Patient led measure of outcomes

[Kitty]

I think for something like a long term study of pwME's fluctuations over time that would not be useful. It doesn't take into account worsening, and, depending on the activities picked, may turn out to be to limited, with further improvements or worsening not registering.

That's why I think there should be at least one 'wish list' activity, and probably two. One that's a bit out of reach at the start of a study and another that's further away.

A longer term study to needs to capture easily-missed phenomena such as cutting out activities to leave capacity for the basics, and putting in place layers of protection (e.g. sensory avoidance and reducing the possibility of calls, messages or home visits because you're not well enough to cope).

One way to make a list of six or seven items less tiresome to use might be to set all the options at "No" by default. You only have to check the boxes you can answer "Yes" to, which means the more unwell you are, the less box ticking you have to do.

 

[hotblack]

If there was a list of preferred terms, would a voice activated recording device work.

Could do. Despite my love of technology I was thinking a piece of paper and putting a mark in a couple of boxes once a day may be enough. But others may prefer a spreadsheet.

I don’t think getting too hung up on technology etc at this point is going to be helpful if we want to see if we can get a framework that works. If we can and researchers see use it they may have the budget/resource to develop apps or whatever as needed.

Minimal viable product and paper designs are quite common in tech prototyping

 

[NelliePledge]

I agree based on my experience with free Visible that remembering to do it daily is likely to be an issue but I guess even a couple of times a week would be better than not tracking at all?

Paper would be just as much of an issue on remembering as a simple app I think. And I’m more likely to know where my phone is than a piece of paper.

When I do the 7 day am/pm blood pressure monitoring for my annual review I send it in through an online form, they send text reminders. So possibly that could be a useful function in an app. I get reminders in visible but the texts work better at getting my attention.

 

[Trish]

I agree based on my experience with free Visible that remembering to do it daily is likely to be an issue but I guess even a couple of times a week would be better than not tracking at all?

That's why I keep going on about wearables.
If we have daily tracking of steps, resting heart rate and ideally time upright too with no effort apart from remembering to charge the device once a week, then the other tracking can be done less often.

 

[hotblack]

It might be interesting to each try designing our own descriptors of deterioration, no change, significant improvement and recovery.

My version for me in my current state:

This is really interesting! Thanks for sharing.

 

[NelliePledge]

Yes good point.@Trish

 

[bobbler]

Could do. Despite my love of technology I was thinking a piece of paper and putting a mark in a couple of boxes once a day may be enough. But others may prefer a spreadsheet.

I don’t think getting too hung up on technology etc at this point is going to be helpful if we want to see if we can get a framework that works. If we can and researchers see use it they may have the budget/resource to develop apps or whatever as needed.

Minimal viable product and paper designs are quite common in tech prototyping

I’d also warn that you would need to allow people to retrospectively fill it in. As I often with sleep being messed around and suddenly fatigue ability would feel miffed if I missed a deadline because I thought I’d wait 2hrs to end of day then slept thru it.

On the other hand I’d do neither of these - paper nor voice note - and only the app works because I have a peak in labelling to prompt gaps I’ve missed.

So I think for any measures that are a lot of times eg not thinking back to how many showers in a week/month or if you brushed teeth those days (and even then an app to be able to note the day in when I remember because yes I will annoy myself thinking was it thurs of Tues I did that, even if it seemingly doesn’t matter) then tech that does that specific for you , but you test it just for the validity of that measure would be the way. Eg @Sasha mentioning feet on floor or angle of calves (I can be in bed and whether they are flat can make a difference)

Be warned I miss even pressing the button on x once a day when I’m ill so a lot will need to be retrospective. And the more it can be made into prompts not thinking of words the better (doesn’t mean an app couldn’t translate people’s own terms into a survey box tick instead of writing down of course)

 

[Kitty]

Minimal viable product and paper designs are quite common in tech prototyping

I remember making a working phone app that used photos of icons I'd drawn in biro on Post-It notes.

But paper forms would add a lot of costs to a trial. In theory you should be able to scan them, in practice there'll be problems. They'll have creases from being sat on by cats, wrinkles from getting damp, or battered edges from being kept in a perfectly neat pile but for quite a long time. A pile of audience response forms that only spent 10 minutes in people's hands would jam up our scanner beautifully.

I agree based on my experience with free Visible that remembering to do it daily is likely to be an issue but I guess even a couple of times a week would be better than not tracking at all?

Yes, daily is a pain. Registering once a week whether or not you did X, Y and Z is less so, but the questions have to be chosen or phrased to reflect severity at the start of a trial—for instance, taking a shower once a week would indicate improvement for some people, and significant worsening for others.

