The requirement of randomization in experimental design was first stated by R. A. Fisher, statistician and geneticist, in 1925 in his book Statistical Methods for Research Workers. Earlier designs were systematic and involved the judgment of the experimenter; this led to possible bias and inaccurate interpretation of the data. Fisher's dictum was that randomization eliminates bias and permits a valid test of significance. Randomization in experimenting had been used by Charles Sanders Peirce in 1885 but the practice (...) was not continued. Fisher developed his concepts of randomizing as he considered the mathematics of small samples, in discussions with "Student," William Sealy Gosset. Fisher published extensively. His principles of experimental design were spread worldwide by the many "voluntary workers" who came from other institutions to Rothamsted Agricultural Station in England to learn Fisher's methods. (shrink)
Randomised clinical trials involve procedures such as randomisation, blinding, and placebo use, which are not part of standard medical care. Patients asked to participate in RCTs often experience difficulties in understanding the meaning of these and their justification.
The growing prevalence of health care ethics consultation (HCEC) services in the U.S. has been accompanied by an increase in calls for accountability and quality assurance, and for the debates surrounding why and how HCEC is evaluated. The objective of this study was to evaluate the effectiveness of HCEC as indicated by several novel outcome measurements in East Asian medical encounters.
Peter Urbach has argued, on Bayesian grounds, that experimental randomization serves no useful purpose in testing causal hypothesis. I maintain that he fails to distinguish general issues of statistical inference from specific problems involved in identifying causes. I concede the general Bayesian thesis that random sampling is inessential to sound statistical inference. But experimental randomization is a different matter, and often plays an essential role in our route to causal conclusions.
In standard medical care, physicians select treatments for patients based on clinical judgment, considering which treatment is best for the individual patient, given the patient's history and circumstances. In contrast, investigators conducting randomized clinical trials select treatments for participants based on a random selection process. Because this process represents a significant departure from the norms of standard medical care, it is widely assumed that potential research participants must understand randomization to give valid informed consent. This assumption, together with data (...) that many research participants do not understand randomization, implies that randomized clinical trials often fail to obtain adequately informed consent. Before accepting this conclusion, and before initiating extensive efforts to improve research participants' understanding of randomization, we should assess the plausible, but rarely analyzed assumption that participants need to understand randomization to give valid informed consent for randomized clinical trials. (shrink)
To give substance to the rhetoric of ‘learning health systems’, a variety of novel trial designs are being explored to more seamlessly integrate research with medical practice, reduce study duration and reduce the number of participants allocated to ineffective interventions. Many of these designs rely on response adaptive randomisation. However, critics charge that RAR is unethical on the grounds that it violates the principle of equipoise. In this paper, I reconstruct critiques of RAR as holding that it is inconsistent with (...) five important ethical principles. I then argue that these criticisms rest on a faulty view of equipoise encouraged by the idea that a RAR study models the beliefs of a single rational agent about the relative merits of the interventions being studied. I outline a view in which RAR models an idealised health system in which diverse communities of fully informed experts shrink or grow as their constituent members update their expert opinions in light of reliable medical evidence. I show how a proper understanding of clinical equipoise can reconcile this conception of RAR with these five ethical principles. This analysis removes an in-principle objection to RAR and sheds important light on the relationship between clinical equipoise and transient diversity in the scientific community. (shrink)
In clinical and agricultural trials, there is the danger that an experimental outcome appears to arise from the causal process or treatment one is interested in when, in reality, it was produced by some extraneous variation in the experimental conditions. The remedy prescribed by classical statisticians involves the procedure of randomization, whose effectiveness and appropriateness is criticized. An alternative, Bayesian analysis of experimental design, is shown, on the other hand, to provide a coherent and intuitively satisfactory solution to the (...) problem. (shrink)
Objective: To examine lay persons’ ability to identify methods of random allocation and their acceptability of using methods of random allocation in a clinical trial context.Design: Leaflets containing hypothetical medical, non-medical, and clinical trial scenarios involving random allocation, using material from guidelines for trial information leaflets.Setting and participants: Adults attending further education colleges , covering a wide range of ages, occupations, and levels of education.Main measures: Judgements of whether each of five methods of allocation to two groups was random in (...) a medical or non-medical scenario. Judgements of whether these allocation methods were acceptable in a randomised clinical trial scenario, with or without a scientific justification for randomisation.Results: The majority of our group of participants judged correctly that allowing people their preference was not random, and that the following were random: using a computer with no information about the individual , tossing a coin, drawing a name out of a hat. Judgements were split over allocating people in turn . Judgements were no different in medical and non-medical scenarios. Few of the correctly identified random methods were judged to be acceptable in a clinical trial scenario. Inclusion of a scientific justification for randomising significantly increased the acceptability of only one random method: allocation by computer.Conclusions: Current UK guidelines’ recommended description of random allocation by computer seems warranted. However, while potential trial participants may understand what random allocation means, they may find it unacceptable unless offered an acceptable justification for its use. (shrink)
