Development and measurement properties of the PEM/PESE activity questionnaire (PAQ), 2023, Davenport et al

BACKGROUND:
Existing instruments often are inappropriate to measure the effects of post-exertional malaise (PEM) and post-exertional symptom exacerbation (PESE) on activities of daily living (ADLs). A validated questionnaire to measure self-reported ability with ADLs would advance research and clinical practice in conditions like myalgic encephalomyelitis and Long Covid.

OBJECTIVE:
Determine the measurement properties of the PEM/PESE Activity Questionnaire (PAQ).

METHODS:
The PAQ is adapted from the Patient Specific Functional Scale. Respondents rated three self-selected ADLs on two 0-100 scales, including current performance compared to (1) a ‘good day’ and (2) before illness. Respondents provided a Burden of Functioning rating on a 0-100 scale, anchored at 0 being the activity took “No time, effort, and resources at all” and 10 being “All of my time, effort, and resources.” Respondents took the PAQ twice, completing a demographic questionnaire after the first PAQ and before the second PAQ. Descriptive statistics and intraclass correlation coefficients were calculated for each scale to assess test-retest reliability. Minimum detectable change outside the 95% confidence interval (MDC95) was calculated. Ceiling and floor effects were determined when the MDC95 for average and function scores crossed 0 and 100, respectively.

RESULTS:
n = 981 responses were recorded, including n = 675 complete surveys. Test-retest reliability was generally fair to excellent, depending on function and scale. MDC95 values generally indicated scale responsiveness. Ceiling and floor effects were noted infrequently for specific functions.

CONCLUSION:
The PAQ is valid, reliable, and sensitive. Additional research may explore measurement properties involving functions that were infrequently selected in this sample.

Open access, https://content.iospress.com/articles/work/wor220553

 

Development and measurement properties of the PEM/PESE activity questionnaire (PAQ)

"Measuring Functional Deficits in PEM/PESE: An In-Depth Look at the Development and Properties of a New Survey Tool

Are you, or someone you know, experiencing post-exertional malaise (PEM), also called post-exertional symptom exacerbation (PESE)? PEM/PESE refers to a delayed worsening of symptoms after physical, cognitive, or emotional stress. PEM/PESE is the hallmark feature of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and those with long COVID who meet the diagnostic criteria for ME/CFS.

Until recently, we lacked a tool for patients to self-report the effects of PEM/PESE on their ability to function. Quantifying the impacts of PEM/PESE can be helpful for assessing medical treatments, documenting disability, or simply tracking function over time.

The recently-published PEM/PESE Activity Survey (PAQ) by Workwell Foundation seeks to measure the impacts of PEM on a person’s ability to function."

https://workwellfoundation.org/deve...s-of-the-pem-pese-activity-questionnaire-paq/

 

The questionnaire is available here, https://pacific.qualtrics.com/jfe/form/SV_aUZNZUGGbcZnQBU

Todd Davenport on Twitter
"The PEM/PESE Activity Questionnaire (PAQ) is one of the first tools developed to measure disability associated with PEM/PESE. It is now available for personal use. Please help us continue to validate it by participating in an ongoing IRB-approved study."

"You can email a copy of your results to yourself and to two others you choose, like a clinician or a research team. It's a little barebones at the moment, but as I get better at the survey platform, things will continue to get more polished. Thanks for using the PAQ!"

 

I gave up on the questionnaire because
1. triggering options included toxic mould but not viral/other infection (only specific ones that most will not be familiar with)
2. none of the "what's most important to you" questions really fitted what I would have put, and the questions/answers about it seemed not to fit my experience. (there were probably more questions that I didn't get to).

My conclusion was that they first need to pilot this with patients and translate it into plain English - mostly I would put it at undergrad level.
ADDED- Fog Index score of 20 (17 is college graduate)
This is for the very first answer option:
"marked physical and/or mental fatigability in response to exertion, which may be minimal such as activities of daily living or simple mental tasks can be debiitating and cause a relapse."

METHODS:
The PAQ is adapted from the Patient Specific Functional Scale.
But the PAQ is flexible and lets people choose what matters to them - the authors adaption of it does not work for me.

I'm a big fan of getting meaningful scales - and I like the approach of comparing vs when last well and a 'good day'. But the questions and options need more work, IMO.

Maybe ask some people with ME to join them in setting up the next study? It often makes for better research.

 

I answered all the questions except the one about severity. I had to skip it because it bore too little resemblance to my illness.

I'd describe myself as moderately affected, but there's no way I could manage some of the things it suggested moderately ill people can do. The severe categories were much more accurate in some respects, yet it'd be nonsensical to compare my physical function with people who actually are severely ill.

This part of the questionnaire needs to be much more flexible. Part of the issue is that it doesn't take into account cognitive vs physical capacity; these aspects need to be separated because they're not the same in everyone. Cognitively, I am on the boundary of severely affected.

There's a third category that's something about capacity to deal with stress and external demands. I could sit at a desk for two to three hours a day some of the time, but I couldn't work at a desk. It would be too cognitively difficult, it would carry responsibility, there'd probably be deadlines involved, and I can't cope with obligation. Some pwME people are more physically limited than I am, but seem to have more capacity in this respect.

