Is the NIH/CDC going to use the right PEM definition for all their future research? Do patients need to act? Deadline 31 Jan

The NIH/CDC are deciding on the “common data elements“, CDE, that must be collected in all studies they fund. And they’ve asked for patient comments on the draft proposals by 31 January.

Perhaps the most important of the CDEs covers PEM, as the cardinal symptom of the illness. The PEM subgroup has proposed a questionnaire that will play a key role in defining PEM and I have concerns about its suitability. I’d like to hear what other patients think.

I should say that its choice - the PEM subscale from Lenny Jason’s DePaul Symptom questionnaire, DSQ, - seems to be the only credible option right now. And the subgroup recognises many of the DSQ’s limitations. However, I feel we need a clear plan to urgently develop a better option.

The questionnaire will help define who has PEM and who does not (in some cases it will be the only measure used). If it’s inaccurate, it risks including patients who don’t have the illness, weakening research, and excluding patients who should be, making recruitment harder and leading to smaller studies, again weakening research.

The DSQ assesses post-exertional malaise by asking patients to rate severity and frequency of five symptoms over the last six months:

• Dead, heavy feeling after starting to exercise.
• Next day soreness or fatigue after non-strenuous, everyday activities.
• Mentally tired after the slightest effort.
• Minimum exercise makes you physically tired.
• Physically drained or sick after mild activity.

My feeling is that these question circle around PEM rather than really capturing its essence. How do others feel about this?

A 2017 study by Murdock found that the PEM subscale did differentiate patients from healthy controls, but only when the volume was turned up quite high (score >20/40). A really good PEM measure should do that more easily – PEM is not a normal, healthy experience.

Interestingly, the subgroup itself has adopted the Institute of Medicine’s definition of PEM (now the National Academy of medicine, NAM). As it put it:

Post-exertional malaise is defined as an abnormal response to minimal amounts of physical or cognitive exertion that is characterised by (my summary, full text in a later post):

• A flare of some or all of an individual’s symptoms,
• with an immediate or delayed onset and
• with prolonged recovery.
• Severity and duration of symptoms are out of proportion to the initial trigger.
• There is a loss of functional capacity and/or stamina.

Which to me looks like a good description of PEM and substantially different from the DSQ. So, on the one hand, the subgroup have adopted the NAM definition, while on the other they have selected the DSQ to measure it, and that doesn’t seem to stack up.

Of course, what’s actually needed is a short, validated questionnaire that can be used in studies. We don’t have that yet for anything but the DSQ.

For now, it’s the right call to use the DSQ, but I believe we need a clear plan to urgently develop a replacement that starts with patient input. Only patients experience PEM, so their input is crucial.

I’ve started this thread to hear if other people share my concerns. I’d appreciate any answers to (any of) the following three questions:

1. Does the proposed five items questionnaire fall short of properly capturing post-exertional malaise?
   
2. If so, should we be asking for a timetabled plan to urgently develop a better alternative?
   
3. I also have concerns that the way symptoms are scored could give misleading results - see next post. Again, is this an issue for anyone else?

IMPORTANT. This is not intended as a criticism of Professor Leonard Jason, who has been a lonely pioneer of case definitions and post-exertional malaise in particular. Without the research of him and his team there would be nothing to use now. Likewise, it’s not a criticism of the working group (co-chaired by Lily Chu and patient advocate Mary Dimmock). This has produced many good recommendations (including using the DSQ for PEM at first) and has highlighted many of the weaknesses of the DSQ.

This, though, is about how PEM will be defined in every N I H and CDC study, every biomedical study, so it needs to be as good as it can be, as soon as possible, to produce the best possible research.

Can anyone help with more info e.g. more on the DSQ, especially the PEM questions, which I believe were discussed on PR in the past? I’ve read the research myself years back but am too ill now to revisit. Just doing this has been a huge effort.

@Woolie is a questionnaire expert, @Adrian and @Graham know a lot too.

 

The way the symptoms are scored might give misleading results

My other big concern with the questionnaire, which again relates to its accuracy, is about how PEM is scored. I think we need to score in a way that reflects how patients try very hard to avoid PEM. Here’s the questioin

“For the following questions, we would like to know how often you have had each symptom and how much each symptom has bothered you over the last 6 months.“ Each is scored 0-4, and a PEM diagnosis requires a score of at least 2 on both frequency and severity for any of the 5 items. These scores corresponds to at least moderate severity and PEM half of the time or more.

