The alliance study is now published in the prestigious Journal of Consulting and Clinical Psychology (if I can boast a bit!): Anker, M., Owen, J., Duncan, B., & Sparks, J. (2010). The alliance in couple therapy: Partner influence, early change, and alliance patterns in a naturalistic sample. Journal of Consulting and Clinical Psychology, 78, 635–645. Congrats again to the crew!

A quick and dirty summary: N = 500 (total sample) n = 236 (subsample attending 4 or more sessions).  This study further supports both the feasibility and the importance of the feedback (PCOMS) intervention. The alliance significantly predicted outcome over and above early change, demonstrating that the alliance is not merely an artifact of client improvement but rather a force for change in and of itself. The study also found that those couples whose alliance scores ascended attained significantly better outcomes than those whose alliances scores did not improve. Together these findings suggest that therapists should not leave the alliance to chance but rather routinely assess it and discuss it with clients in each session.

Here is a discussion I had with the principal and onsite investigator, my friend, colleague, and partner in crime here at the Project, Morten Anker:

This study is significant for several reasons. First, as the Norway Feedback Study, it brings academic credibility to the use of the measures. Next it offers an antidote to the alliance nay-sayers who dismiss the alliance as only correlational despite the over 1000 studies that support the association between the alliance and outcome (this is like dismissing the association between cigarette smoking and lung cancer!) and often assert that the alliance is merely an artifact of early change. In other words, they argue that the alliance is confounded with early change and does not make any contribution beyond the client’s experience of improvement (clients experience positive alliances because they experience change). It is worthy question but it is often wielded in a way that undermines the importance of the alliance and the fact that it is second most replicated finding in the literature (the dodo bird verdict is first). This is one of only 6 studies that systematically examine this issue. Noteworthy is that we took a very stringent perspective of change. We looked at the clients who experienced reliable change so it was significant change, not just a little change. The alliance was predictive of outcome over and above early change even when that change exceeded 5 points on the ORS. Finally, the study, as mentioned, demonstrated the feasibility and benefit of routine alliance assessment.

On another note, a very interesting blog was sent to me by the author, a consumer. If you ever have any doubts about why collaborative monitoring of outcome and the alliance is a good idea, check this out from a consumer who says: “I’ve had a number of conversations with those harmed by therapy. A common thread is how resistant the profession seems to be to feedback from clients. To that end is my modest entry into the fray:” http://disequilibrium1.wordpress.com/2010/10/10/a-disgruntled-ex-psychotherapy-client-speaks-her-piece/

I recently did an exchange with a reviewer of The Heart and Soul of Change: Delivering What Works (2nd Ed.):

The Heart and Soul of the Dodo: A Review of The Heart and Soul of Change (2nd Ed.)

Thomas L. Rodebaugh

“The time has come,” the Walrus said, “To talk of many things.”

In The Heart and Soul of Change: Delivering What Works in Therapy, considerable attention is paid to establishing that Saul Rosenzweig was the original articulator of the dodo bird hypothesis: All psychotherapies work about equally effectively. Let us look closer at the source of the quotation, found in Alice in Wonderland, “Everyone has won, and all must have prizes!” (Carroll, 1865 and 1871/1998, p. 49).

In the story, an assortment of animals and the protagonist, Alice, have become drenched in a sea of Alice’s own tears. The ensuing “Caucus-race” (Carroll, 1865 and 1871/1998, p. 48) is the dodo’s invention to motivate the creatures to dry themselves off. It is not actually a race to be won, which is also demonstrated by the pitiful prizes: Each animal receives a single comfit (a candied, dried fruit). Because the animals eat all of those, Alice herself receives a thimble. More precisely, she keeps a thimble, because the comfits and the thimble were her own to begin with.

The dodo bird’s statement is not meant to be a hypothesis: It is meant to quiet the animals. Taken literally, the declaration regarding winners and prizes is clearly intended as nonsensical. The dodo, otherwise best known as a dead bird, is thereby made immortal as a purveyor of nonsense. Rosenzweig’s use of the dodo as a witty epigram some 74 years ago was inspired; that the dodo should live on as a metaphor for psychotherapy research so many years later strikes me as truly strange.

