Finally I have become part of the machine. – Brian Eno
There is currently a movement within psychology promoting what are called “evidence-based” treatments as offering the best means of addressing patients’ emotional difficulties therapeutically. Here the term “evidence” refers to data gleaned from statistical scientific research, data that is thought to provide the most dependable guide to selecting techniques yielding the highest recovery rates for the greatest numbers of patients.
Psychotherapists are now being subjected to pressure from insurance companies and certain factions of academic researchers within their field to adopt “evidence-based” approaches. Increasing numbers of insurers, mental health professional organizations, and psychotherapy training programs now embrace and aggressively promote “evidence-based” treatments as the new standard for what are often referred to as “best practices”. Some insurers even require that therapists in their networks agree to utilize only treatments considered “evidence-based”.
I wish to sound a warning about this trend in psychotherapy research, a development which I see as unhelpful and even harmful to emotionally disturbed individuals. My objections to the shift toward “evidence-based” therapies are two-fold. First, this trend is based in philosophically dubious assumptions about human nature, adhering to a model of the mind’s structure that seriously misreads what I take to be its essential nature. These wrong-headed assumptions then become the justifications for treatment approaches that do not address the actual essential features of human emotional life, and, hence, are not likely to produce treatment approaches that provide genuine help to suffering persons. My second objection is based in a discovery I made while reviewing some of the research literature on “evidence-based” therapies. It is one that took me completely by surprise: namely, that “evidence-based” treatments demonstrably do not provide optimal therapeutic results, even when these outcomes are evaluated by the standards of its own criteria for success.
Yes, you read that last sentence correctly.
I will explain each of these objections in turn. First, let us examine the nature of the philosophical assumptions about human nature and emotional suffering that is embedded within “evidence-based” research into therapeutic effectiveness. We should begin by exploring what this model of treatment defines as “evidence”. As used in this context, the term refers to measurable, quantifiable behavior only. This approach reflects a philosophical presumption about how we come to know something as “true” called objectivism. When applied to psychological topics, objectivist approaches invariably favor quantifiable behaviors as providing the most reliable data for judging a treatment’s effectiveness, because behavior is objectively observable, hence largely immune from observers’ distorting biases, it is thought. Here the entire realm of human subjectivity, the patient’s private thoughts, wishes, and fantasies, is either ignored completely or assigned a place of secondary importance in determining therapeutic success.
It is vitally important to note that adherence to objective behavior as the cardinal standard for assessing success carries with it the unstated assumption that behavior expressing psychological “health” looks the same in all people. Specifically, the “evidence” that “evidence-based” treatments use to determine success are culled from research studies of large samples of subjects. Those therapies that show the largest number of subjects displaying observable, quantifiable behaviors deemed beforehand to be “functional” or “healthy” are then selected as constituting “best practices”. Hence, this means that treatment success is narrowly determined by assessing if and how completely a given patient’s behavior changes to look more or less like everyone else’s: the more a patient acts like most other people in his or her age group and social surroundings, the more effective the treatment is thought to have been.
We can immediately infer serious flaws in such an objectivist approach to human nature, when applied uncritically to the analysis of behavior and emotional problems. For one, it is doubtful that someone whose behavior conforms to social norms is necessarily living a life that is emotionally fulfilling, fruitful, or, for that matter, even lucid, rational, and/or balanced. As psychoanalyst and social philosopher Erich Fromm said, “The fact that millions of people share the same mental pathology does not make these people sane”. Conformity to normative definitions of good psychological functioning as the main criterion for assessing a therapy’s helpfulness is not only illogical, but possibly psychologically damaging to the patient, inasmuch as conformity can, and often does deaden one’s ability to respond to life with a sense of initiative, vigor, purpose, and generosity that is felt to issue forth spontaneously from within the self, that is, from one’s subjectivity.
