Mental Health Services Research, Vol. 1, No. 4, 1999

History and Evidence-Based Medicine: Lessons from the History of Somatic Treatments from the 1900s to the 1950s Joel T. Braslow1

This paper examines the early history of biological treatments for severe mental illness. Focusing on the period of the 1900s to the 1950s, I assess the everyday use of somatic therapies and the science that justified these practices. My assessment is based upon patient records from state hospitals and the contemporaneous scientific literature. I analyze the following somatic interventions: hydrotherapy, sterilization, malaria fever therapy, shock therapies, and lobotomy. Though these treatments were introduced before the method of randomized controlled trials, they were based upon legitimate contemporary science (two were Nobel Prize-winning interventions). Furthermore, the physicians who used these interventions believed that they effectively treated their psychiatric patients. This history illustrates that what determines acceptable science and clinical practice was and, most likely will, continue to be dependent upon time and place. I conclude with how this history sheds light on present-day, evidence-based medicine. KEY WORDS: evidence-based medicine; history of psychiatry; somatic therapies; biological psychiatry.

INTRODUCTION

measured up to the new scientific standards. These facts have led us to view these pre-RCT therapies and the doctors who used them as mired in a prescientific age, where personal conviction, local context, and social and cultural values played as large a role as science in the care and treatment of patients. The recent proliferation of practice guidelines based upon evidence from RCTs has reinforced the apparent contrast between our therapeutic age and that of the first half of the century (American Psychiatric Association [APA], 1997; APA Steering Committee, 1996; Lehman & Steinwachs, 1998; Veterans Health Administration, 1997). However, despite our methodological and therapeutic advances, the past has much to teach us, particularly about how a given era's scientific treatments can be transformed into what physicians' believe to be effective medical practice, even though we may later learn that a particular remedy that once "worked" in fact possessed little or no therapeutic value. Focusing on the period from the 1900s to the 1950s, the aim of this essay will be to sketch briefly the science and actual use of treatments used before the introduction of the RCT. I examine somatic and

Over the last couple of decades, researchers, clinicians, and policy-makers have urged clinicians to base their practices upon scientific evidence, the most robust of this evidence being the randomized controlled trial (RCT). Introduced into clinical medical science in the late the 1940s (Medical Research Council, 1948), psychiatric researchers first began employing the RCT in the 1950s (Elkes & Elkes, 1954; Elkes & Healy, 1998; Rees, 1956). The RCT quickly gained status as the "gold standard" of therapeutic efficacy. For psychiatrists, the RCT dramatically altered their therapeutic landscape (Healy, 1997). With the exception of electroconvulsive therapy, nearly all psychiatric treatments used in the first half of the century have been discarded. Though not the primary reason why physicians abandoned these therapies, the evidence supporting these therapies no longer

1

UCLA Departments of Psychiatry and History, and VISN 22 Mental Illness Research, Education and Clinical Center of the Department of Veterans Affairs, Los Angeles, California.

231 1522-3434/99/0400-0231$16.00/0 © 1999 Plenum Publishing Corporation

232 biological remedies for the severely mentally ill and show how the science and practice of each intervention reinforced each other. The treatments that follow largely have been abandoned and a few, such as lobotomy, have been thoroughly discredited. Yet my purpose is not to point to the misguided efforts of biological psychiatry. One could certainly write a history of similarly benighted efforts at psychological healing. Further, to see this history as a series of failed attempts at treating severe mental illness would be missing the point. Instead, this history illustrates that accepted science and therapeutics are dependent upon time and place. Indeed, this review of past biological therapies is important precisely because biological therapies have proven to be so effective that we often forget the evolving nature of science and the practice that flows from that science.

HYDROTHERAPY The first widely acknowledged effective somatic therapy of the twentieth century was hydrotherapy. Introduced into state hospitals throughout the United States in the late nineteenth and early twentieth centuries, this therapy consisted of a number of devices and techniques that employed water. The two most frequently used forms of hydrotherapy were the continuous bath and the wet sheet pack (Baruch, 1920; Finnerty & Corbitt, 1960; Wright, 1940). The "pack" required little in the way of sophisticated equipment. A sheet was dipped in water ranging from about 40 to 100°F and then the patient was snugly wrapped within this wet sheet. Very agitated patients were given colder sheets and more frail patients were placed in warmer sheets. Patients generally remained bundled for several hours at a time. Attendants often wrapped a blanket around the patient and the sheet. Finally, if the patient resisted the wet pack, the attendant placed a third sheet over the patient, securely tying him or her to the bed. Patients went through several stages while in the pack; first they were cooled, but over time the pack eventually heated. At times, physicians ordered a rubber sheet to be wrapped around the wet sheet to enhance the heating effects. Continuous baths required more elaborate devices than did wet packs. The baths most often consisted of a tub with an inlet for hot water and an outlet to drain the water. Attendants placed the patient in the hammock to which he or she was fastened. Attendants then covered the tub and patient with a

Braslow

canvas sheet that had a hole for the patient's head to go through. A series of valves and temperature gauges allowed the attendant to regulate both temperature and water flow. A single treatment could last anywhere from hours to days. These treatments provided early twentieth-century physicians with what they believed to be a genuinely therapeutic and biological approach to the treatment of severely mentally ill patients. In particular, they found these interventions especially effective means by which to therapeutically control psychotic patients. While late nineteenth-century asylum doctors had a variety of drugs by which to sedate and calm severely agitated patients (Ackerknecht, 1979), they rarely considered these medications as having true therapeutic value (Chapin, 1891-1892; Drapes, 1889; Mabon, 1888-1889; Macleod, 1900). Emphasizing the nontherapeutic nature of pharmacological treatments, such as bromides, chloral hydrate, hyoscine, paraldehyde, sulfonal, and narcotics, the wellknown British psychiatrist, Henry Maudsley (1895, pp. 554-555) wrote: Mechanical restraint, except under surgical necessities, was formerly abandoned, not because its use was sure to become abuse, but because it was deemed better for the patient to let him have the relief and self-respect of pretty free exercise than to keep him tied up like a mad dog . . . but it may be doubted whether its coarse bond did as much harm as has been done by the finer means of chemical restraint which have been used to paralyse the brain and render the patient quiet.

