In a number of pieces here on HealthCentral, I have challenged the notion of depression as a single disease. By treating all depressions as the same, researchers and clinicians alike are failing in the performance of their duties.
What woke me up to this was a journal article I came across in 2004 by Gordon Parker of the University of New South Wales. In the article, Dr Parker noted that in clinical trials drug companies lump all depressed patients together, regardless of age, gender, or symptoms. Not surprisingly, trial results for the test treatment are only slightly better than the placebo - good enough for an FDA indication but eminently unsuited to foster any real understanding.
As Dr Parker wrote in 2007:
Depression is a diagnosis that will remain a non-specific "catch all" until common sense brings current confusion to order. As the American journalist Ed Murrow observed in another context: "Anyone who isn't confused doesn't really understand the situation.”
Similarly, in 2013, Nassir Ghaemi of Tufts attacked the prevailing view of depression, noting that the DSM concept is so wide as to be virtually meaningless. According to Dr Ghaemi:
The traditional analogy of antidepressant usage for depression to insulin usage to diabetes may be wrong-headed: Depression is not a single disease entity …
Now we have convincing support from recent research by Eiko Fried of the University of Leuven in Belgium and Randolf Nesse of Arizona State University. In an article in the October 2014 Journal of Affective Disorders, the authors report how they broke down data from STAR*D, an NIMH-underwritten real-world trial of antidepressants.
Some background …
Over the early 2000s, STAR*D recruited about 3,700 “depressed” patients. In the first round of the trials, about half the subjects showed short-term improvement on the antidepressant Celexa.
In the second round, other options were applied on the non-responders, including a different class of antidepressant and various meds combos. Some of the patients responded to these new options, but no particular treatment showed an advantage over the other.
The third round produced the real surprise: A failure on two consecutive antidepressants virtually predicted failure on a third. Twelve-month results were totally discouraging: Those who showed promising short-term outcomes tended to relapse over the long-term.
The principal researchers in the study struggled mightily to put a positive spin on the results, but their own numbers told a different story. It is tempting to use the findings to condemn antidepressants wholesale. More realistically, I submit, we need to question our whole notion of depression. Until we know what we’re dealing with, we will have no idea how to go about treating it.
Mining the data …
In recruiting the patients for their study, the STAR*D researchers screened their subjects using the QIDS rating scale, which tests for 12 symptoms. Taking a second look at the data, Fried and Nesse found, on average, the subjects exhibited six symptoms, the most common being sad mood, loss of energy, and concentration problems.
In theory, the different combinations of symptoms would yield more than a thousand unique profiles, and that is what Drs Fried and Nesse found - 1030 with an average of 3.6 individuals per profile (nearly half with just one person per profile). Even their attempts at lumping together similar symptom profiles showed wide diversity (what the experts refer to as heterogeneity).
Making sense of the data …
The quick takeaway is that not all depressions are the same. Which is why - to freely editorialize - STAR*D was doomed from the get-go. Back in the early 2000s, it seemed like a good idea to treat 3,700 individuals with the same drug.
A decade later, it is easy to see the absurdity in that proposition. Perhaps “depression” is one disease with numerous manifestations. But if that were the case, we would have a magic bullet for all seasons. At the least, we would expect better results from our antidepressants and talking therapies.
More likely, we are looking at different illnesses that differ in cause and biology, and - by implication - treatment. It may be that antidepressants and other therapies work well for certain subsets of patients, but we have yet to single out these populations.
To date, we have failed to identify reliable biomarkers for depression. Nor have we made credible attempts to differentiate one depression from another based on what we can observe and investigate.
Until we do, we remain mired in ignorance. In the authors’ words, we need to acknowledge that major depression “is not one coherent condition with a single cause.”
Further reading ...
Antidepressants for Depression
Finding Out What Works Best for YOU