Is it true that more than half of medical consultations are for MUS? A look at the evidence.

[ME/CFS Skeptic]

It was a military hospital that served the military, their families and retired military. They claim the demographics were pretty representative of the population

Yes and 61% were women, so perhaps that's not the issue. But I was thinking that 38% is pretty low for these common symptoms (fatigue, headache, chest pain etc.). They only had to meet one, and yet the majority didn't. Perhaps the fact that it was a military hospital, could explain this. The demographics may be similar but possibly the health complaints were not.

 

[Jonathan Edwards]

It was a military hospital that served the military, their families and retired military. They claim the demographics were pretty representative of the population.

OK, though I am not sure I am convinced their assurances are worth much.

 

[Hoopoe]

I wonder what they say about severity of symptoms. People often see a doctor for minor problems that are probably too mild to be diagnosable.

 

[roller*]

but she argues...

The Up to Date system provides global access to concise summaries of current research and consensus on every condition. That system notes that more than 50% of patients are not suffering from a medical condition. Even if this one paper is utterly unscientific, even if Wessely inflated it and I inflated it and you can prove all of that beyond all doubt, what then? It's a drop of rain in the sea.

... then indeed 50% of patients to clinics are MUS according to the probably top notch harvard database uptodate

1) why, was the 30y old kroenke paper cited when the recent intl harvard data confirm its 50%?
2) is this kroenke paper stored in the uptodate-database at all?
3) how did the harvard system calculate that 50%. is there an infographic perhaps, or figures by country..?

 

[Barry]

If there is something I learned since becoming sick, it is that you always have to check by yourself.

I used to be in the RAF, and every single day of my apprenticeship they hammered home over and over: Don't assume. Check!

 

[Barry]

This debate about the 50% figure is not deserving of your time and energy, which are valuable. You'd get a lot more bang for your buck if you focused on the substance of the article.

But it's a crucial metric, and any value given needs to at least be in the right ball park. Which is why people have picked up on it here Diane. It is deserving of our time and energy.

 

[Diane O'Leary]

So Up to Date Collates the info and uses a 50% figure? Does that tell you the source of any potential meme that leads to the eventual figures for the diagnostic label MUS?

If we look back over the life of the term MUS is it on an upward curve or a downward one or has it remained steady?

When I stated the issue of the term MUS suiting the UK medical establishment I wasn't being exclusive. I am pretty sure overstating of such a diagnostic label could exist equally all around the world and would conveniently suit who ever picked up the bill for medical care be it in a socialised system or a private insurance based one.

There is also the issue of empire building and even if the figure is claimed to be the same all around the world that doesn't exclude potential deliberate marketing of the term to make it take on a life of its own via policy and education.

One only has to look at the US DSM as one example to understand how this could work.

Things can take on a life of their own built on poor foundations. If a figure is similar all around the world for a certain claim it doesn't tell you automatically what the motive was or wasn't for that.

Would you comment on the US DSM in terms of its categorisation of "illnesses" and the potential implications that could have in meme building and how it could affect the ability to collect accurate information on categorisation of diseases/diagnsotic labels etc?

Or perhaps comment on the US medical insurance industry?

Lots of great questions and insights here.

* Up to Date summarizes current consensus on every medical condition in a way that's useful right there in the room with the patient. So they don't do their own research and they never go out on a limb. They played no role in the 50% figure that's commonly accepted. That figure was accepted long before anybody imagined a database of this kind.

* I'd say estimates of MUS prevalence have remained steady, granting that you can find any number you like in the research. I expect they were highest in the 80s and 90s, but that research continues to drive current understanding and practice.

* I'm having a lot of trouble understanding why everyone is so convinced that diagnosis does take place most of the time for outpatients. Can you or anyone else explain why people think that?

* I don't see any health system or insurer building an empire by noting their budgetary proportions. MUS are one of their central concerns because "I don't know" is extremely common and extremely expensive in medicine. Nobody is getting rich or famous off that idea. Wessely and co have gotten rich and famous by offering a very cost effective way of managing that very real budgetary challenge.

* Important to note that there's no MUS in the DSM, as you probably know. MUS is an "abridged" form of DSM "somatization", so DSM's involvement in this is indirect. As for DSM's categorization of mental illnesses - I think many of their decisions are made by crap shoot, by personal power plays and utter nonsense. Sometimes that reality is catastrophic for diagnostic accuracy and information collection (and patients).

