01Sep09j Criticism of selection procedures
Below is a part of the article "David and Goliath: When Empirical and Clinical Standards of Practice Meet" by Larry E. Beutler, Ph.D. (2000), taken from http://www.education.ucsb.edu/~bobr/leb/goliath.htm
The article claims that psychosocial treatments are often selected, practiced and governmentally approved based on procedures that are political or commercial in nature (broad beliefs? cost-effective?), or that do not stand the test of scientific rigor (e.g. treatments accepted though clinical experience is unreliable).
The "American Psychiatric Association" is especially criticized in this respect. The first part of the article (given below) discusses the reign of dubious treatments primarily in light of the recovered memories & satanic ritual abuse saga. Note the following passage: "Even in the case of the deprogramming therapy of Braun, the treatment continues to be practiced though the courts did, in 1998, act to remove Braun's medical license (Gruman, 1998), a decision that is under appeal by Braun at this writing."
[Beginning of article]
In 1995, the Public Broadcasting System (PBS) presented an expos on a therapy that addressed the sequelae of satanic abuse (Bikel & Dretzin, 1995). The broadcast reported on the treatment program developed at the Multiple Personality Disorder Treatment Unit at Rush-Presbyterian Medical Center in Chicago under the direction of Bennett Braun, a psychiatrist, and Roberta Sachs, a clinical psychologist. This treatment was based on the theory that the symptoms of multiple personality disorder (MPD; now, dissociative identity disorder [DID])were indicators that the patient so diagnosed may have been satanically abused (Braun, 1990; Young, Sachs, Braun, & Watkins, 1991).
Although Braun's deprogramming therapy was exceptional among the psychotherapies, his theory earned him many accolades, and his viewpoints were accepted as factual by a substantial minority of professional practitioners during the late 1980s and early 1990s. This theory asserted that generalized distress accompanied by ambivalence and dissociation provided reliable evidence that a patient had come under the influence, albeit subsequently repressed, of a coven of witches, one of several families of witches who dated to the middle ages. These families had preserved their satanic rituals and evil powers by systematically programming their children -- through a process of human sacrifice, cannibalism, and sexual rituals -- to pursue a life-long mission of slaying non-believers. Most remarkable, in this theory, because of the use of hypnosis and other powerful persuasion-based procedures, no one -- neither the witches who carried out the programming nor the children who were the recipients of it -- had any memory of this plot or of the rituals necessary to ensure that the plan of destruction was successful.
The witches in these satanic covens, operating on their own programmed histories, allegedly were induced to awaken from a hypnotically induced state of repression when a priest or priestess in the coven presented them with certain satanic signals or cues, usually letters of the Greek alphabet. Having awakened, they then assumed the role of priest and priestess themselves, to initiate the human sacrifices and sexual rituals necessary to program their children or kill the uninitiated. Having done so, they then lapsed again into a state of programmed repression, assuming the mantel of normal bankers, homemakers, professors, teachers, plumbers and the like until their hidden programs were again activated. MPD (or DID) was a side effect of this process.
The PBS story documented the case of Mary, who was identified by Braun as being a high priestess in an ancient coven. Her identity as a witch who had killed others and who had programmed her children, by subjecting them to rituals of human sacrifice, cannibalism, and sex, to murder her enemies, was secured and confirmed only by the presence of two major symptoms, dissociation and depression, along with a strong suspicion of incest. As a result of Braun's diagnosis, Mary was subjected to several years of deprogramming therapy, an intensive inpatient treatment that included withholding food, unusually high doses of medications, four-point restraints, and deprogramming groups that included several therapists who attached her denials until she confessed to being a witch. These treatments were then extended to her husband and to her two sons. Ultimately, her treatment cost her insurance carrier over two million dollars and resulted in the dissolution of her marriage and her estrangement from her two sons. It also resulted in her bringing suit against Braun for malpractice.
