Maybe We’re Trying Too Hard.
Updated: Jan 20
Investors have taught me more about how to think about therapy than I’d like to admit.
Dean Williams, an investor, has an excellent article called “Trying Too Hard” in which he makes the case that a simple investing strategy is the best. Listening to him and other investors, the good ones, the ones who produce market returns, you get the sense that they know how to work with their ignorance and they realize the limits of their knowledge. Because they understand the limits of their knowledge they recommend that most people are best served Passive Investing — buying a tiny bit of every major US company and waiting for at least ten years. Turns out passive investing outperforms 95% of guru investors, analyst, and money managers. The fancy investors you see on TV are trying too hard.
The magic of passive investing is hard for people to really believe. How can waiting beat 95% of market gurus? Everyone thinks, “well, if I could just beat the market once or twice, I’d be set for life.” They don’t realize they’d be better off dead. Literally. I once read a story where dead investors did better than money managers. Why? Because the dead didn’t jump in and out of the market. They were the ultimate passive investors.
The thing is therapy and human psychology, like the field of investing, is not something we understand. Worse yet, the gurus who purport to understand human psychology are unaware of their ignorance. Working in a nebulous and murky area, where even the experts are blind, is something good investors have mastered. That’s why psychotherapists can learn a lot from them.
The Uncomfortable Realization.
Psychotherapy and counseling is still in its infancy. Despite being founded over 140 years ago, we know surprisingly little about how change happens. Sure, we’ve developed newer more sophisticated ideas about human nature since Freud, but we’ve shown little to any improvement in how good we are at helping people change.
Recent literature reviews show that outcomes in psychotherapy haven’t changed in 40 years, rates of mental illness haven’t declined in at least 20 years, all models are equally effective, and therapists do not become more effective with time and experience. All of this leads to the uncomfortable realization that despite how hard we’re trying, we’re not improving as a field.
Some will push back saying, “well of course not. Therapy isn’t one size fits all. It’s about tailoring your intervention to the client in front of you.” The problem with this is, just like a shoe tailor must first read the size of the foot, to tailor therapy we must first read the personality, motivations, and needs of the client. Something the vast majority of us can not do. Since the publication of Being Sane in Insane Places the fallacy of our ability to read clients has been on display. We just aren’t good at it, a fact further confirmed in the lie detection research of Maureen O’Sullivan and Paul Ekman, the outcome research of Bruce Wampold, and the expertise research of Robyn Dawes.
Consider signing up for my newsletter. You'll get more goodies like this.
Charisma and Charm.
Despite working in the dark, therapy is remarkably effective. It’s just as effective as the flu vaccine, helping 50% of people. And we know that this change happens despite the model. Now that doesn’t mean 50% of people recover. We know that 50% of people in therapy experience some change and 20 % of them recover.
This seems to be true regardless of where you sit in the therapy hierarchy. See the therapy hierarchy generally sits like this:
Model Trainers and therapy Ph.Ds
Fully licensed professionals
Interns/ Associate licensed professionals
Students
We first discovered that the "therapy hierarchy" didn't hold up in 1978 when Drs. Strupp and Hadley published their landmark article. They compared the results of a group of college professors to professional counselors and found both helped clients the same amount. And, since the field hasn’t improved in at least 40 years, this holds true even today. For instance one of the most evidenced based models, Multi-Systemic Therapy, uses lay counselors to work with kids at risk of being incarcerated. The lay counselors who do the work get weekly supervision in the model, and overall, produce good results.
This means the “hierarchy” of therapists is remarkably flat. Research study after research study has shown that there’s generally no difference between Ph.Ds. and licensed professionals and students. If anything the research points in the other direction, the longer you’re in the field the more complacent you become and your outcomes drop slightly over time. The newer you are in the field the harder you work and your outcomes tend to be slightly stronger. The ultimate example of this is what happened with the original Solution Focused Brief Therapy team. In the 90s the SFBT team had a group of independent researchers research their effectiveness. Turns out SFBT was no more effective and no more brief than any other model. Furthermore, Insoo Kim Berg, the model creator was not the most effective therapist, rather one of the interns was.
This is the paradox of therapists outcomes. On the one hand just because someone is higher on the hierarchy doesn’t mean they have better outcomes than you (#InsooKimBerg). On the other hand some therapists are more effective than others (#randomintern). David Ricks was the first to study this in his article Supershrink: Methods of a Therapist Judged Successful on the Basis of Adult Outcomes of Adolescent Patients, but it’s been confirmed over and over. Currently two of the top therapists are an Australian woman name Eri Vlass and a Canadian man named Allan Abbas, MD. When you look at their outcomes they both have amazing results. Allan Abbas even teaches courses on his model of therapy. But there’s no evidence that his students do better than average. Whatever he’s doing he doesn’t know how to transfer it.