In contrast to drugs, physicians found hydrotherapy to be a genuinely scientific means by which to act upon their psychiatric patients' biology in order to effect mental cures. Indeed, researchers had discovered a variety of biological mechanisms through which hydrotherapy "worked." Some, for example, asserted that the remedy relieved "cerebral congestion" through its influence on the peripheral vascular system (Kellogg, 1887). Others argued that hydrotherapy helped eliminate "toxic impurities" that might cause insanity (Foster, 1899; Jagielski, 1896; Shepard, 1900). "It is extremely likely," a physician wrote in explaining the physiology of hydrotherapy, "that the excretory function of the skin and kidneys is stimulated" (Strecker, 1917, p. 1797). A body of research based on precise measurement of parameters such as blood pressure, pulse, respiratory rate, and differential blood count lent further support to the science of hydrotherapy (Adler & Ragle, 1913; Niles, 1899; Peck, 1909).

History and Evidence-Based Medicine Practitioners readily employed these physiological explanations in their everyday treatment of patients. Testifying in 1920 on behalf of a California state hospital accused of misusing hydrotherapy, a psychiatrist at the University of California Hospital in San Francisco declared that hydrotherapy "is the only scientific treatment for the acute excitement of the insane that has yet been discovered." Echoing the published literature, she went on to describe its physiologic effects: "Packs act by increasing the elimination by the skin, helping to rid the system of toxins and poisonous matter in the constitution." Furthermore, hydrotherapy brings "blood to the surface and relieve[s] the congestion in the brain and spinal cord, which in most cases seems to cause the excitement" (In the Matter of the Investigation of Agnews State Hospital, 1919, p. 531). While the scientific luster of hydrotherapy faded over the following three or four decades, psychiatrists commonly prescribed it until they gradually replaced it in the 1940s and 1950s first with electroconvulsive therapy and later with antipsychotic drugs.

233 by severing the vas deferens, the interstitial cells of the testicles would undergo hyperplasia. This in turn would lead to the increased production of beneficial hormones that "rejuvenated" the individual's mind and body (Benjamin, 1925; Money, 1983; Wolbarst, 1922). Psychiatrists logically adduced that vasectomy might also benefit psychiatric patients, given that a variety of reports demonstrated testicular abnormalities in the insane (Editorial, 1915; Epitome, 1915; Gibbs, 1923a, b, 1924; Mott, 1922; Tiffany, 1921). As with their use of hydrotherapy, practitioners used this scientific evidence to support their clinical practice as illustrated by the following conversation between a state hospital physician and his patient transcribed in 1928 (Stockton State Hospital [SSH] 32735, 1928, clinical conference, p. 1): Patient: I have these spells, I get a little melancholy, and then things don't break quite right and I get quite nervous. Doctor: Have you ever been sterilized? Patient: No. Doctor: You had better let us operate on you while you are here? Patient: That will certainly be all right with me and with my wife also. Doctor: We will do that then. Patient: Doctor, will that bring better composure to the nervous system? Doctor: It is supposed to, it has in a number of cases, we do not guarantee it, but in a number of cases it has had marked beneficial effects. It cannot hurt you and does not interfere with your sexual life in any way, we just cut a little duct and you absorb your own secretions. Patient: It has always been all right with me, and my wife did not want to take the responsibility of signing it. I have spoken to my wife about it and have told her I wanted it to get through this time. Doctor: Well, it cannot hurt you and it might have a marked beneficial result. Patient: I will be very much obliged to you, sir.2

STERILIZATION Sterilization was another major intervention introduced in the early years of the twentieth century (Dowbiggin, 1997; Reilly, 1991). First introduced in only a few states by 1910, by 1950, 26,000 American psychiatric patients had been sterilized, 11,000 of them in California (Robitscher, 1973). Though rightfully portrayed as a dark chapter in American psychiatry, physicians use of this surgery also illustrates the close relationship between a therapeutic practice and its supporting scientific evidence. Granted, most sterilization laws were passed in the first 25 years of this century largely at the urging of a small but influential group of eugenicists. In California, for example, physicians could sterilize patients "afflicted with hereditary insanity or incurable chronic mania or dementia" (Laughlin, 1922). Yet physicians did not necessarily sterilize patients for eugenic reasons; some, in fact, opposed eugenics even though they readily employed the surgery. The reason for this apparent contradiction was that physicians thought that the operation had therapeutic value in itself, especially for their male patients. For men, vasectomy was supposed to diminish anxiety and depression and increase vitality. Like hydrotherapy, physicians based their therapeutic use of sterilization upon contemporary science. A body of evidence supported the belief that,

MALARIA FEVER THERAPY In the first quarter of this century, general paralysis of the insane, a tertiary form of syphilis, posed 2

In order to protect patient privacy, all hospital numbers and identifying characteristics have been changed. I have retained the original spelling and punctuation in all of my quotations from the medical record. Records from Patton State Hospital are still retained by the hospital, which is located in Patton, California. Records from Stockton State Hospital were at the Stockton Developmental Center until its closing in February 1996. With appropriate permission from the California Department of Mental Health, one can obtain the actual patient record numbers from the author.

