No One Knows What Therapy is For

And that’s a huge problem for insurers, systems, and patients

In the late nineteenth century, the “father of therapy” Sigmund Freud described therapy as a way to turn “hysterical misery” into plain old, ordinary human unhappiness.

American philosopher and psychoanalyst Jonathan Lear said that therapy “aims at helping the patient achieve practical wisdom by deepening self-understanding and liberating the mind to think creatively and freely.”

Actor Jon Hamm said that therapy “gives you another perspective when you are so lost in your own spiral.”

And a random person on the street would mostly likely say the goal of therapy is to feel better – or at least emotionally stable.

If you ask ten different people what the goal of therapy is, you’ll get ten different answers. And it might sound like a harmless academic debate until you realize that we’re hurtling full-speed towards “value based care” with no shared definition of “value”.

If you think a lack of clarity is harmless, you haven’t been paying attention to how the industry works. Without a shared agreement on what therapy is actually supposed to achieve, “value” is going to be defined by whoever holds the checkbook—aka insurers and large networks—instead of the person in the chair or the one sitting across from them.

Expecting a system to pull everyone together to improve care outcomes while we can’t even define what those outcomes are is just an invitation for the most powerful members of that system to choose the cheapest possible definition of “progress” to justify paying for less and less.

Progress, as measured by a marketing tool

My own understanding of the wicked problem that is defining (and measuring) success in mental health came from working with Dr. Russell Dubois. Russell, now Vice President at BetterHelp, was (at the time) the Head of Clinical Success at Blueprint, who were (at the time) focused on measurement based care.

As Russell explained it to me, the mental health industry’s biggest hurdle is the fact that a working definition of quality care is basically non-existent. Because there is a total vacuum where a definition should be, health plans have stepped in with their own best guesses. They track things like how many days it took you to get an appointment or whether your therapist remembered to hand you a screener. These are just logistical data points that have almost nothing to do with whether you’re actually getting better.

As an industry, we’re essentially tracking how fast the scores on the screeners move without ever stopping to ask if the person at the front of the line is actually getting what they need, mostly because the system hasn’t even agreed on what “need” means. Do you need a better understanding of yourself, as Jonathan Lear said? Or just a reduction of your “hysterical misery” so you can just have just regular unhappiness, a la Freud? Nobody knows.

The clearest example of what fills the vacuum when we lack clarity is our weird obsession with the PHQ-9. It’s the industry’s favorite ruler, but it was built by Pfizer for primary care doctors and psychiatrists to flag physical symptoms when determining whether to prescribe medications. Because it’s so ubiquitous, patients often fill it out rotely, just clicking boxes to get it over with. So the entire system hinges on data that doesn’t measure what the client came in for, often doesn’t get answered accurately, and may be essentially useless to the therapist.

This is Howard Kroplick. Howard has a bachelor of engineering degree from Stony Brook University and a MBA from the Wharton School. Howard, while working at Pfizer, commissioned the creation of a small screener that inadvertently became the pillar of our whole mental health system.

I just want that to sink in: our primary measure of psychological wellness is a worksheet that was designed to sell antidepressants.

In other words, you might come to therapy so you can finally grapple with the trauma you experienced as a child, but your therapist is going to be measuring whether you do your hobbies more days a week- or whether you’re eating and sleeping better- because those are what shows up on the PHQ-9.

What does everyone want therapy to do?

As a therapist, I know we have to spotlight the things that actually motivate people to show up for therapy in the first place. Most people aren’t walking into a session because they want to “reduce their PHQ-9 score by five points.” They come in because they are lonely, or they’re grappling with trauma, or they’ve lost their sense of self-efficacy. Often, the “depression” is just the surface-level symptom of these deeper issues. If you solve the loneliness or address the trauma, the depression often lifts as a result. But for that to happen, the patient has to be able to use their own personal “why” as their motivator.

