A talk on definitions might not seem to be the occasion for one of the more passionate and exciting presentations of the conference, but it was. Lenny Jason is convinced that getting the definition right is absolutely critical for the future of Chronic Fatigue Syndrome, and he’s passionate about it. By the time he was done I was convinced, and I imagine everyone else in the room was as well. It may have been the best presentation of the conference.
How bad is the situation right now? Jason started off by showing a house of cards on which everything — the research, the search for biomarkers, treatments — is endangered by a vague understanding of ME/CFS.
Building the right definition is just one part of the process. Rigorously characterizing every important aspect of this illness so that everybody is on the same page is another. It’s about making ME/CFS understandable and above all ‘studyable’. There’s no mystery how to go about this. It’s been done in other fields — it just takes work.
Chronic Fatigue Syndrome, he noted, is decades behind other illnesses in the hard work of standardizing its core aspects. ME/CFS is where psychology was in the 1950s when diagnoses were made without using standardized criteria. That meant a person could go to one doctor and get one diagnosis and walk across the hall to another doctor and get another one. Doctors each had their own way of determining if someone was depressed.
In 1972 a St. Louis group devised ‘operationally explicit criteria’ — standardized questionnaires — that all doctors would follow henceforth. That ‘simple’ practice led to an immediate and dramatic improvement in physician-to-physician consistency.
A standardized interview process was the key.
Operationalizing Core Aspects of Chronic Fatigue Syndrome
We hardly have the ‘easy stuff’ done at this point.
Life-Long Fatigue – Take the seemingly simple issue of determining whether someone has had life-long fatigue. That’s an important issue because people with ME/CFS are considered to have a fatigue onset at some point but not ‘life-long’ fatigue. That means you probably want to weed the people with life-long fatigue out of research studies. But it turns out that, if you simply ask someone if they’ve had life-long fatigue, a significant number of people are going to answer incorrectly.
Jason stated that, when asked, 20 of 217 people with Chronic Fatigue Syndrome (Fukuda) said they had life-long fatigue. Asking them two more questions indicated, however, that nine did not. Asking another question knocked almost 50% of them out. Operationalizing the life-long fatigue question (it took four questions) put 60% of the people who initially thought they had life-long fatigue (or about 5% of the ME/CFS population) back into a potential ME/CFS study.
Fatigue — a core characteristic of ME/CFS — is another difficult issue. About 10% of severely impaired individuals with ME/CFS do not experience much fatigue, apparently, because they’re rigorously staying within their energy envelope. Unless that’s made clear, those individuals could fall out of a research study.
And then there’s onset. There’s gradual onset, gradual onset punctuated by infectious episodes, sudden onset that comes completely out of the blue, and sudden onset preceded by a series of infectious episodes. There are infectious events, psychological stressors of various sorts (work, personal), activity levels, and injuries to account for. There’s defining the difference between gradual and sudden.
My own ME/CFS, for instance, came on over about a month. Is that gradual or is that more sudden or is it in between? Then I got better for about three months, and then I gradually got worse and it stuck. I had punctuated episodes of gradual onset not associated with infection or trauma or severe stress. That’s a pretty complicated scenario, which makes the fact that one of Jason’s students is doing her doctoral dissertation on onset no surprise. If you can’t get a life-long stress question answered correctly, how are you going to get onset right? Should that be “fatigue” instead of “stress”?
After listening to Jason it became clear that you can probably throw just about every existing onset study out the window. Jason asserted that this sloppy approach to ME/CFS is a dead weight on the ME/CFS field that simply has to be removed.
Comparing and Contrasting the Definitions
Studies comparing and contrasting the different definitions provide something of a wakeup call. They’re critical to determining which definition is best, but the high variance in study results – with one study finding that 50% of Fukuda patients meet the Canadian Criteria while another finds 90% – indicates something is awry here as well.
DePaul Symptom Questionnaire Provides Help
Jason’s DePaul Symptom Questionnaire, which includes both symptom severity and frequency (i.e., is your headache severe and infrequent, severe and frequent, moderate and occurs often, etc.), is a significant step forward. Earlier studies by the Jason team found that simply asking if the symptoms in the Fukuda definition were present resulted having some people with major depression being diagnosed with ME/CFS. The symptoms in Fukuda were simply too similar to those found in depression for that not to happen.
