Patient registries

Some details on the german biobank/registry: https://classic.clinicaltrials.gov/ct2/show/NCT05778006?term=Scheibenbogen&draw=2&rank=1, https://www.bundesgesundheitsminist...g/handlungsfelder/gesundheitsversorgung/mecfs, https://mecfs-research.org/wp-conte...nielaSchindler_MECFS-Register-und-Biobank.pdf. Here are some details on funding: https://www.gesundheitsforschung-bm...men-des-postinfektiosen-chronischen-15511.php.

Of course there are also now the two dutch ones (Lifelines and NMCB) but those are the worst examples and can only serve as an example on how not to do it.

 

We obviously have all the responses from our participants to our primary questionnaire, and there will be a secondary questionnaire that will explore additional areas of interest, but we have never called our cohort a registry.

There is also The Netherlands ME/CFS Cohort and Biobank (NMCB) consortium which is in it's early days.

I'm sure you have thought of this but it's probably worth looking at what benefit registries have delivered for other disease areas, given that insufficient time has passed for any ME/CFS one to prove it's worth.

 

Apparently there is now also an Austrian Biobank setup by the WE&ME Foundation (formerly TEMPI Stiftung) https://www.weandmecfs.org/research...t-austrian-biobank-for-me-cfs-patient-samples.

 

Looks like there's some movement on a possible NZ registry. ANZMES just put out a press release which briefly mentions the possibility of setting one up

Also, have you seen this guide?
"So You Want to Build Your Disease’s First Online Patient Registry: An Educational Guide for Patient Organizations Based on US and European Experience, 2023, Wicks et al" https://doi.org/10.1007/s40271-023-00619-w
It provides a good overview of types of registries and of the process of setting them up, as well as this piece of advice

I think it's worth discussing how to get the best out of a registry. Setting one up is not a trivial exercise and it's not cheap. So you want to get significant extra value out of any new registry, on top of what you can get from the existing ones or from other sources

 

I know this was a while ago but can you still recall or dig up, without going to too much trouble, what you wrote about establishing a patient registry?

Like you said, the most likely scenario is that it would be tacked onto the Long Covid registry though I'm not sure how they stand regarding long-term funding.

But, to get a hang of the thinking behind the LC registry, this is from their enrolment page

So the immediate aims are primarily about assessing the impact or burden of LC, plus some symptom monitoring. Or maybe I should say that was the immediate aim. I'm pretty sure I read somewhere they've already written a report for the Ministry of Health about their findings but it hasn't been officially released. Though it must be floating around somewhere as some media seem to have access to it. I haven't tried to track it down

Recruitment bias would look to be an issue here. Diagnosis, too, since getting one seems to be just as hit and miss as it is for ME.

A non-representative cohort can still tell us a lot about gaps in support in a qualitative way and that would be worth documenting in NZ as some of those gaps will be specific to NZ

But is there a realistic way of getting a more representative cohort?

Monthly questionnaires make sense in the context of early illness, to track how it's developing. Probably doesn't make sense for established ME, at least for the types of questionnaires mentioned. The one exception would be symptom tracking but for that you might want to go daily with an app like Solve. Could they partner with Solve? How is Solve going these days anyway, is the data they're generating being used in useful ways?

I wonder how many other researchers have used this opportunity?

The recruitment bias and self-report diagnosis issues would be a stumbling block for some types of study

But for ME a registry might help attract some new researchers. ANZMES appears to be having trouble finding people to give grants to. Conceivably one reason for this could be that potential researchers don't know how to get hold of a cohort to study on a tiny budget

Theoretically a good idea. Practically double trouble. Trouble 1 is that a lot of people's health records (NHI linking) are such a mess they're somewhere between useless and totally misleading. Trouble 2 is that a lot of people are extremely wary of sharing sensitive statistical information (IDI linking) with anyone, de-identified or not, and that reluctance very likely is not evenly distributed across population groups

All quotes above are from https://www.lcregistry.auckland.ac.nz/about/

 

So what would be the purpose of an ME patient registry in NZ, if it was similar to the Long Covid one?