I think the frequency question looks at things the wrong way, and am pretty sure there are plenty of mecfs patients with PEM who experience it less than half the time - and so who would be excluded from every NIH/CDC study.

The problem is that PEM is a function of disease level, but also of pacing and life circumstances. As the report note elsewhere, “many ME/CFS patients use strategies to avoid and/or to minimize the impact of PEM”.

The subgroup is clearly aware of the problem, with the case report form emphasising the need for researchers to interpret the questionnaire finding:

“In making this [PEM] determination, the researcher or clinician will need to consider whether there are other conditions, such as overwork, that could result in a false positive DSQ PEM subscale response. On the other hand, the researcher or clinician should also consider whether the study participant responded negatively because, for instance, they carefully manage their energy expenditures with pacing to avoid episodes of PEM.

In addition to asking questions about workload and pacing, the researcher may also ask what happens to the study participant if/when they engage in physical or mental activity and whether there are activities they avoid because it exacerbates symptoms.”

This is all very good, and important.

But I think we need a questionnaire that allows the patient to give more appropriate information in the first place. This could be with a question on what happens if/when they engage in physical or mental activity. Or to what extent the patient has to change their life to avoid PEM – I suspect that for most patients it’s “a great deal”, and any questionnaire should capture that.

A major reason for a mandatory questionnaire across studies is to standardise approaches, whereas relying on researcher judgement to interpret questionnaire scores will do the opposite.

Again, my views – what do other patients think?

Thanks for your patience with my long posts. Please bear in mind that the consultation document ran to 28 pages. And thanks for all work done by the PEM subgroup.

 

Background information that many will want to skip J

Various extracts from the proposals that highlight what a thorough job the PEM subgroup has done, and illustrates some of the issues covered in my first post.

PEM will be an essential part of the case definition of ME/CFS:

The “PEM Determination” is a Core CDE instrument to be used across all research studies as a common method for ascertaining and recording the presence or absence of PEM as a case defining symptom in each individual study participant.

Why the DSQ was chosen:

While not perfect, the PEM subscale of DSQ is the best choice at this time for a standard method of identifying the existence of PEM in a research study participant... has been tested and used in several ME/CFS studies, has been evaluated in multiple diseases, includes a PEM subscale, and has been used by different researchers.

The recommended core method for assessing PEM is a 2-step process in which the study participant responds to the DSQ PEM questions and the researcher then evaluates those responses in light of other information (e.g. study participant interview, physical examination, objective testing) about the study participant to determine whether the study participant experiences PEM or not.

Limitations

There is a dearth of studies asking participants about their experiences of PEM in an open-ended manner, which is needed….

The instrument does not assess the full range of symptoms that could be exacerbated by PEM and only one item addresses the sometimes delayed onset/ prolonged duration of PEM ("Next day soreness or fatigue after non-strenuous, everyday activities.")

● The studies published to date primarily evaluate the DSQ in ME/CFS compared to healthy controls, not in other fatiguing conditions or conditions where there could be diagnostic ambiguity. Recent research has shown a difference between ME/CFS and MS study participants.

● The use of the DSQ PEM subscale as a stand-alone instrument has not been separately validated although as noted above, its use has resulted in a high prevalence of PEM being reported in ME/CFS study participants.

Need for further research

● The instrument will need to be further evaluated to ensure it accurately reflects the symptom of PEM. (…Further research is needed to address these limitations.)

6. Unmet needs;

1. Further development and refinement of the Core PEM Instrument [includes DSQ] and CRF.

> In my view this needs to be much stronger and should say replacement or development. It may be a new PEM questionnaire would be no more effective; we need to find out urgently. And the whole process needs a tight timetable.

Full version of the NAM (formerly IOM) PEM definition

 

Consultation:

Anyone can write in, anytime – up til Jan 31st deadline. Should we combine to offer comments on PEM? Anyone at S4ME have an opinion on this? @Andy @Valentijn. I guess it depends on how the discussion goes.

“We are in the process of finalizing these documents and would like to urge you to help by contributing your expertise during this public review period.

Please send all comments to NINDSCDE@emmes.com by January 31, 2018.

After this open comment period, the ME/CFS Working Group will review and revise the recommendations. Version 1.0 of the ME/CFS CDE recommendations will be posted at the end of February 2018.”

