Accurate and Objective Determination of ME/CFS Disease Severity with a Wearable Sensor. Palombo et al. 2020

[wigglethemouse]

Depends over what time period they looked to correlate this.

They measured for 3 days after the test. There is some good discussion about things that could affect measurement of PEM using the NASA Lean Test. Good discussion for other studies that look into PEM to consider.

UpTime Before vs. After NASA Lean Test
Interestingly, UpTime for the control group alone decreases after the Lean Test; however, this change is believed to be due to weekend UpTime trends rather than the effects of the NASA Lean Test. (Days 1, 2, and 3 after the Lean Test are Thursday, Friday, and Saturday, respectively.) Furthermore, the ME/CFS groups’ UpTime spikes could have been a direct result of participating in the NASA Lean Test. A 5-10% increase in UpTime equals roughly 1-2 hours of upright activity. This increase could easily be the amount of time required to take the Lean Test and drive home from BHC.

With the results of Figure 4, we find ourselves forced to reject the hypothesis that activity decreases after the 10-minute NASA Lean Test. This finding can be explained in a few different ways. On the first day of each trial, the subject traveled to and from the BHC to be equipped with the Shimmers. Due to the extreme sensitivity of ME/CFS patients, this travel alone could have unintentionally induced PEM. With patients experiencing PEM throughout the entirety of the study (rather than just during days 4 through 6), we would expect to see constant UpTime scores. Future studies should consider home-visits to reduce this effect.

Whatever the reason, it is clear that the 10-minute NASA Lean Test had no statistically significant effect on UpTime. A better experiment design would track each subject for a more extended period before and especially after the 10-minute NASA Lean Test, thus establishing more accurate baseline UpTime scores for each subject. However, limitations in funding and time prohibited these design improvements. Further investigation may provide deeper insight into the causes and effects of PEM.

The authors felt that day of week affected the result, especially among controls, and this variation needs to be controlled for.

Furthermore, we have identified that weekdays differ from the weekend in terms of activity—especially for the control group. Regularly occurring trends made it difficult to accurately determine what portion of observed changes were caused by the 10-minute NASA Lean Test. Future studies should randomize each subject’s start day to remove day of the week as a confounding factor. This randomization was not performed in our study to minimize costs due to a lack of external funding at the onset of the study.

 

[Amw66]

They measured for 3 days after the test. There is some good discussion about things that could affect measurement of PEM using the NASA Lean Test. Good discussion for other studies that look into PEM to consider.






The authors felt that day of week affected the result, especially among controls, and this variation needs to be controlled for.

Thanks for the additional info .
We did a home version of the 10 minute lean test monitoring HR ( no BO sensor on our wearable). Pretty convincing OI, but need BP to explore this further.
Had we done this out of home situation PEM may have been a consequence.

My daughter has a routine , but feet on the floor would not have been a reliable indicator as she spends most of the day sitting on her bed ( feet up ) . She is semi reclined on a back support for a chunk of the day.

Lower resting HR was noticeable for a few days afterwards - always a harbinger of shit.

 

[Mithriel]

How long I have feet on the floor is definitely a sign of how well I am, distinct from problems with OI. It is much better than some of the scales where siting in a chair is seen as being better but they do not describe what sort of chair.

I use a heart rate monitor to tell how my ME is going and a device round my ankle which measured how long my feet were on the floor would be useful information for me day to day. It would be a great aid to pacing.

 

[wigglethemouse]

Dr Shad Roundy answered a couple of my questions. Good news is the work will continue. They believe it could be an important tool to encourage more investment in treatments.

We have a lot of data that we have not fully analyzed yet. We will be looking at other measures that might be useful in describing disease severity including activity levels.

Our real goal is to develop a robust digital biomarker for ME/CFS so that people developing treatments have an objective method to demonstrate to the FDA that their treatment is effective.

As I understand it, the lack of a robust method to demonstrate the effectiveness of a treatment is preventing more widespread investment in ME/CFS treatments. So, we are definitely planning on continuing our work here.

Once the work is a little more mature we will probably share all the algorithms, but I don’t think we are at that point yet.

