The Hospital Anxiety and Depression Scale (HADS) - a discussion

ME/CFS Skeptic

ME/CFS Skeptic

Senior Member (Voting Rights)
Thought it might be helpful to have a thread on the use of this scale. The HADS is often used in research into ME/CFS because it was, as is said, designed to measure anxiety and depression in patients who are physically ill.

Back in 1990 Lewis & Wessely did a study on the usefulness of HADS to detect anxiety and depression, compared to the General Health Questionnaire (GHQ) and the Clinical Interview Schedule (CIS). Yet, one of the creators of the HADS, Snaith responded that this is not how the scale should be used. He wrote:
The study by Lewis & Wessely [...] provides example of the misuse of the Hospital Anxiety and Depression (HAD) scale. As they point out, there is no provision of score values for the sum of the two subscales. This is because the HAD was produced in order to distinguish between the constructs of depression and anxiety and not to provide an alternative to the General Health Questionnaire (GHQ) as a screening test for the presence of a global psychiatric disorder."
Click to expand...

I know little about HADS and its origin (EDIT: and I don't think it is normally used as Lewis & Wessely did in 1990], but thought it might be useful to have a thread with references that are relevant to understanding it's use.
 
Last edited:
I'd not happened across the HADS until filling out a pre-appointment questionnaire for the local CFS clinic. Some of the questions are ambiguously worded, in my opinion, if you put them alongside other questions regarding to fatigue, particularly "I feel as if I am slowed down..." and "I still enjoy the things I used to enjoy...". If I answer those literally, in the context of my ME, then the answers are at the extreme negative end because I am fatigued/in pain and because I can't do the things I used to enjoy.

I presume the reason for the scoring will come out in the appointment but they struck me as open to misinterpretation.

As an engineer who's spent his career capturing, processing and analysing data I find how this type of information is assessed interesting.
ETA: I've just read a little more about it and maybe I'm missing the point. I'm having a brain off day today.
 
Michiel Tack said:
Thought it might be helpful to have a thread on the use of this scale. The HADS is often used in research into ME/CFS because it was, as is said, designed to measure anxiety and depression in patients who are physically ill.

Back in 1990 Lewis & Wessely did a study on the usefulness of HADS to detect anxiety and depression, compared to the General Health Questionnaire (GHQ) and the Clinical Interview Schedule (CIS). Yet, one of the creators of the HADS, Snaith responded that this is not how the scale should be used. He wrote:


I know little about HADS and its origin (EDIT: and I don't think it is normally used as Lewis & Wessely did in 1990], but thought it might be useful to have a thread with references that are relevant to understanding it's use.
Click to expand...
The broader use of questionnaires in medicine is a growing problem, it's turning the whole practice into pseudoscience. It simply has no place other than as hint, it should never be considered a reliable and formal measure of anything. Questionnaires can be used to show anything a dishonest (or confused) researcher wants to show and there is zero accountability since they are highly subjective, not only in their results but in the very choice and inflection of questions they choose to ask.

This is a debate medicine needs to have among itself. The deliberate use of highly imprecise measurements that are somehow laundered into credible findings always leads to failure. It's already created several human rights disasters and their growing use will only make things worse. In honest hands there is value but the potential for fraud and abuse is immense and it's already been normalized.
 
ringding said:
ETA: I've just read a little more about it and maybe I'm missing the point. I'm having a brain off day today.
Click to expand...
No I think you're seeing it clearly. The problem is misuse. The questionnaires don't say anything by themselves, it's the interpretation that can lead to problems and it's already such a clear problem that their continued use is problematic. Unfortunately they are so open to interpretation that any combination of answers can be used to support any assumption a researcher or clinician may have.

It's just not good enough to say "it shouldn't be used this way", that's how it is used by some and that's already leading to massive problems. It basically creates blind spots that are wholly unaccountable and subject to magical thinking and confirmation bias. In our case it literally lead to a decades-long systemic failure of a misleading body of "evidence" built entirely on may bes and could bes. That's total failure, full stop.

Tools are only as useful as how they are used by practitioners. These tools have been misused for decades, pretending otherwise, that in perfect conditions they still have value, is beyond unacceptable, has been for a long time.
 
