Fatigue in selected primary care settings: sociodemographic and psychiatric correlates, 1996, Hickie, Lloyd et al

An old but influential paper:
I B Hickie 1, A W Hooker, D Hadzi-Pavlovic, B K Bennett, A J Wilson, A R Lloyd

Abstract
Objectives: To determine the prevalence and sociodemographic and psychiatric correlates of prolonged fatigue syndromes among patients in primary care.

Design: Prospective questionnaire survey.

Patients and setting: Adults over 18 years attending three general practices in metropolitan Sydney and one on the Central Coast, north of Sydney.

Results: Of 1593 patients, 25% had prolonged fatigue, while 37% had psychological disorder. Of the patients with fatigue, 70% had both fatigue and psychological disorder, while 30% had fatigue only. The factors associated with prolonged fatigue were concurrent psychological disorder, female gender, lower socioeconomic status and fewer total years of education. Patients with fatigue were more likely to have a current depressive disorder.

Conclusions: Prolonged fatigue/neurasthenia syndromes are common in Australian primary care settings, and are commonly associated with current depressive disorders. Such syndromes, however, do not fit readily into current international psychiatric classification systems.

https://onlinelibrary.wiley.com/doi/abs/10.5694/j.1326-5377.1996.tb122199.x
In the Medical Journal of Australia

Paywall

 

This paper has been cited in support of this statement:

in an article in the Australian Family Physician by John Murtagh - reproduced on the Royal Australian College of GPs website:
https://www.racgp.org.au/getattachment/9842f4ab-a0a8-4c67-8546-271853a0d0f7/attachment.aspx

Content from that article has been used in guidance to doctors and patients, suggesting that fatigue is mostly no big deal and can be fixed by lifestyle factors
e.g.
Question 1: What are probability diagnoses?
Answer 1: Stress and anxiety, Depression, Viral/postviral infection, eg. EBV, Sleep related disorders, eg. sleep apnoea
...
Question 5: Is the patient trying to tell me something?
Answer 5: Highly likely. Psychogenic and lifestyle disorders are very common in this presentation in general practice

So, I'm keen to understand how reliable that '50 to 80% of fatigue cases are mainly due to psychological factors' really is. If anyone can share the Hickie paper with me, I'd appreciate it. (Edit, I've accessed it via academia.edu)

 

Back to the 1996 paper:
A series of questionnaires were administered to people attending GP clinics in Sydney.

GHQ - General Health Questionnaire - "used to detect cases of psychological disorder"
"symptoms include anxiety and depression; patients were defined as having a "psychological" disorder if they experienced more than five of the symptoms "more than usual" "
I wonder what exactly the 'usual' is that the participants are supposed to be comparing their symptoms to.

SOFA - The Schedule of Fatigue and Anergia - (I bet the person who came up with that acronym thought they were very clever, and hilarious)
A scale of 10 physical and psychological symptoms used to identify cases of severe, chronic and disabling fatigue present for at least a month. The SOFA was derived from symptoms reported by patients 'who met clinical criteria for chronic fatigue syndrome' and is reported to be good at separating people with CFS from healthy people.

The authors modified the scale, asking the participants to rate the frequency of the 10 symptoms over the past two weeks. If they answered 'most of the time' or 'a good part of the time', they were regarded as positive for that symptom. A patient with three or more positive responses was regarded as having 'fatigue'. This modified use seems all fairly arbitrary and unvalidated.

General practitioner diagnosis - the general practitioner provided the reason the person attended the consultation. The authors coded the responses as either 'a clear medical condition (e.g. hypertension), a non-specific complaint (e.g. headache, nausea, fatigue), a clear psychological conditions (e.g. depression, anxiety disorder), or the person as accompanying another person to the surgery'.

Psychiatric evaluation - participants identified by the GHQ as having a psychological disorder and participants identified by the SOFA as having prolonged fatigue were given a psychiatric evaluation. The interview covered questions about depression, anxiety disorder and somatoform diagnoses.

 

Around a quarter of people approached to participate refused. 1593 people participated. 75% of the sample was female.

25% of the participants reported three or more symptoms of 'fatigue', from the SOFA questionnaire. (But, actually, these were supposedly symptoms of chronic fatigue syndrome). 70% of these also had a 'psychological disorder'.