The dodo is a strong force in The Heart and Soul of Change. The book is a series of chapters by different authors but maintains a structure largely focused on the dodo bird hypothesis, its historical context, the research that can be taken to support it, and its implications for practice. Much of the rest of the book consists of further demonstrations that the dodo bird hypothesis is the most sensible interpretation of the data, set alongside critiques of empirically supported therapies (ESTs) and policies that support their adoption. Some later chapters focus primarily on what should be the next steps given that the dodo bird’s viewpoint is better supported than is a viewpoint that emphasizes ESTs.

Any adherents to ESTs who stumble upon the book might be forgiven for thinking they had accidentally landed in the mirror world described in Lewis Carroll’s other famous adventure for Alice: They are likely to cry foul, that evidence has been distorted and conclusions have been drawn contrariwise. Most (but not all) of the authors opine that ESTs offer no advantage and have been massively overblown and overpromoted.

Yet supporters of ESTs will probably already have to hand several recent challenges to the dodo (e.g., Ehlers et al., 2010). Among these counterpoints, I find particularly lucid Siev and Chambless’s (2007) demonstration that one must examine specific treatments for specific disorders to uncover differences between treatments. Supporters of ESTs might question why such findings are not responded to in this book. Certainly at least Siev and Chambless’s meta-analysis was available at the time of the writing of the chapters. Such apparent stacking of the deck does little to persuade people already inclined to support ESTs.

This book is clearly not aimed at such readers; neither is it, despite the title, primarily aimed at individuals looking for a how-to book regarding common factors in therapy. Although a chapter by Norcross, “The Therapeutic Relationship,” presents an excellent summary of these factors and the research that has investigated them, very little evidence is given as to how these factors can be better brought to bear in therapy. That is, although it seems clear that (for example) a stronger therapeutic alliance is desirable, there appears to be little systematic research available to establish that any particular intervention (e.g., a type of therapist training) necessarily improves alliance (although feedback, dealt with below, is held up as an exception to this general rule).

In fact, in another chapter, Wampold indicates that piecemeal investigations of one of the common factors cannot be conducted successfully: “The presence or absence of a common factor cannot be manipulated” (pp. 72–73). If this were accurate, then true experiments regarding common factors would be impossible and their causal role would remain unclear to the many researchers and clinicians who rely upon strong causal inference to understand the nature of treatment (cf. Borkovec & Miranda, 1999).

For whom, then, is the book intended? People who are amenable to the dodo bird hypothesis or find support of ESTs misguided are most likely to find the book palatable, and presumably this is the target audience. It seems likely that many of the authors would like policy makers to read the book, although I am not sure how likely that outcome is. Although it might seem a curious recommendation, I suggest that those who most strongly believe that ESTs are valuable could benefit from reading this book. I do not think this book will likely sway many such readers, but I do think it will be very helpful in illuminating the concerns of the researchers and clinicians who find adherence to ESTs misguided.

As most readers will have probably already guessed, I myself am convinced of the value of ESTs, at least for some disorders. Nevertheless, I can see many of the authors’ points. Although the repetitive dismissal of ESTs and related research, found chapter after chapter, seems excessive (like beating a dead dodo), my primary disappointment in the book is that it contains so little information regarding what changes an individual practitioner could make that are known to improve outcomes. In short, readers looking for guidance in employing the common factors (aside from feedback) might do better to read the Norcross chapter and follow it with seminal work by previous authors (I have my own favorites: Rogers, 1961; Wachtel, 1993) rather than read the entire book.

The major concept put forward for improving the common factors is gathering systematic feedback from clients, focusing on avoiding or mending ruptures in the therapeutic relationship; two full chapters (and additional space in other chapters) are devoted to demonstrating that such feedback is valuable and can have effects in community mental health organizations. These chapters appear longer on promise than on specific guidelines on what works and what does not.
Much additional research needs to be done, but the point regarding the general value of feedback is well taken and should be well considered by any practicing clinician. Devotees of cognitive therapy might nevertheless find perplexing the news that “of course, one need not choose between giving feedback and using empirically supported treatments. They can work in concert” (see Lambert’s chapter, “‘Yes, It Is Time for Clinicians to Routinely Monitor Treatment Outcome,” p. 249). Feedback from clients in each session has long been emphasized by cognitive therapists (Beck, 1995).