Coyne called for HADS to be abandoned a while ago since the results were very inconsitent. Some references here:


https://www.rug.nl/research/portal/files/6774405/Coyne_2012_J_Psychosom_Res.pdf
https://www.rug.nl/research/portal/files/6780761/Coyne_2012_J_Psychosom_Res_2.pdf

I think this is based on a systematic review from Cosco https://pdfs.semanticscholar.org/011a/45b1e6e0ac3759133cac0a0df65edc9e8cbd.pdf who concluded

The heterogeneous results of the current review suggest that the latent structure of the HADS is unclear, and dependent on statistical methods invoked. While the HADS has been shown to be an effective measure of emotional distress, its inability to consistently differentiate between the constructs of anxiety and depression means that its use needs to be targeted to more general measurement of distress.
Click to expand...
 
It is also worth reading the questions especially the depression ones - the answers are my guesses over potential reasons but they say there are different reasons for appearing depressed on the HADS scale than actually being depressed!

"I still enjoy the things I used to enjoy" - No I am too ill to enjoy them (i.e. not depression)
"I laugh and see the funny side of things" - Perhaps but in a way that uses less energy
"I feel cheerful" - Perhaps but this is where the points of emotional distress and the overall stuggles with daily living perhaps have a big impact.
"I feel I am slowed down" - again a response to illness (and PEM?)
"I have lost interest in my appearance" - I'm ill and find it hard to do the basics (washing etc so yes I try not to care).
"I look forward with enjoyment to things" - or dread because I can't manage simple things anymore but want to try.
"I enjoy a good book or radio/tv program" - Yes but it uses a lot of energy so not as often as I would like
 
Adrian said:
It is also worth reading the questions especially the depression ones - the answers are my guesses over potential reasons but they say there are different reasons for appearing depressed on the HADS scale than actually being depressed!
Click to expand...

Perhaps we could survey patients and ask what they associate the question answers with to see the reasons responses would change with ill/not ill rather than depression/not depression.
 
Michiel Tack said:
That seems like the opposite of what Snaith was saying.
Click to expand...

What Snaith seems to be saying is that HADS was designed to tell the difference between anxiety and depression rather than measure some general health notions. For this to be the case the underlying latent structures with the questionnaire would need to clearly show a separation when applied to a given set of patients. The review by Cosco seems to suggest that isn't the case and the latent structures vary with the analytic technique. So it suggests that HADS isn't good at doing that. I would point out that Cosco's work is much later and in light of HADS usage.

There is another issue which is why I listed and commented on the depression questions which is when looking at say depression vs fatigue or depression vs general illness then is there semantic confusion in the questions that will fail to pick out the depression vs illness. So in this case HADS could show something is wrong but has no power to discriminate why.
 
Adrian said:
Perhaps we could survey patients and ask what they associate the question answers with to see the reasons responses would change with ill/not ill rather than depression/not depression.
Click to expand...
Good idea. I think it would be worthwhile to interview patients to see what they mean when they fill in depression questoinnaires such as HADS. This shouldn't be too hard or expensive for a research team to do.

The argument that ME/CFS has much higher rates of depression than other illnesses has been one of the core arguments of the BPS view and one that seems to prevail many years after. It could be true of course, but I don't think this has received the amount of critical thinking it deserves within the research community. 1) I think that most studies didn't use controls that were equally ill and impaired and there seems to be a correlation between impairment and depression. 2) Controls often didn't have to face the stigma and disbelief that comes with ME/CFS. 3) As you say, questionnaires such as HADS seem to have difficulty discriminating between depression and fatigue-related symptoms.

Adrian said:
I would point out that Cosco's work is much later and in light of HADS usage.
Click to expand...
Snaith was saying what he wants HADS to do, Cosco et al. pointed out what HADS actually does. That's how I interpreted it.
 
Michiel Tack said:
Good idea. I think it would be worthwhile to interview patients to see what they mean when they fill in depression questoinnaires such as HADS. This shouldn't be too hard or expensive for a research team to do.
Click to expand...


I was wondering if we could do something from the forum - I will think about this a bit further. The issue is that researchers still use HADS as a meaurement instrument particularly Crawley (if I remember correctly) and things like PACE used it.
 
Very early on in the CFS/BPS debacle (before I had internet) an American patient who had been a researcher (well educated anyway) looked at the various questionnaires used. She found that those using fatigue, headaches, and things like that gave very high scores of depression in people with CFS but if all the questions about physical symptoms were removed the numbers dropped drastically.

