Trial By Error: CBT Provides No Benefits to Advanced Cancer Patients, Study Finds

By David Tuller, DrPH

Since 2008, the National Health Service (NHS) in England has been rolling out a program known as Improving Access to Psychological Therapies (IAPT). Initially focused on patients with mental health issues like depression and anxiety disorders, IAPT was then expanded to include those who are also simultaneously suffering from “long-term conditions” and so-called “medically unexplained symptoms” (MUS). IAPT has cited 50% “recovery” rates from its interventions, but these claims are not especially credible, as Liverpool consultant psychologist Michael Scott has documented on his informative blog, CBT Watch.

Meanwhile, MUS experts who promote IAPT—some of them close academic colleagues of the lead PACE investigators–have regularly exaggerated the estimated costs of these ailments to the NHS. Specifically, they have misquoted a seminal study in their field of expertise and claimed that MUS patients accounted for 10 % of total NHS costs (in England). The actual figure from the study was that people of working-age with MUS accounted for 10 % of costs for that age group, not total NHS costs. The mistake more than tripled the apparent economic impact of MUS. Last year, two major journals—the British Journal of General Practice and BMJ Open–corrected papers they had published after I alerted them to these errors.

There is plenty of evidence that the IAPT’s hopes and claims exceed its current grasp. A study published last September by the Journal of British Psychiatry, for example, has found that interventions provided by IAPT therapists proved no more effective with a core patient group than treatment-as-usual. The study is called “Effectiveness of cognitive–behavioural therapy for depression in advanced cancer: CanTalk randomised controlled trial.”

(I came across this trial while reading Scott’s recent article, Ensuring IAPT Makes a Real World Difference, in the journal Mental Health in Family Medicine. Other references in Scott’s article reinforce concerns about IAPT.)

Most of the investigators were from University College London. In contextualizing their findings, they noted that “the benefits of CBT for people with advanced cancer had been previously unclear because of underpowered trials, poor diagnosis and measurement of depression, lack of detail about interventions and concerns about generalizability.”

Their own conclusion was pretty stark. Here’s what they wrote:

Delivery of CBT through the Improving Access to Psychological Therapies (IAPT) programme has been advocated for long-term conditions such as cancer. Although it is feasible to deliver CBT through IAPT proficiently to people with advanced cancer, this is not clinically effective…[O]ur results suggest that resources for a relatively costly therapy such as IAPT-delivered CBT should not be considered as a first-line treatment for depression in advanced cancer. Indeed, these findings raise important questions about the need to further evaluate the use of IAPT for people with comorbid severe illness.

The study included 230 participants with advanced cancer and depression; they were randomly allocated to receive CBT or treatment-as-usual and assessed at multiple time points up to 24 months. The main outcome measure was a well-known instrument called the Beck Depression Inventory-II, which has been used previously with patients with advanced cancer. The analysis found no statistically significant effect on depression and on several secondary outcome measures.

Those randomized to receive CBT did not access the intervention to the degree available. They each received an average of 4.7 sessions of CBT out of the 12 sessions possible. More than a third—36%–had no CBT sessions at all. Nevertheless, the investigators determined that the outcome would have been similar even if participants had accessed all the sessions. “The per-session effect of CBT was too small to scale up to a clinically significant change even if the full 12 sessions were delivered,” they wrote.

The limited uptake of CBT itself should raise further questions for IAPT. The investigators did not discuss possible reasons for this phenomenon beyond noting that “our clinical experience was that physically ill people had difficulty in managing the demands of CBT.”

The investigators made efforts to create fertile trial conditions for demonstrating the possible effectiveness of the CBT intervention. The study used only IAPT therapists already skilled in delivering so-called “high-intensity” CBT. Moreover, the investigators developed specialized materials and a training module to ensure that the intervention matched the needs of these advanced cancer patients. According to the study,

A treatment manual informed by previous work…was developed for the trial by members of the research team. CBT therapists attended a 1-day course on how to use the manual and adapt their standard CBT work for people with advanced cancer. This included adapting techniques within the constraints of physical illness, working with realistic negative thoughts, dealing with fears about death and dying, and including carers in sessions where appropriate.

Despite these favorable conditions, the trial produced null results—which in turn raise serious questions about the current NHS strategy for IAPT. Seems like UK health officials might have some rethinking to do.

Comments on this entry are closed.

  • Richard Vallee 15 January 2020, 9:40 pm

    “More than a third—36%–had no CBT sessions at all. Nevertheless, the investigators determined that the outcome would have been similar even if participants had accessed all the sessions.”

    Oooooh, a self-burn. Those are super rare.

    I agree but this is as frank an admission as it gets that the whole thing is entirely useless.