So there are real dangers to eliminating or minimizing considerations of human subjectivity in the therapeutic endeavor. Now, this is not at all to say that behavior has no important role in determining therapeutic progress. Quite the opposite: someone who acts in ways that most people agree are, for instance, hurtful to others and/or themselves, should be sensitively engaged in a discussion about the meaning(s) of these behaviors. My only objection to the objectivism in “evidenced-based” research is that it tacitly assumes that we know something to be “true” only when we can count, sort, and categorize it. In this vision of reality anything subjective is assumed to provide little or perhaps nothing of value to the therapist’s understanding of the patient, ipso facto. Now, this is like saying that Melville’s “Moby Dick” or the “Star Wars” films adds nothing to our understanding of life because the tales they tell didn’t “really” happen; that is, that these dramatic creations are incapable of revealing truth about the human condition because they do not report objective occurrences. Not only do I reject such an assertion as utterly simplistic, as likely do most readers of this essay, but I will go so far as to say that the most enduring and profound truths about humanity are often revealed only by envisioning and formulating them imaginatively, in the dream-like privacy of the world of mind. In this case one appropriates aspects of the objective data of experience, and then retreats from the realm of consensual meaning-making altogether, withdrawing into the self where new, perhaps unprecedented meaning is crafted (or, perhaps, realized) through the action of reverie. Reverie is the playful and unpremeditated joining together of features of one’s experience of existence that, at first blush, may appear completely unrelated, but which over time are revealed to be intimately bound together. As an aside, I will say that the analogy to sexual intercourse might be apparent here…which is why the Judeo-Christian scriptures quite often and accurately refer to a man’s sexual pleasure with a woman by saying that “he knew her”.
In this way unprecedented new understandings of Being are often fertilized (yes, sorry, another sexual reference), nurtured, and, one hopes, eventually introduced into the objective social world. This is to the benefit of other people, who may come to realize some new stance to be taken in relation to a difficult aspect of existence. When this happens, people’s lives in the objective world are usually enhanced in some measurable way, which should cause us to pause before dismissing something as “only” a product of the imagination, and hence as impractical. In fact, I would say that any genuine contact with the world is possible only by an act of imagination, in that imagination alone has the capacity to tease out and make plain the meaning(s) of phenomena, hence, how a phenomenon will effect us pragmatically, and the possible behaviors and/or attitudes that we can generate in response to that phenomenon. To put it more simply, the objective world does not come with a set of definite meanings attached to it. The things that happen in the world are devoid of meaning until some human observer ascribes a certain significance to them. Things only “mean” something because they mean something to a particular person, that is to say, to a subject. The objective world is real, but only takes on existence in a complete sense when encountered and interpreted by a particular human consciousness. In a sense, the subject creates and animates objective reality.
Observable behavior is critically important in forming judgements about how to best help patients, once it is understood by an interpreting subject. However, “evidence-based” research makes the error of assuming that the meaning(s) and desirability of actions exist in some pristine realm, as self-evident truths detached from human interiority, that is, apart from the subjective realm in which significance and connotation are assigned to them by an individual. Proceeding thoughtlessly from this bias, “evidence-based” research aims to create an unassailable list of truly effective therapies by eliminating or minimizing considerations of human subjectivity, which are presumed to be roadblocks to our ability to clearly judge reality, in the investigative process. That is to say, “evidence-based” models of treatment view human subjectivity as an obstruction to arriving at a “true” understanding of a person’s mental state. In contrast to this view, I say that any given behavior is only made comprehensible when interpreted with an eye to what it reveals about the intangible and unique meaning-making structures bestowed on them by an individual. That is to say, behavior alone does not reflect any given, static set of meanings to which we can confidently point when determining its relationship to a given patient’s difficulties. For example, someone may be pleasant and accommodating toward others, which an objectivist worldview takes as evidence of satisfactory social adjustment. But this inference is exposed as relative, not absolute, when one considers that objectively-observable pleasantness may imply, not goodwill toward others, but other things, up to and including the very opposite of goodwill, as when seemingly generous behavior is employed as a defense masking unconscious anger and disillusionment.
