validity of conventional analyses and its complexity in implementation. However, both the statistical and clinical validity of minimization were demonstrated in recent studies. Minimization random allocation system integrated with blinding function that could facilitate the implementation of this method in general clinical trials has not been reported. SYSTEM OVERVIEW: The system is a web-based random allocation system using Pocock and Simon minimization method. It also supports multiple treatment arms within a trial, multiple simultaneous trials, and blinding without further programming.

METHODS

This system was constructed with generic database schema design method, Pocock and Simon minimization method and blinding method. It was coded with Microsoft Visual Basic and Active Server Pages (ASP) programming languages. And all dataset were managed with a Microsoft SQL Server database. Some critical programming codes were also provided. SIMULATIONS AND RESULTS: Two clinical trials were simulated simultaneously to test the system's applicability. Not only balanced groups but also blinded allocation results were achieved in both trials.

DISCUSSIONS AND CONCLUSIONS

Practical considerations for minimization method, the benefits, general applicability and drawbacks of the technique implemented in this system are discussed. Promising features of the proposed system are also summarized.

Defining the effect and mediators of two knowledge translation strategies designed to alter knowledge, intent and clinical utilization of rehabilitation outcome measures: a study protocol [NCT00298727]

BACKGROUND

A substantial number of valid outcome measures have been developed to measure health in adult musculoskeletal and childhood disability. Regrettably, national initiatives have merely resulted in changes in attitude, while utilization remains unacceptably low. This study will compare the effectiveness and mediators of two different knowledge transfer (KT) interventions in terms of their impact on changing knowledge and behavior (utilization and clinical reasoning) related to health outcome measures.

METHOD/DESIGN

Physical and occupational therapists (n = 144) will be recruited in partnership with the national professional associations to evaluate two different KT interventions with the same curriculum: 1) Stakeholder-Hosted Interactive Problem-Based Seminar (SHIPS), and 2) Online Problem-Based course (e-PBL). SHIPS will consist of face-to-face problem-based learning (PBL) for 2 1/2 days with outcome measure developers as facilitators, using six problems generated in consultation with participants. The e-PBL will consist of a 6-week web-based course with six generic problems developed by content experts. SHIPS will be conducted in three urban centers in Canada. Participants will be block-allocated by a minimization procedure to either of the two interventions to minimize any prognostic differences. Trained evaluators at each site will conduct chart audits and chart-stimulated recall. Trained interviewers will conduct semi-structured interviews focused on identifying critical elements in KT and implementing practice changes. Interviews will be transcribed verbatim. Baseline predictors including demographics, knowledge, attitudes/barriers regarding outcome measures, and Readiness to Change will be assessed by self-report. Immediately post-intervention and 6 months later, these will be re-administered. Primary qualitative and quantitative evaluations will be conducted 6-months post-intervention to assess the relative effectiveness of KT interventions and to identify elements that contribute to changing clinical behavior. Chart audits will determine the utilization of outcome measures (counts). Incorporation of outcome measures into clinical reasoning will be assessed using an innovative technique: chart-stimulated recall.

DISCUSSION

A strategy for optimal transfer of health outcome measures into practice will be developed and shared with multiple disciplines involved in primary and specialty management of musculoskeletal and childhood disability.

Acute normovolemic hemodilution in moderate blood loss surgery: a randomized controlled trial

BACKGROUND

The risks associated with allogeneic blood transfusion are increasingly recognized. More blood is cross-matched for moderate blood loss surgery than any other indication. The role of acute normovolemic hemodilution (ANH) as a blood transfusion strategy was evaluated in a prospective randomized controlled trial.

STUDY DESIGN AND METHODS

A total of 155 patients undergoing elective hip surgery were randomly assigned to either "ANH" (n = 78) or "standard transfusion" (n = 77). ANH on induction of anesthesia was to a target hemoglobin (Hb) level of 110 g per L with return of autologous blood on wound closure. Allogeneic blood was prescribed by an objective transfusion trigger based on an Hb level of less than 80 g per L. Transfusion requirements and postoperative complications were recorded.

RESULTS

Allogeneic transfusion was necessary in 22 (29%) standard transfusion patients and 15 (19%) ANH (odds ratio [OR], 0.6; 95% CI, 0.28-1.3; p = 0.23) with 63 and 33 units transfused, respectively (p = 0.1). Significant postoperative complications occurred in 30 (38%) standard transfusion patients compared with 14 (18%) of those randomly assigned to ANH (OR, 0.3; 95% CI, 0.14-0.65; p = 0.009). The major difference between the groups was the frequency of infective complications.

CONCLUSION

Despite modest allogeneic transfusion requirements in hip surgery, ANH reduced postoperative complications.

Minimization method for balancing continuous prognostic variables between treatment and control groups using Kullback-Leibler divergence

This paper proposes a method for balancing prognostic variables between treatment and control groups in design of clinical trials. It assumes that some of prognostic variables are continuous and others are categorical and that they are independently distributed. The proposed method uses the Kullback-Leibler divergence (KLD) as the index of difference in distribution between two groups. It sequentially allocates each subject to a group using a biased coin method so as to reduce the estimate of KLD. That is, when first i subjects have been allocated to two groups and the (i+1)th subject is enrolled, the KLD is estimated if the (i+1)th subject was to be allocated to either of the groups, and the subject is then allocated with a certain probability, e.g. 0.80, so as to make the KLD small. Simulation studies based on the hypothetical prognostic variables and on the actual data of hyperlipidemia patients were carried out in order to compare the proposed method with the Pocock-Simon method, which transforms the continuous prognostic variables into categorical variables by dividing the whole scale into several categories. The p values of homogeneity test of means and variances were used to evaluate the achieved balance. The observed p values in the proposed method were better than those in the Pocock-Simon method. In addition to the balance, the precision of parameter estimates assuming analysis of covariance model was examined. The results showed the precision of estimators tended to be more stable in the proposed method than the Pocock-Simon method.