234 one of the greatest challenges for psychiatrists. Not only was the illness nearly invariably fatal, it also had a high prevalence. In Europe, for example, some institutions reported that up to 45% of their male patients suffered from this disease (Diefendorf, 1906). In America, physicians reported lower, although substantial, rates of paresis. In the 1910s, approximately 20% of male first admissions in New York State mental hospitals had a diagnosis of paresis, a figure that did not decline until 1925. Women were less likely to be admitted with paresis and had an admission rate of approximately one third that of male paretics. In the United States as a whole, about 9% of all first admissions during the 1930s had the diagnosis of general paralysis (Grob, 1983; Valenstein, 1986). In 1917, a Viennese neurologist, Julius Wagner von Jauregg, discovered that he could halt the progression of paresis by injecting patients with blood infected with benign tertian malaria. Once infected, von Jauregg's patients experienced a series of fevers (up to 106° F) and chills, which he then terminated after several weeks with quinine. Providing what appeared to be the first successful remedy for paresis, malaria fever therapy spread rapidly throughout the world, becoming one of the first somatic treatments for a mental illness widely acknowledged by the scientific community. Employing pre-RCT clinical scientific standards, numerous researchers replicated von Jauregg's findings using historical case controls, open trials, and clinical observation (Driver, Gammel, & Karnosh, 1926; Barnacle, Ebaugh, & Ewalt, 1936; Rose & Solomon, 1947). Scientists put forth a number of explanations for the efficacy of malaria fever therapy. Through animal experiments, some found that the fever itself destroyed the syphilitic spirochetes (Delgado, 1922; Schamberg & Rule, 1927), while others attempted to demonstrate that an enhanced immune response was responsible for the treatment's effectiveness (Bennett, 1938b; Delgado, 1922; Solomon, 1923). In 1927 von Jauregg received the Nobel Prize, the first ever awarded for a psychiatric intervention. Even as late as the early 1960s and after the introduction of penicillin, physicians continued to recommend the use of malaria fever therapy (Walshe, 1963). While by present-day standards we cannot be certain of the efficacy of malaria fever therapy, the treatment nonetheless dramatically altered the ways in which physicians dealt with their neurosyphilitic patients. Prior to the introduction of the remedy, physicians' views of their neurosyphilitic patients re-

Braslow flected prevailing cultural values in which individuals afflicted with syphilis were seen as immoral transgressors, perhaps even deserving of their often hopeless condition. Physicians rarely gave these patients any choice in their therapeutic regimen and, not unexpectedly, individuals afflicted with neurosyphilis were loath to admit themselves voluntarily into a state hospital for treatment. In a progress note written in 1923, several years before the hospital in which he worked had begun using malaria fever therapy, a California state hospital physician wrote the following about his paretic patient: "An extremely vulgar paretic who has led an immoral life. Had been treated for syphilis. I think her judgment is better than her behavior. This is the place for her" (Patton State Hospital [PSH] case 25806,1923, continuous notes). After fever therapy was introduced, physicians often described their patients more sympathetically and even invited them to participate in therapeutic decisions. Transcribed over a decade later at the same hospital, the following conversation between a doctor and his patient suggests that the malaria treatment had altered the relationships between doctor and patient (PSH case 29324,1937, clinical conference, p. 5): Patient: Good morning. Doctor: Do you want malaria? Patient: Well yes, I want anything to make me better. Thank you very much.

Furthermore, when patients refused the treatment, physicians acquiesced to their patients' wishes. Finally, unlike the pre-malaria era, patients voluntarily admitted themselves specifically for treatment with malaria fever. Taken together, these elements suggest that therapies influence far more than disease processes. Even the most biological of interventions can change the doctor-patient relationship and the very ways in which physicians' view their patients (Braslow, 1995). SHOCK THERAPIES Introduced in the 1930s and known collectively as "shock" therapies, these treatments consisted of three distinct, albeit overlapping, remedies: insulin, Metrazol, and electroconvlusive therapy (ECT). Insulin differed the most from the other two treatments in that it actually produced a state of physiologic shock but no seizures, while Metrazol and electricity produced grand mal seizures or convulsions but no physiologic shock. For this reason, the latter two were also known as the convulsive therapies. While insulin

History and Evidence-Based Medicine

and Metrazol have long since been consigned to the history of medicinal curiosities, ECT continues to be one of psychiatry's most effective interventions. On November 3,1933, Manfred Sakel reported to the Vienna Society of Physicians his new therapy for schizophrenia, termed insulin shock treatment (Insulinshockbehandlung), in which he gave massive doses of insulin to induce a profound state of hypoglycemic shock. Sakel had been using low doses of insulin in the late 1920s to quiet patients with delirium tremens as well as to improve their appetites. However, it was not until the early 1930s that he attempted to induce hypoglycemic comas in psychotic patients (Sakel, 1937,1938; James, 1992). Sakel and most subsequent practitioners of the therapy believed that these comas had especially beneficial effects on patients with schizophrenia. Nonetheless, practitioners and researchers never articulated a coherent, generally accepted theoretical explanation as to why insulin shock "worked." Subject to a number of modifications, the most commonly accepted method required daily injections of progressively higher doses of insulin until a comatose state was reliably produced. The patient then underwent daily injections at this "coma" dose. These daily treatments generally lasted several hours with termination of the coma by administration of a sugar solution via a nasogastric tube or an intravenous glucose solution. A complete course of insulin entailed about 50-60 "coma" days. Not surprisingly, given that patients were often brought to the brink of death just before doctors resuscitated them, the procedure was extremely labor-intensive, requiring the diligent attention of nurses and doctors to guard against a patient slipping too close toward an irreversible comatose state. Yet despite even the closest surveillance, patients died at a rate of 1-2% from complications such as hypoglycemic encephalopathy, heart failure, aspiration pneumonia, and cerebral hemorrhage (Kinsey, 1941; U.S. Public Health Service, 1941). In spite of these difficulties, the treatment spread rapidly. In a 1941 U.S. Public Health Service survey, for example, 71% of 305 public and private institutions reported that they used insulin shock therapy (U.S. Public Health Service, 1941). Its widespread application, however was short-lived and was quickly replaced by the much easier to administer ECT (Bennett, 1966). ECT had its origins in the work of Ladislas von Meduna. Believing that "a certain biochemical antagonism exists between the convulsive state and the

