Much has been made of the
Texas Medication Algorithm Project.
I have written about it earlier (
1,
2), as have others (
3).
A
lawsuit has been filed alleging that TMAP was a sneaky way to convince state mental health programs to switch their patients to newer, much more expensive medications.
TMAP defenders, on the other hand, say that TMAP was simply allowing patients access to the state of the art, most effective medications.
What is TMAP? Essentially, TMAP is a program that used “expert consensus” to develop treatment guidelines for patients (depression, bipolar, and schizophrenia) in the public mental health system. On one hand, I can see why care should be improved – not many people seriously argue that patient care is very good in most public mental health systems.
According to the TMAP model, medication treatment is provided in stages according to these guidelines. If you are not responding to treatment #1, then you move to treatment #2, and if that does not work, then to treatment #3, and so on. Naturally, the “objective experts” who developed said guidelines stuck the newer, more expensive medications on the top of the list for treatments, especially as the guidelines have been revised during the past couple of years.
TMAP was unfurled in the mid 1990’s and similar programs have since been sweeping across many states.
Was this a way to backdoor newer medications onto patients? Well, I think that is probably the case, but the issue I am going to address here is one that I think is even more important…
Does TMAP Work? Do TMAP patients show more improvement than patients who were not on the TMAP treatment regimen? The TMAP team has produced some evidence in which they claim that TMAP treatment works better than “treatment as usual,” which was standard state mental health care. In this case, we’ll discuss TMAP for bipolar patients.
The Study: Some patients received TMAP, which included both a standardized medication algorithm as follows:
Manic:
Stage 1 – Depakote or Lithium or Tegretol
Stage 2 – Depakote + Lithium OR Tegretol + Lithium
Stage 3 – Depakote + Lithium OR Tegretol + Lithium
Stage 4 – Depakote + Tegretol
Stage 5 – Add atypical antipsychotic to mood stabilizer
Stage 6 – ECT
Stage 7 – Other (e.g., Lamictal, Neurontin)
Depressed:
Stage 1 – Wellbutrin or SSRI + mood stabilizer
Stage 2 – Wellbutrin or SSRI or Effexor or Serzone + mood stabilizer
Stage 3 – Mood Stabilizer + two antidpressants
Stage 4 – Mood Stabilizer and MAOI antidepressant
Stage 5 – ECT
Stage 6 – Other (e.g., Lamictal)
Note that in the 2005 revision of this standard, atypical antipsychotics are featured much more prominently. But when the study on bipolar patients was conducted, this was not the case.
Those who received “treatment as usual” (TAU) received whatever care they would normally receive.
The Results: On some measures, TMAP patients did modestly better than TAU patients. This could be interpreted as evidence that these strict treatment algorithms that involve a high frequency of prescribing Depakote and Tegretol, and to a lesser extent, newer antipsychotics, are a good idea for patients. However, one would be fooling oneself to buy this conclusion. Why?
The (Huge) Caveat: The TMAP patients all received: Group education, consumer to consumer discussion groups, individual patient education from the physician, referrals to therapy groups and more. These interventions were rolled out exclusively for the TMAP group. Why does this matter? Well, patients in the TMAP group were likely getting more time with their physician, which is likely going to boost their relationship with the physician, which will likely lead to better outcomes regardless of the medication taken. In addition, the patient education groups provide additional support for patients, which has been shown to improve outcomes.
Even the study authors, to their credit, admit this is a gigantic potential issue:
At this time, the relative contributions of different elements [i.e., medication versus the extra patient care] of the “disease management package” to the obtained results has not been evaluated.
A Better Idea for TMAP: If you wanted a study that would have compared the effects of a) extra patient education and support, b) use of medication algorithms that favored use of newer medications and c) “treatment as usual” – regular care in the state mental health system, then why not have a study that looks like this:
A) Treatment as usual (No extra patient support)
B) Extra Patient Support + TMAP Algorithms
C) TMAP Algorithms (No extra patient support)
D) Extra Patient Support + Treatment as usual
If C’s outcomes are better than A’s outcomes, then you can shout about the evidence base of your algorithms. If B is greater than D, then you can also do your evidence-based practice speech. However, the TMAP bipolar study as was actually conducted was A versus B, – that is a pretty lame comparison. Was it the patient support (which is my guess) or was it the algorithms?