 

[Ravn]

"With the added perspective of large-scale studies, this sensor-based platform could provide a recovery path for individuals struggling with ME/CFS."

I hope it means something like "this could be a useful objective measurement of physical activity, in trials of therapies for ME..."

Weird phrasing all right. Based on the article it looks like something got lost in translation between mental concept and the written expression of it.

This is a Preprint. Can they still fix it?

Good news is the work will continue. They believe it could be an important tool to encourage more investment in treatments.

That's good to hear. This could be a useful addition to the toolkit. Of course no single device is ever going to be perfect for everyone in every situation - actually what I would really, really like is for somebody to invent a headband that can tell me when I've had enough mental exertion for a session - but having a wider range of physical activity measures can only be good, both as an assistance to individual pacing and as outcome measures in clinical trials.

For myself, I'm pretty sure the UpTime device would register a difference between better and worse periods, ranging from better end of severe to worse end of severe. For pwME who don't spend any time with their feet down the device would just register 'very severe' all the time - and that would be correct. Not exactly news to the pwME but a form of objective documentation nonetheless.

 

[wigglethemouse]

This is a Preprint. Can they still fix it?

He did reinforce that this is a preprint and still to get reviewers comments. Seemed to welcome input.

 

[FMMM1]

Other than the last sentence of the conclusion, which seems a bit odd, it does seem useful.

"With the added perspective of large-scale studies, this sensor-based platform could provide a recovery path for individuals struggling with ME/CFS."

I hope it means something like "this could be a useful objective measurement of physical activity, in trials of therapies for ME..."

ETcorrect: typo

Yea this type of objective measurement would have shown that PACE was crap.

 

[Mij]

This sensor does not measure the severity of ME. It's not that simplistic and you can't base it on upright time. There is a difference between measuring levels of disability and severity.

"Severely ill ME/CFS patients spend less than 20% of each day with feet on the floor". My best friend died from complications of ME and was very severe, much sicker than me, but his upright time was longer than mine.

 

[Wonko]

If my maths is right then I probably spend, on average, less than an hour with my feet flat on the floor (by which I mean standing upright, moving around etc.) - as there are often 24 hours in a day that would equate to under 5%.

Of course it could be made more complicated by working it out based on an average 16 hour day (around 6%), or the number of hours I am actually awake (which is usually more than 16).

I don't really see how they could make it time out of bed as then they wouldn't capture, or would inaccurately capture, a lot of people.

Physically I am mainly housebound, and 'limited', but still manage to get the odd thing done, some aspects of my cognition are 'limited', others seem MIA, and others have actually improved (as far as the rest of me can tell anyway).

Moderate physical, probably moderate/severe cognitively (at least when 'challenged' by anything not totally routine).

To me (N=1) the idea that 20% foot down time, or less, equates to 'severe' is 'not accurate'.

20% could possibly equate to 3.2 hours up and moving around a day (assuming a 16 hour day) - that IMO is not 'severe'.

 

[Trish]

If my maths is right then I probably spend, on average, less than an hour with my feet flat on the floor (by which I mean standing upright, moving around etc.) - as there are often 24 hours in a day that would equate to under 5%.

The 'feet on the floor' description includes sitting with feet on the floor. This could include sitting in a wheelchair or on a sofa provided the feet are not raised level with the knees.

Our severity of different symptoms varies, so those with severe OI as part of their severe ME would have very limited time with feet down, whereas others with less OI but severe cognitive dysfunction and other symptoms but milder OI might have less trouble sitting but still be severely affected.

That's why to get a full picture of severity levels we need something like this feet down measure alongside measures of activity and cognitive function.

 

[Wonko]

I rarely, very rarely, as in almost never, sit with my feet on the floor.

They tend to only hit the floor when I am am about to make an attempt to stand, or when I am doing something that requires me to stand.

So, yes, I agree, other measures are needed to gauge severity level, and as I was hinting, severity level isn't just one thing - I can do many simple short duration tasks whilst cognitively being slightly stupider than my fridge.

This may, indeed will, give an uninformed, or biased, observer the impression that my capabilities are higher than they are in reality.