SOFA questions (to diagnose fatigue) seem to include:
excessive tiredness after activity
sleep difficulties
poor concentration
poor memory
muscle tiredness
pain



37% reported 5 or more symptoms of so-called psychological disorder in the GHQ

17% were classed as having both fatigue and a psychological disorder

*****
Only 1.3% of patients (participants?) were given a diagnosis of fatigue, malaise, lethargy or chronic tiredness by their GP, and only 0.3% were given a formal diagnosis of chronic fatigue syndrome. So, the diagnosis of fatigue via the SOFA questionnaire seems extremely permissive (reminiscent of many approaches to identifying Long Covid).

 

"Fatigue was more common in women, patients of a lower socioeconomic status and patients with fewer years of formal education."
Patients from the practice in the poorest area had the highest rates of fatigue and psychological disorder.

On the 'fatigue' classification using the modified SOFA:
People were asked to say if they had suffered from each symptom over the last two weeks, and the symptoms included pain, and sleep difficulties.
Poorer people might be more likely to have multiple conditions, they may well be working more than one job, maybe physically exhausting jobs, they might not have help with housework and live in an uncomfortable house, maybe a noisy crowded house, they might have much longer commuting times, more children. It's entirely possible that poorer people are more fatigued.

But, also, there could have been a difference in the nature of consultations being had by people of lower economic status as compared to people of higher economic status. For example, people of lower economic status might only go to see the doctor when they have a really bad problem, with the pain keeping them awake, and the lack of sleep causing cognitive issues - or they might only see the doctor when they have a few issues that they want to get sorted all at once. Whereas the people of higher economic status might just go in for their flu injection, or Pap smear, or annual check up, or to renew a prescription. So, the poorer people would be more likely to tick a lot of the SOFA boxes about their previous two weeks and be classed as 'fatigued', whereas the richer people might have been fatigued, but wouldn't necessarily also have pain and cognitive difficulties as well and so wouldn't be so likely to be classed as 'fatigued'.

That idea is supported by the report in the paper that the modified SOFA only classified 42% of the participants ticking the 'major fatigue symptoms' SOFA boxes as fatigued, but 92% of people ticking the 'muscle tiredness and pain' SOFA boxes as fatigued.

Therefore, remarkably, the modified SOFA isn't great at identifying people with fatigue, but might be fairly decent at identifying people who have had pain and generally felt pretty rubbish in the last two weeks.

I don't think that the study reliably tells us about people with fatigue.

 

Oh, here are the questions in the modified SOFA - the measure of fatigue:

1. I feel tired for a long time after physical activity
2. My concentration is poor
3. My muscles feel very tired after physical activity
4. I get headaches
5. I need to sleep for long periods
6. I get muscle pain after physical activity
7. I sleep poorly
8. I have problems with my speech (e.g. feeling "lost for the word")
9. My memory is poor
10. I get muscle pain even at rest

Incredibly, I'm not sure that any of those questions really describes the heavy dead feeling of limbs and the feeling of having to use a lot of will power in order to move that I associate with fatigue.

(Edit - that list isn't even great for identifying CFS, which was what it was actually designed for.)

 

So, to the GHQ-30, which supposedly detects cases of psychological disorder.
That was the one that has 30 items, and people were defined as having a psychological disorder if they experienced at least five of the symptoms 'more than usual'.

1. Could not concentrate on whatever you are doing?
2. Lost much sleep over worry?
3. Been having restless, disturbed nights?
4. Not been managing to keep yourself busy and occupied?
5. Not been getting out of the house as much as usual?
6. Not been managing as well as most people would in your shoes?
7. Not felt on the whole you were doing things well?
8. Not been satisfied with the way you've carried out your task?
9. Haven't felt warmth and affection from others?
10. Haven't been finding it easy to get on with others
11. Haven't spent much time chatting with people?
12. Haven't felt that you are playing a useful part in things?
13. Haven't felt capable of making decisions about things?
14. Felt constantly under strain?
15. Felt you couldn't overcome your difficulties?
16. Been finding life a struggle?
17. Not been able to enjoy your normal day to-day activities?
18. Been taking things hard?
19. Been getting scared or panicky for no good reason?
20. Could not face problems
21. Found everything getting on top of you?
22. Been feeling unhappy and depressed?
23. Been losing confidence in yourself?
24. Been thinking of yourself as a worthless person?
25. Been feeling that life is entirely hopeless?
26. Not been feeling hopeful about your own future?
27. Not been feeling reasonably happy, all things considered?
28. Been feeling nervous and strung-up?
29. Felt that life isn't worth living?
30. Found at times you couldn't do anything because your nerves were too bad?