Such verbal feedback does not match the technical and statistical sophistication of the processes reviewed in this book, but the same intent is there. That Lambert needs to point out that ESTs and feedback are, in fact, compatible speaks to a very strange disconnect, the fissures of which seem to run throughout the book.

Perhaps my underwhelmed reaction to this book speaks merely to the effects of my allegiances. Of course, the authors and editors have allegiances of their own, although I wonder if they are as uniform in those allegiances as it might seem at first glance. Upon a closer inspection, it seems to me that a range of understandings of the dodo hypothesis is expressed across chapters.

In the weakest form, the argument seems to assert merely that ESTs may have been overemphasized by some and that common factors deserve more research. In its strongest form, the argument seems to assert that (a) anything that therapists and clients can believe is a therapy will work as well as any other such treatment; (b) common factors explain virtually everything about the way therapy works, yet there is probably little that could be mandated that could improve their effects; and (c) naturalistic tracking of outcomes is perhaps the sole exception to (b) and can also conclusively demonstrate that therapy is useful. In the strongest form, then, therapy and therapists are treated as a set of black boxes: There is no way to systematically alter the functions of these boxes, yet one can select therapists and therapist/client dyads on the basis of results.

I find myself concerned that some readers, perhaps most particularly those who see ESTs as a magnifier of the bureaucratic nightmare of insurance company requirements, might too easily endorse the strong dodo hypothesis. The position might seem attractive because it basically implies that therapists should be allowed to do whatever it is they do.

However, this position strikes me as pregnant with unwanted consequences. If good therapy entails a special quality (in the therapist, client, or both) that cannot be systematically varied (that is, caused to be present in some courses of therapy but not others), then one might wonder why anyone should research psychotherapy at all.

It seems to me that rather than the (strong) dodo hypothesis, we would be better off listening, but just for a moment, to the walrus hypothesis: The time has come to talk of many things. The field of psychotherapy needs more research, using many approaches, at all levels; it does not need an excuse to leave well enough alone.

However, research is not the only consequence of the strong dodo hypothesis. Practice, too, could suffer. If being a good therapist cannot be systematically taught, who would want to pay for years of training? One might wonder: Why not let anyone, with any level of training, try out being a therapist? One could simply select those people who are able to get the best results while accepting a minimum wage (perhaps the minimum wage) as payment.

It seems to me that the strong dodo hypothesis supports a form of essentialism that will not do science, practice, or policy any good at all. Neither supporters of ESTs nor their detractors want to see the therapeutic practice of clinical psychology go the way of the dodo.

Some Therapies Are More Equal than Others? A response to the review of The Heart and Soul of Change: Delivering What Works in Therapy (2nd ed.)

Barry L. Duncan

Rodebaugh (2010) candidly admits his allegiance to empirically supported treatments (EST), which perhaps explains the myopic lens used to examine the book. The dodo verdict (“Everybody has won and all must have prizes.”) still perfectly describes the state of affairs in psychotherapy—all bona fide approaches, in spite of vociferously argued differences, appear to work equally well. It is the most replicated finding in the outcome literature. Commenting on the dodo verdict’s ubiquity is hardly “stacking of the deck” when the findings that contradict it are less than would be attributable to chance alone. Importantly, saying that the dodo verdict persists in no way suggests that specific treatments for particular problems are not helpful.

While we take a critical stance toward claims of model superiority and confirm the veracity of the dodo verdict across modalities and populations, we do not denigrate model and technique nor specific effects, but rather propose that model/technique are essential components of a common factors perspective. We offered a way to understand how the alliance, expectancy, and model/technique are interdependent and overlapping. Technique is the alliance in action, carrying an explanation for the client’s difficulties and a remedy for them—an expression of the therapist’s belief that it could be helpful in hopes of engendering the same response in the client. Indeed, you cannot have an alliance without a treatment, an agreement between the client and therapist about how therapy will address the client’s goals. Similarly, you cannot have a positive expectation for change without a credible way for both the client and therapist to understand how change can happen.