In itself this showed that the questionnaires were useless. The BPS people ignored it. I think it was this research that SW dismissed as "not being done by a doctor" and she replied that she lectured doctors. :)

Also remember that Behan did research looking at muscle biopsy and found something like 70% had dysfunctional mitochondrial function. At around the same time SW released a paper which had something like 85% depression. He said that the higher figure he found showed that mental stress was most important to look at.

Even without any discussion of the pros and cons of the actual research, depression is a yes/no condition either present or not but looking at one strand of muscle you do not know if it is a damaged fibre or not. You can have symptoms even if only some of the fibres are diseased.

So the exact numbers in the trials can't be compared. I felt at the time that depression could only be the cause of CFS if everyone had it while even if only one muscle strand had been shown to be damaged it was a pointer for further research.
 
Mithriel points out physical symptoms of ME are often those that also occur in depression

But

HADS focuses on non-physical symptoms so that it can be used to diagnose depression in people with significant physical ill-health. Any overlap, for instance impaired concentration secondary to pain rather than depression, is usually easy to separate on an individual basis. HADS does not include all of the diagnostic criteria of depression (Diagnostic and Statistical Manual of Mental Disorders, Fourth/Fifth Edition (DSM IV/V)) or all those required by the Health and Work Development Unit (HWDU) National Depression and Long Term Sickness Absence Screening Audit [4].
https://academic.oup.com/occmed/article/64/5/393/1436876​

Do I sit relaxed and at ease?
Hell no! I have severe ME and little skeletal muscle support. I slump painfully when I sit up.

McCue 2006 found it used a 3 factor model, including autonomic anxiety.

http://www.academia.edu/download/3435973/mccue.etal06.pdf

They concluded it valid in CFS / Fukuda without assessing the scores in CFS for patients without anxiety or depression.

McCure also states:

Establishing the underlying factor structure of the HADS for use in individuals with CFS is important because the instrument has been recommended as a screening tool of choice for CFS (Deale & Wessely, 2000), with little psychometric support for such an assertion.
​
HADS is a favorite tool for Crawley.
 
daftasabrush said:
But

HADS focuses on non-physical symptoms so that it can be used to diagnose depression in people with significant physical ill-health. Any overlap, for instance impaired concentration secondary to pain rather than depression, is usually easy to separate on an individual basis. HADS does not include all of the diagnostic criteria of depression (Diagnostic and Statistical Manual of Mental Disorders, Fourth/Fifth Edition (DSM IV/V)) or all those required by the Health and Work Development Unit (HWDU) National Depression and Long Term Sickness Absence Screening Audit [4].
https://academic.oup.com/occmed/article/64/5/393/1436876
Click to expand...

Well if they mean non physical then they should say so loud & clear... It should say (for example) 'If you were physically able to read a good book or watch tv... would you enjoy it'?
or
If you had no reason to fear being inappropriately sectioned for being physically ill, losing your home when they take away your benefits, or be forced into undergoing 'treatment' that would increase your suffering 50fold in order to keep your benefits& thus a roof over your head... would you "get a sort of frightened feeling like something awful was about to happen'.

Sorry that last one was facetious, but honestly! No one should be using HADS with ME patients, they use it knowing the problems with it because it gives them the results they want. THEY are the ones creating the 'frightened feeling'.... so it's pretty twisted to then use one's response as a reason to justify the 'treatment' that creates the frightened feeling in the first place.

:banghead:
 
Language is so tricky anyway that too much weight is being put onto the answers to questions which are easily misinterpreted.

Nigel Speight once reported that when he asked his young patients how their walking was now a lot of them said it was fine. He did not think they could be that much better so he asked for details. The usual answer was that it was no problem now that they had a wheelchair (especially if it was electric!)

Same with the PIP assessment form. They ask things like "Do you use aids to walk more than 50 metres?" No could mean that you don't need aids or that you can't manage that far even with aids.
 
Mithriel said:
Language is so tricky anyway that too much weight is being put onto the answers to questions which are easily misinterpreted.

Nigel Speight once reported that when he asked his young patients how their walking was now a lot of them said it was fine. He did not think they could be that much better so he asked for details. The usual answer was that it was no problem now that they had a wheelchair (especially if it was electric!)

Same with the PIP assessment form. They ask things like "Do you use aids to walk more than 50 metres?" No could mean that you don't need aids or that you can't manage that far even with aids.
Click to expand...
giphy.gif


By far the most ironic thing in all this is how, despite the psychosocial belief system, most ME patients seriously downplay their symptoms. They have no idea.
 
You must log in or register to reply here.
Back
Top Bottom