235

schizophrenic process" and that convulsions ameliorated psychosis, Ladislas von Meduna in Budapest developed a method to artificially induce convulsions. In early 1934 and after animal experimentation, he created convulsions first with intramuscular injections of camphor, but later switched to Metrazol (pentylenetetrazol) (Meduna, 1938). This new treatment gained wide and rapid acceptance, rivaling that of insulin. Compared to insulin, an individual Metrazol treatment was easier to administer, required less observation, took much less time, and produced fewer complications. As with insulin, most physicians used Metrazol on patients diagnosed with schizophrenia. By the late 1930s, however, an increasing number of researchers found that it had a greater efficacy on patients with depressive disorders than on patients with psychotic disorders (Bennett, 1938a). Aware of the success of Metrazol convulsive therapy, the Italians Ugo Cerletti and his co-worker Lucio Bini began work in 1936 on developing a method to produce electrically induced convulsions in psychiatric patients. By 1938, they had perfected a safe technique on dogs, and shocked their first human subject in April 1938 (Alverno, 1990; Endler, 1988; Harms, 1955; Impastato, 1960; Kalinowsky, 1980). Having fewer complications and easier to administer than either metrazol or insulin, ECT spread rapidly and eventually replaced both other shock therapies. According to the previously mentioned 1941 U.S. Public Health Service survey, 42% of 356 psychiatric institutions surveyed had electroshock machines just 3 years after the first human electroshock trial. The early use of ECT provides an excellent example of how divisions between somatic and psychological interventions are often arbitrary. Of particular interest is that state hospital physicians, though commonly portrayed as employing ECT as a means of patient control, often used the treatment as a means of enhancing a patient's accessibility toward psychological interventions. State hospital doctors frequently recommended both psychotherapy and electroshock simultaneously on many of their patients, believing that the two modalities acted synergistically (Gordon, 1948; Millet & Mosse, 1944; Selinsky, 1943). In a typical passage, a California state hospital physician wrote in 1952: "Transfer for psychotherapy (EST [electroshock therapy] also suggested)" (SSH case 68621, 1952, continuous notes, p. 1). While patients at times resisted ECT, this was not invariably the case, especially when physicians also attended to their patients' psychological needs. "I don't know doctor," a grateful patient told his ward physician in 1950, "I

236 had the electric shocks and that's the greatest thing ever happened in my life. I am telling you, that's the greatest thing that ever happened to me" (SSH case 63564,1950, clinical conference, p. 1).

LOBOTOMY Of all therapeutic interventions introduced prior to the 1950s, lobotomy is perhaps the most infamous, although interestingly the practice faced its harshest criticism long after physicians stopped using it (Pressman, 1998; Valenstein, 1986). Ironically, the rationale for lobotomy rested on relatively stable scientific ground. Its most important justification came from John Fulton's physiology laboratory at Yale. Fulton, beginning with his appointment as chairman of the physiology department in 1929, devoted much of his scientific energies to understanding frontal lobe function. In 1935 Fulton and his younger colleague, Carlisle Jacobson, delivered a paper at the Second International Neurological Congress. Also attending the conference was Egas Moniz, a Portuguese neurologist and the inventor of cerebral angiography. Moniz learned of how they had destroyed the frontal lobes of two chimpanzees, Lucy and Becky, a procedure that resulted in dramatically altered behavior. Moniz later used Fulton's and Jacobson's findings as part of his justification for proceeding with his brain surgeries, performing his first lobotomy in 1936. Moniz' surgery entailed drilling two holes into the top of the scull and then injecting alcohol into the frontal lobe white matter of the brain. Later, Moniz replaced the less predictable alcohol injections with a device called a leucotome, a rod-shaped instrument with a steel loop that crushed the white matter (the loop was eventually replaced by a band that cut instead of crushed) (Moniz, 1937, 1956). Over the following two decades, surgeons devised numerous modifications to Moniz's original surgery, although they all had the basic aim of severing frontal white matter fibers. While few would dispute that lobotomy acted directly on the brain, the exact mechanism by which it worked was never agreed upon. Moniz proposed that psychiatric pathology was the consequence of neuronal pathways becoming "fixed" within the white matter. He believed that by severing the frontal fibers these pathological associations became disrupted, creating less fixed and more normal patterns (Black, 1982; Damasio, 1975). Walter Freeman, the major proponent of lobotomy in the United States,