 

I'm glad that you brought this up, @Simon. I agree that the Jason questionnaire doesn't very well reflect PEM, both as it's defined in the IOM document and as I and many others experience it. I took part in his questionnaire study and found that the questions forced me to respond in ways that didn't reflect my experience.

Lenny Jason is a great asset to this community but I share your concerns that this questionnaire doesn't reflect PEM properly and as such, could be damaging to research.

I've been surprised that there has been so little discussion so far on these common data elements. The NIH and CDC will be using them in all the research that they fund, going forward, so it's crucial that we help them get things right. This is hugely important.

Perhaps people have been daunted by the length of the document and the short time-frame but we need to be looking at this.

 

Thanks for looking into this @Simon M. It's lot of info so I will have to come back and re-read a few times to get my head around it.

First glance: I share your concerns. I am very much opposed to questionnaire that require "judgement" I behalf of the researcher. These have been used to our detriment in the UK IMO.

I agree that most patients who have been ill for any length of time will be pacing (even if they don't recognize the term or its not done consciously) to avoid PEM. The questionnaire needs to get to the bottom of that.

 

In my opinion these are simply too vague. They could be a very, very loose description of PEM, but equally they could describe someone with anaemia or hypothyroidism and I'm sure many other conditions as well.

I agree the NAM descriptors are much better.

If PEM is going to be treated as a core symptom then it is vital that the questionnaire accurately captures PEM and captures it regardless of whether the patient has learned to pace themselves or not. It is better to delay and get the right tools in place than rush ahead with the wrong ones.

 

PEM does not always show up within 24 hours, till the last few years i was typically going a few days, once a whole week

 

Looking at each of these (as a patient with CCC/ICC/IOM ME/CFS of several decades):

• Dead, heavy feeling after starting to exercise [Quite the reverse - I feel good initially if I start to exert myself. Even later, I never feel 'dead and heavy'.]
• Next day soreness or fatigue after non-strenuous, everyday activities. [My everyday activities are now paced so that I don't provoke any symptoms.]
• Mentally tired after the slightest effort.[Nope. Mental fatigue isn't really a thing for me.]
• Minimum exercise makes you physically tired. [Yes but that's exertion intolerance - not sure I'd call it PEM. Is fatigue malaise? I wouldn't say so.]
• Physically drained or sick after mild activity. ['Physically drained'? No. 'Sick'? I never push myself to that extent. In order to feel 'sick' I'd have to engage in what for me, these days, isn't 'mild activity' although it once would have been.]

So I think I'd score zero on this scale, even though PEM is a limiting factor on me that keeps me extremely severely disabled.

 

My brain is not up to much today, so I apologise for the brief response to your thoughtful and detailed posts.

Yes, yes, and yes.

Agreed, this is much better.

Ideally yes, very much so. There are a lot of patients here, so a combined, agreed, submission ought to count for something.

 

I think this issue of the delay is crucial. Otherwise, there could be very little to distinguish PEM in ME from normal post-exertional fatigue.

 

I agree that a response from S4ME raising concerns would help add weight. There could be a poll, if necessary, to get a sense of the scale of concern.

 

Wow

Wow, I thought I was doing well reading to the blue part but there’s more!
Sorry i won’t be able to read more before replying.

1) yes. It seems to just be talking about fatigue in general and too easily interpreted as normal healthy if unfit responses. Does not relate to my experience well as I would have been able to say yes to them sometimes before I ever experienced PEM.
2) um, is there a simpler solution? I like the second list. Is there already a question form of the IOM/NAM list?
3) will try to read more tomorrow/later. I find it hard to juggle so many questions without breaks. If I have an answer it makes reading on harder as I want to remember and read at the same time. And this is the start of a good day. Sorry.

 

Yes.

Yes but I can see it both ways. The trouble is less about use of judgement (which unfortunately is always going to be present in some form - patients included) but if more than one different person is using judgement on different subjects/patients. With any method, consistency is the most important part in letting later researchers reinterpret results given new information.
Your approach is (without reading the consult doc) good: ensure you’re asking the right/best questions from the start.

Thanks. It’s good to have a chance to help/participate.

 

If we can facilitate something then we certainly will. If you all continue the discussion then the finer details of a submission by S4ME can be settled later, closer to the closing date.