Aims and background: Little is known about how participants perceive prevention trials, particularly trials designed to prevent mental illness. This study examined participants’ motives for participating in a trial and their views of randomisation and the ability to withdraw from a randomised controlled trial for prevention of depression. Methods: Participants were older adults reporting elevated depression symptoms living in urban and regional locations in Australia who had consented to participate in an RCT of interventions to prevent depression. Participants rated their (...) agreement with various statements describing motivations for enrolment in the trial and opinions regarding randomisation and withdrawal. Results: The majority of participants expressed a triad of altruistic motivation for participation, relative lack of concern about randomisation and commitment to the trial. Certain subgroups of participants, such as women and those with higher depression scores, reported higher levels of concern about specific issues. Conclusions: The findings suggest that participants enrolled in prevention trials for mental illness are likely to hold positive attitudes towards research trials. The identification of relationships between key person factors and trial-related attitudes enabled profiling of participant groups, which can inform recruitment strategies and interactions of participants and research projects in future prevention trials. (shrink)
Urbach (1985) has concluded that the use of randomization in the design of clinical and agricultural trials is both inappropriate and ineffective. It is argued here that it is appropriate, as it eliminates the dependence of inference on the unknown precise physical model that underlies a set of observations, and effective, in that it is relatively simple to apply in practice compared with any competing method. Furthermore, it has been proven in practice.
A benefit of randomized experiments is that covariate distributions of treatment and control groups are balanced on average, resulting in simple unbiased estimators for treatment effects. However, it is possible that a particular randomization yields covariate imbalances that researchers want to address in the analysis stage through adjustment or other methods. Here we present a randomization test that conditions on covariate balance by only considering treatment assignments that are similar to the observed one in terms of covariate balance. (...) Previous conditional randomization tests have only allowed for categorical covariates, while our randomization test allows for any type of covariate. Through extensive simulation studies, we find that our conditional randomization test is more powerful than unconditional randomization tests and other conditional tests. Furthermore, we find that our conditional randomization test is valid unconditionally across levels of covariate balance, and conditional on particular levels of covariate balance. Meanwhile, unconditional randomization tests are valid for but not. Finally, we find that our conditional randomization test is similar to a randomization test that uses a model-adjusted test statistic. (shrink)
Outcome-adaptive randomization has been proposed as a corrective to certain ethical difficulties inherent in the traditional randomized clinical trial using fixed-ratio randomization. In particular, it has been suggested that OAR redresses the balance between individual and collective ethics in favour of the former. In this paper, I examine issues of welfare and autonomy arising in relation to OAR. A central issue in discussions of welfare in OAR is equipoise, and the moral status of OAR is crucially influenced by (...) the way in which this concept is construed. If OAR is based on a model of equipoise that demands strict indifference between competing interventions throughout the trial, such equipoise is disturbed by accruing data favouring one treatment over another; OAR seeks to redress this by weighting randomization to the seemingly superior treatment. However, this is a partial response, as patients continue to be allocated to the inferior therapy. Moreover, it rests upon considerations of aggregate harms and benefits, and does not therefore uphold individual ethics. Issues of fairness also arise, as early and late enrollees are randomized on a different basis. Fixed-ratio randomization represents a fuller and more consistent response to a loss of equipoise, as so construed. With regard to consent, the complexity of OAR poses challenges to adequate disclosure and comprehension. Additionally, OAR does not offer a remedy to the therapeutic misconception—participants’ tendency to attribute treatment allocation in an RCT to individual clinical judgments, rather than to scientific considerations—and, if anything, accentuates rather than alleviates this misconception. In relation to these issues, OAR fails to offer ethical advantages over fixed-ratio randomization. More broadly, the ethical basis of OAR can be seen to lie more in collective than in individual ethics, and overall it fares worse in this territory than fixed-ratio randomization. (shrink)