Braslow believed that the efficacy of lobotomy resided in severing the fibers between the thalamus and the frontal lobes. He argued that the thalamus imparted the pathological emotional content to ideas and that a surgeon had to destroy these fibers in order for lobotomy to succeed (Freeman & Watts, 1947). Unlike the shock therapies, the diffusion of lobotomy into physicians' practices took place slowly. Though introduced into the United States in 1936 by Freeman and James Watts (1937), lobotomy would not reach its golden age for at least another decade. This comparatively slow diffusion was, in part, due to its apparent lack of efficacy on patients with chronic schizophrenia (Freeman & Watts, 1936). For example, in his original report, Moniz found that the surgery worked best on those with agitated depressions and worst on those with psychosis. However, encouraged by positive reports of lobotomy on schizophrenia in the early 1940s (Strecker, Palmer, & Grant, 1942), state hospital physicians slowly began trials of the treatment, although the total number of lobotomies performed remained relatively low. For example, between 1940 and 1944, physicians reportedly had performed 684 lobotomies. After the war, however, the fortunes of lobotomy turned sharply for the better. By 1949, for example, spurred on by Freeman's tireless efforts to expand the surgery to as many state hospitals as possible, physicians had operated on 5,000 patients in a single year. In that same year, Moniz was awarded the Nobel Prize for his work on lobotomy. Fulton's research and personal effort further reinforced the acceptance of lobotomy (Fulton, 1951,1956; Pressman, 1988,1998). By 1951, a total of nearly 20,000 lobotomies had been performed in the United States (Kramer, 1954). As quickly as it rose, the fortunes of the surgery turned for the worse. After the introduction of the antipsychotic drug chlorpromazine in 1954, doctors quickly abandoned the surgery in favor of this new drug, and by the 1960s doctors rarely performed the surgery (Barahal, 1958; Robin, 1958). Though it would be easy to dismiss the practitioners of lobotomy as, at best, misguided, this interpretation is perhaps too simplistic, especially when evaluated in the light of everyday clinical dilemmas faced by physicians of the 1940s and early 1950s. Far from suggesting that lobotomists were unusually sadistic or deluded, examination of the everyday experiences of these physicians illustrates the ways in which science and local context and needs shape physicians' assessment of whether a treatment works. For state hospital physicians working in over-

237

History and Evidence-Based Medicine

crowded and understaffed institutions, lobotomy provided a scientifically based means by which to treat their most psychotic and uncontrollable patients. It is worth emphasizing that though physicians often used lobotomy as a "last resort," after all other therapies had failed, they nonetheless believed in its status as one of their most scientific of treatments. Writing to a prospective surgical candidate's family, a state hospital superintendent explained the surgery and its rationale: All forms of medical and psychiatric treatment up to this time have not been of more than temporary benefit. Unless a more drastic therapy is carried out, there will be little hope of any improvement . . . The treatment suggested is a delicate brain operation performed by a qualified neurosurgeon, which involves cutting certain nerve pathways controlling the basic emotions. This is known technically as psychosurgery or prefrontal leukotomy.

As this letter makes clear, while lobotomy was admittedly a "drastic" remedy, physicians did not employ it simply out of desperation or a need to do something for their otherwise intractable and most difficult patients. Doctors who recommended the surgery saw it as an unimpeachably scientific procedure. In state hospitals, physicians used lobotomy almost exclusively as a means of therapeutically controlling extremely psychotic patients. In a typical progress note recommending lobotomy, a ward physician wrote (SSH case 70456,1953, continuous notes, p. 2): She is a senile woman who shows her years, and at the rate she is going she will probably wear herself [out] before long. She is gradually deteriorating physically and something should be done about it now. Lobotomy is recommended in this case, primarily as a means of terminating the disturbed behavior, hoping that it may affect favorably the long term course of the illness before she becomes critically ill.

In this case, as in many others, the lobotomy successfully eliminated the patient's disturbed behavior. In so doing, it reaffirmed her physician's belief in the effectiveness of the treatment. Local context, in this instance, state hospital overcrowding, defined what physicians counted as the most significant aspects of disease, namely unmanageable psychotic behavior. Not surprisingly, then, interventions that quelled these symptoms, such as lobotomy, were seen as effective. The way in which local context, combined with scientific evidence, reinforces physicians' determina-

tion of the effectiveness of a treatment is particularly well illustrated by how Stockton physicians interpreted a common, though potentially troubling, outcome of lobotomy in which a lobotomized individual became unmotivated, apathetic, and indifferent. The following conversation between a doctor and his recently lobotomized patient exemplifies this outcome (SSH case 54919, clinical case conference, no date): Doctor: Hello, Joan. Patient: Hello. Doctor: Why are you wearing your hat? Patient: I don't know. Doctor: Joan, did you have an operation? Patient: Not that I know of. Doctor: Do you feel differently now? Patient: No.

Whether this neurological sequela was seen as an untoward side effect or evidence of the treatment's effectiveness depended largely upon context. In state hospitals, where controlling behavior measured therapeutic success, physicians' were less likely to see this well-known outcome as necessarily an unwanted consequence of the surgery (Aldrich, 1950; Cohen, Novick, & Ettleson, 1942; Ewald, Freeman, & Watts, 1947; Freeman & Watts, 1937). Summarizing a recently lobotomized patient's progress at a clinical case conference in 1954, the patient's ward physician observed (SSH case 61399,1954, clinical case conference), "Lobotomy through the eye was done and apparently it has had beneficial effects . . . It is quite definite that the operation helped him. His comments almost sound like a testimonial, as though he has been coached by somebody to make these statements about his improvement at the hospital. In any case, he is dull, somewhat apathetic, and he answers questions in single words." Not only did apathy not negate a successful therapeutic outcome, but, at times, was seen as a precondition for the treatment's effectiveness. According to one practitioner at the same California state hospital, "Maybe that [apathy] is what cures them" (SSH case 51465,1949, clinical conference, p. 3). While scientific pronouncements (and scientific evidence) certainly shape physicians' decisions to use a particular treatment and to perceive whether it works, they are not the sole determinants, as illustrated by the fate of lobotomy. In the mid 1950s, physicians abruptly abandoned lobotomy in favor of chlorpromazine. The meteoric rise in the popularity of chlorpromazine (by 1955, over two million prescriptions in the United States had been written) and

