Photo by Remy Gieling on Unsplash
Since 2008, the NHS has treated over a million people annually through IAPT (Improving Access to Psychological Therapies). It's been called a "world-beating standard."
During this same period, recorded common mental health disorders increased from 17.6% in 2007 to 22.6% in 2023/4. One in five young people now has a probable mental health condition.
We're treating more people than ever. They're getting sicker than ever.
Something's not working.
CBT dominates NHS mental health provision because it offers something irresistible to a stretched system: quick, measurable work that can be standardised and delivered at scale. Six to twelve sessions. Structured protocols. Questionnaire scores that show "recovery."
The economic case was explicit from the start. Richard Layard, the economist who championed IAPT, argued it would pay for itself through reduced welfare costs. Get people off benefits, back to work. £650 for therapy versus £650 per month in disability payments. Simple maths.
But the service figures raise harder questions: 60% of people referred don't complete treatment. Of those who do, about half "recover" - meaning their questionnaire scores drop below a clinical threshold. Not necessarily thriving. Not necessarily understanding themselves better. Scoring differently on standardised measures.
Some are back within a year. Same patterns, different symptoms.
CBT can treat your mind like faulty software that needs debugging. Negative thought? Challenge it. Catastrophising? Here's a worksheet. Anxiety? Let's restructure those cognitions.
It's compelling because it offers control. You can master techniques, complete homework, track your progress. For certain specific problems - a phobia of flying, panic attacks with clear triggers, some forms of OCD - this can be genuinely helpful.
But most human suffering isn't a technical problem. It's not about thinking errors that need correcting.
It's about patterns laid down before you could speak. Relationships that taught you who you had to be to survive. The gap between who you are and who you learned to perform. The rage you weren't allowed to feel. The grief that has nowhere to go.
CBT can struggle to touch that layer. It may teach you to manage symptoms while the underlying patterns continue largely untouched.
And that management isn't temporary. It's a life sentence of self-correction - monitoring your thoughts, challenging your cognitions, policing your own mind indefinitely. A person left to supervise their own thinking patterns forever isn't healed. They're employed.
Here's something curious: knowing the techniques doesn't necessarily make them feel sufficient when your own life falls apart. You can spot every cognitive distortion, understand the protocols inside out, and still want a space that goes deeper than symptom management.
Because when you're lying awake at 3am, you don't need someone to challenge your thinking errors. You need to understand why you keep ending up in the same painful place.
When your third relationship implodes in exactly the same way, you don't need a thought record. You need to understand what you're unconsciously recreating and why it feels so familiar.
When you've spent years being competent and capable and suddenly can't get out of bed, challenging negative thoughts may not reach what feels most wrong.
Psychodynamic therapy starts from a different premise: your symptoms make sense. They're not errors to be corrected but communications to be understood.
Depression might be protecting you from rage you're not allowed to feel. Anxiety might be one way your body has learned to say no. That pattern of choosing unavailable partners? It might be the kind of love that feels familiar.
The exhaustion behind your competence. The performance your relationships have become. The gap between who you appear to be and who you actually are.
This work takes time because we're not just fixing surface problems - we're trying to understand the machinery that creates them. It's not efficient. It can't be standardised easily. It may not be fully captured in questionnaire scores.
But it aims at the patterns beneath symptoms rather than only managing the symptoms themselves.
The mental health crisis may not only be a treatment failure. It may also be a conceptual one. We've mistaken emotional suffering for a technical problem. We've confused symptom management with healing. We've prioritised getting people functional over helping them understand what might actually be wrong.
CBT can be useful for what it was designed to do: reduce symptoms in a structured, measurable way. The system is, to a significant extent, working as intended.
But human suffering isn't a productivity issue. Trauma isn't a thinking error. The patterns that keep us stuck are not always created by faulty cognitions - they may be survival strategies that made perfect sense at the time.
You may not be able to worksheet your way out of patterns that were formed before you had language. You can't easily challenge thoughts about feelings you don't know you're having. You can't always restructure cognitions when the problem is what you learned about who it's safe to be.
The NHS measures "recovery" through questionnaire scores. Did your PHQ-9 drop below 10? You may be recorded as recovered from depression. Never mind that you might be back at your GP six months later. Never mind that the pattern that contributed to the depression may still be running.
We're measuring symptom reduction, not understanding. Functionality, not meaning. Return to work, not quality of life.
Calling this success can obscure the fact that mental health continues to worsen.
CBT isn't useless. For some people, in some situations, it's exactly what's needed. Learning to manage panic attacks or challenge specific phobias can be life-changing.
But we need to be honest about what we're doing. Teaching people to cope isn't the same as helping them heal. Managing symptoms isn't the same as understanding roots. Getting people functional enough to work isn't the same as helping them build lives that feel worth living.
The question isn't whether CBT or psychodynamic therapy is "better." The question is: what are we actually trying to achieve?
If we want to manage symptoms and hit targets, CBT may do that efficiently for some people.
If we want to understand why people keep ending up in the same patterns, why the same problems keep recurring despite everyone's best efforts, why rising access has not ended the problem - that may require something different.
It requires being willing to sit with not knowing. To tolerate messiness. To value understanding over efficiency. To work at the pace of actual change rather than the pace of questionnaire scores.
Most importantly, it requires admitting that the current system - despite treating more people than ever - isn't enough for many people.
The production line is running perfectly. We're just not sure what it's producing anymore.