* US insurers face the same financial challenge as the NHS (plus they want to make a boatload of money). They know diagnostic uncertainty is as common as diagnostic clarity, and they know the patients in this group threaten their profit margin. Like the NHS, they have very reason to embrace the comfy solution offered by Wessely and co. Psych management of "I don't know" is soooo much cheaper.

 

[Hutan]

Another paper
The resource utilisation of medically unexplained symptoms
Lee et al 2016
This was a fairly recent paper by New Zealand authors who appear to be supporters of the concept of MUS.

Analyses from the United Kingdom have estimated that MUPS is the presenting problem for around 22% of primary care attendances and 8% of inpatient bed days.4,5

The references appear to be documents aiming to influence health policy. I haven't checked them, no doubt the estimates have a poor basis.
4. Personal Social Services Research Unit, London School of Economics and Political Science. Mental health promotion and mental illness prevention: the economic case. Report, Department of Health, London, January 2011.
5. Personal Social Services Research Unit, London School of Economics and Political Science. No health without mental health: a cross-government mental health outcomes strategy for people of all ages. Supporting document – the economic case for improving efficiency and quality in mental health. Report, Department of Health, London, January 2011.

 

[Diane O'Leary]

But it's a crucial metric, and any value given needs to at least be in the right ball park. Which is why people have picked up on it here Diane. It is deserving of our time and energy.

I'm still having a lot of trouble understanding that. Most of the time when people talk about MUS they mean as the BPS camp sees MUS - i.e. psychosomatic symptoms. In that context yes, of course, without doubt the 50% figure is completely ridiculous. Totally unsupported by science. So if we're discussing how common it is for symptoms to have psychological causes, the metric is crucial and wholly deserving of detailed time and attention. Much of the time policy makers understand it this way.

The thing is that MUS can also mean just symptoms that are not yet diagnosed. In this very simple sense we're just asking how often doctors get an answer - and I don't think there's much reason to quibble about that. Diagnostic science is not very effective. After many years of reading the stuff on prevalence, both from the BPS camp and from neutral parties just working out figures on diagnostic success, I don't see any reason to imagine that doctors are getting an answer most of the time. There's just too much evidence to show that usually they don't know what's happening.

This is the challenge of thinking and writing about MUS. The BPS camp has destroyed clarity about the difference between psychosomatic symptoms and unexplained symptoms - so we can't just chat about MUS as if we're all on the same page. Every time you talk about MUS you have to start by clarifying which group of symptoms you're talking about. Some studies address "I don't know", while others address psychosomatic symptoms. Some people in this thread are talking about "I don't know", while others are talking about psychosomatic symptoms. Sometimes people start with one then slide into the other.

The BPS program has succeeded because of this ambiguity. The only way to defeat this problem is to state clearly, every single time, what exactly you're talking about.

 

[Trish]

The 50% figure makes sense if you think about it. The vast majority of Sx that people take into a medical appointment are benign and self-resolving. And diagnostic science is just not that advanced.

Really? Is that a serious argument?

That's just not tenable. There is far, far too much research to support it, and this research is global. The Up To Date system, for example, is the global leader in research review and summary (made in the US), and they use the 50% figure.

I'm having a lot of trouble understanding why everyone is so convinced that diagnosis does take place most of the time for outpatients. Can you or anyone else explain why people think that?

And I'm having trouble understanding why you insist that it is so because everyone has always said it's so. Since when was that a scientific argument? Lots of things have been claimed to be true by medical authorities for generations that turned out to be nonsense.
Just because some website that doctors use says it's so doesn't make it so.

You have not yet provided us, either in your paper, or here, with a single piece of research that shows the claim is true. All the research you list either doesn't address the question at all but has been misinterpreted to pretend it does, or shows much lower figures for undiagnosed symptoms or patients.

I have no idea why it matters to you so much to insist the 50% figure is correct. You don't need it for your ethical argument about misattribution of unexplained symptoms to psychogenic causes. That would still be a serious problem for a significant number of patients whether the scale of undiagnosed cases or symptoms is 3% or 50% or anywhere in between. By misquoting statistics you undermine your credibility, and therefore give ammunition to those who want to argue against your main thesis.

