A cross post from Furious Seasons.
There was a fascinating exchange of letters in this month's American Journal of Psychiatry concerning just how much depression doctors should accept in their patients and the implications of such decisions.
What prompted the initial letter was the federally-funded STAR-D trial, which showed that current depression treatments--including some psychotherapies--are no where near as robust as doctors (and presumably patients) would like. What the trial showed, in short, was that various anti-depressants had anywhere from an 8 percent to 30 percent chance of success in remitting symptoms of depression.
That leaves a large subset of people who do not get relief using current therapies and that raises a host of practical issues for the mental health field. This situation affects millions of Americans.
Two other practical issue before I get to those: Both letter writers accept that STAR-D trialed drugs and therapies produced remission in 67 percent of people in the trial. That is wrong for two reasons: one, each arm of the trial was a discrete experiment, having no impact statistically on subsequent arms of STAR-D and to engage in additive probabilities the way these academics are is embarrassing. Two, the STAR-D trial, though the first long-term, real-world, independent study of depression treatment in this country, was still only a measure of how well people did in a snapshot in time. For instance, of the approximately 30 percent of people who took Celexa in the trial's first step and achieved remission of symptoms we have no idea how they are doing now, well after their comparatively short trial. Remission is hardly a static state because depression is simply too episodic a disorder. You can have no depressive symptoms for two years, say, on or off-meds and then suddenly have symptoms reappear later.
Those caveats aside, several psych researchers at Brown University wrote the AJP to offer commentary on an editorial by John Rush of the University of Texas. Rush's editorial on the STAR-D results appeared in the AJP last February. You can read it here.
What the researchers at Brown had issues with was the insistence of Rush and others that complete symptom remission is the appropriate goal of treatment. "It is not clear, however, that this recommendation is in the best interest of our patients," they write in their letter.
I've noted previously how obsessed the psych world is with complete symptom remission in all forms of mental illness and how useless such a goal is for patients. Yes, it's a laudable public health goal, but within the limits of current technologies and practices does not seem to be achievable for many patients. If complete symptom remission is your polestar, then that could spell all sorts of practical problems for patients, as I noted two years ago:
"I think there's a certain power to be gained from the admission that complete symptom remission is a false god and that what we really ought to be doing is getting symptom remission where we can and letting people live where we cannot instead of pounding them to death with meds and wasting billions in the process."
The trouble is that many psych docs have gotten in the habit of reasoning that if Prozac doesn't work, for example, and a patient doesn't have symptom remission on Celexa, then they need to put the patient on multiple anti-depressants (the argument goes that if you bomb the brain's serotonin receptors from slightly different "angles" then you'll cover the problem) or perhaps an anti-psychotic plus an anti-depressant (ick).
That's part of what the Brown researchers pick up on in their letter:
"Advocating for 'more complex regimens' even earlier in the treatment algorithm may cause more harm than good. Very few studies have assessed either the safety or the effectiveness of complex polypharmacy trials. Polypharmacy increases the likelihood of side effects, drug interactions, cost increases, and noncompliance. Polypharmacy, nonetheless, is becoming more commonly used in routine clinical practice, presumably in part because of the setting of remission as the goal of treatment. Focusing too much on symptom remission in treatment-resistant patients may aggravate an already difficult-to-manage illness. Patients may feel even more discouraged if they do not respond to complex treatment trials. Such discouragement may lead to noncompliance with treatment. The STAR*D trials reported substantial rates of attrition despite the extra staffing, attention, patient education, and free care usually associated with clinical trials."
Not only is the practice of polypharmacy a concern for the researchers, so is the idea that all symptoms must be remitted:
"There is ample evidence to show that patients who continue to experience residual symptoms of depression are at higher risk for multiple adverse outcomes. Such findings are used to justify the push for remission. The correlation of adverse outcomes with residual symptoms, however, does not prove causation. Persistence of depressive symptoms in spite of optimal treatment may be an indicator rather than a cause of a form of depression that is not likely to respond to treatments currently available. The idea that there are some patients with a form of depression that is not responsive to available treatments is consistent with our current nosology, which groups together many different types of depression."
I think it's safe to conclude that there are forms of bipolar disorder and schizophrenia, for example, that are unremittable and where doctors have got to start asking themselves why pound the patient to death with meds when we ought to be asking how they can live decent lives with what they've got. Those are perhaps trickier questions with those two disorders, but I'm sure most of you know what I am getting at.