Identification of gaps in support in the NZ system would be one though you'd still not get any reliable quantitative data. Also not sure if a registry would be the most efficient way of generating this sort of more qualitative data, going through the patient support orgs may be enough. Although a registry might reach people who don't engage with the patient support orgs

Playing the long game, a credible and well publicised registry could have valuable indirect effects such as

More accurate diagnostic coding of ME if the existence of a registry they know they can refer to encourages more doctors to consider an ME diagnosis in the first place and then to get it right. In this case ideally the registry would be linked to quality patient support resources as well as medical education resources, for synergistic effect. The registry itself may still not provide good prevalence data but health record research may become more accurate

Growing the research ecosystem by enabling new researchers on tight budgets easier access to a cohort to study​

Playing the long game requires stable, long-term funding. I'm not sure indirect effects such as the above would be convincing arguments in potential funders' eyes?

What would be convincing? An international partnership with Solve? Other suggestions?

 

@Hutan - I don’t know if it would be worth speaking with Miriam Rodrigues….
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5611797/
She works out of Auckland Hospital. She may have some good ideas. I wonder if she knows Paula Lorgelly…

 

Thanks @Hutan. Excellent letter, too.

Do you know what if anything came of the application? If you're not bound by any confidentiality considerations

I expect Mona and Paula will be talking to each other about this?

The ANZMES link is unclear to me, too. The registry was mentioned as part of the grant they awarded to Mona Jeffreys and Kahurangi Dey

At first glance it's an odd sounding mix of food insecurity and a registry. Of course you could ask questions about the food insecurity of registry participants. But that would mean establishing the registry first and a $25,000 grant isn't going to do that I wouldn't have thought. Unless Mona got some other funding and the grant is more for Kahurangi? Or they didn't get other funding but are doing some groundwork with the help of the grant in the hope that this will open up more funding later? Guesses guesses guesses

Talking about guesses, anyone have any idea what it would cost to establish and then maintain a registry? Just a ball park figure?

 

Has anyone found out more about the potential NZ patient registry? Like, what's actually being proposed and what stage it's at?

Taking another look at the bolded bits in the quote below, what are the most important questions we would want a national patient registry, in NZ or elsewhere, to answer? In an ideal world, everything. But in a highly funding-constrained world, what would be the priorities?

 

Some thoughts about the example questions in the previous post's quote, and their relevance to ME (I'll break it into several posts)

Would be important information to have, we seem to be 'missing' whole chunks of the population e.g. non-Caucasians, but given the diagnostic mess - including suspected high numbers of undiagnosed pwME - is this a question for a registry or does it need a good epidemiological study?

Other patient registries often have fewer issues with recruitment bias because the condition they're monitoring falls under a medical specialty, has more diagnostic certainty and every patient diagnosed gets automatically referred to the registry. They may choose to not participate and there may be some bias in who does and doesn't but nowhere near the level of bias there'd be in an ME registry

People volunteering to be contacted for trials would be a simple and useful addition to a registry but not by itself sufficient reason to build one as there are other ways to find potential trial participants

/continues.../

 

/...continued/

This could be a useful question. Not so much for the mod/severe issue, we already stratify by severity, and the bit we don't know - how many in each category - would be another question for that elusive good epidemiological study

But a large enough registry could help make sense of the many diagnostic criteria. With the right questionnaires people diagnosed with ME or CFS could be matched against several key criteria. This would facilitate studies looking at people with and without PEM, or with and without any other symptom of interest.

Longitudinal follow-up could look at if people stay in the same set of criteria over time, or if people fitting certain criteria are more likely to recover. Lots of opportunities here but it would take a large cohort and very carefully constructed questionnaires (though DecodeMe and CureME probably have good ones to get started with)

Then there's the whole issue of some pwME accumulating a long list of comorbidities whereas for others the same symptoms are attributed to ME. Does one group get better support, medical or financial, than the other?