    Looks like IAPT can best be described as the most expensive way to figure out that CBT may be good at convincing health administrators of what CBT could do if it worked and not much else.

    “Is CBT an effective method of persuasion to convince health administrators of the value of funding CBT researchers and massive wastes of public funds?”

    Turns out, yes, yes it is. To the secondary question of whether it is of any use, the answer appears to be a very loud sad trombone sound, all to the low, low price of… how many billions by now?

    Extraordinary claims still do not work when even ordinary evidence can’t be provided. Bummer. Very costly bummer. Bit of a pickle for the retort that CBT is useful in cancer so it’s just as good with ME. Predictably, it’s useless for both. Now on to watching for when the sunk cost meets the austerity fan.

  • CT 16 January 2020, 5:28 am

    As I understand it, (from Scott’s paper which I can access), this study used the better trained High Intensity IAPT therapists that work at Postgraduate diploma level in CBT, not the low intensity PWPs (Psychological Wellbeing Practitioners) who complete an IAPT training course based around CBT. So these patients were being given the best psychotherapy that IAPT can offer.

    They were patients with advanced cancer, so this doesn’t rule out these therapists being able to help patients whose cancer hadn’t advanced so far. But I find it hard to understand how anybody could think that the IAPT programme was feasible for patients with cancer, let alone advanced cancer. It doesn’t come across as a patient-friendly experience – battling to get appointments, being kept waiting for ages, forms to fill in on every visit – adding to the stress that people are already under. But perhaps patients with LTC and MUS are now being prioritised so they have less of this, while those with boring old anxiety and depression (that won’t save the NHS heaps of dosh) are being pushed to the back of the queue? Because that’s what IAPT is all about, saving the NHS money. Wasn’t it always (economist) Lord Layard and (Oxford/ex-IoP psychologist) David Clark’s plan that IAPT would target those with physical health conditions in order to reduce expenditure on physical health services and the NHS’s secondary care budget? (And it may succeed in doing that, but for all the wrong reasons.) This isn’t parity of esteem for mental health services for these patients, if it was it wouldn’t pit mental health funding against physical health funding and wouldn’t consider cutting physical health spending up front before the economic results of the psychological interventions were fully known. This strikes me as 2 feeble men pulling out all the stops now to avoid facing the uncomfortable truth – that they have sold the UK government a pup because their theory is flawed – and that, rather than helping patients and delivering parity of esteem, IAPT is helping to facilitate the worst kind of austerity cuts in the NHS.

  • CT 16 January 2020, 6:04 am

    In case people are struggling to see the relevance of this blog for people with ME, they need to understand several things:

    a) David Clark, who is the co-architect of IAPT, came from the ‘Oxford School’ that promoted CBT for a whole host of things, including for ‘chronic fatigue’ (Michael Sharpe’s chapter), in the 1997 book – “Science and Practice of Cognitive Behaviour Therapy” that was co-edited by David Clark.
    b) The management model for MUS has been built on the management model for ME, or CFS as they like to call it. -Their intention to do this can be seen in the last sentence of the Abstract of this 1997 Wessely, Chalder, Sharpe (and Palmer) paper -https://www.ncbi.nlm.nih.gov/pubmed/9218987
    c) ME patients in the UK are now being directed to IAPT for CBT and GET and coded with ‘somatization disorder’ ICD-10 F45.0
    d) if you’re outside the UK you may think this won’t affect you, but Layard and Clark and people from the Oxford School and beyond are intent on selling the IAPT or similar MUS management models (there are variants) across the globe, and seem to be having some success in this.

  • @MECFSNews 16 January 2020, 6:27 am

    The rationale for CBT is that negative thinking patterns cause depression.

    Was there a reason to think that depression in cancer patients is caused by negative thinking patterns? As opposed to it being caused by an accurate appraisal of the current situation these patients find themselves in?

  • Lady Shambles 16 January 2020, 6:28 am

    If people with advanced cancer are unable to cope with the strain of attending CBT sessions due to the physical weight their disease already burdens them with, then how can it possibly be argued that people with the LTC ‘ME’ be helped either? For the very same reasons CBT would be impractical for pwme at the very least and that’s without factoring in that CBT when used for pwme is not intended to help with associated depressions (if they exist) but is assumed to be able to tackle the underlying aetiology of the disease, ie that people with ME have unhelpful illness cognitions than can be reversed using CBT (oh!… if only that were true!). Of course in reality people with properly diagnosed ME have the unique characteristic whereby expenditure of energy makes the illness worse, so attempting to get to any appointment is pretty much impossible (except in extremis) and for those who are severely affected the notion of CBT appointments, be they in person or via some sort of device, would be a complete non starter. And anyway we already know from the dodgy PACE Trial that CBT provides no benefit for people with ME so I’m not really sure why I’m trying to reiterate that old chestnut again..other than to compare the use of CBT between two groups of patients. As others have commented, the way cancer is used as the cynical default model pressing the case for using CBT in any disease (where the sleight of hand about its use being complementary as opposed to curative when applied to ME is never mentioned) is now, given these results from UCL, somewhat implausible.