It should be noted that most medical treatments are subject to specific empirical tests before they are or can be offered as treatments in the United States. These tests are designed by the Food and Drug Administration (FDA) to assure that the treatment is first of all safe and second, effective. Laypersons and alternative medicine practitioners frequently criticize the FDA standards of proof for being so demanding that they often exclude from the U.S. market treatments that are routinely available without prescription in European and South American countries. Yet, from a scientific perspective, the criteria used by the FDA are rather lax and sometimes have been applied in such a way that evidence of harm has been overlooked.
The FDA criteria require only that two independent, controlled studies have been published, demonstrating statistically (not necessarily clinically) significant finding when the treatment is compared with an inactive or placebo treatment. Critics point out that this standard does not weigh negative findings if positive ones are available. As a result, for example, in 1998, the FDA was forced to remove three drugs from the market because they had dangerous side effects that were overlooked by adherence to the FDA's criteria. Redux, a popular dietetic drug, along with Posicor and Duract, was removed because the FDA criteria, in relying only on evidence of two studies in which there were positive effects, overlooked evidence that they caused serious heart and liver damage.
In the PBS expos (Bikel & Dretzin, 1995), Braun spoke of evidence underlying his theory and practice. However, the evidentiary base to which Braun referred does not even approach the modest standard set on medical treatments by the FDA. It was based on supposedly clinical evidence, supplemented by a review of books and early writings on satanic cultures in Europe and by a plethora of letters from believers. The so-called research did not include systematic research to determine the validity of the assumptions of satanic abuse, nor did it include studies of whether memories of this type could actually be suppressed by the hypnotic techniques of the coven, tests of the potential negative effects of treatments, or an assessment of how effectively the proposed deprogramming therapy either reduced symptoms of dissociation or improved one's quality of life. In short, Braun's research did not include what psychologists would conventionally consider to be minimally necessary to certify any medical treatment as either safe or effective.
Braun's treatment was composed of psychosocial methods of deprogramming and psychotherapy. In this domain, two fundamental myths have always been accepted and applied whenever new treatments are introduced: First, psychosocial treatments are not strong enough to cause real harm, and second, all psychosocial treatments are all pretty much the same -- a position that was labeled the "dodo bird verdict" by Luborsky, Singer, and Luborsky (1975)-or, in other words, all have won and all must have prizes. Notably, these myths have been extended to and implicitly incorporated by managed care and service delivery programs when treatment decisions are made. If nothing else, the Rush-Presbyterian Medical Center experience should testify that neither of these assumptions should be accepted as universally true.
Although empirical comparisons of most structured treatments do result in similar mean effects (e.g., Wampold, et al., 1997), there is good evidence that hidden within these mean effects are widely disparate outcomes (e.g., Howard, Krause, Saunders, & Kopta, 1997). Specifically, psychotherapeutic interventions, as a rule, increase the possibility of both benefit and deterioration (Lambert, 1992), and some practices (and practitioners) are consistently associated with a deterioration of patients' condition--they are psychonoxious and sometimes dangerous (Bergin, 1963; Bergin & Lambert, 1978). If psychologists, or the health care enterprise, accept the propositions that some treatments do in fact cause harm, and that all treatments are not equally damaging, then they are faced with the problem of identifying what treatments work best, for whom, and under what circumstances. That is, there would be costs.
On one hand, health care programs would then be obliged to ensure that their cadre of practitioners was trained in these procedures and that they were practiced correctly. On the other hand, practitioners would have to face the possibility of giving up favored positions and theories for which scientific evidence of effectiveness is lacking and adopting others that might be less interesting to increase the scientific credibility of their work (Beutler, Kim, Davison, Karno, & Fisher, 1996).