We attempted to unite the warring factions via a more sophisticated understanding of change (interconnected factors, not disembodied parts or a tiresome specific v common factors polemic) as well as APA’s more contextual definition of evidence based practice. As the APA Task Force noted, the response of the client is variable and therefore must be monitored and treatment tailored accordingly to ensure a positive outcome. Proponents from both sides of the common versus specific factors aisle have recognized that outcome is not guaranteed, regardless of evidentiary support of a given technique or the expertise of the therapist. Monitoring outcome with clients, what has been called practice based evidence, has been shown to significantly improve outcomes regardless of the treatment administered. There are now nine RCTs showing the significant benefits of feedback (Duncan, 2010).

Rodebaugh’s assertion that one must examine specific treatments for specific disorders to uncover differences between treatments ignores the many direct comparisons that have not yielded any differences for specific disorders, like the TDCRP, Project Match, the Youth Cannabis Project, to mention a few (see Duncan et al., 2010). Consider the study we didn’t cite (Siev & Chambless, 2007). Although it is hard to imagine many therapists who would solely do relaxation training with panic, CBT beat relaxation alone on primary measures (although a closer look at the five studies reveals that one was significantly more positive than the other four, and two found very little difference). But even accepting this investigation at face value, that CBT is better than relaxation for panic (but not GAD) on primary measures only, hardly seems like any definitive overturn of the dodo verdict.

Nowhere in the book is there any suggestion that the dodo verdict implies that we should “leave well enough alone” regarding research, or perhaps the most egregious comment, that anything goes in the consulting room—or that there is little point to training. Quite the contrary, the book advocates for a shift toward research and training about what works and how to deliver it, and away from a sole reliance on comparative, “battle of the brands,” clinical trials. For example, my colleagues and I recently explored the relationship of the alliance to outcome and found that it predicted outcome above early treatment change and that ascending alliance scores were associated with better outcomes (Anker, Owen, Duncan, & Sparks, 2010), a strong argument for continuous alliance assessment. The book also calls for a more sophisticated clinician who chooses from a variety of orientations and methods to best fit client preferences and cultural values. Although there has not been convincing evidence for differential efficacy among approaches, there is indeed differential efficacy for the client in the room now—therapists need expertise in a broad range of intervention options, including ESTs, a point made by several authors.

Dismissing the book on the basis that some therapies are more equal than others is reminiscent of another set of animals in another classic story. It’s time to transcend the polemics and instead focus on what works with the client in my office now.

A Response to Barry L. Duncan

Thomas L. Rodebaugh

Let me emphasize that my reaction to The Heart and Soul of Change: Delivering What Works in Therapy was not uniformly negative. Further, I did not intend my review to be completely negative. I found the book useful overall; some chapters were particularly helpful. It would be a shame if the current debate were to overshadow that point.

The current format demands brevity. A point-by-point response to Barry L. Duncan (all the way down to Animal Farm) is untenable. The interested reader might re-examine my original review; my answers to some of Duncan’s statements are already implied there.

Allow me to focus on the term bona fide, upon which the current version of the dodo bird hypothesis rests. Bona fide treatments are treatments that are intended to be therapeutic. Intended by whom? Duncan expresses doubt that “many psychologists” would use relaxation treatment alone to treat panic disorder. I know one psychologist who would do so. I have informally polled my colleagues, who state that they have encountered others. Perhaps it is important that many psychologists believe that a treatment should work before it be considered bona fide. How many?

Without precise definition, whether something is bona fide is a subjective judgment. Studies could be dismissed because particular authors believe a treatment not to be bona fide or because they believe the researchers probably did not believe them to be bona fide, even if the researchers actually thought otherwise. I have had only modest experiences with clinical trials, but even I have seen many variations in level of belief at different levels of study teams. Sometimes therapists seemed to clearly believe more or less in particular conditions than did the principal investigator(s). Is it the therapists, investigators, or psychologists at large who count? Unless we define what level of belief is needed in the individual clinician or researcher, or how many psychologists must have such belief, our resulting decisions cannot be consistent (cf. Ehlers et al., 2010, for similar concerns).

Duncan seems to dismiss the idea that his argument indicates that “anything goes” in treatment. I can see his point, if bona fide means that “many psychologists” believe a treatment should work. We could thus be saved from endorsing ludicrous, fringe treatments. All the more reason to stringently define bona fide and thus reduce confusion among psychologists interpreting this literature.