Since the modern era, the discourse of punishment has cycled through three sets of questions. The first, born of the Enlightenment itself, asked: On what ground does the sovereign have the right to punish? Nietzsche most forcefully, but others as well, argued that the question itself begged its own answer. The right to punish, they suggested, is what defines sovereignty, and as such, can never serve to limit sovereign power. With the birth of the social sciences, this skepticism gave rise (...) to a second set of questions: What then is the true function of punishment? What is it that we do when we punish? From Durkheim to the Frankfurt School to Michel Foucault, twentieth century moderns explored social organization, economic production, political legitimacy, and the construction of the self_turning punishment practices upside down, dissecting not only their repressive functions but more importantly their role in constructing society and the contemporary subject. A series of further critiques_of meta_narratives, of functionalism, of scientific objectivity_softened this second line of inquiry and helped shape a third set of questions: What does punishment tell us about ourselves and our culture? What is the cultural meaning of our punishment practices? These three sets of questions set the contours of our modern discourse on crime and punishment.What happens now_ now that we have seen what lies around the cultural bend and realize that the same critiques apply with equal force to any interpretation of social meaning that we could possibly read into our contemporary punishment practices? Should we continue to labor on this third and final set of questions, return to an earlier set, or, as all our predecessors did, craft a new line of inquiry? What question shall we_children of the twenty_first century_pose of our punishment practices and institutions?This essay suggests that the form of the question never really mattered. In all the modern texts, there always came this moment when the empirical facts ran out or the deductions of principle reached their limit_or both_and yet the reasoning continued. There was always this moment, ironically, when the moderns took a leap of faith. And it was always there, at that precise moment, that we learned the most: that we could read from the text and decipher a vision of just punishment that was never entirely rational, never purely empirical, and never fully determined by the theoretical premises of the author. In each and every case, the modern text let slip a leap of faith_an ethical choice about how to resolve a gap, an ambiguity, an indeterminacy in an argument of principle or fact.It is time to abandon the misguided project of modernity. Rather than continue to take these leaps of faith, it is time once and for all to recognize the critical limits of reason, and whenever we reach them, to rely instead on randomization. Where our facts run out, where our principles no longer guide us, we should leave the decision_making to the coin toss, the roll of the dice, the lottery draw_in sum, to chance. This essay begins to explore what that would mean in the field of crime and punishment. (shrink)
A key role in inference is played by randomization, which has been extensively used in clinical trials designs. Randomization is primarily intended to prevent the source of bias in treatment allocation by producing comparable groups. In the frequentist framework of inference, randomization allows also for the use of probability theory to express the likelihood of chance as a source for the difference of end outcome. In the Bayesian framework, its role is more nuanced. The Bayesian analysis of (...) clinical trials can afford a valid rationale for selective controls, pointing out a more limited role for randomization than it is generally accorded. This paper is aimed to offer a view of randomization from the perspective of both frequentist and Bayesian statistics and discussing the role of randomization also in theoretical decision models. (shrink)
The job of scientists is to try to distinguish what is true from what is false by means of observation and experiment. That job has been made difficult by some philosophers of science who appear to give academic respectability to relativist, and even postmodernist, postures. This chapter suggests that the contributions of philosophers to causal understanding have been unhelpful. It puts the case for randomised studies as the safest guarantee of the reliability of scientific evidence. It uses the case of (...) hormone replacement therapy to illustrate the importance of randomisation, and the case of processed meat and cancer to illustrate the problems that arise in the absence of randomised tests. Finally, it discusses the opposition to randomisation that has come from some philosophers of science. (shrink)
There are two uses of randomization in efforts to control systematic bias in experimental design: (a) Alchemical uses seek to convert unavoidable systematic errors into random errors. (b) Hygienic uses seek to reduce the prospect of the experimenter's involvement with the implementation of the experiment contributing to bias. A few remarks are made at the end of the paper about the hygienic use of randomization as a preventative against sticky fingers. The bulk of the discussion addresses the alchemical (...) applications. The thesis is that attitudes towards the cogency of Fisher's alchemical use of randomization ought to depend on views concerning statistical deduction or direct inference. (shrink)
Randomization is the “gold standard” design for clinical research trials and is accepted as the best way to reduce bias. Although some controversy remains over this matter, we believe equipoise is the fundamental ethical requirement for conducting a randomized clinical trial. Despite much attention to the ethics of randomization, the moral psychology of this study design has not been explored. This article analyzes the ethical tensions that arise from conducting these studies and examines the moral psychology of this (...) design from the perspectives of physician-investigators and patient-subjects. We conclude with a discussion of the practical implications of this analysis. (shrink)