Another example of this phenomenon is what psychoanalysis refers to as the “flight into health”. This describes the behavior of a patient who suddenly makes a dramatic rebound after initially complaining of struggling with dark emotional states. The patient may broach the idea of terminating treatment at this time, pointing out that they have benefitted greatly and are now ready to resume normal life. Now, the observable presentation of such a person is often fully consistent with what we know of the behavior of those who are psychologically well-adjusted. But psychoanalytic theory suggests that such an abrupt recovery should not be accepted at face value. Rather, it must be analyzed over time, to tease out if it is perhaps symptomatic of an unconscious resistance to delving further in therapy into the hidden psychic origins of an emotional problem, which, arguably, is what creates the more lasting and substantive structural change in the personality required for a genuine recovery. The therapist who cannot or will not make the inferential leap beyond observable behavior, into considerations of the unconscious conflicts possibly driving their patients’ superficial presentations (such as, in this case, the dread of change) is ill-equipped to facilitate real resolution of their emotional difficulties. All this to say that the fact of human subjectivity can and constantly does act as a confounding factor in drawing anything like clear and consistent causal relationships between behavior and actual psychological states. Turning the human subject into an object not only promotes an unimaginative and soulless picture of the human condition, but one that is fatally flawed.
Attention to the subjective realm demands that the therapist approach so-called “symptomatic” behavior, such as drug abuse, social isolation, or compulsive behavior, with a certain humility, one borne of the awareness that he or she cannot quickly determine and classify the relative “health” of any given action. Factors in the person’s subjectivity means that the “true” or “actual” meaning of a given behavior is far from certain, and may remain so for an indeterminate length of time, even permanently. Subjectivity cannot be assessed solely by objective standards. We can and should draw upon social norms and other, objective standards when faced with the task of understanding an individual’s emotional struggle. But this must always be accompanied by an equal respect for the interior realm, the immaterial and hence opaque dimension of “mind” that is incompletely accessible to the kinds of measurements by which we come to know the material, physical world (and in certain, albeit rare, cases may be grasped only by completely ignoring objective, standardized categories of good adaptation). This perspective assumes that human subjective life possesses a certain fundamental existential freedom, one that liberates it from slavish dependence on the interpersonal and social realms when settling on the meaning-making schemes that will inform and guide one’s attitudes and actions in the world. The element of free will in the interior realm means that a person’s behaviors do not necessarily correspond logically to the taken-for-granted assumptions by which we typically infer the existence of a particular mental state, including someone’s motivations. The correspondence between these factors is not written in stone, but is perpetually tentative and often opaque. So we “miss the boat”, interpretively, by adopting to a naively simplistic view of the correspondence between observable behavior and inner life.
Psychologist Theodore Millon, a cognitive behaviorist, has established a popular description of personality types, including his interpretations of the kinds of environments and childhood experiences that predispose an individual to certain personality pathologies. Take his understanding of the etiology of the narcissistic individual, for example. As a cognitive-behaviorist, Millon defines the narcissistic “style” according to observable behaviors reflecting common ideas about how a pathologically self-preoccupied person engages with others, i.e., in an aggressively manipulative, demeaning, dismissive, and openly arrogant manner designed to ensure that other persons “mirror” and accede to their inner needs and wishes. Based on this model of cause-and-effect, Millon reasons that the adult narcissist got that way because he or she was overindulged in childhood, permitted to engage in usury, exploitative behaviors that impair social adjustment in adulthood; in short, they are spoiled. His proscription for treating this condition consists in repeated confrontation of narcissistic behavior, which he presumes can result in a measurable decrease in objective instances of maladaptive interactions.
