A Psychiatrist Reads the Washington Post

There are an endless number of ways that the appearance of conflict of interest can be spun to make any organization look bad.  The obvious question is why that always seems to occur with psychiatry?  The arguments all follow the general form that a financial benefit resulting from work related to the pharmaceutical industry disqualifies those experts from writing objective research about medication or rendering opinions about the treatment of psychiatric disorders in general. That is the theme of the latest article from The Washington Post entitled “Antidepressants treat grief? Psychiatry panelists with ties to drug industry say yes."  It is an old story with little variation and I add some commentary based on the organization of the article.

"In what some prominent critics have called a bonanza for drug companies, the American Psychiatric Association this month voted to drop the old wording against diagnosing depression in the bereaved, opening the way for more of them to be diagnosed with major depression and thus, treated with antidepressants.”

This statement assumes that this practice is not occurring right now. In fact, it is widely known that the diagnosis of depression is not rigorously made in primary care settings. It is highly likely right now that patients suffering from grief as well as psychological adaptations to acute stress are being treated with antidepressants. There is no reason to believe that the patients being treated in primary care resemble the patients with a diagnosis of major depression in clinical trials of antidepressants.

"The change in the handbook, which could have significant financial implications for the $10 billion  US antidepressant market, was developed in large part by people affiliated with the pharmaceutical industry, an examination of financial disclosures shows.”

The previous statement talks about a "bonanza for drug companies" and builds on this image in the second statement. It ignores the fact that most commonly prescribed antidepressants are currently generics and available for as little as four dollars per month. The only two major antidepressants at this time that are not generics are Cymbalta (duloxetine) and Vibryd (vilazodone).  Where does the "10 billion dollar" figure come from?  If you read the entire article on page 5, that figure was from IMS America a company that tracks total prescriptions from American retail pharmacies.  Anyone knowing the applications for antidepressants would know that they are prescribed for many conditions other than depression including headaches, hot flashes, and chronic pain. The total retail sales figure is unlikely to reflect either drug company profits or the amount of depression being treated.

A little arithmetic is always instructive. If we assume that a physician prescribes a generic antidepressant for a patient that costs four dollars per month that translates to a total cost of $48 per year. The $10 billion/year figure quoted here would represent 208 million prescriptions or 66% of the entire population of  the U.S. taking antidepressants 12 months out of the year.  Even if we take $2 billion out of the $10 billion figure for Cymbalta and Vibryd, that results in 53% of the population taking antidepressants 12 months out of the year. Those figures are 5-8 times higher than any actual estimation of antidepressant use.  The $10 billion dollar figure is certainly eye-opening but there is plenty of evidence that it is not remotely accurate and will not have the purported impact on the pharmaceutical industry.

"About 80% of the prescriptions for antidepressants are written by primary-care physicians and others, not psychiatrists, a fact that makes the APA handbook particularly important. Faced with a patient complaining of depression-like symptoms, a general practitioner may be likely to rely on the Association's handbook for advice.”

This statement reveals the authors lack of knowledge about the practice of medicine and about the DSM that he is criticizing. The DSM is strictly a diagnostic manual and it contains no treatment recommendations. Primary care physicians are not avid readers of the DSM and that has probably led to the practice of using a DSM-based checklist – the PHQ-9.  This practice has not been promoted by the APA or the pharmaceutical industry (although the PHQ-9 is copyrighted by Pfizer pharmaceuticals).  Using a checklist to make a rapid diagnoses (in minutes) and rapidly treat large numbers of patients is promoted by managed care organizations and HMOs. That is probably the single greatest factor contributing to antidepressant prescriptions but it is ignored by the author - probably because it challenges his contention that this is all driven by conflict of interest in psychiatry rather than the business world.  It is cheaper for HMOs to treat depression with medications rather than detailed psychiatric assessments and psychotherapy.

"The Association itself runs on a budget of about 50 million a year, and for years industry funding has been critical to its operations. Today, about 14% of the Association's budget comes from pharmaceutical companies, mainly in the form of advertising at annual meetings and publications."

The author does a good job of providing no context here. Is the APA any different from other medical specialty organizations? Does advertising create a conflict of interest? Is any other print media outlet held to that standard? There is information available in those areas.   An Institute of Medicine report focused on conflict of interest showed that the APA's revenue from the pharmaceutical industry was in the middle of the pack with regard to medical specialty societies. As an example, the year that report was done the APA reported that medical companies supplied 28% of their annual income.  The American Academy of Family Physicians reported that 42% of their annual income was from pharmaceutical companies (p 220).  That same report (Recommendation 6.1) noted that increasing work for the pharmaceutical industry correlated with a 7% reduction in real physician wages and recommended that there was nothing wrong with “consulting arrangements based on written contracts for expert services to be paid for at fair market value”.   Depending on the expert involved, restricting the amount to $10,000 per year could practically mean anywhere from 2 to 10 presentations per year or about 2 1/2 weeks of contract work. 