Braslow

238 the equally rapid decline in lobotomy had little to do with scientific evidence (Feyhan, 1955; Freeman, 1958). It was not until the late 1950s that a number of large-scale studies demonstrated the questionable efficacy of lobotomy and not until the early 1960s that large-scale randomized controlled trials showed the efficacy of chlorpromazine (Psychopharmacology Service Center, 1964; Casey, Lasky, Klett, & Hoilister, 1960). At the same time, physicians did not stop using lobotomy because they decided it was ineffective. Instead, they found chlorpromazine a more effective intervention within the state hospital context.

CONCLUSION Though nearly all of the psychiatric treatments (with the exception of ECT) introduced before the 1950s have been abandoned, the aim of this overview has not been to chronicle the failures of biological psychiatry. Whether biologically or psychologically oriented, science and therapeutic practices are bounded by time, place, and culture and, as such, are subject to reevaluation over time. The concept of the schizophrenogenic mother and the psychoanalytic treatment of patients with psychotic disorders are examples of an era when psychosocial reductionism partook in mainstream psychiatric science and, if taken out of their contexts, could be viewed as exemplars of the folly of psychological approaches to psychiatric illness. Similarly, many of the treatments discussed in this essay could be interpreted as biological reductionism run amok. However, instead of a story about the excesses of biological psychiatry, this history shows how dependent a successful, scientifically based therapy can be upon time and place. What does this teach us about our present-day efforts at evidence-based medicine? First, this history should encourage a sense of humility despite our scientific and therapeutic advances. Every generation believes in what they deem as "evidence" and, as this history illustrates, what counts as evidence is not fixed, but evolves over time. Second, this history should encourage us to ask critical questions about our contemporary methods of producing evidence and treating patients, since, if history is any guide, these methods will no doubt be subject to revision. While no one would seriously consider abandoning the RCT, researchers have begun to question its generalizability and utility for informing policy decisions. Recent efforts to combine efficacy and effectiveness

paradigms are perhaps the early stages of the creation of a new mode of making evidence (Wells, 1999; National Advisory Mental Health Council's Clinical Treatment and Services Research Workgroup, 1998). This history also raises difficult ethical questions. On the one hand, looking back from our contemporary vantage point, one could rightly view such treatments as lobotomy as brutal and inhumane. It is a history that reminds us that science and good intentions in the care of the severely mentally ill can, at times, have disastrous consequences for those whom physicians seek to help. On the other hand, we should not judge these physicians too harshly. Each intervention partook in legitimate contemporary science with, perhaps, the exception of sterilization. Further, acting to heal their patients, practitioners found these interventions effective within their social and cultural practice context. As the context of both care and science has changed, the ways in which physicians in everyday clinical practice judge the effectiveness of an intervention also has changed. And our moral evaluation of how we intervene upon those afflicted with severe mental illness has shifted as well, and will, no doubt, continue to evolve.

REFERENCES Ackerknecht, E. (1979). The history of drug treatment of mental diseases. Transactions and Studies of the College of Physicians of Philadelphia, 1, 161-170. Adler, H. M., & Ragle, E. H. (1913). Note on the influence of hydrotherapy on the red blood cell count in the insane. Boston Medical and Surgical Journal, 169, 425-426. Aldrich, C. K. (1950). Problems of social adjustment following lobotomy. American Journal of Psychiatry, 107, 459-462. Alverno, L. (1990). The origins of electroconvulsive therapy. Wisconsin Medical Journal, 89, 54-56. American Psychiatric Association. (1997). Practice guideline for the treatment of patients with schizophrenia. American Journal of Psychiatry, 154, 1-63. American Psychiatric Association Steering Committee. (1996). The Expert Consensus Guideline Series. Treatment of schizophrenia. Journal of Clinical Psychiatry, 57, 3-58. Barahal, H. S. (1958). 1,000 prefrontal lobotomies—A five-to-10year follow-up study. Psychiatric Quarterly, 32, 653-678. Barnacle, C. H., Ebaugh, F. G., & Ewalt, J. R. (1936). Treatment of dementia paralytica: Comparative study of combined artificial hyperpyrexia and tryparsamide versus therapeutic malaria: A preliminary report. JAMA, 107, 1031-1036. Baruch, S. (1920). An epitome of hydrotherapy for physicians, architects and nurses. Philadelphia: Saunders. Benjamin, H. (1925). New clinical aspects of the Steinach operation. Medical Journal and Record, 122, 452-457, 515-518, 592-594. Bennett, A. E. (1938a). Convulsive (pentamethylenetetrazol) shock therapy in depressive psychoses. Preliminary report of results obtained in ten cases. American Journal of the Medical Sciences, 196, 420-428. Bennett, A. E. (1938b). Evaluation of artificial fever therapy for

