I just wrote this essay, i'm gonna call it:"The middle ground is preferred" by J.B.

Controlled trials are a kind of organized comparison, used in clinical research that is specifically meant to measure the sameness and difference between therapeutic agents. Understood this way, controlled trials have no business being labelled as treatment. It is a tool used not in the pursuit of healing people but instead in the gathering human biological knowledge (Freeman, 131). Placebo-controlled trials require ethical analysis as they are a form of research that by practice exclude one arm of a trial from a therapeutic treatment. B. Freedman argues for an ethical system that outlines placebo-controls as the result of clinical researchers pursuing scientific knowledge rather than healing people, which is first and foremost the duty of clinicians. Therefore he concludes that when at all possible active-controls, which use known effective treatments instead of placebos, is ethically required. F. G. Miller on the other hand argues for a middle ground. He recognizes that placebo-controls are effective and in some situations necessary for valid clinical research. The delineation of ‘treatment’ from ‘research,’ determining the scientific benefit of placebo-controls, and when placebo-controls are necessary are central to determining which ethical system is better. Miller’s middle ground is the better ethical system as it sufficiently respects the duty clinicians have to research subjects while recognizing and utilising the effectiveness of placebo-controls.
‘Treatment’ is distinct from ‘research’ in many respects important to understanding how a placebo-controlled trial maybe be ethically utilised. Simply put, the function of treatment is to help and heal people. Treatment is individually tailored to a patient by a doctor and general medical practice requires that the best accepted remedy for a particular ailment be applied. Treatment is contrasted with research, which is a scientific study - the function of which is to gain medical insight. A controlled trial is used to compare two therapeutic agents when there is genuine disagreement within the medical community as to which therapeutic agent is superior. This disagreement is called clinical equipoise (Freedman, 117). There is no debate as to whether it would be ethical to use treatment X when there is a consensus among the medical community that treatment Y is much superior. Obviously it is not, so clinical equipoise is a requirement for the use of a controlled trial. Assuming clinical equipoise, the qualifying situations in which placebo-controlled trials can be ethically implemented need to be outlined.

Placebo-controlled trials utilise placebos for the control arm of a trial. One half of the research subjects receive a placebo, which is to say, they do not receive a therapeutic agent. The other arm of the trial does receive a therapeutic agent and the results are compared. Active-controlled trials utilise the recognized standard therapy for the control arm of a trial. One half of the research subjects receive the standard therapeutic agent, the other arm receives the therapeutic agent being researched, and the results are compared. Purely for the purposes of research, placebo-controlled trials are generally preferred. As Freedman notes, this view generally holds four propositions. First, Double-blind, placebo-controlled trials are the best, cleanest means of comparison. Second, active-controlled trials are second to placebo-controlled trials as tools for research. Third, placebo-controlled trials are unethical practices in certain situations. Finally, a balance between science and ethics is required to determine when placebo-control trials can be used ethically (Freedman, 131). Proponents of this viewpoint will argue that it would be ethically justified for research subjects to forego known effective treatment in favour of a placebo unless it would result in death or irreversible morbidity (Miller, 143). In opposition to this viewpoint there are those that deny the use of a placebo-controlled trial whenever there is an available treatment to be used in an active-control.

Part of the argument that supports the use of placebo-controlled trials is the proposition that they are more effective as a research tool than active-controlled trials. What is being assessed is the net therapeutic advantage of a therapeutic agent. The net therapeutic advantage is the “measure of a treatment’s direct impact upon the event causing the disease and specific benefit with respect to a patient’s short- and long-term prognosis... discounted by the treatment’s detrimental side effects,” (Freedman 132). The proposition that placebo-controlled trials are the best, cleanest method of comparison usually presumes that the net therapeutic advantage of placebos is zero. However this is not the case, as there have been many documented examples of placebos being an active agent with measurable positive and negative effects worthy of note. One example is that red pills will stimulate a stronger placebo effect than blue pills. What is of importance is that the combination of the method of administration, the placebo itself, the consent form or a combination of these will generate positive or negative effects that disrupt the idea of a placebo being used as a neutral bench mark (Freedman, 134). For this reason, some have offered amendments to the placebo-controlled trial through a number of practices, including double-blinding where neither the clinical researcher nor the research subject know if a placebo is being used or not, to maintain an equal effect in both arms of the trial. Freedman articulates these measures as “investigators ethically regressing to reliance upon deception,” and that these measures defeat the practical advantage of placebo-controlled trials in terms of simplicity, speed and economy (Freedman, 134). However, if the double-blind placebo-controlled trial can effectively measure the placebo effect and subtract it from both arms of the trial to show the true drug effect then it appears that it is still the most effective, cleanest means of comparison. Furthermore, the use of active-controls may be more dangerous than placebo-controls in particular instances. The difference in the rate of response is lower in treatment-controlled trials than placebo-controlled trials and in order to determine the efficacy of a therapeutic agent in a comparison trial it requires many more research subjects (Miller, 141). Sometimes the kind of malady in question and the kind of standard treatments required do not leave a large enough pool of research subjects to conduct a valid comparison. It is important to note that methodologic criteria may call for placebo-controls.

