How Long Does Therapy Take?: What Research Says About Therapy Length

Last week I published a blog about the relapse crisis. Basically, of the clients who recover, about half of them relapse shortly after therapy. Much of this comes from research on short-term therapies for depression where clients are seen for 8-18 sessions. Today, I'm answering a reader's question that perfectly captures the tension between what we see in research studies and the reality of deep, meaningful therapeutic work. The question has been lightly edited for clarity and to omit identifying details.

The Question:

"I’m curious about the research about depression you mentioned. 18 sessions seems so few! I’ve known people who’ve experienced serious trauma. Some of them have done over 100 sessions—many twice a week. One person in particular states her therapist was her lifeline during this period of her life.Is there any research on therapy beyond just 18 sessions? For most of my clients 18 sessions wouldn’t even begin to scratch the surface! I can only guess the same for participants in those research studies."S.N.

The Answer: How Long Does Therapy Take?

Thanks for your question! It is such an important point, and your experience with friends and clients highlights why the question of "how long does therapy take?" is so complex and personal.

The rule of thumb I use is: most change on standardized measures (like depression surveys) peaks around sessions 25-35. After that you get diminishing returns. So yes, people absolutely do need more support than the 8 or 18 sessions you often see in studies.

Your experience also points to another key idea: it’s likely that people need change that can’t be measured on these quantitative surveys. I’ve had this happen many times. After 20 sessions, a client is no longer showing up on a survey as “depressed,” but they are still hurting. The survey says they're "recovered," but the person in front of me is still clearly suffering.

My sense is that once clients score well on quantitative measures, we need to move to using qualitative measures.

Basically, after a client scores well on the HAM-D (a quantitative measure), we need to ask them, “How do you feel? Is there still work to be done?” (a qualitative question).

All that being said, our research on “how long does therapy take?" has two big problems, especially when looking at long-term therapy:

1. It's Hard to Study: People may benefit enormously from long-term therapy, but it’s incredibly difficult and expensive to set up high-quality research studies that last for years. For example, a large part of what made John Gottman so famous was simply the fact that he’s followed couples for decades.

2. Survivorship Bias: Most people simply don’t stay in therapy very long (often due to cost, time, or feeling "good enough"). The rule of thumb is most clients in community clinics come for about 5 sessions, and most clients will see a therapist at a group practice for 8-12 sessions. The ones who do stay for 100+ sessions are a specific group, which makes it hard to generalize their results to everyone.

   
   

If we really wanted to get into the weeds, I’d say the question of "how long does therapy take?" is, technically, the wrong question.

For example, in education, we often move students from one grade to the next, NOT based on whether they've mastered the material, but simply because it’s the end of the year and it’s time to move on. We’ve built the entire system around this idea of age-based grade levels.

Sometimes in therapy we do something analogous. There’s a temptation to equate “effectiveness” with length of time in therapy. But we are probably better served looking at process markers. For instance, maybe we should ask, “How many stages of change has this client moved?” or “Has this client moved from the precontemplation stage (where they don't yet see a problem) to the action stage (where they are actively working to change)?”

Notes:

After responding to this question via email, I double checked my reference. Here’s what I found:

“Using the stricter change criteria, 50% of patients starting treatment in the dysfunctional range required 21 sessions to meet clinically significant change criteria. However, for 70% of patients to meet these criteria, more than 35 sessions were necessary. Numerous other studies in different countries have replicated this dose-response pattern, despite the application of different outcome measures and analytic approaches (Barkham et al., 1996)."

• p 96 Bergin and Garfield's Handbook of Psychotherapy and Behavior Change, 7th ed (2021).

I hope this serves you.

Best,

Jordan (the counselor)

---

Jordan Harris Jordan Harris, Ph.D., LMFT-S, LPC-S, received his Doctor of Philosophy in Marriage and Family Therapy from the University of Louisiana Monroe. He is a licensed professional counselor and a licensed marriage and family therapist in the state of Arkansas, USA. In his clinical work, he enjoys working with couples. He also runs a blog on deliberate practice for therapists and counselors at Jordanthecounselor.com