Psychology and Psychiatry

Since the mid-twentieth century, American psychiatry has been characterized by increasing efforts to appear both medical and scientific, in terms of the reliability of its diagnostic criteria, the biological specificity of its treatments, and the methods by which these treatments are legitimated. Such efforts suggest the image of a laggard field attempting to play catch-up with its more scientific medical colleagues, but such a characterization ignores major transformations in the science of medicine as a whole over this time period. These transformations, most notable among them the development of the randomized controlled trial (RCT), coincided both with psychiatry's brief psychoanalytic deviation from a biological approach to mental illness and with the advent of psychotropic drugs. Together these developments created the conditions and need for many of the changes that have characterized subsequent psychiatric history.

The RCT.

Prior to the mid-twentieth century, physicians rarely resorted to experimental methods as a means to prove whether or not a treatment worked. Instead the determinant of legitimate therapeutic knowledge was expert clinical opinion, exercised through historical case controls, open trials, and clinical judgment. These means of evaluating treatments have since been replaced by the RCT.

The basic elements of the RCT—blinding, controls, randomization, and placebos—each have their separate histories. Research psychologists have actively employed experimental methods since the mid-nineteenth century, using randomization and controls much earlier than did the clinical sciences, psychiatry included. From the point of view of medical science, however, the formal birth of the RCT was in 1946, when these features were brought together in the streptomycin trials of the British Medical Research Council.

The design of the RCT is intended to ensure that perceived treatment outcomes are in fact due to the treatment under investigation, rather than to external factors or bias. Thus a basic RCT consists of two groups, an experimental group (given the treatment under investigation) and a control group (given another treatment or a placebo). Patients are randomly assigned to these groups to prevent their individual characteristics from biasing the results, and all participants—researchers, clinicians, and patients—are blinded as to which group a given patient is in, so that they do not bias the results of the experiment.

The RCT and psychiatry.

Like all scientific methods, the RCT presupposes certain facts about the nature of the world, and thus circumscribes the questions that can be asked and the answers that can be extracted from nature. The RCT views treatment outcomes as data that are independent from the subjective opinions of both doctors and patients. Thus the RCT has arguably supported the turn toward a biological view of psychiatric illness and cure, including the development of discrete diagnostic categories and diagnostically specific treatments.

The influence of the RCT on psychiatry has been practical as well as philosophical. As much as any other medical professionals, psychiatrists wanted better methods of determining whether the methods they employed actually worked. Chlorpromazine was one of the first psychiatric interventions to undergo RCT evaluation, with successful outcomes. However subsequent evaluations of psychiatry's older somatic treatments ended in dismal failure, no doubt reinforcing psychiatrists' enthusiasm for the new pharmaceutical cures. Many other forms of psychotherapy have fared poorly as well when subjected to RCTs, though many psychiatrists have been more skeptical of these outcomes, given the ill fit between the reductionistic design of the RCT and the more context-and relationship-dependent nature of psychotherapeutic cures. In spite of these reservations, by the late 1960s and early 1970s, most psychiatrists and clinical scientists had accepted the RCT as the best means of judging whether a treatment works.

Critiques of the RCT.

Since the mid-1990s, the RCT has come under increasing scrutiny. A growing number of researchers have argued that the method favors biological treatments over psychological ones, and that it cannot assess the role that psychosocial factors (for instance, contexts and doctor-patient relationships) and individual factors (for example, the meanings a patient gives to a particular remedy) play in shaping how well the intervention works. Others contend that the clinical experiment is so unlike the unpredictable world of actual clinical practice that it may not provide a reliable gauge of whether a treatment will work in actual practice. Some critics have also challenged whether the RCT actually succeeds in eliminating bias. A number of literature surveys have found that the greatest predictor of an RCT's outcome is who funded it. Beginning in the late 1990s, a series of editorials and articles in major medical journals such as the Journal of the American Medical Association and the New England Journal of Medicine have wrestled with the problem of how financial interests shape, direct, and, at times, subvert the science of clinical evaluation, lamenting that the RCT, no matter how well it is executed, is vulnerable to the very biases it was designed to expunge.