What's deeply ironic about this challenge coming from researchers at Brown University is that the school is also home to Peter Kramer, the bestselling author of Listening to Prozac and Against Depression who is very much of the full symptom remission crowd. As the Brown researchers note, eschewing polypharmacy and complete symptom remission creates an obvious question:
"What should be the goals of antidepressant treatment? One goal should be to achieve the greatest symptomatic relief possible, with the recognition and acknowledgment that this may not mean remission for a significant minority of depressed patients. For these patients, in particular, more attainable goals may be to improve their quality of life and psychosocial functioning in spite of persisting depressive symptoms.
"There are ways to help patients manage their persistent illness more effectively. Disease-management programs for chronic and remitting/relapsing illnesses are available that help patients focus on improving their psychosocial functioning and quality of life in the face of persisting symptoms. Pharmacologic treatment and disease management are not mutually exclusive. Ongoing medication and psychosocial treatment trials should be pursued concurrently in order to engender hope, since some patients may benefit over time or take longer to achieve remission. Clinicians should also feel comfortable, however, to address with patients the reality that they are suffering from a chronic illness. We can do a great deal of good by facilitating more effective coping strategies rather than promising something that we cannot deliver."
While I'm not sure what role meds should have in what I'll call unremittable depression (I think that choice best rests with the patient), I admire the honesty of the researchers and their practicality. I'm not so sure I buy their argument that doctors have a huge role to play in helping patients achieve a reasonable level of psychosocial functioning. Isn't that what social workers and counselors are for? Isn't it something for the individual to work out on their own?
Quibbles aside, I think the initial letter opens up a huge can of worms for doctors who treat depression. Into the breach steps John Rush with his reply.
"Since STAR*D did not evaluate all available treatments for depression, we cannot conclude that the 33% who did not reach remission after four treatment steps would not have benefited from other medications, psychotherapies, or somatic treatments."
Oh, really? The differences between Celexa and Lexapro, for example, are so slim as to be almost meaningless, as they are between many SSRIs. Rush's bit of posturing is scientifically correct but in the real world is nonsense. So is another assertion of his.
"[T]he decision to scale back the goals of treatment to less than remission seems unwise until at least four treatment attempts. On the other hand, some patients may well be unable to reach and sustain remission. Clinicians must decide when to no longer pursue remission as the goal of treatment by making further treatment changes. However, patients who partially benefit from medication may further improve their well-being and quality of life when psychosocial interventions or other rehabilitative efforts are put in place. On the other hand, given the undisputed advantage of remission, both functionally and prognostically continued efforts may well be worthwhile in selected patients. The decision to switch from remission to improved quality of life should be a collaborative one between patient and doctor."
It's nice of Rush to admit that there are people who may have unremittable depression and that alternative strategies may be required. But why the emphasis on doctors--and by doctors Rush means psychiatrists--deciding when treatment has failed? Whose bodies and lives are we talking about here anyway? Who appointed doctors emperors of our lives and souls and depressions? No one. Would you trust someone to tell you how to live your life psychosocially whose offers of treatment have already proven to miserable failures? Not if you are smart.
Rush concludes his reply thus:
"Whether "more complex regimens" (i.e., medication combinations) are more burdensome, risky, or effective is an empirical question that deserves study. Many psychiatrists now use combination medications, but few controlled trials have actually evaluated this practice. Some studies do suggest better efficacy and little additional side-effect burden for selected combinations. Whole sale polypharmacy is not to be recommended. Carefully conducted randomized trials pitting monotherapy against drug combinations are needed to directly assess whether both acute and longer-term outcomes can be enhanced without undue patient burden."
Nice that he doesn't support wholesale polypharmacy, but his insistence upon further empirical trials to investigate medication combinations is a bit silly. We already have plenty of real world evidence that such polypharmacy is fraught with risk and offers few results.
I'm glad to see that some researchers are willing to be honest about the practical problems faced by a significant group of patients with depression. However, that does leave a big question on the table: how much depression is acceptable for a patient? Is there a scale that can properly measure this? Does one simply not shoot for as low a result on the Ham-D scale, for example?
I can't speak to how you'd measure such a thing scientifically, but then I find most of the depression rating scales unreliable to begin with. What I do know is that this is another instance where doctors ought to begin listening very closely to their patients instead of arrogantly assuming that all symptoms must be destroyed. I don't think the latter approach is healthy for real people who are the ones who should be making these real decisions.
The trouble is that I fear whatever answers might be hammered out to this question will be driven not by science or patient input, but by the same politics and power dynamics that have always won arguments about mental health treatment in the past. And that means the pharma companies and doctors will be in charge and patting one another on the back at the end of the day because, at bottom, they know they don't have to listen to patients and consumers because we have so little power. They'll find a way to tip their hats to the notion that we should have some input and then find a way to undercut our interests because doctors are simply so much smarter and more well-intentioned than the rest of mankind. So are drug companies.
Can't you tell? That's why current depression treatments are working so well already.