So far, when I've asked you to justify your belief that the 50% figure is correct you have provided either appeals to authority or to supposed 'common sense'. You have not pointed to a single piece of research that credibly demonstrates anywhere near the 50% figure for either symptoms or patients being undiagnosed.

Would you like to try again?

 

[Diane O'Leary]

Another paper
The resource utilisation of medically unexplained symptoms
Lee et al 2016
This was a fairly recent paper by New Zealand authors who appear to be supporters of the concept of MUS.


The references appear to be documents aiming to influence health policy. I haven't checked them, no doubt the estimates have a poor basis.
4. Personal Social Services Research Unit, London School of Economics and Political Science. Mental health promotion and mental illness prevention: the economic case. Report, Department of Health, London, January 2011.
5. Personal Social Services Research Unit, London School of Economics and Political Science. No health without mental health: a cross-government mental health outcomes strategy for people of all ages. Supporting document – the economic case for improving efficiency and quality in mental health. Report, Department of Health, London, January 2011.

Can I ask what conclusion you think we might draw from this? For example, do the authors understand MUS consistently - or do they shift (as most do) between understanding MUS as undiagnosed symptoms and understanding them as psychosomatic? How do those shifts bear on their final figures? Are they asking about proportion of patients or the proportion of cases (given that only roughly half of patients present for new Sx care)? Do the policy makers in NZ understand MUS to mean undiagnosed symptoms or psychosomatic symptoms?

Because of these ambiguities, you can find prevalence studies that verify any number you happen to prefer. I see that a great many of you have a strong preference for lower numbers.

 

[dave30th]

though I am not sure I am convinced their assurances are worth much.

I'm not sure why you would be so suspicious.

 

[Trish]

Because of these ambiguities, you can find prevalence studies that verify any number you happen to prefer. I see that a great many of you have a strong preference for lower numbers.

Really? So where are the prevalence studies that support the 50% figure you prefer?

I don't have a preference for particular numbers, high or low. I have a preference for credible scientific evidence.

 

[Esther12]

I think some of the frustration in this thread is stemming from people starting with different assumptions about the way a discussion on these sorts of matters should take place. It could be that there's some sort of cultural consensus on S4ME that means we can forget our way can seem a bit alien and weird to those who are used to doing things in different ways. Worth trying to be as welcoming as possible (although I know I can be bad at those niceties myself).

 

[Diane O'Leary]

Really? Is that a serious argument?

And I'm having trouble understanding why you insist that it is so because everyone has always said it's so. Since when was that a scientific argument? Lots of things have been claimed to be true by medical authorities for generations that turned out to be nonsense.
Just because some website that doctors use says it's so doesn't make it so.

You have not yet provided us, either in your paper, or here, with a single piece of research that shows the claim is true. All the research you list either doesn't address the question at all but has been misinterpreted to pretend it does, or shows much lower figures for undiagnosed symptoms or patients.

I have no idea why it matters to you so much to insist the 50% figure is correct. You don't need it for your ethical argument about misattribution of unexplained symptoms to psychogenic causes. That would still be a serious problem for a significant number of patients whether the scale of undiagnosed cases or symptoms is 3% or 50% or anywhere in between. By misquoting statistics you undermine your credibility, and therefore give ammunition to those who want to argue against your main thesis.

So far, when I've asked you to justify your belief that the 50% figure is correct you have provided either appeals to authority or to supposed 'common sense'. You have not pointed to a single piece of research that credibly demonstrates anywhere near the 50% figure for either symptoms or patients being undiagnosed.

Would you like to try again?

I have asked you many times why you believe doctors do get it right most of the time. Can you answer that question, with evidence?

First, your question is too confused for me to answer clearly. Are you asking me, again, to run through the section of the article where I support this conclusion? Or are you actually asking me to provide you with a bibliography - a literature search - for the higher rates that you feel like disputing? If it's the latter we have three problems: (a) you have provided no reason for anyone to believe the claim that doctors actually know what's causing symptoms most of the time, so I think the onus is on you there; (b) you can look that up yourself if you're just looking for references; (c) if you're looking for an assessment of the reality underneath the references a literature search would be extremely time consuming. Researchers are all using terms differently, shifting their use of terms within the same paper, conflating different kinds of figures. If we're talking about what consensus is, there are simple answers and they're in the paper. If we're talking about reality, the answers are complex and very time consuming.