History and Evidence-Based Medicine neuropsychiatric disorders. Archives of Neurology and Psychiatry, 40, 1141-1155. Bennett, I. (1966). Changing concepts in insulin coma therapy. In M. Rinkel (Ed.), Biological treatments of mental illness (pp. 782-792). New York: Page. Black, D. W. (1982). Psychosurgery. Southern Medical Journal, 75, 453-458. Braslow, J. T. (1995). The effect of therapeutic innovation on perception of disease and the doctor patient relationship: A history of general paralysis of the insane and malaria fever therapy, 1910-1950. American Journal of Psychiatry, 152, 660-665. Casey, J, F., Lasky, J. J., Klett C. J., & Hollister, L. E. (I960). Treatment of schizophrenic reactions with phenothiazine derivatives: A comparative study of chlorpromazine, triflupromazine, mepazine, prochlorperazine, perphenazine, and phenobarbital. American Journal of Psychiatry, 117, 97-105. Chapin, J. B. (1891-1892). On the abuse of hypnotics. American Journal of Insanity, 48, 202-208. Cohen, L. H., Novick, R. G., & Ettleson, A. (1942). Frontal lobotomy in the treatment of chronic psychotic overactivity. Psychosomatic Medicine, 4, 96-104. Damasio, A. M. (1975). Egas Moniz, pioneer of angiography and leucotomy. Mount Sinai Journal of Medicine, 42, 502-513. Delgado, H. F. (1922). The treatment of paresis by inoculation with malaria. Journal of Nervous and Mental Disease, 55, 376-389. Diefendorf, A. R. (1906). Etiology of dementia paralytica. British Medical Journal, 2, 744-748. Dowbiggin, I. R. (1997). Keeping America sane: Psychiatry and eugenics in the United States and Canada, 1880-1940. Ithaca, NY: Cornell University Press. Drapes, T. (1889). Hyoscin in insanity. British Medical Journal, 1, 942. Driver, J. R., Gammel, J. A., & Karnosh, L. J. (1926). Malaria treatment of central nervous system syphilis. JAMA, 87, 1821-1827. Editorial. (1915). Sex organ changes in insanity. JAMA, 65, 254-255. Elkes, J., & Elkes, C. (1954). Effect of chlorpromazine on the behaviour of chronically over-active psychotic patients. British Medical Journal, 2, 560-565. Elkes, J., & Healy, D. T. (1998). Towards footings in a new science: Psychopharmacology, receptors, and the pharmacy within. In D. T. Healy (Ed.), The psychopharmacologists II: Interviews by Dr David Healy. London: Altman. Endler, N. S. (1988). The origins of electroconvulsive therapy (ECT). Convulsive Therapy, 4, 5-23. Epitome. (1915). Epitome, asylum reports. Journal of Mental Science, 61, 288-321. Ewald, F. R., Freeman, W. J., & Watts, J. W. (1947). Psychosurgery; the nursing problem. American Journal of Nursing, 47, 210-213. Feyhan, F. (1955). The immediate and long range effects of chlorpromazine on the mental hospital. In Smith, Kline & French Laboratories (Ed.), Chlorpromazine and mental health. Philadelphia: Lea & Febiger. Finnerty G., & Corbitt, T. (1960). Hydrotherapy. New York: Ungar. Foster, G. W. (1899). Hydric treatment of the insane. American Journal of Insanity, 55, 639-665. Freeman, W. (1958). Prefrontal lobotomy: Final report of 500 Freeman and Watts patients followed for 10 to 20 years. Southern Medical Journal, 51, 739-745. Freeman, W. J., & Watts, J. W. (1936). Prefrontal lobotomy in agitated depression: Report of a case. Medical Annals of the District of Columbia, 5, 326-328. Freeman, W. J., & Watts, J. W. (1937). Prefrontal lobotomy in the treatment of mental disorders. Southern Medical Journal, 30, 23-31.

239 Freeman, W. J., & Watts, J. W. (1947). Retrograde degeneration of the thalamus following prefrontal lobotomy. Journal of Comparative Neurology, 86, 65-93. Fulton, J. F. (1951). Frontal lobotomy and affective behavior, a neurophysiological analysis. New York: Norton. Fulton, J. F. (1956). A textbook of physiology. Philadelphia: Saunders. Gibbs, C. (1923a). Relation of puberty to behavior and personality in patients with dementia praecox. American Journal of Psychiatry, 80, 121-129. Gibbs, C. (1923b). Sex development and behavior in male patients with dementia praecox. Archives of Neurology and Psychiatry, 9, 73-87. Gibbs, C. (1924). Sex development and behavior in female patients with dementia praecox. Archives of Neurology and Psychiatry, 11, 179-194. Gordon, H. L. (1948). Fifty shock therapy theories. Military Surgeon, 103, 397-401. Grob, G. N. (1983). Mental illness and American society, 18751940. Princeton, NJ: Princeton University Press. Harms, E. (1955). The origin and early history of electrotherapy and electroshock. American Journal of Psychiatry, 111, 933-934. Healy, D. T. (1997). The antidepressant era. Cambridge, MA: Harvard University Press. Impastato, D. J. (1960). The story of the first electroshock treatment. American Journal of Psychiatry, 116, 113-114. In the Matter of the Investigation of Agnews State Hospital. (1919). California State Archives, Sacramento, California. Jagielski, V. (1896). The therapeutic action of the Turkish bath. JAMA, 27, 413-416. James, F. E. (1992). Insulin treatment in psychiatry. History of Psychiatry, 3, 221-235. Kalinowsky, L. B. (1980). The discoveries of somatic treatments in psychiatry: Facts and myths. Comprehensive Psychiatry, 21, 428-435. Kellogg, T. H. (1887). Hydrotherapy in mental diseases. New York Medical Journal, 46, 427-431. Kinsey, J. L. (1941). Incidence and cause of death in shock therapy. Archives of Neurology and Psychiatry, 46, 55-58. Kramer, M. (1954). The 1951 survey of the use of psychosurgery. In Proceedings of the Third Research Conference on Psychosurgery. PHS Publication 221. Washington, DC: U.S. Government Printing Office. Laughlin, H. (1922). Eugenical sterilization in the United States: A report of the Psychopathic Laboratory of the Municipal Court of Chicago. Chicago: Psychopathic Laboratory of the Municipal Court of Chicago. Lehman, A. F., & Steinwachs, D. M. (1998). Patterns of usual care for schizophrenia: Initial results from the Schizophrenia Patient Outcomes Research Team (PORT) Client Survey. Schizophrenia Bulletin, 24, 11-20; Discussion, 20-32. Mabon, W. (1888-1889). Clinical observations on the action of sulfonal in insanity. American Journal of Insanity, 45,493-499. Macleod, N. (1900). The bromide sleep: A new departure in the treatment of acute mania. British Medical Journal, 1,134-136. Maudsley, H. (1895). The pathology of the mind. London: Macmillan. Medical Research Council. (1948). Streptomycin treatment of pulmonary tuberculosis. British Medical Journal, 2, 769-782. Meduna, L. J. (1938). The significance of the convulsive reaction during the insulin and the cardiazol therapy of schizophrenia. Journal of Nervous and Mental Disease, 87, 133-139. Millet, J., & Mosse, E. (1944). On certain psychological aspects of electroshock therapy. Psychosomatic Medicine, 6, 226-236. Money, J. (1983). The genealogical descent of sexual psychoneuroendocrinology from sex and health theory: The eighteenth to the twentieth centuries. Psychoneuroendocrinology, 8, 391-400.