Clinical Standards as Criteria of Truth
Unfortunately, in determining the value and validity of clinical practice, those of us who practice psychosocial treatments have held themselves to exceedingly low but implicit standards based on so-called clinical evidence. Shared, strong beliefs and the personal sincerity of those who advocate these beliefs have been held as sufficient evidence to ensure the truth of a treatment recommendation. Treatment practices have been validated by reference to clinical experience, but the nature of this experience has usually been unstated, and the beliefs that are attributed to it have varied widely among similarly experienced clinicians. Extensive research evidence supports the conclusion that, among clinicians, personal beliefs, thus developed, are almost always given greater credence than scientific evidence and are considered to be the real or moral truth when the two sources of evidence are at variance (see, e.g., Garb, 1998; Singer & Lalich, 1996). To leave the future of professional practice to the dubious validity of sincere beliefs and appealing clinical theories is to risk that there will be more experiences like the one illustrated in the aforementioned PBS program. History confirms that if one believes a position very strongly and is sincere in proposing this point of view, the viewpoint is bound to be accepted as true by someone. Accepting scientifically credible approaches may well mean relinquishing any commitment to the truth of some theories that have been constructed wholly from clinical experience.
Public Standards of Treatment Effectiveness
In a world that values beliefs over evidence, it may be no surprise that legal and quasi-political bodies, rather than professional clinicians or scientists, often make decisions about what treatments are valuable or effective. Such decisions are made daily by health insurance agencies and legal bodies as they determine what practices will be covered by third-party payers and what ones are defined as malpractice (Aaron, 1996; Beutler, Bongar, & Shurkin, 1998; Nathan, 1998). Three standards have been reified in U.S. professional practices and reinforced in U.S. courts of law, and none of the three requires scientific study or empirical evidence.
Cost-Effectiveness is a criterion developed by health care systems and reflects political concerns with rising health care costs. From this perspective, the treatment that will be considered good enough or effective enough to be covered by managed care is one that serves a large number of people but costs less than the average services offered to those who share a given diagnosis (Aaron, 1996; Munoz, Hollon, McGrath, Rehm, & VandenBos, 1994). At this criterion's logical extreme, the most acceptable, valuable, and cost-effective treatment is one that is offered to many people but used by no one.
The standard of common practice is one that has been applied by legal bodies and by professional organizations who are concerned with minimalist criteria. The definition of malpractice arises from this so-called clinical standard. That is, court systems and peer review groups alike equate good-enough practice with what is most frequently done. This is the concept of the common or usual practice (Malcomb, 1986; Robertson, 1988). Stated another way, a treatment's effectiveness is equated with its popularity; taken to the standard's extreme, no new treatment could ever be judged to be appropriate or good enough.
The principle of the respectable minority was developed by the courts to acknowledge the vagaries in professions that foster divergent perspectives on any given problem. According to this principle, a treatment cannot be held to constitute malpractice if the treatment has an explicit theoretical foundation, a standard of delivery, and a significant number of followers. By case law, a good enough treatment is one that is believed by as few as six professionals (Klerman, 1990; Malcomb, 1986). Under this principles, almost all treatments are good enough.
None of these criteria excludes treatments whose effects are harmful. Indeed, in the medical malpractice case by which the principle of the respectable minority was defined, scientific evidence indicated that the treatment was harmful and was not known to have any positive effects (Hood v. Phillips, 1976; Leach v. Braillar, 1967). Likewise, neither the standard of common practice nor the standard of cost-effectiveness has proposed a successful solution to the situation in which a popular or cost-efficient treatment is ineffective. Thus, when a treatment fails to meet these standards, a practitioner may be censured, but the treatment is neither corrected nor removed from the list of acceptable practices. Even in the case of the deprogramming therapy of Braun, the treatment continues to be practiced though the courts did, in 1998, act to remove Braun's medical license (Gruman, 1998), a decision that is under appeal by Braun at this writing. Interestingly enough, Braun's appeal cites the court's intrusion into matters that are best left to clinical judgment.
It is apparent that another standard of effectiveness is needed to replace the ones based on good intentions and strong belief. This standard must account for the actual benefit induced and the actual danger of deterioration imposed. This alternative is generally thought to be found in the rigors of the scientific method.