Yet ineffective treatments sometimes have a popular following. As Ehlers et al. (2010) have pointed out, critical incident stress debriefing is certainly one example of a treatment that psychologists intended to be therapeutic but seems, upon investigation, possibly worse than useless. The hypothesis is that all (bona fide) treatments have won. To disprove it requires only one that has lost.

And don’t forget to register for the free webinars covering each chapter of On Becoming a Better Therapist: This month’s webinar covers Chapter 3 and will be on September 28th, 6-7:30PM Central. Register now at: https://www2.gotomeeting.com/register/945596986

I have been thinking about and writing about the common factors and their operationalization for many years. Research continues to build a compelling case for the presence of pantheoretical factors in operation that overshadow any perceived or presumed differences among approaches. For example, our alliance article soon to be published in the Journal of Consulting and Clincal Psychology found the alliance to be predictive of outcome over and above early treatment change and our in preparation investigation of therapist effects found that differences among therapists were best explained by their alliance abilities–over gender, discipline, or experience (more on both of these studies later). Some of you may have seen my depiction of the factors shown below:

I am always striving to describe the factors in a way that illustrates their interdependence. Here is my latest effort: Five factors comprise this perspective: client, therapist, alliance, the model/technique delivered, and feedback—all interdependent and overlapping. Technique is the alliance in action, carrying an explanation for the client’s difficulties and a remedy for them—an expression of the therapist’s belief that it could be helpful in hopes of engendering the same response in the client. Indeed, you cannot have an alliance without a treatment, an agreement between the client and therapist about how therapy will address the client’s goals. Similarly, you cannot have a positive expectation for change without a credible way for both the client and therapist to understand how change can happen. And the only way to know whether the common factors are in operation is to obtain real time client feedback about the benefit and fit of services. Feedback overlaps with and affects all the factors—it is the tie that binds them together—allowing the common factors to be delivered one client at a time. Soliciting systematic feedback is a living, ongoing process that engages clients in the collaborative monitoring of outcome, heightens hope for improvement, fits client preferences, maximizes therapist-client fit, and is itself a core feature of therapeutic change.

And I believe the only way to fully understand the importance of the common factors, including feedback, is to see them in action with real clients. Consequently, a new feature has been added to CDOI Members: actual client videos (client idenities are protected) are now available to Members for anytime viewing and learning.

Also, please check out the free webinars about my new book, On Becomng a Better Therapist. They are posted as they occur on the Video page, and the pdf of the slides as well as the videos are posted on the discussion page.

Finally, check out the following video made by Robyn Pope, a certified CDOI trainer of the Heart and Soul of Change Project, of staff at Wesley Community Action.  Wesley is a broad based social service agency that provides culturally sensitive and socially just services in non traditional settings. Robyn solicits staff reactions about the use of the Outcome Rating Scale and Session Rating Scale, including how they have grown professionally and personally as well as the challenges they have faced in implementation.

Some of you may remember that I (along with Jackie and the support of many others from the Heroicagencies List) took issue with a Newsweek article about a study published in the July, 2008 issue of American Journal of Psychiatry. (For the full story, visit http://chemicalimbalance.org) The study looked at two variables: SSRI prescription rates and suicide rates and compared these in various age groups. The authors suggested that a drop-off in prescribing caused by the Black Box warning led to increased suicide rates.

Newsweek was one of the media outlets that enthusiastically supported the study’s claims. Their article, “Trouble in a Black Box” was written by Tony Dokoupil.

We sent an email to the author and the editor of Newsweek pointing out the problems with the article. The problem was that an examination of the study revealed that the “parallel” development was not parallel at all. A graph in the study clearly demonstrated that the precipitous drop in prescriptions occurred after the increase in suicides. As you read this exchange, keep in mind that everything we pointed out was eventually acknowledged by both the New York Times and The Boston Globe – but not Newsweek.