 

[NelliePledge]

As a moderate person something around frequency of going out of the house is a good indicator of my best level of functioning. But would need some subtlety between “essential” and “social” activity to capture how discretionary for that person . Eg funeral of close family, dental appointment would be at the essential end with football match, family lunch at the social end.

At the other end of my capacity levels lower frequency of showering is a good indicator of my worst.

I reckon just those 2 would give a pretty decent indication of where I am.

 

[Kitty]

I reckon just those 2 would give a pretty decent indication of where I am.

Yes, same for me. But in a trial of a treatment that worked, it could become less relevant or too crude to capture the reality.

For instance, to healthy people the ability to meet up with a friend or take a shower whenever you want isn't an event, it's daily routine. To someone who's improved but still pacing it could begin to look like that, even though it isn't really because they keep needing to pull back a bit.

We need to take into account the possibility of needing to capture the nuance of recovery that's very significant but not complete.

It would involve identifying patterns that show the difference between choosing not to do things because of lack of time or energy (which is normal), and choosing not to do them because they'd probably trigger PEM (which isn't).

 

[hotblack]

But paper forms would add a lot of costs to a trial.

A good point, although so does developing a website or app.

Also one of my aims here is for us as patients to try and find a something which works for us as patients. Get that basic framework right. What can we track ourselves that we feel is meaningful for us and would indicate to us if an intervention has worked. Because that is what most of us say solutions others develop don’t do right.

How is important but that should for now be up to individuals to do however they find easiest. Or maybe we can revisit it in more detail once we get the what working?

I do like the wider discussion on what a good trial may look like too. But for us to try any of that seems beyond our current resources.

 

[hotblack]

We need to take into account the possibility of needing to capture the nuance of recovery that's very significant but not complete.

Exactly! It’s a tricky one. We’re not trying to capture the entirety of our experience of ME/CFS. But equally we do need to have some range in there. I really don’t know where the right balance is.

 

[hotblack]

I’d also warn that you would need to allow people to retrospectively fill it in.

So tell us what you want to record and how. That’s what we’re looking for, how would you approach this? What would you do? You’re the boss @bobbler and you are in control here Take my suggestion and modify it to work for you

 

[NelliePledge]

Yes, same for me. But in a trial of a treatment that worked, it could become less relevant or too crude to capture the reality.

For instance, to healthy people the ability to meet up with a friend or take a shower whenever you want isn't an event, it's daily routine. To someone who's improved but still pacing it could begin to look like that, even though it isn't really because they keep needing to pull back a bit.

We need to take into account the possibility of needing to capture the nuance of recovery that's very significant but not complete.

It would involve identifying patterns that show the difference between choosing not to do things because of lack of time or energy (which is normal), and choosing not to do them because they'd probably trigger PEM (which isn't).

I agree about trialling a treatment

But for monitoring my state for non trial purposes for my own information or for eg should any form of individual review with health professionals take place I think to be able to say I could make it out of the house once a week last year but this year it is once a month and I could shower four or five times a week but recently once or twice. Would be informative. Equally being able to shower every day or go out twice a week in my car.

 

[NelliePledge]

I agree that the tracking of steps heart rate time upright should definitely be the standard for recording and would suggest adding hours slept to that, possibly sleeping pattern

Then the optional 2 or 3 activities chosen by the individual to reflect their circumstances.

 

[NelliePledge]

I wonder if the people at visible would be willing to engage with this @hotblack might be worth asking

 

[Kitty]

I do like the wider discussion on what a good trial may look like too.

Oops, I think I might have misunderstood the thread—I thought it was about how we measure outcomes. Ignore my waffle!

 

[bobbler]

So tell us what you want to record and how. That’s what we’re looking for, how would you approach this? What would you do? You’re the boss @bobbler and you are in control here Take my suggestion and modify it to work for you

sorry I thought I had but didn't mention that I'm thinking like the app/visible type thing, mainly ticking a few boxes on a scale for not too many features, but being able to select those first. that has a shortened funcap monthly too.

I can give you (an) answers to a question of name 3 things you guess you can do, as I have done again in the next paras, but I'm pointing out there is another level too. I can't think that creatively on my abilities other than going by what I've tried. I have so much on my plate asking me to self-assess from all sorts of angles and it is about the hardest level of thinking you can get when there are lots of different overlapping conditions and of course ME being situation-dependent. So I do have insight into how blinking hard this is in reality. And how complex it is when something mightn't be a 'cure' or even 'specific to ME' but a case of fishing out benefit/risk or whatnot for things that might help and exist already.

but making sure that it can be done retrospectively (within limits, depending on what the data is you mightn't remember if you had a headache 5 days ago) if you don't get it done on that day before falling asleep then wake up and its past eg midnight cut-off.

and that if anything has to be specifically taken (eg hr) on a certain day then it does it automatically at a certain time as a back-up, so if there is eg a better time for it to be taken because of what someone has that day re: sleep-wake or whatnot then (depending on how relevant that bit it) the person can either add to it or override it. BUt it is there in case you forget.