The problem is that insurance companies don’t want to pay for “self-exploration”. They don’t see it as their responsibility to fund your journey toward personal meaning, especially in an era where they are looking to be financially responsible for less and less

Payors think they should be responsible for paying only for therapy that:

• Suppresses symptoms by “treating” mental illness- like how a doctor treats a broken leg, fixing the “acute” issue so the patient can return to work.

• Mitigates risk by ensuring you aren’t a danger to yourself or others, which could result in higher care costs if you take drastic actions.

• Returns users to function, which is just a clinical way of saying they’re stable enough to be a productive member of the economy again (and take their meds, which will keep them out of the hospital.)

To an insurer, “value” is the absence of a claim. If you stop showing up because your PHQ-9 score dropped from a 15 to a 9, they consider that a success. But as any therapist can tell you, a client whose scores have dropped might just be better at faking it, or they might have simply given up on answering the screener questions truthfully. They aren’t necessarily better; they are just no longer “clinically significant” on a spreadsheet.

People who show up to therapy- we can call them clients, or patients, or consumers- have their own set of goals, and the goals will vary based on the person. A patient might want therapy that:

• Helps them makes sense of trauma, grief, or loss..

• Helps them feel more grounded in relationships, work, or parenting.

• Restores a sense of agency or self-trust they feel they’ve lost.

• Supports them in building a life that feels tolerable, or meaningful, or at least coherent again.

A therapist’s goals for therapy usually line up with the patient’s, but there are moments when they don’t. A client might come in wanting to be a better partner, while the therapist starts to notice patterns of coercion or abuse in the relationship. In that situation, the therapist isn’t there to help the client adjust to something that’s harming them. The work shifts toward helping the client notice what’s happening, think clearly about safety, and make their own decisions with more information than they had when they walked in.

A therapist might also aim to provide therapy that:

• Increases insight, even when that insight is uncomfortable or disruptive.

• Expands a person’s emotional range rather than simply dampening distress.

• Strengthens the client’s ability to tolerate ambiguity, conflict, and uncertainty.

• Interrupts patterns that are self-protective but ultimately self-defeating.

• Prioritizes long-term psychological health over short-term symptom relief.

And, although we don’t allow it to be more than a peripheral force in the US, society itself has its own incentives (and stands to benefit the most from truly improving mental health- too bad we don’t let society pay!)

If we let it enter the marketplace, society would want therapy that:

• Reduces incarceration by addressing mental illness, addiction, and trauma before they escalate into criminalized behavior.

• Lowers long-term healthcare spending by preventing repeated hospitalizations, ER visits, and chronic disability.

• Increases labor force participation.

• Supports stable families and caregiving systems.

• Produces adults who can regulate themselves, relate to others, and contribute without constant institutional intervention.

So the real task is finding the overlap. Where do the goals of patients and clinicians line up enough with what payors are willing to reimburse that therapy can actually be covered? That overlap is where value should be defined, if it exists at all.

The harder question comes next. Once we decide what belongs in that space, how do we measure it?

Better Measures We’re Not Using

So what we need is a unicorn definition of success. Working to create that definition of success needs to deliver savings for payors, meaning for patients, durable psychological change for therapists, and downstream stability for society. That shouldn’t be hard, right?

A lot of people have tried to build measures that capture something broader than symptom reduction, but they haven’t stuck. Here are a few of the top contenders:

1. Recovery Assessment Scale (RAS)

What it is:
A questionnaire that asks people how recovered they feel, not how many symptoms they have. It focuses on things like hope, confidence, agency, and whether symptoms still dominate their lives.

Who made it:
Recovery-oriented researchers and clinicians working with people with serious mental illness.

Why it’s a better alternative:
It actually measures what many people mean when they say they want to “get better”: feeling capable, hopeful, and able to live their life, even if symptoms haven’t vanished.