Second, please try to keep in mind that you have slaved over these figures for a day or two and feel frustrated that you're not getting ready answers. I have slaved over them in that same way, with that same frustration, day in and day out, year after year after year. No one in a professional setting would ever imagine that they could uncover a finding in one day that a colleague who's worked on something for years has failed to notice.

Third, on what possible basis can you support the very serious charge that I am misquoting statistics? Do you really think you're going to be better at determining that than all the people I listed who were involved in reviewing the paper and then critiquing it in teams? You have every authority to challenge my arguments, but if I were actually misquoting statistics that paper would not have been published. I thought we established that.

Here is a big picture view:

The 50% figure is so important because it makes sense of the astonishing rate at which people who are suffering cannot secure medical care. My work is not about ME. It's about MUS, because ME is just one of a great many groups of patients who suffer from this problem. I have slaved over these figures for all these years because I can see beyond any doubt that the number of people wrongly denied medical care is outrageously high, and I'm trying to make sense of how this is possible, how it came about. Outside the realm of contested conditions there are these realities -

I rarely talk with someone about what I do - either in a professional context or in a personal context - without them telling me a story about being unable to get medical care out of their doctors for a medical problem. This is an everyday problem that leaves everyday people frustrated and confused.

51% of patients with autoimmune disease report having been denied medical care based on misdiagnosis of psychosomatic symptoms. That's 12 million people in the US alone.

Studies show ties between mistaken psychosomatic diagnosis and diagnostic delay for rare disease. Those people now wait 7 years for diagnosis on average, and they are in no way rare. People with rare disease are as common as people with diabetes.

In so many areas, women are denied access to health care that men routinely enjoy because doctors routinely construe their symptoms as psychosomatic.​

Having studied research in all of these areas, along with the research on MUS, I have actually found an answer to my question, and it is the foundation for the book I am now writing. I know why this is happening to so many patients, and it's is not just a hunch. It's happening because doctors are trained to accept that unexplained symptoms can arise up to half the time - and then, thanks to the BPS camp, they're also trained to ignore the difference between unexplained symptoms and psychosomatic symptoms.

Once you see this, you can make statistical sense of the great wide range of patients who suffer from this problem. That is why it's important.

 

[Daisybell]

My understanding of this is that the conflict here is arising from many S4ME members wanting to have certainty about the facts. We are very used to the facts being twisted and misrepresented, selectively reported, and also to the conflation of theories to facts.
The 50% figure has that sense of illusion about it. It’s presented as ‘the accepted norm’. I think the question, certainly for me, is why?
Has that figure come about from sound research? From misunderstanding? Is it really true? If researchers are changing their terminology and conflating different kinds of figures, then the answer seems likely to be no, it’s not true. So - what pieces of research have been completed which actually hold up to really robust scrutiny?

Perhaps it doesn’t matter from a medical ethics perspective - I don’t have any idea about that! But long-term forum members here have got very used to getting their teeth into the nitty-gritty and not letting go until facts emerge. That’s really not hostility towards you and I’m sorry if you feel that way. But being persistent with questions and seeking out the ‘truth’ has been a huge help to us in many areas of advocacy, and I think that’s what we are after here. There is often dissection of minutae at great length, and sometimes over and over!

 

[Trish]

I have asked you many times why you believe doctors do get it right most of the time. Can you answer that question, with evidence?

I don't have any belief in any figure, large or small, relating to this issue. I simply want to see what data is used to justify the 50% figure, and you have told us you are expert in this, so I turned to your paper, and when the studies you quoted as showing a higher than 50% rate turned out not to do so, as I read them, I asked you if you could help us out with research evidence to back up the higher end of the range.

See your own next comment for the reason I thought you would be a good person to ask.

Second, please try to keep in mind that you have slaved over these figures for a day or two and feel frustrated that you're not getting ready answers. I have slaved over them in that same way, with that same frustration, day in and day out, year after year after year.