Dear Editor:
Please find below our letter in response to the article, “Trouble in a Black Box.” Our examination of the study forming the basis for the article revealed a glaring inaccuracy–the study’s results do not match the findings reported in the Newsweek article. Given that very few individuals read or understand research, we believe it important for Newsweek readers to be aware of this discrepancy to evaluate the necessity of the Black Box warning:

Tony Dokoupil’s Trouble in a ‘Black Box’ (July 16) importantly addresses the risks and benefits of prescribing antidepressants to children. However, the referenced study is far from “compelling” evidence for removing the FDA Black Box warning and such an interpretation of its findings is misleading. An inspection of this industry funded study reveals that the precipitous drop in SSRI prescriptions did not occur, as reported, from 2003 to 2005 but rather from February to October of 2005 (over 85% of the drop in the last 6 months of the reported time). The so-called “parallel development” of increased suicides occurred between 2003 and 2004—and therefore had no relationship to the drop in prescription rates reported in this study. Given that the decrease in prescription rates and increase in suicides occurred in different time periods, it begs the question of how such unsubstantiated statements could be made by the experts cited in the article.

Only 3 of 15 clinical trials have shown antidepressants to be superior to a sugar pill on primary measures. Children and parents in those 15 studies reported no advantage of antidepressants over a sugar pill. Data from the FDA and its British counterpart demonstrate that children and adolescents taking antidepressants are twice as likely to experience suicide-related events. Given the meager results and increased risk for suicide-related events (as well as other serious adverse events), antidepressants are not a good first choice for youth struggling with depression—a conclusion reached after an extensive risk/benefit analysis conducted by the American Psychological Association’s Work Group on Psychotropic Medication

Dear Dr. Duncan:
Thank you for responding to my recent story, “Trouble in a Black Box.” In answer to your concerns, I writing to let you that the important “parallel development” we had in mind was that child and adolescent use of antidepressants dropped (for the first time since coming on the market) while suicides rose significantly for the first time since the late 1970s. The fact that antidepressant use dropped most steeply in 2005, the year for which suicide data is still forthcoming, may merely foreshadow trouble–which is why the FDA is concerned. “The evidence is very compelling,” they say.
Tony Dokoupil

Dear Mr. Dokoupil:
Thanks for your note. I understand the proposed relationship between the increase in suicides and decrease in prescriptions that your article and the cited experts were asserting. But the evidence for such a relationship is far from “compelling” when the two developments occurred over different time periods. Your article and the cited experts gave the impression that the “parallel development,” as the word “parallel” suggests, occurred during the same period and therefore were related to one another. This was quite misleading given that an inspection of the study and its graphs revealed something quite different. Had you only commented that it could be a foreshadowing, and offered other explanations, then the article would have not so misrepresented the data. I would appreciate your clarification for Newsweek readers or that my letter be published.

Dear Ms. Lichtschein and Mr. Dokoupil:
I would greatly appreciate knowing your decision regarding our letter. It is of course your perrogative to print or not print any letter you receive. In this case, however, we believe it is critically important for you to get the facts straight because of the unfortunate misrepresentation of the data reported in your article. It is particularly troubling given the bold and even outrageous comments made by the cited experts which, at times, bordered on hysteria and fear mongering, far removed from an objective interpretation of the facts. Perhaps an interesting story would address how the drug company affiliated researchers responded to the findings as a “parallel development” while understanding full well that the precipitous drop in their study occurred after the increase in suicides.

Dear Dr. Duncan,
We appreciate your concerns, but don’t have plans to run your letter. We feel that the story adequately expresses the available data, which concerns the simultaneous snapping of two 15 year trends. The fact that the sharpest drop in antidepressant use occurs in the year for which suicide data is still forthcoming is significant, but more significant in this context is the reversal of a steep, longstanding trend toward increased SSRI use. We also sought comment from experts with interests on all sides of the issue.
Tony Dokoupil

Dear Mr. Dokoupil:
Thank you for your response. Just for the record, you are stating that you will not print our letter because “the story accurately expresses the available data.” We have, in fact, shown that, based on data from the cited study, it does not. The “simultaneous snapping” of the two trends is clearly neither simultaneous nor “parallel” as your story and letter depict. Nor is there enough available evidence to make any definitive statement about youth suicide trends for the time frame mentioned. Perhaps you “sought” consultation from experts on both sides. However, you did not publish both sides. Nowhere in the story are there opposing points of view or other possible interpretations for any trends the existing data might foreshadow. We must assume that, based on your refusal to publish a valid and important counterpoint, it is your intention to keep counter voices from your readers. Not only does this do a disservice to your readers, it walks a perilous line. It reflects how a major media source can, through biased reporting, create unjustified fear and potentially influence the repeal of a warning label implemented, after extensive scientific debate, to protect children. It is our belief that any observer of this process may likely view it as we do—representative of neither balanced nor ethical journalism, especially as it involves a life-and-death issue impacting our nation’s youth.