How does this happen? How do so many gurus claim, “learn my model. It will make you a better therapist!”
The truth is regardless of the field, unless gurus formally track their results, they just don’t know how good they are. This has been shown very clearly in the work of Daniel Kahneman, Nassim Taleb, Anders Ericcson, Gary Klein, and Ian Harris.
And no almost no one is tracking. Even the great John Gottman, who has done the most in-depth research in couples, doesn’t have outcome data on his model. When he says his Sound House Relationship Model is effective he reports the survey results from his psychoeducational weekends where he teaches people his model skills. Which begs another question: How is it that the man known for observing and coding couples behavior and tracking couples long term, hasn’t applied any of those same skills to researching the change process? Why is he using self report surveys?
If these gurus have average outcomes then why do we call them experts? Mostly because of charisma and charm. Dan Siegal or Ester Perel are considered experts because they are incredibly charismatic and they have a platform, not because they are better than you and I. Add to this they make much of their living selling a product — certifications, books, online courses, — and it’s clear they are incentivized to play up their wins and ignore their losses.
You can’t measure change, can you?
If you’ve made it this far you’re probably asking the next obvious question. “But isn’t change dependent on the client? You can’t quantify a feeling or use a metric. It’s an experience between the therapist and client, built on trust. You can’t measure that.”
That’s the exact right question to ask.
You’re right. These assessments are flawed and miss things and often give false positives or false negatives. It’s an imperfect system.
Here’s the question. Do you believe you can ask a client, “has therapy helped?” If you believe you can ask that question, then you can measure change, if imperfectly. These assessments are just ways of asking clients, “has this been helpful” in a standard way. You could literally put that on a piece of paper and use that as your assessment. In fact, one of the most popular outcome monitoring surveys basically does just that. And, having this question on a sheet of paper is actually a kinder way to see if we’re helping clients than just asking. Most client’s I’ve had are kind people who don’t want to insult you to your face, which means they aren’t going to give negative feedback. In general, clients prefer answering on a sheet of paper. It’s less personal.
Also, we know assessments are better than the alternative. What is the alternative? Well, the way models used to be created was some guru just made up a model based off of his (and it was usually a “his”) experiences and thoughts about human nature. The history of medicine is littered with examples of how harmful that was.
So, yes, assessments are flawed. And they are better than the alternative.
Let’s take the pressure off.
So where does this leave us? In a field where we know so little, what do we do?
This is where investing has taught me the most. Just like in investing where the fancy mathematical models don’t improve your results, in psychotherapy the fancy psychological models haven’t improved our results. And just like in investing where most of your outcomes come from the simple method of passive investing, in psychotherapy most of the outcomes come from the simple art of active listening. This is good news. We know from the studies of lay counselors that with minimal training we can teach a lot of therapists to help a lot of people.
I do think there’s more than that. I can’t deny the power of experiential approaches. If you’re working with an individual those experiential interventions may be something like EMDR, Focusing, Hypnosis, Empty Chair work or exposure, and if you’re working with couples or families those experiential interventions are role plays, enactments or restructuring where members practice acting differently under the guidance of a therapist. But this is not an exact science. We don’t know before hand which methods will work with which clients, or why certain methods fall flat. So, because on the whole all methods work about the same, you’re free to pick the one you enjoy the most.
So that’s it. We offer active listening and experiential interventions. The rest of the change is out of our hands and probably dependent on the client. I mean that quite literally. Outcome research shows that the biggest factor in change is the client.
Once I realized this, each session became remarkably easier. It took the pressure off. We show up. We do our best. We realize our limitations. We leave the rest to rest. It’s a liberating way to work.
Where we go from here.
While right now we know very little about therapy and human psychology, eventually we will find better models and better training methods. That will be a glorious day. Our ability to relieve human suffering, to banish depression, to calm forever anxiety, or release people from addiction will be one of the greatest gifts and lasting legacies that we mental health providers will bestow upon humanity. And you and I are a part of that. Yes, in our own small way all of us mental health providers are working to enact that vision.
And despite knowing so little today, there are some things we can do now if we want to help more people. We can track our outcomes, record and watch our sessions, and practice doing instead of reading more books or articles. These are the three obvious areas to target if we want to improve.
But I want to give a word of warning. Whatever we do in the coming years to improve, the pressure needs to be at the institutional level. Grad programs need to automatically record every session, EHRs need to automate outcome tracking surveys, and Continuing Education training needs to be heavily focused on deliberate practicing. The alternative, to ask clinicians to spend hours doing this on their own time and for no pay, is quite simply asking them to try too hard.
Fin-
PS After writing this post I discovered this podcast episode by Farnam Street. For people wanting to understand the investor mindset described above this is a must listen.