However, if those same symptoms were assessed according to severity, it was pretty easy to differentiate the two. People with ME/CFS had much more post-exertional malaise, cognitive problems, and unrefreshing sleep than people with major depression. Adding a couple of more symptoms such as shortness of breath, self-reproach (low in ME/CFS/high in depression), and others to the questionnaire made the diagnosis bulletproof: nobody with major depression was diagnosed with ME/CFS and vice versa.
A large 2014 Jason study found that a third of the healthy controls meet the criteria for Chronic Fatigue Syndrome when they were simply required to have a symptom; i.e., they experience ‘fatigue’. Only 5% met the criteria and 95% of people were identified correctly when people with ME/CFS were required to have a higher frequency and severity of symptoms.
The DePaul Symptom Questionnaire also has algorithms built into it that can tell whether a patients meets the Fukuda, Canadian, ICC or London criteria. With this one advance any researcher who used this questionnaire could be able to tell how different types of patients fared on his study.
The questionnaire is now being used in Canada, Mexico, England, and Norway as well as by the CFIDS Association Biobank and the CDC in the USA.
The Next Research Definition
Consensus definitions have played a very important role (for good or bad) in ME/CFS and they still will in the doctor’s office, but in the research field their time has clearly come and gone. Factor analyses and data mining can now determine the symptoms that best discriminate between people with ME/CFS and those without. A machine learning program indicated that most discriminative symptoms for ME/CFS were in order of importance: (central?) fatigue, neurocognitive problems, post-exertional malaise, unrefreshing sleep, and joint pain.
Lenny chastised the ME/CFS community a bit for not learning from the past and producing definitions (CCC, ICC) that have too many symptoms to work as a research definition. The eight symptoms in the first definition for CFS — the Holmes definition — ended up capturing people with high percentages of psychiatric illnesses. The Fukuda definition with its reduced symptom requirements was produced to fix that problem, but then the CCC and ICC with their long symptom lists bumped into the Holmes problem.
Jason asserted that just three core symptom domains — post-exertional malaise, sleep and cognitive functioning — were all that was necessary to discriminate between ME/CFS and other disorders, and avoid the problem of oversampling people with psychiatric illnesses. One Jason study found that simply asking people if they experienced ‘extreme tiredness’, had trouble focusing on one thing at a time, and if they had a dead feeling after exercise was sufficient to accurately classify people with ME/CFS.
His analyses show that adding more symptoms to the definition doesn’t improve the accuracy or specificity of the definition. He proposed using symptoms such as pain, autonomic nervous system functioning, neuroendocrinology, etc., to subtype patients — but only the three core symptoms would be used to define ME/CFS.
Jason was asked in the question period about the Oxford definition which simply requires long-lasting fatigue to meet the criterion for Chronic Fatigue Syndrome.
The Oxford definition is to the UK what the Empirical Definition is to the CDC; the UK and some European researchers are the only ones that use it. Nobody outside that small clique has ever used it. I can’t imagine it’s ever once been used in the U.S., yet there it is. One has to ask why UK behavioral researchers, in particular, have so stubbornly stuck to it.
Jason said the Oxford definition probably applies to about five percent of the population and probably includes people who are simply overworked, have substance abuse, or are depressed. My guess is that each of these problems is probably more amenable to CBT and GET than ME/CFS and UK behaviorists would lose a good chunk of their funding if their definition was more tightly constructed.
The Virgin’ ME/CFS Subset
Jason believes that studying young people is critically important because they don’t have the all the medications, the medical history, and the trauma that make this disease more difficult to study as patients get older. If you can catch them early before the misdiagnoses, the drug regimens that don’t work, and the depression that often sets in, you have a cleaner shot at this illness.
The good news is that the Jason team is building structured interviews specifically for ME/CFS, the DePaul Symptom Questionnaire with its symptom and severity fields is light years beyond the Fukuda definition in being able meaningfully characterizing the symptoms found in ME/CFS, and perhaps most importantly, Jason’s ability to statistically characterize the symptoms found in ME/CFS means he’s provided the groundwork for a research definition that works.
The foundations being laid by Dr. Jason and others will support more rigorous and effective research efforts that feature people with ME/CFS and only people with ME/CFS.
Dr. Jason will be speaking on Diagnostic and Criterion Issues in ME/CFS at the IOM meeting on May 5th.