Due to the inherent interpretive limitations of the cognitive-behavioral model, Millon cannot account for the existence of narcissistic behavior that do not conform to popular understandings of it as only observably aggressive. For example, in addition to those narcissistic characters that are openly arrogant are those who are anything but rancorous, but rather behave in a timid, withdrawn manner. These introverted narcissists harbor the same longing as the aggressive, extraverted type for others to “mirror” their minds. However, they may have adopted a passive stance toward others, mutely waiting for them to intuit and respond to their unarticulated wish to be understood. Or they may have given up the fight, silently nursing unconscious resentment toward others’ perceived empathic failures from a position of social isolation. Further, these individuals nurture the same angrily demeaning perception of others as the aggressive type, though may be viewed by those in their circle as simply shy and/or inexplicably depressive.
A further shortcoming of Millon’s theory of the origins of narcissism is what I consider to be its proscription that the aggressive narcissist be reeducated through confrontation. While tactful confrontation is often appropriate in all treatments, Millon’s assumption that narcissism stems from being “spoiled” in childhood fails to grasp or address the aggressive narcissist’s unconscious experience of their parents’ repeated dismissal of their need for empathic “mirroring” of their minds, their blooming aspirations and wishes for acceptance of their authentic expressions of self during childhood. Because the cognitive-behavioral view of human nature adheres to a correspondence theory of truth, it tacitly presumes that arrogant behavior only reflects overweening pride, that is, an “oversized ego” that must be cut down to size. While grandiosity is certainly a dimension of narcissistic arrogance, the cognitive-behavioral therapist fails to account for and interpret the presence of the unconscious self-hatred that I believe is a primary impetus for narcissistic behavior. I see this technique as unhelpful and even harmful in addressing the central dynamics of the narcissistically-injured personality, in that it actually inflicts additional narcissistic injury on the patient by aggressively underscoring his or her moral failure to understand and accept others. That is, it is a coercive technique, thus one likely to create further shame and self-loathing in the patient.
Now to the second of my two objections to “evidence-based” therapies. It is one that is likely to be shocking and possibly unbelievable to those heavily committed to research-based, objective validation of treatment methods; as I noted above, it is a finding that took me completely by surprise. Stated simply, “evidence-based” treatments are largely ineffective, even when their outcomes are judged by the medicalized, objectivist premises about human nature that are described above. That is, these therapies do not live up to their own standards for assessing what is and is not a superior treatment, standards that are philosophically dubious in the first place, as I have argued.
This finding is documented by Jonathan Shedler, Ph.D., in his recent review of the literature entitled “Where is the Evidence for ‘Evidence-Based’ Therapy?” (2015). Shedler analyzes in depth the “evidence” that these research paradigms rely upon to support their classification of certain therapies as objectively superior to others, and also cites the work of other social scientists who have conducted extensive literature reviews about this topic and have summarized the overall trends that they reveal. His conclusion is damning indeed, begging the unsettling conclusion that the entire enterprise of identifying “evidence-based” treatments is largely a sham whose believability is compromised from the start by distorted research methodologies yielding interpretations of data that are hopelessly skewed and misleading. (Shedler seems to be a gentleman, and so generally uses more tactful language when presenting his case, though I doubt he’d disagree with my harsher characterization of his findings.)
To begin with Shedler’s grand conclusion: “Research shows that ‘evidence-based’ therapies are weak treatments. Their benefits are trivial. Most patients do not get well. Even the trivial benefits do not last” (p.48). He starts by citing flawed research paradigms as creating biased research samples and bogus conclusions in “evidence-based” investigations. Hence, for example, he notes that in typical randomized controlled trial for these treatments, “about two thirds of the patients get excluded from the studies a priori” (p. 52). Shedler reveals that generally those subjects accepted to be participants in “evidence-based” research have met an unrealistically narrow set of criteria for inclusion. Applicants who are excluded from the outset are overwhelmingly identified as having more than one diagnosis, suffer from a personality disorder, are notably unstable in some way, and/or report feeling suicidal. This accounts for two-thirds of all potential study candidates, on average. In other words, the very kinds of patients that constitute the bulk of most clinician’s caseloads are never counted.
