NHS Talking Therapies use two psychometric tests to monitor outcome, the PHQ-9, a measure of the severity of depression and GAD-7, a measure of the severity of generalised anxiety disorder. But nevertheless proclaim a 50% recovery rate, (IAPT Manual, February 2023 ) across a heterogenuous set of disorders. Depression and GAD are but two of the multitude of disorders that present to the Service. An assessment of clientele by Hepgul et al (2016) using a diagnostic interview, covering 16 disorders, found that just over a third, 37%, fell into the depression or GAD category. Thus the Service has pontificated on outcome for the two thirds of its population that it has not used a disorder specific measure for. Matters are even worse in that specific phobia, body dysmorphic disorder and illness anxiety disorder were not assessed by Hepgul et al (2016). Making the spectrum of disorders even broader. NHS Talking Therapies is telling more than it can possibly know, this is blatant marketing.
But Hepgul et al (2016) showed how misleading the two tests used in isolation could be. 72% of the sample scored over 10 on the PHQ-9 but only 52% were found to be currently suffering from depression by the diagnostic interview. Sole reliance on the PHQ-9 would mean 1 in 3 people treated unnecessarily for depression.Whilst 78% of the sample scored over 8 on the GAD-7 only 66% were found to have GAD, still an overdiagnosis resulting in overtreatment, albeit not as glaring as for depression.
The fallacies of using a psychometric test in lieu of diagnosis are elaborated in detail in Clinical Psychology Science and Practice ( 2023). Bovin and Marx (2023) highlight that this reification of a psychometric tests leads to both a missing of those with a disorder and the unnecessary treatment of those without a disorder.
Though their focus was on another test, the PCL-5, used with those who have been traumatised, their findings were that cut-offs varied with the particular population addressed and the prevalence of the disorder in the particular community. Without attention to these details, clients will be misdirected.
Listening is the way forward
Bovin and Marx (2023) suggest that a test should only be used as an adjunct and involve a discussion with the client as to the meaning of each item. But this is highly likely to introduce extra bias in the UK, as clinicians are penalised if they do not achieve 50% of patients falling below caseness. They further point out that an essential part of a diagnostic interview is the assessment of social and occupational impairment, a psychometric test cannot do this. Use of a diagnostic interview involves taking the time to listen to the client, if this does not happen treatment will likely be built on sand.
A psychometric test takes a photograph of an individuals functioning over typically the last 2 weeks, but most disorders naturally take a fluctuating course. A ‘video’ of how they have been functioning over say the last 8 weeks is likely to be much more informative about their real-world functioning. But the distillation of this video with the client, takes prolonged listening. This has become anathema in the UK & US, targets have become everything with nobody questioning whether they are meaningful. Speed has become everything but without any understanding of where you are heading – headless chickens suggest themselves. The name of the game is apparently to take a photograph of the client as soon as they hit a better patch, claim success and reimbursement and close the case. In the DSM-5-TR the diagnostic criteria specify a 6 month timeframe for GAD, social anxiety disorder and illness anxiety disorder.
Dr Mike Scott