Discussion in 'General ME/CFS News' started by Ember, Dec 22, 2012.
Can you explain why you think that the evidence favours overlapping sets over nested sets?
Hi Ember. First, Leonard Jason has made a few studies on patient cohorts using various definitions. I don't recall all the details, I would have to go back and read them, but they are worth reading.
Second, each definition has its own sets of criteria. Some are mandatory, some are optional. CFS, ME etc. are syndromes. That is they are sets of symptoms. While we currently think of them a group as distinct unitary entities, we have no idea if that is correct or not. The Light's data on exercise recovery indicates there are two different groups in strictly defined CFS. The Pacific Fatigue Lab research show that, contrary to regular fatigue, strictly defined CFS has a decline in energy production and work effeciency for a period following activity. Its not just fatigue. Then we have a host of secondary data on immune states, cytokines etc.Some of our mitochondrial issues may be unique too, and of course now we have things like possibly unique EBV antibodies in a distinct subset. This requires further investigation. Even data like that of Chia's findings on enteriviral presence in the gut, or response rates to rituximab, or even 37kd RNase L, have results in about a two third to three quarter split versus the rest. One of the things I really want to know is if any of these findings cluster in the same patients.
One thing that is not clear to me now is that OI in general can separate post viral fatigue and ME. There is at least one study (I posted a link a day or two ago) that shows that post viral fatigue patients have OI at around the same rate as CFS.
Take all these data points, and plot them on a multidimensional graph. Are they all the same syndrome ... highly unlikely, a point I will argue from another perspective in a few paragraphs. The only perspective that makes sense is to look for clusters. That is what definitions like CCC and ICP are based on ... clusters arising from clinical experience. Those clusters will have symptom overlap, but not total overlap. They are overlapping sets.
Then we have the trigger/cause problem. There are many different triggers. Do these lead to a single common path, or multiple paths representing multiple disease states? We do not know. For that we have to look at the physiology, and we need to understand how triggers initiate the syndrome. We just don't know yet. The hypothesis I am currently investigating is that ME is all viral induced. What we call a trigger is the tipping point, not the cause.
Something that has been said very many times, and I agree with, is this. Fatigue is the single most common symptom there is, with the possible exception of pain. Both are warning signals. Historically the process of medical discovery has been to take the huge pool of medically unexplained syndromes and remove disease from that pool, one at a time, and to continue this process inside individual diseases. Once upon a time we had cancer. Then we had breast cancer. Now we have specific breast cancer subsets due to genetic issues. This is a process of continual differentiation, leading to more specific diagnoses and more specific treatments. I think ME and CFS will go the same way, but I can't predict just how many subgroups there will eventually be.
Another way to think of it is this. Run the process in reverse. Suppose that we had yet to discover a whole host of diseases that are now known and hence not medically unexplained. Since they are not discovered, and since they probably cause fatigue, they fit in unexplained fatigue syndromes. Now take this mixed group. Try to find a common biomarker ... no chance. Try to find a common cause ... no chance. Try to find a single treatment that works on everyone ... no chance. Yet we want to call this group Chronic Fatigue Syndrome (Oxford)? Its a nonsense. To assert its all one thing is to assert that every possible disease that will be known in the future is known now. (chortle)
In the time I have been aware of the research there have been quite a number of disease, many genetic, that have been differentiated from CFS. I was in a study to examine cortisol binding globulin mutations. I don't have them, but some do. Prior to the genetic research of an extended family with "CFS", these patients were often diagnosed with CFS ... its a great big wastebasket of too hard cases.
We now have post SARS syndrome too, which would have been called CFS prior to SARS. We don't know if its really CFS or something different. I suspect that a lot of post pathogen fatigue syndromes are the same, but this requires more research to demonstrate this.
One next to final point. I have said overlapping sets are the best way to think of it, but I have not addressed the issue of how much overlap there is. In each case it could be a little or a lot. The problem is that some research simply presumes its either a lot or total overlap. It is not good enough to make this presumption, it should be a priority in research.
My final point is about the Oxford definition and the view that our illness is perpetuated by dysfunctional beliefs. The only thing that holds Oxford together so far as I am concerned is the dysfunctional belief model or its variants. It that model is correct, and there is NO objective evidence it is, and it applies to the vast majority of Oxford CFS patients, then the Oxford CFS definition might make sense. If its wrong, then the Oxford definition will implode under its logical inconsistencies. Given that all the psychosomatic explanations rely on that wastebasket diagnosis idea (over generalization), and further irrationally assert that anything left in the basket is a psychogenic illness (the psychogenic fallacy), I predict the whole thing will implode horribly at some point, though die hards will be claiming its has to be right long after the world moves on. I would also like to add that something that is logically consistent only under one intrepretation, but illogical under others, fits the fallacy of begging the question.