 

[Sasha]

I'd love one of these to monitor variation in my condition as I try out new things. Quite expensive, at 359 Euros. Here's the product.

 

[Trish]

I rarely, very rarely, as in almost never, sit with my feet on the floor.

My total feet on the floor per day would probably average half to one hour split into about 10 bits of one to five minutes each for bathroom and kitchen necessary activities. The rest is legs horizontal on sofa or bed.

On their definition that makes my ME severe, but cognitively it's only mild most of the time - I can spend quite a bit of time on the forum and watching TV or listening to the radio when I'm not crashed.

 

[wigglethemouse]

I'd love one of these to monitor variation in my condition as I try out new things. Quite expensive, at 359 Euros. Here's the product.

It will be interesting to look at the data when the paper is released post reviewing as then they will make the Excel data available. The algorithms they are using are not ready for sharing - still work to do.

 

[Sean]

I'd love one of these to monitor variation in my condition as I try out new things. Quite expensive, at 359 Euros. Here's the product.

But surely it would be too much of a burden for patients?

 

[Sasha]

But surely it would be too much of a burden for patients?

indeed!

 

[FMMM1]

The 'feet on the floor' description includes sitting with feet on the floor. This could include sitting in a wheelchair or on a sofa provided the feet are not raised level with the knees.

Our severity of different symptoms varies, so those with severe OI as part of their severe ME would have very limited time with feet down, whereas others with less OI but severe cognitive dysfunction and other symptoms but milder OI might have less trouble sitting but still be severely affected.

That's why to get a full picture of severity levels we need something like this feet down measure alongside measures of activity and cognitive function.

I need to make an effort to even scan this paper. It seems they can separate people with ME, from healthy controls, i.e. based on activity levels. However, I wonder if it would be possible to do the same for cognition. Have any studies shown that it is possible to separate people with ME, from healthy controls, based on cognition? I would have thought you would need to have a "before ME cognition test" and an "after ME cognition test" i.e. to establish whether people with ME have a decline in cognitive function compared to healthy controls.

REQUEST FOR ASSISTANCE
I'd like to do some brain storming here --- looking for ideas --- I'm involved in this https://europeanmecoalition.wordpress.com/

Is there any cheap way to measure activity levels using a mobile phone? E.g. to use activity levels as a way of selecting a group (cohort) of people with ME for a (GWAS) gene study?

@Michiel Tack

 

[Peter T]

I would have thought you would need to have a "before ME cognition test" and an "after ME cognition test" i.e. to establish whether people with ME have a decline in cognitive function compared to healthy controls.

Yes for best results you would need to be able to go back in time to test people before the onset of their condition, though Covid-19 does present a theoretical opportunity to do a prospective study of an entire population, with enough subsequently developing the condition to actually get some meaningful results.

However if you have large enough sample size such that your long Covid group is big enough to represent a reliable profile of the entire long Covid population you can then meaningfully compare to a normal population. The current pandemic also offers that opportunity.

 

[FMMM1]

Yes for best results you would need to be able to go back in time to test people before the onset of their condition, though Covid-19 does present a theoretical opportunity to do a prospective study of an entire population, with enough subsequently developing the condition to actually get some meaningful results.

However if you have large enough sample size such that your long Covid group is big enough to represent a reliable profile of the entire long Covid population you can then meaningfully compare to a normal population. The current pandemic also offers that opportunity.

Thank you for your reply.

Could you combine a physical stressor (I think this used tilt test) with a test of cognitive function? E.g. identify a group of people with ME using a physical stressor and then assess the cognitive performance (of the people with ME and the control group) before and after the physical stressor.
If I'm talking crap then please point it out and explain why - I'd rather hear it in this forum than elsewhere!

 

[wigglethemouse]

Is there any cheap way to measure activity levels using a mobile phone?

There are activity tracker apps. On the iPhone it is "built in" on the Health app. A phone is not particularly cheap though......

The cheapest is a step counter/pedometer. Here is one that has a 30 day memory for $25 - you can manually log the data when you feel able too, say once a week into a free app such as Google Sheets.

https://www.amazon.com/gp/product/B078HWTK5P/

Amazon product ASIN B078HWTK5P