What do people think of SF-36?

[Utsikt]

Could someone increase their score on the SF-36 by 50 points on the 100 point scale just by reframing how they answer?

After 6 months of LC I spent 5 weeks at a rehab center. I exercised every week day, sometimes multiple sessions. They never taught me about PEM or delayed symptoms. I was told to rest between the exercise sessions, but I had to rest way more than the 75 year old open heart surgery patient.

I went from 400 to 600 meters on the 6MWT. I ran uphill intervals the last week. I had loads of fun and terrible symptoms (but that had been my life for 6 months anyways so at least I got to do something now). And I would have said myself that I could do a lot more. I crashed for a few months after getting home to my parents again.

It took me 18 months to realise that I could in fact not keep up with exercising 2-3 times a week. My health had deteriorated to the point of becoming bedridden, which took me another 6 months to realise and accept (so I crashed repeatedly from overexertion from trying to make my own food and walk a total of 3-400 meter daily.

All of this is to say that you might not need any direct brainwashing to get people to answer better on SF36 or any other subjective scale. All you need is to tell them that they have permission to do more - and they will happily oblige simply because they want to. And the perceived effect can last for a long, long time.

 

[Trish]

Could someone increase their score on the SF-36 by 50 points on the 100 point scale just by reframing how they answer? For example, if you were asked to answer thinking about your worst times over the last month, and then later asked to answer on what you could do right now, ignoring consequences.

I could, very easily.

I could have done easily when my ME/CFS was mild and I was still working and crashing quite often. When crashed my SF-36 physical function would have been around 10, everything was very difficult. When not crashed and going to work I could do all but the top long walks and runs, though some with some difficulty. In an optimistic mood I could have scored about 70.

So yes, easily a change of 50+.

Now, I score about 10 to 15 on my best days and zero on my worst days, so no, I couldn't change that much.

Graham's video really brings home how easy it is to change scores by assessing twice in the same few minutes when nothing has actually changed but the temporary mindset. All you need to do is re-rate everything you scored as 'limited a lot' to 'limited a little', and everything you scored as a little limited as no problem.

 

[rvallee]

I could have done easily when my ME/CFS was mild and I was still working and crashing quite often. When crashed my SF-36 physical function would have been around 10, everything was very difficult. When not crashed and going to work I could do all but the top long walks and runs, though some with some difficulty. In an optimistic mood I could have scored about 70.

So yes, easily a change of 50+.

Now, I score about 10 to 15 on my best days and zero on my worst days, so no, I couldn't change that much.

Graham's video really brings home how easy it is to change scores by assessing twice in the same few minutes when nothing has actually changed but the temporary mindset. All you need to do is re-rate everything you scored as 'limited a lot' to 'limited a little', and everything you scored as a little limited as no problem.

Yeah, early on my spread was very wide as well. Then it compacted, and dropped. And then some more. Now the spread is very tight, but it's very easy to find people who can make it swing by that much, accurately, or reportedly. Just find them early and mild, who by far make up most participants in trials, and voilà: massive improvement.

 

[bobbler]

This is an important question. And I seem to remember some important points about frames of reference that were prompted by another paper in the last year (might have been just before that reply by Sarah Tsyon on another thread) that I think @Hutan is also mentioning in her comment.

I'm just dropping by at this point because I'm reading David Tuller's piece on the physios4FMD research: Trial By Error: Null Results in Physiotherapy Trial for Functional Motor Disorder | Virology Blog

And it made me think of one important thing. Which is that for these people I can well imagine the next step is that if you don't get your results you want then change the measures (like we saw explicitly suggested in the Crawley et al, 2013 paper - another arm of PACE but done 'in clinic' but it didn't get any results for physical function so suggested it needed PROMS developing to 'change what was being measured', and of course we have seen this enacted slowly through the literature over the years since that by that same group).

And of course we know about ceiling and floor effects in the Chalder fatigue questionnaire


There is something rather important about us being able to remind everyone else not just about the coercion that is present in research done to people with things deemed as 'functional' and looked into by certain individuals who might also have powers over their being assessed as 'non-compliant' affecting all sorts etc. who are trapped in a rather more powerless situation than would be acceptable for any other condition

But also about whether any claimed, fished around for supposed 'change' is meaningful. And by that it means meaningful in the sense of whether other people would find it acceptable that people who don't want to hear about or acknowledge the harms or how much had to be sacrificed for that person to apparently improve how many times they could wave their hand on a test on that given day, actually doing so not because their health improved but because they felt coerced or misled to give up other more important activities like reducing shopping, reading, showering in order to perform on something useless day-to-day.