Behind the attempt to maintain the effectiveness of the placebo-controlled trial with a double-blind (and other means) is a scientific mindset. If Placebos can be used as a bench mark even if they act positively or negatively, then the clinical researcher has the cleanest means of comparison. Freedman argues however that this mindset is improper for the clinical researcher, who is supposed to be focussed on healing rather than scientific inquiry. Freedman is right to note that clinical researchers who use placebo-controlled trials are holding the value of scientific inquiry over the value of healing in particular cases. However, the information being pursued is meant to increase the effectiveness of medical practices. The utility of the research can provide a lot of good. Utilitarianism - which states ethical acts are those which promote the greatest amount of good, or the least amount of evil, for the greatest amount of people - ethically supports the use of placebo-controlled trials. What may cause some harm to a few people, in the form of a placebo, will be outweighed by the (possible) benefits of clinical research to be used in medical practice forever hence forth. However, this ethical justification can lead to dehumanizing acts which need to be guarded against. Freedman may be making a valid deontological point. As an action kind research should never knowingly deny a research subject treatment known to be effective. This view states that using a placebo-control where a treatment-control is useable is always unethical as it denies that person medical aid and as Freedman argues it the clinician’s job to heal people. Thus, the qualifying cases for an ethical placebo-controlled trial according to Freedman are where there are no standard treatments, the standard treatments are shown to be equal to a placebo, the standard treatment is a placebo, the standard treatment has been seriously called in to question or where standard treatment is not available (Freedman, 135). Essentially it is required that all possible attempts at using the preferred effective treatment be made before the use of a placebo is justified. In this way the focus is on the treatment of patients rather than on pure scientific inquiry. In opposition to this, and the better ethical system for justifying placebo-controlled trials, is the middle ground offered by Miller. That clinical researchers are motivated by scientific inquiry is not unethical itself. What is required is a clear delineation between research and treatment, as Miller points out “this argument conflates clinical research with clinical care,” (Miller 141). Research is not treatment, but is itself an ethical pursuit. The deontological perspective is unnecessary if it is adopted solely to protect against medical atrocities. In situations where scientific research requires or strongly favours the use of placebo-controls it ethically requires their use. It would be unethical to produce invalid research. The middle ground states that clinical research must adhere to several universal requirements. Among them are that research must be “scientifically valid and must minimize the risks to which the research participants are exposed,” (Miller, 142). The cases where utilitarianism would justify dehumanizing or life threatening research trials are checked against this premise. Miller concedes that when a life-prolonging or life-saving treatment is available it would be unethical to deny a research subject that treatment in favour of a placebo. He specifically states that “research participants in the placebo group should not be substantially more likely than those in the active-treatment group to die..” (Miller, 142). By applying strict standards of precaution the safety of research subjects can be ensured enough to quell ethical debate concerning the use of placebo-controlled trials when they are required by methodologic criteria. These precautions include things such as patients who are at increased risk are excluded and rescue medications are prepared and implemented in cases where research subjects experience serious symptoms (Miller, 143). With these considerations, the middle ground is the better ethical system as it allows for the protection of research subjects which appears to be the core concern for those who oppose placebo-controls, while recognizing and utilising the effectiveness of placebo-controls when necessary.


tl;dr placebo-controlled trials are preferred when valid scientific research requires it, probably because active-controls are impossible AND in cases where the research subject is not at risk of severe symptoms.

This post was edited by hellstruelight on Feb 14 2011 12:53am