IAPT Outperformed by Anxiety UK

In a recent study by Than et al (2022,) outcomes for attendees of the Charity, Anxiety UK have been compared with the results of attending the Improving Access to Psychological Therapies (IAPT) service. The authors claim a higher recovery rate, 62.4%, for Anxiety UK compared to 51.1% in IAPT. Anxiety UK clients attended a mean of 8.53 sessions compared to a mean of 6.9 sessions in IAPT. The results apply for treatment completers only, defined as attending two or more treatment sessions. But only 4.4% attend only one treatment session in Anxiety UK, but for IAPT in the year running up to the pandemic the figure was 44.7%. It seems that Anxiety UK is massively more able to engage clients than IAPT, strangely a point that Than et al (2022) completely missed.

Anxiety UK operates outside the NHS and does not use IAPT’s stepped care model of low and high intensity interventions. Raising the obvious question of why bother with such a distinction? The minimum access standard set for IAPT for 2020/2021 was 25%, suggesting that potentially 75% will not access the service. This makes the case for the provision of non-commissioned services external to the NHS.

But there are reasons to proceed with caution:

1. Both Anxiety UK and IAPT have marked their own homework. There has been no independent assessment of these two service providers.
2. They have adopted the same metric, changes in score on two psychometric tests,the PHQ-9 and GAD-7 between the first and last administration of these measures. Neglecting that clients a) may score lower on re-administration, because they don’t want to feel that they have wasted their time in therapy b) may not want to appear ungrateful by scoring near their initial score and c) present initially in therapy at their worst and there is some naturally occurring resolution of difficulties (regression to the mean).

3. There can be no certainty that the self-report measures are measuring what they purport to measure. The PHQ-9 is a measure of the severity of depression and only has validity if the individual has been reliably diagnosed (using a standardise semi-structured interview) as suffering from depression. Similarly, the GAD-7 is a measure of the severity of generalised anxiety disorder (GAD) and only has validity if the individual has been reliably diagnosed (using a standardise semi-structured interview) as suffering from GAD. It is perfectly possible to score highly on both measures but to have neither disorder, for example a Ukrainian refugee with a sub-syndromal level of post -traumatic stress disorder/ adjustment disorder and for whom psychological treatments might be misplaced as opposed to watchful waiting/support. The danger of relying entirely on self-report measures is that normality is pathologized.

4. Despite the claim of Than et al (2022) measuring outcome by a change of score on on self-report measures is not the most ‘client centric’ form of evaluation. It is essential that an outcome measure must be intelligible to the client, changes in psychometric test scores lack any clear meaning to clients. What clients do clearly understand is whether or not they are back to the former self’s post-treatment or at least back to their best selves and the duration of those gains. Inquiry into these domains by an independent observer is the only way of reliably determining whether there has been real world benefit from treatment. Both service providers have jettisoned, rigorous independent assessment.

5. Neither service provider has run fidelity checks to guarantee that the alleged CBT, clinical hypnotherapy, counselling etc was actually delivered. They both claim NICE compliance, but this necessitates matching a protocol with a reliably diagnosed disorder. Given that neither make diagnoses the claim of compliance has to be fraudulent. It appears an exercise in impression management with Anxiety UK but in IAPT it is arguably more insidious , a mechanism by which funds are secured.

6. The Anxiety UK authors totally ignore the controversy in the British Journal of Clinical Psychology Scott (2021) about the validity of IAPT’s chosen metric and that most plausibly only the tip of the iceberg recover Scott (2018). One service provider might be concerned to demonstrate a better performance than another but the crucial question is what is actually happening at the coal face? Nevertheless, I suspect working conditions are much better at Anxiety UK.

Dr Mike Scott