    Btw: thank heaven for the rare psychologists like Michael Scott who don’t cow tow to the top-down dogma.

  • CT 16 January 2020, 6:31 am

    @MECFSNews

    – Listen to Trudie Chalder, Professor of CBT -https://newseu.cgtn.com/news/2020-01-05/-You-can-t-experience-happiness-if-you-don-t-feel-sadness-too–MXlMqxEgZG/index.html

  • CT 16 January 2020, 6:52 am

    Lady Shambles said: “thank heaven for the rare psychologists like Michael Scott who don’t cow tow to the top-down dogma.”

    I agree. Of course I can’t speak for ME sufferers, but I think most or all would recognize that it’s important for patients to have access to good mental health care and good quality talking therapies if they are struggling with their mental health as a consequence of trying to cope with their illness or for any unrelated reason. The problem with IAPT is that it isn’t good care. While the PWPs and High Intensity therapists no doubt do their very best for their clients, they are working in a high-volume, high-throughput programme that prioritizes results over individual patient care, and they pay the price for that too. Meanwhile patients in severe pain and suffering are being turned away by hospital services and directed to psych services such as IAPT because their symptoms are unexplained by the basic or standard tests and assessments that are on offer, and because doctors across the UK have been taught that such ‘medically unexplained symptoms’ equate to psychological or psychiatric problems. A dreadful state of affairs.

  • CT 16 January 2020, 7:39 am

    This looks to me like the same RCT (just one author added) with free full text available -https://www.ncbi.nlm.nih.gov/pubmed/31097078

  • SC 16 January 2020, 7:50 am

    Fraud,for financial gain in medicine seems to be on the rise. I read about a cardiac stent study recently that has been highlighted as misleading where the authors redefined what a heart attack is to make stent data look better, the data was withheld and the study leads had a financial interest in stents (Sounds like they used David Tullers reporting on PACE to get ideas). It’s only recently UK Cardiac Consultants realised they had been misled/fooled. Medical fraud is not limited to Psychology or Psychiatrybut it’s disappointing to see medicine ignoring the problem. At best it’s a waste of money and at worst patients are paying with their lives. IAOT seems to have the hallmarks of a money making scheme rather than delivering credible results.

  • CT 16 January 2020, 8:18 am

    It seems that the rationale for this study -https://www.ncbi.nlm.nih.gov/books/NBK541197/#s1-1 – on the use of CBT for advanced cancer patients was based on this 2009 review:

    Price A, Hotopf M. The treatment of depression in patients with advanced cancer undergoing palliative care. Curr Opin Support Palliat Care 2009;3:61–6. 10.1097/SPC.0b013e328325d17a

    ….and to be in line with NICE recommendations on depression and the current UK agenda of widening access to psychological therapies for patients with depression (presumably including co-morbid depression) via IAPT. I wonder what other psychological services are being or have been disbanded to make way for IAPT? I can’t believe that there were previously none available for patients with advanced/terminal cancer.

  • SC 16 January 2020, 2:06 pm

    CT, some really great points, as they say in US in white collar crime cases just follow the money and all will become clear. It’s interesting to see the flawed PACE trial results infecting mainstream illnesses. If only the Lancet and BMJ had listened to David Tuller.

  • mehul 17 January 2020, 2:17 am

    Great information to say the least. I really do appreciate everything so much from this great website.

  • Barry McInnes 17 January 2020, 11:35 am

    Hi David and thank you so much for this analysis. IAPT in the UK is an utter shambles. My own analyses have consitently found that only 1 in 4 that start therapy show recovery http://therapymeetsnumbers.com/iapt-2019-still-failing-to-thrive/
    More recently a survey of more than 550 current and past IAPT workers indicated that 41% had been asked to manipulate performance data. It’s hard to know how we can trust even the published data.

  • CT 18 January 2020, 4:19 am

    Thanks Barry. ‘An utter shambles’ is clearly not the right place to send people with severe and/or complex physical health problems and needs. From both angles – patient and therapist – that must be a recipe for disaster. (‘Shambles’ is an interesting choice of word, given its archaic meaning!)

  • Steve Hawkins 23 January 2020, 5:19 pm

    In full, the acronym is ‘CBTTCTTGTSE’

    ‘Clever Buggers Think They Can Teach Their Grandmothers To Suck Eggs’.