It gets better. Says Shedler, “Of the one-third that do get included, about half show improvement. So we are now down to about 16% of the patients who initially sought treatment. But that is just patients who show ‘improvement’. If we consider the percentage that (study outcome data indicate) get well and stay well, we are down to roughly 5%” (pg. 53). That’s right: 5%. And this figure includes only patients with less severe psychological problems. No one would buy a car that is shown to run dependably only 5% of the time. But “evidence-based” researchers seem to have no qualms about using this finding to uphold the supposed desirability of their chosen theories and techniques.
Shedler goes on to report that these studies of “evidence-based” approaches almost never compare their preferred treatments to what he calls “legitimate alternative therapies” (p. 54) . He writes, “The control group is usually a foil invented by researchers who are committed to demonstrating the benefits of CBT” (p. 54). CBT stands for cognitive-behavioral therapy, an objectifying, medicalized variety of psychotherapy that sees humans as stimulus-response organisms, and attempts to “cure” psychological disorders by suppressing or replacing problem behaviors, and the thoughts that are supposed to maintain them, in a mechanistic way.
A brief detour is called for here. In the 1960’s CBT was lauded as ushering in a new era of truly scientific psychological treatments, that is, approaches that rely on attending only to the quantifiable dimension of human behavior to define their mission, rather than things like self-understanding and insight. These latter ideas were dismissed as fuzzy and imprecise concepts of an earlier, more ignorant era, with Freud reviled for setting into motion the emphasis on coming to terms with one’s subjective perceptions and goals in creating and maintaining emotional suffering. In doing so, CBT became the enemy of the millennia-old Western devotion to the notion of human subjectivity, that is, that we are driven by forces and desires that pitch our lives in certain directions.
Hence, for an extended period in the late 1950’s though the mid-1970’s, CBT proponents insisted that the notion of “personality” was outmoded and unverifiable, and that attempts to describe human nature should be replaced by examination of measurable behavior only. For example, as noted above, this is the paradigm that Millon uses to support his theory of personality, which arguably is not a genuine theory of personality at all, inasmuch as it dismisses the relevance, and possibly the existence, of the subjective dimension of human nature that we take to be the origin of the experience of being a coherent “self”. Although it remains with us, the cognitive-behavioral crusade failed to achieve its grand goal, in part because it runs contrary to what almost everyone, with the exception of severe psychopaths and chronic psychotics, feels to be most true about their existence: that they desire, fantasize, and weave meanings in highly idiosyncratic, purposeful ways, and that these factors constitute the very “stuff” of that which directs their lives, for good or ill. This is why Freud focused so obsessively on sex and aggression: we are never so attuned to our wishes and motivations when aroused by passionate love or hatred.
So, back to Shedler’s account of “evidence-based” researchers’ use of phony control groups. He describes a widely publicized NIMH (National Institute of Mental Health) study of PTSD (post-traumatic stress disorder). The study concluded that PTSD patients benefited significantly more from the use of “evidence-based” “prolonged exposure therapy” than from psychoanalytically-informed approaches. (“Prolonged exposure therapy” attempts to desensitize traumatized persons to triggering stimuli by having them purposefully expose themselves to it, in the form of memories and/or actual contact with environments like those in which the traumatizing event occurred.) In reviewing the “fine print” in the NIMH account of this study, Shedler discovered that the control group, in his words, “received exactly two days of training in psychodynamic therapy from another graduate student – a graduate student in a research laboratory committed to CBT. In contrast, the therapists who provided CBT were trained for five days by the developer of this form of therapy, world-famous clinician and researcher Edna Foa” (p. 53, italics his).
As if the above was not enough to undermine the study’s validity, Shedler goes on to state that, in his words, “The so-called psychodynamic therapists were also forbidden to discuss the trauma that brought the patient to treatment…When the patients brought up the trauma, the therapists were instructed to change the topic” (p.54, italics his). He does not cite a rationale for this bizarre tactic. Yet, obviously, if anyone engaged a patient like this in the real world, it would be deemed gross incompetence or worse.
