“Escitalopram and Problem-Solving Therapy for Prevention of Post-Stroke Depression.” JAMA May 2008.
OK, so I recently got around to looking over a recent JAMA study. Right up my alley. Meds versus therapy. I am so curious! Did therapy prevail? Should the health care system be directing patients to counselors, versus to the pharmacy, for yet another clinical problem?
Well, the short answer seems to be “No.”
Darn. Meds wins this one. But maybe I can call, “FOUL!” So I begin a careful reading. Especially alert for the subtle-yet-profound influence of Big Pharma.
The issue: Given that the prevalence of depressive episodes post-stroke is fairly high, can preventive / prophylactic depression intervention prevent some of this depression incidence? If so, does medication or therapy, or both, work? Better than placebo? If both medication and therapy work, which works better?
The article is Pubmed ID 18505948.
The article tells that this study recruited people who had recently had a stroke, and randomly assigned them to one of three treatment groups, each with a different, competing treatment for prevention of depression in a follow-up treatment surveillance period of one year. Some cases of depression should be expected in each group. The degree that either the medication or therapy group would have a lower portion of depressive episodes compared to the placebo group would be the degree that the intervention worked – that it prevented depression.
Well, it worked. Both meds and therapy worked. Fro the group receiving placebo, 22% had a depressive episode. That can be seen as a base rate. For the group receiving escitalopram, 9% had a depressive episode. For the group receiving therapy, 12% had a depressive episode. So, preventive depression treatment worked to prevent depression, both meds and therapy. And meds worked a bit better than therapy.
That is good news for Forest pharmaceuticals, who sell escitalopram as Lexapro, still under patent. Forest can seek the FDA to approve escitalopram for post-stroke prophylaxis of depression, thus creating a whole new market for this antidepressant – anyone who has had a stroke.
Darn. Meds beat therapy in this one. But let’s dig a little deeper.
For many reasons, it is a challenge to run a therapy study. Therapy is always individualized to some degree, despite efforts to have it be standardized. Also, with a pill, you know the specific dose received. But if a therapy client doesn’t really ‘work’ at therapy, and doesn’t happen to get as much out of therapy, then the dose can really differ. For one person, eight hours of therapy might be life-changing, while for another person, it might just be tedious. In contrast, if you give someone a pill, and they take it, you know they got one dose of the pill.
So, I looked over the training and standardization of the delivery of therapy. Pretty good. Problem-Solving Therapy may not be the most intimate, life-changing brand of psychotherapy, but it can be effective for depression. In this study, it was trained and supervised to be orthodox. Good.
One thing that is missing, however, is information on the therapists. Were they licensed? Were they graduate students? Were they master’s level? Doctoral level? Were they psychiatrists?
Well, I search and finally find some information. Not in the article. But tucked in the back between the study and the references, is a set of notes. Including a note thanking two people for administering the therapy. One of these people has a bachelor’s degree noted after their name, and the other has a master of arts degree.
180 patients receiving therapy, and some received therapy from a master’s level person, and some from a bachelor’s level person? Are you kidding me? Is there anyone out there looking to pay $100 an hour for psychotherapy from a person with a bachelor’s degree? This was funded by NIH? No info about license on either.
So, medsvstherapy, and meds won. But there is no evidence that therapy was even provided by a counselor or therapist. What if this were a physical therapy intervention, but the person doing the physical therapy was not a physical therapist? What if this was yoga versus surgery for back pain, but the yoga instructor was not a yoga instructor? What if this was a comparison of two surgery techniques, but the person performing the surgery was not a surgeon?
The person with the bachelor’s degree clearly is not in the game. There is some outside chance that they could legitimately be legit to do therapy – if they were a seasoned, trained minister, for example. But that would be some rare kind of exception. So, let’s just figure that at least one therapist is not a therapist. What about the second noted therapist, with the MA?
This could be a master’s level counseling degree, and the person could be credentialed as a master’s level counselor. Or it could be a master’s degree in some other area. The study just does not say.
So, with a few minutes, I decided to figure out if I could find evidence of licensure. Other published literature establish this person as a colleague at Iowa’s psych research shop for years, sometimes as co-author, sometimes as research assistant. Thus, the most likely jurisdiction of licensure would be Iowa.
Iowa has a listing of all licensed professional occupations here:
You can see who is licensed here:
Well, this person is not a licensed marital / family therapist. Not a Mental Health Counselor. Not a Health Service Provider in Psychology. Not a Psychologist.
It is possible that this person, who provided this therapy, was a licensed counselor of some kind. But the topic is avoided in the study, and there is no indication of licensure in Iowa, despite some hunting on the web I was able to do.
In some U of Iowa document on the web, I did find this person listed as a “Research Assistant II” for 2006 – within the time span of recruitment, 2002 – 2007. Doesn’t sound very likely that this person is a counselor or therapist.
So, meds versus therapy, but not therapy provided by an actual counselor. Were they afraid that therapy might win if it were delivered by an actual counselor?
Maybe the original proposal would say. The proposal submitted to NIH for funding. Well, I don’t have the original proposal. However, the article does have the NIH grant number, allowing the original study abstract to be pulled up on the government’s website listing all research grants: CRISP. Just google CRISP and you will find this.
So, I enter the lead author’s name, and quickly locate the study’s abstract from when it was submitted for funding. NIH R01 grant 65134. I discover the grant. And I discover a couple things that don’t quite match up with the publication.
First, the proposal reported that citalopram, not escitalopram, would be used in one arm, and problem-solving therapy in a second arm. But there was also supposed to be a third arm: a group receiving another antidepressant, nortriptyline. Commercial name Pamelor.
What happened to the nortriptyline?
What happened to the citalopram?
First: What happened to the nortriptyline?
Was the nortriptyline given, per the NIH grant? Did it win? Why would it be excluded? Hmmm.
I have a theory. A suspicion. Nortriptyline is no longer under patent, unlike escitalopram, so discovery of a new use for nortriptyline would not be a gold mine of a new clinical population. Maybe the authors chose not to report the nortriptyline arm of the study, so that escitalopram, under patent, could be promoted, without the data out there in the published realm concerning the generic?
Second: What happened to the citalopram?
It turns out that citalopram was about to go generic as this proposal was submitted to NIH. However, at the same time, the developers of citalopram were just developing a similar drug, escitalopram, to replace citalopram as a money-maker. Right around the time when this study proposal was funded. So, escitalopram being very similar to citalapram, the researchers may have switched to the newly developed, newly available drug which would be under patent for a while, thus potentially generating more revenue if the drug worked as suspected for preventing post-stroke depression.
So, after this little review, these are my thoughts:
First: problem-solving therapy conducted by therapists might have had similar or better results, compared with escitalopram, for preventing depression post-stroke.
Second: if the nortriptyline arm of the study was conducted, nortriptyline may have had a similar efficacy when compared with escitalopram, but the strategy of not reporting this would leave open the possibility of spreading the knowledge regarding the under-patent medication, not the now-generic medication. The benefit of this, for the pharmaceutical company, would be: a new market consisting of the post-stroke population in general, barring any contraindications.
Now, I am gonna consider how to follow up on this. Cuz in the battle of meds versus therapy, therapy really might have won this one, if only it had a chance.