Thus, while The New York Times and The Boston Globe addressed the problems Newsweek never did. Strike one.

Particularly egregious were the comments in the Newsweek article by Robert Vuluck: “You may induce two suicides by treatment, but by stopping treatment you’re going to lose dozens to hundreds of kids. You’re losing more than you’re saving. That’s the calculus,’ says Dr. Robert Valuck, of the University of Colorado Health Sciences Center.”

While sloppy journalism and lack of fact checking is expected from the media, the bold and even outrageous comments that Valuck made went well beyond just an unfortunate misrepresentation of the data. His comments bordered on hysteria and fear mongering, far removed from an objective interpretation of the facts and offered a conclusion from the data that he must have known, as a study co-author, to be false. This is a serious ethical violation because it created unjustified fear and could potentially influence the repeal of a warning label implemented, after extensive scientific debate, to protect children. This was neither balanced nor ethical science especially as it involves a life-and-death issue impacting our nation’s youth.

Researchers, especially those funded by corporate interests have to be held accountable. We tried to hold him accountable but the University of Colorado Committee on Research Ethics (CRE), specifically, John E. Repine, M.D. who informed us that the CRE did not find sufficient evidence to warrant an inquiry. Strike two.

That’s the past. But these things happen all the time. Case in point. Consider an article just out in the New Yorker:

http://www.newyorker.com/arts/critics/atlarge/2010/03/01/100301crat_atlarge_menand  

It is a review of two new books about antidepressants. It has many redeeming qualities in that it acknowledges the problems with diagnosis and the research suggesting that antidepressants are no more effective than placebo. But it’s all downhill after that, and begins making a not so hidden argument for antidepressants. It purports to do a pro/con analysis of the drugs and psychotherapy but makes two incredulous omissions. One is that it doesn’t discuss, at all, the adverse effects of antidepressants; and two, it fails to mention long term comparisons between psychotherapy and medication. Finally the article cites (and significantly misrepresents) the STAR*D study to support antidepressants, and to refute the claim that SSRIs are no better than placebo.

After I read it, there was no turning back. I had to respond. Do any of you know a cure for that? I included Jackie and my analysis of the STAR*D study from the new Heart and Soul of Change. Given it calls into question their fact checking, I doubt it will be published. Strike three, Barry, yer out!

Here it is:
In the entertaining review, “Head Case,” Menand effectively covered the controversy surrounding psychiatric drugs. Two important issues, however, relevant to assessing the risks and benefits of psychiatric medication were omitted: the significant side effect profile of antidepressants and the poor performance of antidepressants in the long haul (especially as compared with psychotherapy). The STAR*D study illustrates. In the STAR*D, the average remission rate based on the primary outcome measure was 28% and 25% on the first two levels, and 14% and 13% on the last two. At Level 1, 28% experienced moderate to intolerable side effects. At Level 2 (participants augmented or switched), 51% experienced side effects ranging from moderate to intolerable. For all levels, 24% exited due to drug intolerability. Data from the 12-month follow-up of those who either remitted or responded indicated a relapse rate of 58%. Menand’s review omits the significant number of STAR*D drop-outs and claims that a 67% effectiveness rate after all levels. This figure was apparently derived by cumulatively adding percentage rates across levels, a practice statistically meaningless and certainly misleading. Since the rates of effectiveness are calculated from the numbers of participants in each level, average, not cumulative percentages correctly reflect overall improvement. For example, in the first two levels, out of a total of 4,168 participants, 1114 achieved remission, a 27% effectiveness rate, not the 56% rate reported by Menand. When drug studies are reported, original sources must be examined to insure that risks and long term follow up are considered.

Hope to see you soon in New Orleans!