“Other members of the committee have numerous ties to drug companies, too, and not simply conducting research, according to disclosures from last year. One was holding stock in Glaxo Smith Kline, one was a consultant to Servier and another consultant to Pfizer;  one had a grant from AstraZeneca and another a grant from Pfizer and AstraZeneca.”

This is a paragraph from a poorly written section illustrating ties between the 11 member Mood Disorders Work Group set up to draft the guidelines on major depression. There is some explanation of the selection criteria and conflict of interest criteria.  It discusses conflictof interest criteria that the APA designed and made explicit in response to this article.  It provides no context other than an off hand remark by the chairman that he probably regrets making. The article provides no reasonable context for expected reimbursement for experts as consultants to industries or the fact that this is a common practice in many academic departments on any major university campus. In some of those industries, the professional organizations actually make an effort to make sure that businesses are well represented in any process that involves making standards.

"The current handbook-the revised version will be published in the Spring-recommended against diagnosing major depression in the bereaved when the symptoms are milder and of less than two months duration. This is known as the "bereavement exclusion".  (If the signs of depression are severe-the patient has thoughts of suicide, for example-major depression is supposed to be diagnosed)….. The new handbook removes the bereavement exclusion."

There is really nothing new and nothing drastic as anticipated with removing the "bereavement exclusion". To provide a clear example I will quote a text copyrighted in 1982:

"There are many publications that deal with treating psychiatric patients who report recent and remote bereavement. It is possible to find a real or imagined loss in every patient's past. However, for the most part, because there is little evidence from reviewing normal bereavement that there is a strong correlation between bereavement and first entry into psychiatric care, those bereaved who are seen by psychiatrists should be treated for their primary symptoms. This is not to say that the death should not be discussed, but because these people represent a very small subset of all recently bereaved, they should be treated like other patients with similar symptoms but no precipitating cause. A physician seeing a recently bereaved with newly discovered hypertension might delay treatment one or two visits to confirm its continued existence, but treat it if it persists. So the psychiatrist should treat the patient with affective symptoms with somatic therapy but only if the symptoms are major and persist unduly. A careful history of past and present drug and alcohol intake is indicated. Then, the safest and most appropriate drugs to use are the antidepressants. Electroconvulsive therapy is indicated in the suicidal depressed." (Paykel p413-414).

Any psychiatrist worth his or her salt knows the difference between grief and depression and they should know the literature on treating grief, the natural history of grief, and the research on proven non-medical treatment of grief including Interpersonal Psychotherapy (IPT) and grief counseling. When you are seeing a psychiatrist, you are seeing an expert who should know the literature on grief, depression, and the differential diagnosis of depression.  Nothing in this article indicates that.  In fact, quotes are provided to suggest that the APA and psychiatry in general has an interest in redefining “the range of acceptable emotion” rather than using clinical research done by psychiatrists to limit suffering and prevent suicide.

I think the reality here indicates that there is no scandal.  The importance of the DSM-5, the appearance of conflict of interest, and the potential windfall for the pharmaceutical industry appear to be seriously overestimated.  Organized psychiatry is certainly not responsible for what happens  in primary care clinics under the direct guidance of business organizations.  There is a responsibility to establish professional standards for patients referred to psychiatrists for the assessment and treatment of complicated depressions that may occur during bereavement. The suggestion that medications may be useful in some of these situations and the importance of treating depression in bereavement has been around for at least 30 years.

George Dawson, MD, DFAPA

Peter Whoriskey.  Antidepressants to treat grief? Psychiatry panelists with ties to drug industry say yes. The Washington Post, December 26, 2012.

Clayton PJ. Bereavement in Handbook of Affective of Disorders.  Eugene S. Paykel (ed). The Guilford Press. New York. 1982  pages 413-414.

APA Reiterates Stringent Rules on Accepting Pharma Support.  Psychiatric News.  Monday December 31,2012.

Institute of Medicine (US) Committee on Conflict of Interest in Medical Research, Education, and Practice; Lo B, Field MJ, editors. Conflict of Interest in Medical Research, Education, and Practice. Washington (DC): National Academies Press (US); 2009. Available from: http://www.ncbi.nlm.nih.gov/books/NBK22942/