The ‘Patients like me’ website and how it relates to ME/CFS

[Chezboo]

Does anyone know anything about the ‘Patients like me’ website and have any in this community attempted to shape this potential resource into something more purposeful for us, is that even possible?`

The bulk of information I could find here on PR was posted in an old thread from 2010.

http://forums.phoenixrising.me/index.php?threads/patientslikeme-www-patientslikeme-com.2511/


A slightly more recent mention dates back to 2012 where someone says they will petition to have ME/CFS included on the site which seems to have been successful.

http://forums.phoenixrising.me/index.php?threads/patients-like-me.17388/

Today, on the conditions page ‘CFS’ is listed under 'Neurological and Brain'. When you go into CFS it is called ME/CFS and SEID is mentioned. There follows a list of ‘common symptoms reported by people with ME/CFS’ which can be best described as a meaningless assembly of vagueness. It bears no relation to the CCC or ICC and mention of PEM, OI or other common, defining and or debilitating symptoms are likewise avoided.


https://www.patientslikeme.com/conditions/27


My curiosity is piqued, it's seems that it might be a very useful tool. I am wondering why the community has seemingly not used and informed it more over the years? Have there been stumbling blocks, are ‘Patients like Me’ reluctant to engage in a more meaningful way? I don’t even know how it works so perhaps I am not asking the right questions. I wonder does anyone know more?

 

[rosie26]

I received an email 2 weeks ago from someone asking me a question about a tablet experience I had posted there about 4-5 years ago. I had completely forgotten about that site.

 

[Kati]

I have been a member for quite a while and a few months ago I revisited my data and decided to update my condition, symptoms and meds, and I started getting unwanted suggestions like "Have you tried this?..." It was offensive to me, so I am not using that.

 

[undiagnosed]

My main problem with PatientsLikeMe is that they control the data which makes it difficult or impossible to do research. They're a business so they have to make money somehow. I am working on an open source decentralized alternative to PatientsLikeMe that will be more flexible and the community can have direct input or even help with development. It's still early, but you can check out more information here if you're interested.

 

[Mark]

My main problem with PatientsLikeMe is that they control the data which makes it difficult or impossible to do research. They're a business so they have to make money somehow. I am working on an open source decentralized alternative to PatientsLikeMe that will be more flexible and the community can have direct input or even help with development. It's still early, but you can check out more information here if you're interested.

There have been a number of people talking about similar initiatives recently. It's something I've wanted to do for a long time (as indeed did Cort). I'm in the process of organising a Conversation here on PR to get all the different people together to discuss the best way for everyone to move forward on this together, but we're waiting on someone who won't be available for a week or so.

You know what? I'll just mention all those people here and you can all get that together - please copy me in to the conversation...
@JaimeS
@Scarecrow
@ChrisArmstrong
@Cheshire
@sparklehoof

Apologies if I've forgotten anyone, hopefully any others with an interest will register that interest on this thread. One or two of the names above do know of a couple of other people who are interested I think, so feel free to add others to the conversation.

 

[undiagnosed]

There have been a number of people talking about similar initiatives recently. It's something I've wanted to do for a long time (as indeed did Cort). I'm in the process of organising a Conversation here on PR to get all the different people together to discuss the best way for everyone to move forward on this together, but we're waiting on someone who won't be available for a week or so.

You know what? I'll just mention all those people here and you can all get that together - please copy me in to the conversation...
@JaimeS
@Scarecrow
@ChrisArmstrong
@Cheshire
@sparklehoof

Apologies if I've forgotten anyone, hopefully any others with an interest will register that interest on this thread. One or two of the names above do know of a couple of other people who are interested I think, so feel free to add others to the conversation.

I'm not familiar with everyone listed here and what they were planning. To start, I think it'd be helpful if everyone stated what functionality they wanted to have and their skill set regarding how they could contribute to the project.

The functionality I have specified on the p2pdx project page is the following:

• Track medical data including:
  • Lab test results
  • Symptoms
  • Vitals
  • Visit history
  • Medications
  • Demographics
• Data is shared reducing duplication when entering information
• Visualize results with graphing capabilities
• Search data with basic filtering capabilities
• Perform research on dataset with R or Python with SQLite reading helper scripts

I have a software engineering background with experience in front-end and back-end web design. I am exploring a decentralized peer-to-peer architecture rather than the standard centralized client-server architecture for this project. It is a bit experimental as the technology is still evolving, however it ensures that the data is available to everyone. If anyone has any feedback or concerns over this high level architecture decision, feel free to chime in.