Given that the people with a fatigue label were actually largely people with pain, it's interesting to think about how someone with pain that was stopping them being as mobile as normal, for example, might score the GHQ-30, compared to 'usual', i.e. when they don't have pain. I think it would be fairly easy for them to score at least 5, and so qualify as having a 'psychological disorder'.

 

Yes, so the measure of fatigue is actually non-specific and so is the measure of psychological disorders. It's quite unreasonable to start correlating these inaccurate measures and suggest that it tells us anything useful about the presence of psychological disorders in people with fatigue.

And, even if it did, we could not know what caused what. Did having fatigue cause someone to be sad and worried, or even depressed and with an anxiety disorder? Or did being depressed cause the fatigue?

We certainly couldn't conclude from what we have seen so far in this paper that fatigue is caused by psychological issues.

 

GHQ 30

blimey I would imagine the population at large would tick 5 of those

 

Empire building, par excellence.

Hickie is also responsible for the equally bad SPHERE check list.

https://onlinelibrary.wiley.com/doi/abs/10.5694/j.1326-5377.2001.tb143784.x
https://bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-017-1420-1

 

That seems pretty important.

(Bearing in mind that the "fatigue" here isn't really fatigue, it's a range of things, but is better labelled as pain, than fatigue. And, I'm not at all confident in the CIDI diagnoses either, CIDI is the 'computerised version of the Composite International Diagnostic Interview'.)

The discussion goes on to say that the rate of fatigue that they found here is in line with rates in overseas studies. But, they didn't actually measure the rate of fatigue.

It actually doesn't, because the SOFA ratings aren't measures of fatigue, and they are what was correlated with GHQ ratings (that aren't measures of psychological disorders). They go on to say a things about females and people with low socioeconomic backgrounds and fatigue and psychological problems that aren't warranted by their study.

They say that lower socioeconomic background and fewer years of formal education are associated with fatigue and so are probably risk factors, but at least they acknowledge that it is possible that the causation goes the other way, and that people with prolonged fatigue are at increased risk of being poor and not having much education.

 

Out of nowhere, somatoform disorders makes a showing:

There's nothing in this paper to support that contention.

They note that the people who they labelled as having fatigue were not more likely to receive a lifetime diagnosis of depression.

There's this remarkable paragraph, although it too seems to have bene popped into the discussion at random:

They end by saying that patients with prolonged fatigue are at high risk of having an undetected psychological disorder, notably current depression. But, again their SOFA rating, which was used for correlations, was a poor measure of fatigue. And, the sofa measure asked about symptoms in the last two weeks. They did not measure 'prolonged' anything.

*****

All this is the long way round to say that this paper is rubbish. There are so many flaws. It tells us very little about the relationship between fatigue and psychological disorders, but a bit about the direction these authors desperately wanted to push chronic fatigue syndrome in 1996.

*****

And, to answer the question I posed near the beginning of this thread, is the paper an appropriate citation to support the claim that 50 to 80% of fatigue cases are mainly due to psychological factors? (A claim that is quoted in patient and doctor guidance on fatigue management.)

No, not even close.

I don't think even the authors would claim that it does. The only figure even remotely related in this paper is the claim that, of the patients with 'fatigue', 70% had both 'fatigue' and 'psychological disorder'. But I didn't see anything about 50 and 80%. And there's no claim, and certainly no evidence, that psychological factors cause fatigue cases.

It's so utterly wrong, that I'm wondering if I made a mistake with the reference that was given to support the 50 to 80% range. Or if Murtagh did.

 

I've checked, I'm pretty sure I didn't make a mistake.

The title isn't perfectly the same, but the page numbers in the Australian Medical Journal are exactly the same:
MJA 1996;164:585-588

If anyone has the time to check what I'm saying about the incorrect citation in the Murtagh article, I'd appreciate it.

 

This paper, like so many psych papers on ME, is soaked to the gills with the arbitrary assumption of psycho-causation.