Why it hasn’t caught on:
It’s longer than the PHQ-9, not tied to diagnosis codes, and doesn’t fit neatly into insurance reporting. Payors don’t like it because it doesn’t map cleanly to utilization or medical necessity. Systems don’t like it because it requires a different mindset, not just a different form.

2. WHO-5 Well-Being Index

What it is:
A very short questionnaire that asks about positive well-being, like feeling energetic, calm, or interested in daily life.

Who made it:
The World Health Organization, for global use across medical and public health settings.

Why it’s a better alternative:
Instead of tracking what’s wrong, it tracks whether someone is actually feeling okay. Patients tend to find it less alienating and more intuitive than symptom checklists.

Why it hasn’t caught on:
It doesn’t diagnose anything. Insurers and U.S. guidelines prefer tools that line up with DSM categories, even if those tools miss the point. Clinicians are also trained to look for pathology, not well-being, so it often gets treated as “nice but optional.”

3. WHODAS 2.0

What it is:
A measure of how much mental health issues interfere with daily life: work, relationships, self-care, and participation in society.

Who made it:
Also developed by the World Health Organization, and referenced in the DSM as a disability measure.

Why it’s a better alternative:
It gets closer to what payors and society actually care about: can this person function in the world? It captures impact, not just symptoms.

Why it hasn’t caught on:
It takes longer to complete and score, and no one is paid to track functional recovery. Insurers still reimburse based on symptoms and diagnoses, not whether someone can show up for their life again.

4. Recovering Quality of Life (ReQoL)

What it is:
A patient-reported measure of quality of life that includes meaning, relationships, confidence, and day-to-day functioning, alongside some symptoms.

Who made it:
A group of scientists from The University of Sheffield in the UK- with heavy input from people actually receiving mental health care.

Why it’s a better alternative:
It asks the question most patients actually care about: is my life getting better? It balances distress with things like purpose and connection instead of treating symptom relief as the end goal.

Why it hasn’t caught on:
It’s newer, less known in the U.S., and not built into EHRs by default. Payors haven’t adopted it, so systems have little incentive to change workflows to support it.

5. Outcome Questionnaire-45 (OQ-45)

What it is:
A broad measure of overall distress, relationships, and role functioning, often given regularly throughout therapy.

Who made it:
Developed by psychotherapy researchers and commercialized as part of outcome-tracking systems.

Why it’s a better alternative:
It captures change across multiple areas of life, not just mood symptoms, and is useful for tracking progress over time in real therapy settings.

Why it hasn’t caught on:
It’s long, proprietary, and costs money. Small practices don’t want to pay for it when the PHQ-9 is free. Primary care doesn’t want it because it’s too much. Payors don’t require it, so it stays niche.

The World-Changing Magic of Getting This Right

If we solved this problem, the world would change.

Payors would fund care that also happens to help people stay out of hospitals, remain housed, sustain work, and keep from cycling back through crisis every six months.

Therapists would be evaluated and paid not on whether their patients have had trouble concentrating, but on whether their clients can tolerate stress, maintain relationships, and make decisions that don’t repeatedly blow up their lives.

Patients would know what therapy is aiming for and why it’s worth the time and effort.

AND, on top of all that, society would see fewer downstream costs because fewer people would reach the point where intervention is unavoidable.

The irony is, we already track most of those outcomes- but only after things go wrong.

If we actually wanted to measure those outcomes proactively, the whole system would have to agree to try something different at the same time. Payors might have to accept longer timelines, systems would have to change workflows, clinicians would have to work with measures that might take more time to complete.

That kind of shift only happens if there’s agreement on what’s worth measuring, and there isn’t.

We already know there are better tools than the PHQ-9. But because none of them line up cleanly with everyone’s incentives, and none of them come with a clear answer to who absorbs the cost, the effort, or the risk of changing course, nothing moves.

In the absence of agreement, the system does what systems do: it keeps using the tool that’s already embedded and already reimbursable.

And that tool was built by marketers, which we’ve decided is fine.