Why Was TMAP Investigated This Way? This is where it gets interesting but murky. Could such a study have been designed because it was biased to find favorable results for the TMAP intervention (due to TMAP patients receiving extra support not received by treatment as usual)? Then, finding positive results, it becomes a lot easier to sell the program to other states because TMAP is now “evidence based”. Maybe I’m off in Conspiracyville, but one has to admit this is pretty weird stuff.
7 comments:
I may be pretty dumb, but if you wanted to give the algorithm the very best shot at proving superior to "treatment as usual", isn't this pretty much how you'd have to do it?
The problem was that the algorithm was not directly compared to treatment as usual.
It was algorithm PLUS additional patient care versus treatment as usual. Thus, were the generally slight advantages of TMAP due to the algorithms OR due to the additional patient care? That is the problem.
Perhaps I misread your statement -- if you literally mean giving the algorithm the "best shot" at appearing better, then yes, the study as actually conducted was brilliant, as many people likely concluded that the algorithm was the key to success rather than the extra patient care.
For depression I was put on Wellbutrin + Effexor (high dose) + Lamictal. Then the psychiatrist had to add Flomax (anticholinergic effects?)because I could no longer urinate. Oh, and then he threw in Risperdal so I could sleep (because the drug cocktail had me super wired). Turned me into a complete nutcase. I remember taking all my scripts to the pharmacy and the pharmacist handed me my large bag of pill bottles and said, "Good luck."
I am 4+ years now psych drug free (I was told I would have to take drugs the rest of my life). I look back on that period and marvel at where I am today. Full time employment. Graduate degree. Happy and healthy. Taking myself off all that crap and removing myself from the texas mental health system were the best decisions I ever made.
Do I understand you correctly if I read you to be saying that:
- the State of Texas paid a team led by scientists with degrees from institutions of higher learning such as Princeton and Harvard millions of dollars to determine which treatments work best,
- they themselves wrote that they were unsure whether their work allowed any meaningful conclusions to be drawn, but that conclusions were drawn all the same,
- and that this could all have happened because they are hapless.
CS,
It sure looks fishy, doesn't it? Thanks for putting it quite succinctly in your last comment.
If the recent lawsuit has some merit, then Janssen, and perhaps other companies played a key role in throwing together TMAP and this study, and the academics saw a great chance for a huge influx of cash. In academic psychiatry, the amount of external funding you draw in is a huge part of determining your prestige, so of course many academics agreed to sign up. Of course much of this money goes into one's pockets as consulting income, and that cannot hurt one's desire to be involved in such a study.
And, sure, perhaps patients could be helped out in this study, which I'm sure pleased many of the involved academics. I cannot say for sure if these academics were just pawns in a larger game and to what degree they understood that they were not actually evaluating the effectiveness of the algorithms. I bet many, perhaps even all, of the affiliated academics had good intentions, but it sure seems like they missed the boat in terms of desiging the experiment.
Did they forsee that this was the start of spreading TMAP-like algorithms across the nation? Who knows -- but as you pointed out, this sure seems odd at the very least.
Thank you for helping a dolt like me understand the machinations of this claque. I agree wholeheartedly that we ought not jump to conclusions. Doctorates, and alas even Nobel Prizes, are no guarantee against haplessness and even outright stupidity; after all Lenard and Stark, both Nobel Laureates, fought Einstein on Relativity all the way. This out of a belief that that accepting Einstein's theories was not compatible with being a loyal German.
This qui tam case accuses these academics of, perhaps unwittingly, participating in an effort to defraud the State of Texas. What sort of consequences, in terms of licensing, professional prestige, and even criminal proceedings can these academics expect if J&J is found guilty?
Being involved in research that is not very well-designed does not lead to any sort of consequence. And it only took me 7 months to reply to your comment!
Post a Comment