You make my point well for me. I do not claim to know anywhere near as much as you on this topic. As you correctly say, I've only spent a few days on it, and you have spent years. As a former teacher, I spent years on my subject and my pupils turned to me for information. I was putting myself in the role of a somewhat impatient pupil, asking you, the expert, to point me in the direction of research to help me understand the 50% figure in the NHS document. I apologise for my impatient tone.

Third, on what possible basis can you support the very serious charge that I am misquoting statistics?

I was simply quoting in post #65 what you yourself said in your paper about the Kroenke and Mangelsdorff paper and what Wessely et al said about it and pointing out that what you both said, in different ways, did not appear to fit with the data in the paper. I would be very happy for you to point out to me where I have gone wrong.

Here is a big picture view.

The 50% figure is so important because it makes sense of the astonishing rate at which people who are suffering cannot secure medical care. My work is not about ME. It's about MUS, because ME is just one of a great many groups of patients who suffer from this problem. I have slaved over these figures for all these years because I can see beyond any doubt that the number of people wrongly denied medical care is outrageously high, and I'm trying to make sense of how this is possible, how it came about. Outside the realm of contested conditions there are these realities -

So we are in agreement, there are many people who are wrongly diagnosed, and many whose unexplained symptoms are wrongly attributed to psychogenic causes. We are, as you have said yourself, on the same side in this. And I am not disputing that you have far more knowledge than me about how the 50% figure was derived. That is my point - I am asking you, as the expert scholar, to help me out in finding papers that support the figure if you have time and are inclined to do so.

I rarely talk with someone about what I do - either in a professional context or in a personal context - without them telling me a story about being unable to get medical care out of their doctors for a medical problem. This is an everyday problem that leaves everyday people frustrated and confused....

I have no argument with any of that. I have had ME for 29 years, my daughter for 21 years, we are both housebound, mostly bedbound and have never had any consultant level medical assessment or help. You are not telling anyone on the forum anything new here. We live it every day. As I said, we are on the same side.

 

[Hutan]

I see that a great many of you have a strong preference for lower numbers.

I simply googled NZ and MUS and this was the paper that came up with an estimate of MUS prevalence. I didn't select it because it gave a lower prevalence.

Can I ask what conclusion you think we might draw from this?

I thought it was relevant because you (Diane) had mentioned that this view of 'MUS is very prevalent' is commonly held by MUS researchers all around the world and is used to guide health policy . These MUS researchers chose to quote the UK figures suggesting a lower prevalence. And the UK figures came from documents that seemed to be intended to guide health policy.

 

[Hip]

Can someone explain why in this thread the accuracy of the figure of 52% of medical consultations being for MUS is considered so important? What practical relevance is it? Would it change anything if the figure were instead 30%, 45% or 60%?

Apparently the interest relates to this:

My interest in this was stimulated by this post by @Diane O'Leary where she says:

Roughly speaking, 50% of outpatients' symptoms fall into that category (or at least this the figure you generally find in practice guidelines). The health system in the UK actually says 52%, so the majority of the time when people go to a doctor outside the hospital, the doctor concludes that the problem is MUS.

With any medical statistic, there is usually a range of values that come from different studies looking at the same thing. So you are never going to find a precise figure, only a range.

 

[Hutan]

Can someone explain why in this thread the accuracy of the figure of 52% of medical consultations being for MUS is considered so important? What practical relevance is it?

The quote that Trish started this thread with:

The UK document, Guidance for commissioners of services for people with medically unexplained symptoms includes on page 6 the claim that:

Given the wide-ranging symptoms that can occur, patients have high rates of access to a number of outpatient departments. On average, 52% of patients accessing outpatient services have MUS, with the highest rates relating to gynaecology clinics (66%) and the lowest rate (37%) relating to dental services.

suggests that there is a rampant epidemic of MUS. It therefore suggests that there are many people clogging up an under-funded medical system to no good end. And that then provides support for policies that get these MUS patients labelled and out of the medical system as quickly as possible and being given therapies aimed at behaviour change.

The 52% figure, or a more general understanding that MUS is highly prevalent is therefore important in the marketing of IAPT and behavioural change therapies.

It's useful to investigate the figures, to understand how they depend on what assumptions are made in the original research. And also to trace where any widely accepted figures have come from to identify errors in interpretation. Armed with that knowledge, we are better able to argue for more useful health care.