In this paper, I will argue that, while the ethical issues raised by cluster randomization can be challenging, they are not new. My thesis divides neatly into two parts. In the first, easier part I argue that many of the ethical challenges posed by cluster randomized human subjects research are clearly present in other types of human subjects research, and so are not novel. In the second, more difficult part I discuss the thorniest ethical challenge for cluster randomized research (...) – cases where consent is genuinely impractical to obtain. I argue that once again these cases require no new analytic insight; instead, we should look to political philosophy for guidance. In other words, the most serious ethical problem that arises in cluster randomized research also arises in political philosophy. (shrink)
Randomization is a generally accepted principle of sound experimental design and common practice among working scientists. But Bayesian statisticians reject it, most often because of decision theoretic argument against randomization. I trace it back to Abraham Wald's Theory of Inductive Behavior and argue that Bayesians should concur with Ronald Fisher 's criticism of Wald's analysis of randomization. The paper ends with a Bayesian argument in favor of randomization: randomization can lead to an increase in expected (...) utility. (shrink)
A formulation of probabilistic causality is given in terms of the theory of abstract dynamical systems. Causal factors are identified as invariants of motion of a system. Repetition of an experiment leads to the notion of stationarity, and causal factors yield a decomposition of the stationary probability law of the experiment into ergodic components. In these, statistical behaviour is uniform. Control of identified causal factors leads to a corresponding statistical law for the events, which is offered as a notion of (...) probabilistic causality. After a suggestion by Feller, randomization is identified as mixing, formulated in above terms. (shrink)
This thesis examines philosophical controversies surrounding the evaluation of medical treatments, with a focus on the evidential roles of randomised trials and mechanisms in Evidence-Based Medicine. Current 'best practice' usually involves excluding non-randomised trial evidence from systematic reviews in cases where randomised trials are available for inclusion in the reviews. The first paper challenges this practice and evaluates whether adding of evidence from non-randomised trials might improve the quality and precision of some systematic reviews. The second paper compares the alleged (...) methodological benefits of randomised trials over observational studies for investigating treatment benefits. It suggests that claims about the superiority of well-conducted randomised controlled trials over well-conducted observational studies are justified, especially when results from the two methods are contradictory. The third paper argues that postulating the unpredictability paradox in systematic reviews when no detectable empirical differences can be found requires further justification. The fourth paper examines the problem of absence causation in the context of explaining causal mechanisms and argues that a recent solution is incomplete and requires further justification. Solving the problem by describing absences as causes of 'mechanism failure' fails to take into account the effects of absences that lead to vacillating levels of mechanism functionality . The fifth paper criticises literature that has emphasised functioning versus 'broken' or 'non-functioning' mechanisms emphasising that many diseases result from increased or decreased mechanism function, rather than complete loss of function. Mechanistic explanations must account for differences in the effectiveness of performed functions, yet current philosophical mechanistic explanations do not achieve this. The last paper argues that the standard of evidence embodied in the ICE theory of technological function is too permissive for evaluating whether the proposed functions of medical technologies have been adequately assessed and correctly ascribed. It argues that high-quality evidence from clinical studies is necessary to justify functional ascriptions to health care technologies. (shrink)
In a recent paper in this journal, John Worrall (2008) used the example of a series of trials involving extracorporeal membrane oxygenation (ECMO), a technology for the treatment of respiratory failure in newborns, to illustrate the relationship between ethics and epistemology in medical research. One of the issues considered was whether or not it was ethical to perform a particular clinical trial at all, and he showed clearly that the answer was intimately related to epistemological judgments about the weight to (...) be given existing evidence concerning treatment effectiveness. In the case of ECMO, a trial was initiated at the University of Michigan (Bartlett et al. 1985), despite the fact that the researchers had .. (shrink)
Equipoise is widely regarded to be an essential prerequisite for the ethical conduct of a randomised controlled trial. There are some circumstances however, under which it is acceptable to conduct a randomised controlled trial in the absence of equipoise. Limited access to the preferred intervention is one such circumstance. In this paper we present an example of a randomised trial in which access to the preferred intervention, preschool education, was severely limited by resource constraints. The ethical issues that arise when (...) conducting randomised trials in health care are considered in the context of trials of social interventions. In health, education and social welfare, effective interventions are frequently limited due to budgetary constraints. Explicit acknowledgement of the need to ration interventions, and the use of random allocation to do this even in the absence of equipoise, would facilitate learning more about the effects of these interventions. (shrink)