PS I'm assuming here that the measure would be long-term, so 6months AND 12months, 18mnths (and maybe more) because that is when the crash will happen for something we though worked but was eg like a stimulant that just allowed pushing boundaries and then bit in bum with a deterioration longer term. So this longer term takes the pressure off having an over-productive month or where it lucked out with other cumualtive stuff.

BUT that the measures would only be eg for a 2 week period around those points, so people aren't having to be stuck doing an app or whatever for months on end.

As it is , when you talk about measures like shower or toothbrushing I'd be talking about a month accounting period maybe for frequency. I do think it would be interesting to add the app type data to these however because how long you can/have to shower for and how well I'd feel after are key, and because standing and brushing teeth for a consistent time each day would be a whole new ballgame vs struggling through whilst sitting when I can (but the issue is that just lenght of time of brushing could also be misleading)

It gets really complicated once I start thinking about what we'd use that would be useful and then consistently measurable instead of only thinking from the angle of what is important to me and filling in certain things.

Because for example I'd be tempted to suggest my third one would be an on-screen cognitive or reaction type test test. But then up crops issues such as at what time of day would I do that would be impacted by sleep and many other things. And if I'd only do it on every few days for example. And then me having to think whether that or if I'm lying or sitting whilst doing it really matters or is overkill if it is being done eg over a few weeks repeatedly.

I suspect the 'feet up' % is actually the more insightful over the number of steps too as a measure of health. Because if I'm ill or in PEM other symptoms might make me move as well as me feeling awful and unable to move. But I will have an overwhelming need to get my head down when I can. It probably accounted better for the extra strength but not changing the underlying ME/CFS re: steroids.


And then I have to go back to the thinking out loud of when it is trying to solve the whole conundrum instead of just 'testing whether that one thing makes a difference, and if it does then it is a clue in the bank' when I start thinking about all the cumulative stuff causing variation.

I'm not trying to frustrate, actually help by talking it around because you do have to keep coming back to the original 'research question' to check it doesn't get distracted into 'what info can you get' or 'what is the illness vs what is something helping'. This is important stuff.

I think what these measures are is something you are approx on the right track with. I do think that we need to bear in mind a few caveats as I've been trying to think out loud around

1. these things need to remember to focus on what matters with regard differentiation for the actual thing that is being tested. Just my experience over the years of working in eg differentiating things (and yes also many years of market research/insight - strange how those two areas overlapped, but basically I've done a lot of 'how do we find out/measure/validly rank' type stuff), and having to keep eye on prize as to 'what do we need/matters' rather than just ways of differentiating/ranking/getting a figure. My steroid bit goes into that somewhat - lots might change but only certain things maybe are really differentiating at a certain time period. A similar point might be that if we went with hr or something then whether it helped ME or not that (or other drugs) might might it go up anyway in some people.

2. there are some things worth trying where we can't predict where the benefit will be, but it is significant none the less which is why I included the b12 in my examples of thinking through some things in actually context/history. I haven't thought through how we accommodate for that.

I think we need to be open to the possibility that we live with such a symptom load that even if we anticipate x will mean we could do more or change something, if we realised that issue we didn't talk about anymore of a constant thick headache disappearing or breathing better (not having noticed it before) then who knows if it is 'unrelated' but it is worth having if we've done the off-on enough times so it isn't just coincidence. I'm not sure even the longer tranche of questions/symptoms would anticipate that because we mightn't realise our headache was a 4/5 and not a 2/5 because our frame of reference had been skewed by it being there so constantly (and getting worse from there) but once it disappears and the idea of it coming back then it becomes more than a 2/5?

It probably feels like those 2 points are the polar opposite of each other.

The normal approach for both of these would be to run exploratory research so we go into such things with a sense of where the 'good and bad effects' are likely to lie/appear. SO if something makes you more clear thinking it can make sure it includes that.

 

[hotblack]

Oops, I think I might have misunderstood the thread—I thought it was about how we measure outcomes. Ignore my waffle!

It’s not waffle! It’s all really useful. Sorry I didn’t mean to dissuade any discussion.

I think I’ve just got really into the idea of trying something ourselves and tweaking it until it works then letting some researchers run with it.

 

[Sasha]

Oops, I think I might have misunderstood the thread—I thought it was about how we measure outcomes. Ignore my waffle!

I think I may have confused matters with my focus on trials!

It’s not waffle! It’s all really useful. Sorry I didn’t mean to dissuade any discussion.

I think I’ve just got really into the idea of trying something ourselves and tweaking it until it works then letting some researchers run with it.

I have slightly lost the plot, I realise! What uses are we thinking of for this measure?