And in the search for people from certain sectors where there seems to be some norm or pressure to find silver lining claims, and in the context of what we are currently seeing with attitude to disabilities happenning with the government changes to disability benefits, it really is pretty key that we have something that 'frame of references' the sheer debilitation of the illness vs other very different situations or injuries or conditions it now seems. We can't assume anything as common sense or understood.

So I agree on the FUNCAP but I also am starting to think we need both measures in order to triangulate sheer level of disabiltiy into that [vs other conditions people do understand the debility-level for - there is something quite important about a SF-36 42 score vs 58 for 'something they can visualise'].

There is something important about an existing OK'd thing being used over leaving it to the whims of people with other concerns and foci.

I think triangulation / pinning the ME-specific scale vs one that is disability in general is the sort of term I'm meaning here. Given minimisation has been such a problem for our illness(es).

And also to point out that anyone who claims a tiny % increase, even if it wasn't cajoled or based on false hope or people-pleasing or having been fed a coffee that day after having perfected the journey being low exertion compared to the first time of assessment, realises that even if it wasn't at the sacrifice of even bigger downsides then it really isn't meaningful to the extent anyone should be using the term or allowed to use the term 'trying to help'

I'm conscious for example of how important it was as context when looking at the NIH Walitt study to be able to get a sense too of who was post hoc in which groups, and then understand the conditions of and under which the eg effort preference task was being done.

 

[bobbler]

Ahh the other bit that I'm getting faint memories on with this, so will need to look up old posts, is that it was originally only intended to be used as a global score by its designers I think but commonly gets split so the physical function arm and mental health arm are calculated and used separately.

 

[bobbler]

I'm not necessarily a fan of this scale and will probably dissect its use for all sorts of research purposes.

But it is also a nagging thing that time and again there have been many, particularly physio/rehab/bps papers over those key years where it mightn't have been admitted in the abstract but it was the SF-36 physical function that didn't play ball for them back then.

One thing that is intriguing is how it ends up being a measure that is consistently 'not so bad/far off' despite perhaps the sum of parts critique that could be made. I'd guess that implicitly comes from its origins/reason for being (public health for a population, rather than KPI/show results on x, y, z) which might have meant a different starting point when they were thinking of its methodology and what to avoid.

I think we need to be just as cautious of keeping an eye of tweaks and changes to it, how it gets interpreted etc and things being used instead - like PROMS.

We need to make sure it continues if it is used in different ways to be applicable to the type of disability we experience. Even if it is just a finger in the wind measure?

I guess there is also the point that who knows how many of us have comorbidities as many don't really get looked at to certain levels, and how much of that is impacted by/dragged down by one condition like ME/CFS as an underlying thing then affecting other bits (or other way round).

Along with we have an approx idea of severities and the batteries of symptoms and levels in those but that's 'as a package' that level kicks in rather than just on individual symptoms eg whether someone needs a wheelchair or would see their cognitive function as impacted by a certain amount - which all becomes academic when we all know it is about having eg 4hrs a day you can function to any level at eg mild, but 4hrs a month further into severe on top of those limitations.

Hence the laugh (but it isn't funny when you think of the bigotry it actually pictures) of someone thinking they've invented something 'very helpful' that if they got someone to use said 4hrs to practice eg writing, hand waving, walking and that person finds if they prioritise it for said months they can do more of that (because they've learned to reorganise their life around that one 'goal') then it means something. Maybe it is nice to take a holiday from the humdrum getting by to focus on an achievement with lots of support for a change (if the 'quality of life' ever was real rather than a fop to 'thanks for your time trying'). A bit like a healthy person goes to French classes in the evening but to do so didn't tuck their kid into bed or stay late at work that night.

Then the point gets implicitly missed as people assume if you did it for one thing you can build everything back. Because they weren't testing can people still shower once a week (and everything else they did before) and write or walk more. They can't imagine the walking/French goes out of the window not because of motivation, but 'can't' after a year of having to to try and manage the other more important stuff too. Or a step too far causes a deterioration, or just that the steps so far just cause a deterioration that happens at 6months anyway.

But in the healthy person example as said person adds in more night classes the boss, colleagues and partner picking up the slack might be wondering just how much benefit the multi-lingual additions are vs the day-job being sacrificed/done less well for it.

But Rather than just shows if you play games with a disabled person you can't reframe to magic out the 'can't'. That 4hrs + other limitations might bubble out in different individual impacts (situation and other things making life miserable) across individual areas but that overall score says something.

Also Interesting given the term 'holistic' is one that was so popularly used as an ambiguous charading word by eg bps, which in itself would mean looking at/measuring 'the whole'. Which circles back around to my point in para 5 about us keeping an eye on it being used in bits or tweaks etc as if that is 'the same thing'. And the 'sum of the parts' bit I put somewhere.