See previous question

/continues.../

 

/...continued/

Not by itself a question to build a registry on but could be interesting in conjunction with the stratifying question above and it would be an easy question to add to a registry. The symptom/s bothering people the most could be markers of important subgroups, e.g. the always extremely fatigued group vs the group that's only extremely fatigued during PEM could just be different because they're pacing differently or they may have different pathology

Not by itself a question to build a registry on (at this stage) though it could be worth replicating some of the international surveys to get some local figures for advocacy purposes. Different story once new treatments become available, a registry would help give a faster idea of how or even if they work in the real world

/continues.../

 

/...continued/

We have a general idea already, there's much more seeking (and needing) than getting going on. Selection bias would likely skew figures obtained from a registry but they could still be useful for advocacy

See previous question

See previous question

/continues.../

 

/...continued/

An finally:

• What other important questions could a (low-budget) registry answer?
• And which question should be the main priority?
• Or, given there's not been much engagement with this thread, do people think a registry would not be a good idea?
• If so, why do you think it's a bad idea? Apart from the obvious that spending money on a bad idea is usually a bad idea

/end/

 

I think a patient registry is a good idea, the Australian one is run out of La Trobe University, Melbourne. I was on their registry but this is now only really useful for Australians as NZers are now not able to take part in their research. I think selection bias is still an issue for any registry, some people are happy to contribute, some are not. But the more people who join, the more useful it becomes statistically and for advocacy.

It would need to have strict confidentiality and secure IT systems for people with ME to want to submit their data. Though I suppose a low budget anonymised registry could be done but one would have to trust this organisation to keep the data safe and confidential forever. People would have to consent for the different uses of their health information and to who it can be shared.

The Long Covid Registry is based in Auckland. I wonder what funding was required to start and maintain it? It would need continuous funding - something perhaps ANZMES might consider doing. It would be an important tool for that organisation to document all the different things mentioned in Ravn and Hutan's post.

I read on another forum Mona wanted to do a patient registry for ME but didn't have the resource to do it. She is well placed being at Victoria University and is involved in Long Covid research.

Another possible institution that might be suitable is the University of Otago - Wellington Clinical School (or another clinical school, perhaps Christchurch but not Dunedin....) - that has post-graduate research units embedded in their different departments eg General Practice. But one would need a suitable person willing and able to set up such a registry and secure funding.

 

A discussion in a private group raised the question of how to deal with comorbidities in a registry.
Summarising:

• Reasons for capturing common comorbidities: to look for possible commonalities in underlying pathology, to see if symptom subsets correspond with pathology subsets
• The obvious way is to ask people what clinical diagnoses they have
• The obvious problem is that this depends largely on the doctor you happened to see and what diagnoses that doctor 'likes' when it comes to the more contested conditions
• An alternative way would be to not ask about diagnoses at all but instead ask about symptoms (easier said than done as people have such different understandings of what is meant by a particular symptom) and see if grouping people based on those provides any leads

Additional thoughts:

• Asking specifically if people have a clinical diagnosis of MCAS, hEDS etc could mislead people into thinking these are well established diagnoses (and be in for a nasty surprise if they take the idea to their doctor and get short shrift)
• However, recording comorbidities - all types, not just the contested ones - is important. E.g. if a later study wants to use registry data or recruit from the registry for looking at things like HRV or activity levels or employment status then they'd need to know if participants also have for example heart disease, to make sense of the data
• So maybe the best way is a combination approach:
  • ask people to list all of their clinical diagnoses in a free form text box (rather than a list to tick)
  • possibly also ask if they think the diagnoses are accurate rather than lazy misdiagnoses (ok, not in those precise words)
  • tick symptoms on a comprehensive list - free text here would be too unwieldy - with the list covering symptoms from the various diagnostic criteria as well as common comorbidities and misdiagnoses

 

A comprehensive plain language resource covering the very many things that need to be thought through when considering establishing a registry

https://registrybootcamp.org/short-menu

ETA
Also contains links to more in-depth resources like this book & addendum (haven't read either)

Registries for Evaluating Patient Outcomes: A User’s Guide, 4th edition
https://www.ncbi.nlm.nih.gov/books/NBK562575/

Patient- or Participant-Generated Registries - Addendum to Registries for Evaluating Patient Outcomes: A User’s Guide, Third Edition
https://www.ncbi.nlm.nih.gov/books/NBK493594/

Plus links to at least one webinar (I haven't worked my way through the whole resource yet, haven't watched the video yet either)

https://registrybootcamp.org/technical-planning