Scientific evidence has a long history of unraveling the fallacies that exist when the criteria of truth are strong belief and sincerity. One need not go back to the conflict between Galileo and the Roman church to understand the resistance to scientific knowledge when it is contrary to popular opinion, personal belief, or so-called clinical experience. The prince of expounding contrary views has never been small. Remember that Galileo was sentenced to life imprisonment for preaching heliocentrism, and many would have put him to death. A much more recent example was illustrated in 1997, when PBS aired a Scientific American Frontiers segment in which a 10 year old girl developed a simple research project designed to test a cardinal assumption of a therapy that is widely used in medical settings. Therapeutic touch is an accepted practice in many places for enhancing healing and increasing one's response to treatments for cancer and chronic pain. This treatment assumes that the body gives off energy that can be felt by a trained practitioner who moves his or her hands above the body's surface. The energy is sensed through the patterns of warmth and coolness transmitted across the five or six inches of space between the body and the practitioner's hands. Clinical experience has confirmed that practitioners not only can feel, through their skin, the energy that is transmitted from another person but also can sense any disruption in the body's energy flow that are associated with disease. Then, as clinical evidence attests, by the passing of hands over that portion of the body in which the energy is disrupted, the disruption can be repaired, and healing is believed to occur. Unfortunately, for such claims, this young girl demonstrated that touch practitioners were unable to sense the presence of an energy field at all, making it highly unlikely that they were able to sense disruptions in this field, let alone induce a process of repair and healing. However, these findings seem to have been ignored, perhaps because they were made by a layperson. There has been no rush to give up touch therapy.
The Search for Scientific Standards
The development of a standard of evidence based on scientific findings and principles is a logical alternative to the problems encountered by reliance on the criteria of supposedly clinical evidence and experience. Yet, this road, too is uneven. Although professional practice and ethical guidelines have insisted that our practices be in accordance with scientific knowledge, psychologists seem unable to agree on what constitutes reasonable and adequate scientific evidence of a treatment's safety and benefit. Research studies vary in quality, often reach different conclusions, and do so with varying levels of relevance for clinical practice. How much evidence exists? What methodology was used? How strong must be the evidence be relative to the presence of inconclusive and non-supportive evidence? As judged by what criteria? These are among the questions that one must address in order to have reasonable assurance that the scientific evidence indicates that a psychosocial treatment is safe and effective.
Treatment Standards and Guidelines
There have been several efforts to define the nature of treatments that are associated with an increased likelihood of sustained improvement, as defined in one way or another by a review of available scientific evidence. In one form or another and by one term or another, these efforts result in what can loosely be considered to be guidelines for selecting treatments. Some of these guidelines have been privately developed, and some have been developed as extensions of various professional organizations and groups. More importantly is that most of these guidelines are based on subjective and qualitative impressions of research findings; others are based squarely on a political process; still others have attempted to apply the semblance of an objective test of a treatment's efficacy. By any of these means, practice according to the guidelines is always associated with a cost. I will provide a few examples of each, beginning with those that have relied on a quasi-political process.
The American Psychiatric Association (1993, 1997a, 1997b; Rush, 1993) developed a series of standards for treating various disorders such as depression, schizophrenia, Alzheimer's disease, and the like. The development of the guidelines paralleled, in a small way, the procedures used to develop the Diagnostic and Statistical Manual categories. For example, the guidelines for treating depression (American Psychiatric Association, 1993) were based on consensual agreement among a body of psychiatrists who were picked by the association's governance.
The costs of these guidelines has been substantial to the field of professional psychology. The procedures used assured that whatever guidelines were developed were a reflection of the biases of the American Psychiatric Association. Accordingly, the guidelines are very medical in their tone and discourage the use of psychological assessment (Nathan, 1998; Newman, 1993). Although a review of empirical literature was conducted to support each set of guidelines, the results of this review were dependent on the subjective impressions of the various reviewers and were then consolidated by a process of discussion until consensus was reached. The final act of submitting the results to vote was the capstone, implying that research findings can be ignored altogether if there is an active and consolidated electorate. To some, this is indeed a significant cost.