 

Todd Davenport has explained on Twitter that the severity scale is the Bell CFIDS Disability Scale.

I've sent feedback to him, making various points, including this on the "important activities" section,

* "How well can you complete this activity in this moment compared to a "good day?”” assumes a substantive difference between the time point of completing the questionnaire and a “good day”. Also does not seem to allow for the time point of completing the questionnaire being a “good day”.

* “How well can you complete this activity in this moment compared to before you became ill?” can be difficult to complete if length of illness is particularly long. E.g. employment might be important to me, but as my illness started as a child, I don’t have a time to compare to when I wasn’t ill and employed.

There is also the potential for seeming anomalies, such as I am more able to “Create and maintain close, romantic and/or sexual relationships” than I was before I became ill - but you would obviously hope that was the case when comparing an adult in his 50s with an 8 year old child. And even then I can only report being able to "perform at the same level as before I became ill”, not being able to perform at a higher level.

* "What time, effort, and resources does it take to do this activity at the level you rated?"
Which level? This is the third question, where the previous two asked us to rate how well we can complete the particular activity.

 

But does that study have an opportunity for feedback (I bailed out because the questions didn't fit my experience)? Otherwise, I think it's pointless; the questionnaire needs fixing before this becomes a 'validated' instrument that has been developed without proper patient input.

 

Todd seems to be open to feedback but not in the sense of it being a round of specific feedback seeking in order to improve the questionnaire. But then we've seen all of this before with other questionnaires, unfortunately.

 

Some technical issues:

• There's no final option to choose to submit or not, if you don't want to submit you need to bail out before clicking 'next' after rating activity #3 (I didn't so they now have a questionnaire from me with meaningless responses despite my best efforts to answer meaningfully)
  
• The questionnaire has pop-up windows offering the option of continuing without answering but this only randomly works, sometimes you get taken back to the unanswered question and forced to answer it and sometimes you're allowed to continue without
• There's no option to leave feedback in the questionnaire itself

But, to be honest, improving this questionnaire is probably not worth the effort. I think basing it on the PSFS was a mistake. Something designed to measure the impact of a localised problem isn't suited to measuring the impact of something as global as PEM.

https://www.physio-pedia.com/Patient_Specific_Functional_Scale

Are they aware of the existence of the much better FUNCAP questionnaire? I think it would be better if everybody adopted this rather than create their own. Otherwise we again get a whole lot of research findings that can't be compared. This would be a problem even if the questionnaires were all good. Which they aren't

 

Just done this. A few initial thoughts:

1) On the "level of physical function" question: the two closest options for me are ".. 50% of expected. Usually confined to house. Unable to perform any strenuous tasks. Able to perform desk work 2-3 hours a day, but requires rest periods." and ".. 30%-50% of expected. Unable to leave house except rarely; confined to bed most of day; unable to concentrate for more than 1 hour a day". Well, I'm almost completely housebound, and I definitely couldn't perform 2-3 hours of desk work every day, but I'm not confined to my bed for the majority of each day & I'm nowhere near 50% of my pre-illness function - not even close. Also, the pwME that have low levels of cognitive dysfunction and high levels of physical debility - or vice-versa - this scale wouldn't really capture them well.

2) The level-of-function percentages don't seem to align with other severity scales that I have seen.

3) Some of the symptoms aren't really symptoms at all but other medical conditions, for instance:

Migraine and tension-type headache are very different disorders and both are common in the general population, TTH vastly more so than migraine. (I've have migraines since childhood but developed ME as a young adult, never had TTH, and experience non-migrainous headache in PEM.)

4) The list of activities is somewhat eclectic, sometimes confusing, and sometimes just impossible to answer. For instance, if one selects the "carry out my daily routine" option as an "activity.. that is important to you" the following question is "How well can you complete this activity in this moment compared to before you became ill?". But my daily routine today is completely different to my pre-ME daily routine, so that's an impossible question to answer.

5) Some of the activities in the "how well can you complete this activity" list aren't exactly amenable to a 0-100 scale: "lay down", for instance - how can you rate how well you can lie down on a 0-100 scale? It's really binary: either you can move enough in the moment to do it or you don't have the energy to move and have to have someone assist you; there's no quality aspect to lying down, certainly not one that's expressible as a percentage.

6) Many post-viral patients aren't going to be able to categorise their initiating infection as from an enterovirus or herpesvirus. And does the average patient with a glandular fever / mononucleosis onset know that GF is caused by the Epstein-Barr virus?

7) I noticed that the "What time, effort, and resources does it take to do this activity at the level you rated?" question persists even if you select that you cannot perform an activity at all; there's no "not applicable" option.

A lesson, I think, that questionnaires really need to be developed with input from pwME at all levels of severity.

 

Yes, it struck me that sections about symptoms ought to clarify that it means only those that started after ME onset, or that aren't clearly linked to another condition. I answered it that way, but it wouldn't occur to everyone.

Same with Andy's point about this:

I assumed that if you gave the same score for the current day and a 'good' day, it would indicate that the questionnaire day was a good day. But it leaves pwME to work that out, and no one can be sure that the response would be interpreted in that way by the clinician.

 

Thank you for this -