240 Moniz, E. (1937). Prefrontal leucotomy in the treatment of mental disorders. American Journal of Psychiatry, 93, 1379-1385. Moniz, E. (1956). How I succeeded in performing the prefrontal leukotomy. In A. M. Sackler (Ed.), The great physiodynamic therapies in psychiatry (pp. 131-137). New York: Paul B. Hoeber. Mott, F. (1922). The reproductive organs in relation to mental disorders. British Medical Journal, 1, 463-466. National Advisory Mental Health Council's Clinical Treatment and Services Research Workgroup. (1998). Bridging science and service, http://www.nimh.nih.gov/research/bridge.htmtbrl. Niles, R. (1899). Hydrotherapy in the treatment of mental diseases. American Journal of Insanity, 55, 443-447. Peck, R. E. (1909). Rational hydrotherapy vs. Turkish baths. Yale Medical Journal, 15, 455-462. Pressman, J. D. (1988). Sufficient promise: John F. Fulton and the origins of psychosurgery. Bulletin of the History of Medicine, 62, 1-22. Pressman, J. D. (1998). Last resort: Psychosurgery and the limits of medicine. Cambridge: Cambridge University Press. Psychopharmacology Service Center Collaborative Study Group National Institute of Mental Health. (1964). Phenothiazine treatment in acute schizophrenia. Archives of General Psychiatry, 10, 246-261. Rees, W. L. (1956). A controlled study of the value of chlorpromazine in the treatment of anxiety tension states. L'Encephale, 4, 547-549. Reilly, P. (1991). The surgical solution: A history of involuntary sterilization in the United States. Baltimore, MD: Johns Hopkins University Press. Robin, A. A. (1958). A controlled study of the effects of leucotomy. Journal of Neurology, Neurosurgery, and Psychiatry, 21, 262-269. Robitscher, J. (Ed.). (1973). Eugenic sterilization. Springfield, IL: Charles C. Thomas. Rose, A. S., & Solomon, H. C. (1947). Neurosyphilis. American Journal of Psychiatry, 103, 524-527. Sakel, M. (1937). A new treatment of schizophrenia. American Journal of Psychiatry, 93, 829-841.

Braslow Sakel, M. (1938). Nature and origin of the hypoglycemic treatment of psychoses. American Journal of Psychiatry, 94(Suppl.), 24-40. Schamberg, J. F., & Rule, A. M. (1927). Therapeutic effect of hot baths in experimental primary syphilis in rabbits. JAMA, 88, 1217-1218. Selinsky, H. (1943). The selective use of electro-shock therapy as an adjuvant to psychotherapy. Bulletin of the New York Academy of Medicine, 19, 245-252. Shepard, C. H. (1900). Insanity and the Turkish bath. JAMA, 34, 604-606. Solomon, H. C. (1923). General paresis: What it is and its therapeutic possibilities. American Journal of Psychiatry, 79, 623-646. Strecker, E. A. (1917). The continuous bath in mental disease. JAMA, 68, 1796-1799. Strecker, E. A., Palmer, H. D., & Grant, F. C. (1942). A study of prefrontal lobotomy; neurosurgical and psychiatric features in 22 cases with a detailed report on 5 chronic schizophrenics. American Journal of Psychiatry, 98, 524-532. Tiffany, W. J. (1921). Pathological changes of testes and ovaries in dementia praecox. State Hospital Quarterly, 6, 159. U.S. Public Health Service. (1941). Shock therapy survey. Washington DC: Government Printing Office. Valenstein, E. S. (1986). Great and desperate cures: The rise and decline of psychosurgery and other radical treatments for mental illness. New York: Basic Books. Veterans Health Administration. (1997). Clinical guideline for management of persons with psychoses, version 1.0. Washington, DC: Author. Walshe, F. (1963). Diseases of the nervous system described for practitioners and students (10th ed.). Edinburgh: Livingstone. Wells, K. B. (1999). Treatment research at the crossroads: The scientific interface of clinical trials and effectiveness research. American Journal of Psychiatry, 156, 5-10. Wolbarst, A. L. (1922). A report on the Steinach operation in senility and premature senility. New York Medical Journal and Medical Record, 115, 543-545. Wright, R. (1940). Hydrotherapy in psychiatric hospitals. Boston: Tudor.