 

I agree that this is a very bad list. Basically a combination of OI symptoms in response to exertion, and DOMS (healthy response to exertion). So it could have the effect of including OI patients without ME, excluding ME patients without OI, and including healthy people who like to moan a bit about their workouts.

These are much better, though I think still need tweaking:

• A flare of some or all of an individual’s symptoms - There can also be the onset of new or usually dormant symptoms, and the types of symptoms should be listed to distinguish from DOMS in pain-based conditions.
  
• with an immediate or delayed onset and - Immediate onset is more of an OI thing, or other form of exercise intolerance. Many ME patients get immediate symptoms as part of ME, but it isn't PEM, and shouldn't be part of the definition of PEM.
• prolonged recovery - This is too vague - recovery takes at least a day, and typically 2 or more days. The length of recovery also helps distinguish PEM from DOMS.
• Severity and duration of symptoms are out of proportion to the initial trigger - "trigger" is a bit vague, and "exertion" might be better.
  
• There is a loss of functional capacity and/or stamina. - "functional capacity" seems a bit vague. My muscles remain strong, but become fatigued very quickly. And again, while this is an ME symptom, it's more of an immediate/constant reaction to exertion, not part of PEM

So I think it still needs to be more narrowly focused on PEM, and immediate OI and muscular symptoms should be defined separately from PEM, even if they are also symptoms in response to exertion. My suggestion for purely defining PEM:

• The onset or exacerbation of muscular, autonomic, immunological, and neurological symptoms
• beginning hours or days after a normal or trivial amount of exertion,
• with recovery from an episode often taking multiple days.

 

I think it is important to keep in mind the post exertional bit, and not let that be confused with symptoms that occur 'normally' the first time round.

I think (and I do mean think, not certain) that if an averagely healthy (not necessarily terribly fit) person, and a PwME, both do some moderate exercise, then it is the recovery profile from that exercise that is the key differentiator. Following moderate exercise a healthy person will recover in fairly good order, albeit maybe a few aches and stiffness. A healthy person will still feel OK (a bit knackered but happy most likely), and able to basically function normally. The recovery pattern following moderate exercise for a PwME on the other hand, is very different. If my wife overdoes it, bearing in mind she is very good at pacing normally, she feels like cr*p soon after, and it is then a slow climb out of the hole for a day or two after. She feels flu-like and ill; all sorts of aches and pains; feels like someone plugged something into her and sucked all her energy away; thinking seems a superhuman effort; and the very last thing she feels is "tired but happy" ... feels too ill for that. And whereas a healthy person starts to improve almost immediately, my wife will continue to feel bad for quite a long time after stopping the exercise - she may even go further downhill, before it starts to turn around and she starts feeling better ... and it is some while before she turns that corner.

It's not that my wife never experiences these things except after exercise, she does, but then she is invariably pacing herself and doing some exercise anyway. But I do think it is the struggle to recover, and the recovery pattern, from exercise that needs to somehow be got across that is significantly different for PwME, and what PEM is all about. Sometimes the descriptions of PEM seem to muddle that with normal ongoing symptoms, although I'm sure it is a very grey area anyway.

 

Just a thought on the order in which PEM criteria should be listed: If you want to make it clear that PEM is first and foremost debilitating, then I believe any criteria should begin by clearly listing this as the very first point: PEM criteria no 1: a pathological loss of "functional capacity" following exertion or other stressors (untill we think of a better term, I agree @Valentijn it is a bit vague, and the right term would have to also include "brain functionality" ). And then in second place you include various symptom flare-ups and all.

Whenever PEM criteria begin with the idea of symptom flare-ups in general, they have not yet captured that feature, and a sore throat for example is not what causes disability in the PEM experience! I could have 20 different physical symptoms and still be functional, so that would not be PEM in my book. On the other hand, I could have no symptom at all other than a prolonged drastic loss of functionality following exertion, and that would qualify as PEM as I see it.

 

I know when I was less severely ill and didn't know any better I still exercised. I could exercise at that point, but as you say @Barry , the aftermath was very,very different.

The more I tried to exercise the worse it and less able to exercise I became. This is the opposite of my experience of exercise prior to becoming ill.

When healthy the post exercise aches could often be eased by gentle exercise and movement. With PEM I found, when I was still well enough to try, further exercise and movement made the symptoms rapidly worse.