Measuring Depression Treatment by Function

THE SHORT ANSWER

I track depression with the PHQ-9 at every visit, but I judge success by whether patients are back to work, exercise, family life and the other goals we wrote down at the first visit. Symptoms often improve before function does, and partial response is not good enough. If neither moves on schedule, I change the treatment.

When a patient tells me they feel "a little better," I believe them, and then I ask a different question: what did you do this week that you could not do a month ago? The answer to that question tells me more about whether treatment is working than any description of mood. After more than two decades of treating depression, I have come to measure success by function first and symptom scores second, and I want to explain why.

Mood is a lagging indicator

Depression is a disorder of mood, so it seems natural to track mood. The problem is that mood is subjective, variable from hour to hour, and heavily influenced by whatever happened on the drive to my office.

Function is harder to fool. Either you went back to work or you did not. Either you are cooking dinner again or you are eating cereal standing up. These are observable, they are what patients actually care about, and they are the things that depression took away in the first place.

Research supports this emphasis. In a 2006 study in the American Journal of Psychiatry, Mark Zimmerman and colleagues asked depressed outpatients what remission meant to them. The absence of symptoms ranked below the presence of positive mental health, feeling like their usual self, and a return to normal functioning. Patients were telling us, in effect, that they measure recovery by what they can do.

Why I still use the PHQ-9

None of this means I ignore symptom scales. I use the PHQ-9 at every depression visit, and I ask patients to complete it before we sit down. The PHQ-9, developed by Kurt Kroenke, Robert Spitzer and Janet Williams and published in the Journal of General Internal Medicine in 2001, scores the nine DSM criteria for a major depressive episode from 0 to 27. It takes two minutes. A score of 5 or below is generally considered remission; a 50 percent drop from baseline is a response.

The PHQ-9 gives me a number I can compare across visits, and the research on measurement-based care shows that numbers change outcomes. A 2015 randomized trial by Tong Guo and colleagues in the American Journal of Psychiatry found that patients whose treatment was guided by rating scales reached remission more often and faster than patients treated by clinical judgment alone. A 2017 review by John Fortney and colleagues in Psychiatric Services described measurement-based care as a "tipping point" for psychiatry precisely because so few clinicians were using it despite the evidence.

But I pair the PHQ-9 with the tenth question that most people skip: how difficult have these problems made it to do your work, take care of things at home, or get along with other people? That item is the bridge between symptoms and function, and it is the one I watch most closely.

Setting functional goals at the first visit

At every new evaluation I write down two or three functional goals with the patient. Not "feel better." Something I can check: back at work full time, exercising three days a week, going to the grandchildren's games, sleeping in the bedroom instead of on the couch. The goals belong to the patient, not to me. My job is to make them specific enough to measure.

Then at each follow-up, after the PHQ-9, we go through the list. Where appropriate I add a formal functional scale such as the Sheehan Disability Scale, which rates impairment in work, social life and family life on a 0 to 10 scale, or the WHODAS 2.0 that the DSM-5 recommends. These take another minute or two and give me a second number to track alongside symptoms.

What I have learned from doing this for years is that the two numbers often diverge. Symptoms frequently improve weeks before function does. A patient whose PHQ-9 has dropped from 20 to 8 may still not be working, because the habits and avoidance that built up during months of illness do not reverse themselves when the chemistry does. If I stopped at the symptom score I would call that patient a success and move on. Looking at function, I know there is work left.

What the STAR*D trial taught me about "good enough"

The STAR*D trial, published by A. John Rush and colleagues in the American Journal of Psychiatry in 2006, followed more than 4,000 patients through up to four sequential antidepressant treatments. Roughly a third reached remission on the first medication, and the cumulative remission rate after all four steps was about two thirds. Those are sobering numbers, and they have shaped how I practice.

The lesson I draw is that partial response is common and that accepting it is a mistake. Patients who respond but do not remit relapse at higher rates and function worse than those who reach full remission. When a patient plateaus at a PHQ-9 of 9 or 10 and is still not back to the life they had, I do not call that good enough. I change something: adjust the dose, switch, augment, add therapy, or, after two adequate trials, raise the interventional options I oversee at Stedfast Psychiatry, Deep TMS and Spravato. Function is what tells me the plateau is not acceptable.

How this changes the conversation

Measuring by function changes what we talk about in the room. Instead of "how is your mood," the question becomes "you said you wanted to be back at the gym; what happened this week?" If the answer is "I went twice," that is data. If the answer is "I could not make myself go," we talk about what stood in the way, and the answer often points to a specific symptom, low energy or anxiety or poor sleep, that I can target directly.

It also gives patients a role. Mood happens to you. Function is something you do. When I ask a patient to try one small functional step before the next visit, I am treating depression from a second direction while the medication works from the first.

What this looks like in practice

A typical follow-up in my office runs like this. The patient completes the PHQ-9 and, if we are using one, the Sheehan scale in the waiting room or through the patient portal. I look at the trend, not just the latest score. We review the functional goals one by one. If the scores and the goals are both moving, we hold course. If the scores have moved and the goals have not, we talk about what is blocking the return to function, and I may suggest therapy or a behavioral target. If neither has moved by the time I said it should, we change the treatment at that visit.

That last rule matters most. The commonest failure in depression care is not the wrong medication. It is the right medication left unchanged for months while a patient stays stuck, because each visit the patient said they felt "a little better" and nobody checked whether their life had changed. Measuring function is how I avoid that failure.

Frequently asked questions

What is the PHQ-9 and how is it scored?

The PHQ-9 is a nine-question depression scale scored from 0 to 27. Scores of 5 or below are generally considered remission, and a 50 percent drop from the starting score counts as a response. I ask patients to complete it before every depression visit.

What counts as a functional goal?

Something specific and checkable that depression took away: returning to work full time, exercising three days a week, attending family events, sleeping in the bedroom again. Two or three goals, written down at the first visit and reviewed at every follow-up.

What happens if my symptoms improve but I still cannot function?

That pattern is common and it means treatment is not finished. We look at what is blocking the return to function, which often points to a specific residual symptom, and I may add therapy, a behavioral target or a medication change.

More from Dr. Husted

Educational content and not a substitute for evaluation. If you are in crisis, call or text 988.

Why I measure depression treatment by function, not just mood