 

[Chezboo]

Thanks everyone for your replies. It certainly has been educational. In answer to my original questions it seems pwME do not engage (or engage much) with ‘patients like me’ for a number of reasons. Likewise there are a host of other sites based on user rated medications or supplements which @JamieS recently posted about in this thread http://forums.phoenixrising.me/inde...epository-for-me-cfs-found.45254/#post-735972 Again all these sites seem to be of limited (but some) use to pwME for a variety of reasons.

I am no web developer and technical talk of data bases brings on the FEAR rapid! However achieved technically (centralised, decentralised, SQLite, Pythons, snakes of differing origin or otherwise) such a project could have significant potential for us as a community.

I don’t know if this is just mudding the waters (or even already considered/implied) but here’s an idea to throw into the mix. It is not my own, I read about in the old Patients like me Thread (linked above). Samuel was discussing clustering groups of symptoms from the bottom up. Jenny mentions ‘statistical techniques like cluster analysis and factor analysis’ (I don’t even know what these are but they sound mighty useful). Being able to run certain types of analysis may reveal some very interesting data although I understanding running certain types of analysis on data bases might not always be possible depending on how it is set up and its functionality. The idea with this bottom up analysis is that you might see symptoms groups and patterns emerging, likewise with treatment analysis etc. It does seem like quite a huge thing to pull off and a massive commitment. I am not technical so I don’t see how I can help with this. Clearly though, I would love it if someone were able to pull something like this together!

 

[Chezboo]

Just thinking, if people really did want to come together to try and get the most potential from a custom data system perhaps I could help, more in a coordination respect of course than technical! There is some work which could be done before deciding what you want to build, check for all existing and underway data bases been built (thinking of Open Medicine Foundation resource that @JaimeS mentioned and even Ron Davis’ ultra mega big data thingy with the billions of data points.). My comments might not even be relevant to this undertaking but thought it worth mentioning just in case!

 

[undiagnosed]

Thanks everyone for your replies. It certainly has been educational. In answer to my original questions it seems pwME do not engage (or engage much) with ‘patients like me’ for a number of reasons. Likewise there are a host of other sites based on user rated medications or supplements which @JamieS recently posted about in this thread http://forums.phoenixrising.me/inde...epository-for-me-cfs-found.45254/#post-735972 Again all these sites seem to be of limited (but some) use to pwME for a variety of reasons.

I am no web developer and technical talk of data bases brings on the FEAR rapid! However achieved technically (centralised, decentralised, SQLite, Pythons, snakes of differing origin or otherwise) such a project could have significant potential for us as a community.

I don’t know if this is just mudding the waters (or even already considered/implied) but here’s an idea to throw into the mix. It is not my own, I read about in the old Patients like me Thread (linked above). Samuel was discussing clustering groups of symptoms from the bottom up. Jenny mentions ‘statistical techniques like cluster analysis and factor analysis’ (I don’t even know what these are but they sound mighty useful). Being able to run certain types of analysis may reveal some very interesting data although I understanding running certain types of analysis on data bases might not always be possible depending on how it is set up and its functionality. The idea with this bottom up analysis is that you might see symptoms groups and patterns emerging, likewise with treatment analysis etc. It does seem like quite a huge thing to pull off and a massive commitment. I am not technical so I don’t see how I can help with this. Clearly though, I would love it if someone were able to pull something like this together!

Regarding the technical stuff, haha, no problem. The cluster analysis, or more generally, unsupervised learning, you referred to would definitely be possible and in fact was one of the primary analyses I had in mind. There are a bunch of different algorithms that could be applied to this once we gather enough data.

 

[undiagnosed]

Just thinking, if people really did want to come together to try and get the most potential from a custom data system perhaps I could help, more in a coordination respect of course than technical! There is some work which could be done before deciding what you want to build, check for all existing and underway data bases been built (thinking of Open Medicine Foundation resource that @JaimeS mentioned and even Ron Davis’ ultra mega big data thingy with the billions of data points.). My comments might not even be relevant to this undertaking but thought it worth mentioning just in case!

Ya, getting the word out and getting people to actually use the system is just as important as actually developing the system. So, we'll need all the help we can get on that front.

I'm not aware of any existing datasets that are public domain, but if there are I'd be interested to hear about them. I've seen the CDCs Chronic Fatigue Syndrome Wichita Clinical Study, but it looks like you have to write a study proposal and have it accepted to access the data. The ME/CFS Severely Ill-BIG DATA Study looks promising and is similar in concept to this project, although with different types of data available. Last I heard, they were still gathering data for that study. I'm not sure if it will be publicly available or not. Either way, I think the dataset we could create would be complementary and could only help.

 

[parabola]

Something along these lines was launched for Crohns (crohnology)

 

[Chezboo]

Thanks parabola, looking at crohnology and similar things for other diseases is a great idea.