Thanks, Alex. The Jason et al. (2004) study, “Comparing the Fukuda et al. Criteria and the Canadian Case Definition for Chronic Fatigue Syndrome,” is particularly worth reading:
Though I'm not a visual thinker, I can attempt to see the interactive symptom clusters within a definition such as the CCC as overlapping sets. But among the definitions themselves, I think of the ICC cohort nested within CCC cohort, nested within Fukuda cohort, nested withing Oxford cohort. Is there something wrong with my cognitive map? It doesn't preclude the discovery of further nested sets that may overlap with each other.
If the dysfunctional-belief hypothesis were built into the Oxford definition, then I wouldn't consider the other definitions to be sets nested within it at all.
Hi Ember, the dysfunctional belief hypothesis and other similar psychogenic explanations such are hysteria are not built into the Oxford CFS definition. They are however a huge part in justifying such a sweeping overgeneralization. This in spite of the dozens of diseases once though psychogenic but which are now considered physical. These include epilepsy, diabetes, peptic ulcers, rheumatoid arthritis and so on. I do not have a complete list of such illnesses, but I do have several attempts at such lists that I have lurking around and one day will get around to compiling a more complete list with references.
In terms of wording of definitions I can see how you might consider them nested. Similarly if you focus on just one symptom, fatigue ... thats a trap. It might look like its a supergroup, but fatigue is itself composed of many different kinds of things. Its also an overlapping set. However when you inerpret those in the real world, and start thinking of how those symptoms cluster, I don't think its a valid conclusion ... despite that just a few years ago I used to think the same as you.
In terms of the cohorts, Alex, do you think that the definitions are progressively restrictive? My view is shaped to some extent by the fact that I've been diagnosed under three definitions over the years, Fukuda, the CCC and the ICC.
In short ... I am not sure. What do you mean by progressively restrictive? If you mean that each of the definition has become more constrictive over time, even that depends on what you mean. I would have to go over the CCC and ICP again, and compare them to Fukuda and Holmes, but I don't think its that simple based on what I can remember.
The way I would characterize it is that for Fukuda, CCC and ICP each attempt at a definition, with the latests being ICP, is an attempt to better capture the clinical experience. Definitions like Oxford, Australian and operationalized Fukuda are something are something else though.
Wasn't Fukuda less strict than Holmes? I am not sufficiently aware of the London criteria and revisions to see where they fit in, nor the Jason criteria or even Ramsay.
Only with biomarkers can these issues be solved, and even then I think there might be dispute until we understand causes.
I'm sorry, Alex. My question wasn't at all clear. I meant to ask only about the definitions that were under discussion earlier: Oxford, Fukuda, the CCC and the ICC. I've assumed these to be progressively restrictive definitions, each selecting a smaller group or a nested set. I've generally assumed that patients identified by the ICC would be included in the other cohorts as well.
As I see it, the ICC reverses the subterfuge that buried ME so many years ago, hopelessly compromising subsequent research, including the search for biomarkers and causes.
I still cannot say anything definitively. I suspect that for many patients they would indeed fit under all four criteria. This could be seen as nested criteria .. or as overlap conditions. Its the outliers, the ones who don't quite fit that are interesting. We need more research. It also depends on how you want to view definitions. If we go by definition alone, those with ICC/ICP do not have Oxford CFS by definition, the two are exclusiary if I recall correctly. So in short I still can't answer the question. Maybe I could if I had been studying the definitions recently, but I haven't done that.
All these definitions are artificial. None has been properly validated as a single distinct disease entity. With more research this might yet happen, and its more likely with definitions like ICC and CCC as they are more restrictive.
Something interesting happens however if you want to strictly enforce CFS definitions. CFS is a definition by exclusion. So if you are diagnosed with ME under any definition, then strictly speaking you no longer can be diagnosed with CFS (unless you hold that CFS and ME are two separate conditions, and the ME diagnosis does not full explain the symptoms which necessitates a CFS diagnosis in addition). This distinction is however not used in practice. With those supporting the Oxford criteria this cannot be due to the criteria I think, this is due to the psychogenic interpretation that they impose on Oxford CFS.
However, from an individual viewpoint, if you are a person who has a valid diagnosis under various definitions, then for you there would be overlap of all of them. The issue is whether or not that holds generally. I doubt it but I can't prove that. With careful review of the definitions I might be able to show it, but I don't have the time to do that at present.
The exclusions reinforce the ICC/ICP removal of ME from the CFS classification. The hybrid terms (ME/CFS and CFS/ME) are logically inconsistent. We should always refer to ME and/or CFS.
You can also try a Google Site Search
Separate names with a comma.