Shedler is not simply “cherry-picking” particularly questionable research projects to build his case. He describes a comprehensive review of the research of “evidence-based” studies that concluded that these approaches are superior to others. Wampold et. al. (2011) examined controlled trials for depression and anxiety, over 2,500 abstracts total. Over time, they eliminated all but 149 studies. This remaining group appeared to actually legitimately compared “evidence-based” paradigms with other approaches. When these were reviewed in depth, they identified a mere 14 studies that proved to compare “evidence-based” therapies to a control group that received what Shedler calls “anything approaching real psychotherapy” (p. 54). Hence, Shedler’s shocking claim: “Many of the studies claimed to use a control group that received (what they referred to as)’treatment as usual’. But ‘treatment as usual’ turned out to be ‘predominantly treatments’ that did not include any psychotherapy” (p. 54, italics his).
This assertion is so unbelievable that Shedler understandably goes to some lengths to assures the reader that it is not an inference on his part, but a direct quotation from Wampold et. al.’s article. This literature review concludes by stating that there is scant evidence for the supposed superiority of “evidence-based” treatments. Shedler concludes this section by reminding us that this is also, in his words, “the formal scientific conclusion of the American Psychological Association” (American Psychological Association, 2008).
Shedler concludes by noting that research whose findings contradict the presumptive supremacy of “evidence-based” modalities are suppressed, specifically, that they often fail to get published because of a form of what is called “publication bias”. This is a tendency of peer reviewers to favor studies with positive results, rather than those that show no support for the researchers’ hypothesis. He draws on statistical methods that demonstrate the percentage of unpublished studies containing negative finds, and concludes that, if they were published, the actual benefits of “evidence-based” CBT therapies would be reduced by three quarters. This means that only 25 percent of these treatments would have been demonstrated to produce superior outcomes.
There you have it: two compelling arguments for abandoning the conclusions of “evidence-based” studies, the first philosophical in nature, the second derived from data. Either one provides sufficient reason, by itself, to undermine our confidence in this trend in research.
Now, as we (finally) move toward the end of this essay, the question before us is one best posed scientifically…what the fuck? Is it really possible that “evidence-based” research is done with this level of gross incompetence, and without at least a handful of people sensing that something isn’t right and speaking up about it? Well, I suggest that those who design and carry out these monumentally flawed studies are under the spell of a collective worldview that renders them unconsciously poised to ignore, minimize, or rationalize awareness of the serious flaws in their endeavors. This is a topic so complex that it requires its own essay, so here I will simply touch upon some general elements of this intoxifying and apparently immobilizing psychosocial force. First is the natural human desire to have one’s biases confirmed, paired with the fact that researchers are rewarded by their parent institutions for sponsoring studies promising “ground breaking” advances in their fields. And in the background of these factors is the Western cultural emphasis on controlling natural forces, including those constituting the mind, through the scientific method. These factors work conjointly, and unconsciously, to render “evidence-based” studies illegitimate. They seduce proponents of this paradigm to adopt the utterly unscientific bias that their pet theory must be protected at any cost, even at the expense of sacrificing the allegiance to the pursuit of scientific truth, and seriously misleading other professionals and patients about constitutes “good therapy”. The latter offense strikes me as the most alarming, egregious, and disgraceful of the two, in that it causes suffering persons to receive shabby, ineffective treatments.
So, in conclusion, let us reject the notion that we are entering a “brave new world” of psychotherapy techniques, one that has dethroned the alleged flaws and mistakes of those like Freud, Jung, and others. These pioneering thinkers have always been our best guides to effective practice, and remain so.
Shedler, Jonathan (2015). “Where is the Evidence for ‘Evidence-Based’ Therapy?” The Journal of Psychological Therapies in Primary Care (Vol. 4, May 2015).