A New Decade of ME Research: The 11th Invest in ME International ME Conference 2016
Mark Berry presents the first in a series of articles on the 11th Invest in ME International ME Conference in London ...
Discuss the article on the Forums.

what 10 questions would you ask?

Discussion in 'Create A 'PatientsLikeUs' ME/CFS Treatment Program' started by Sherlock, Jun 17, 2011.

  1. Sherlock

    Sherlock tart cherry etc. for joints, insomnia

    Czechosherlockia, USA

    Let's say that this concept went forward in a different way: small scale but able hopefully to be gradually expanded over time to cover any number of things. In that way, progress can be made, some useful data might be garnered and interpreted, and the whole thing won't get bogged down before it ever lifts off.

    This is just a webapp, it can be done with ordinary LAMP. (That's Php for the pages, with MySql as the database -- and with the usual Apache on Linux that's available at most any webhost.) That's all that's needed - though maybe with a Flash frontend for easy use by PWCs, even though the heart is the data and not the looks.

    If that were the case, it in my view would/should have the data being freely available to anybody who wants it. I'm often musing that if data from science studies were open, then individuals might have insights and run queries and can come up with correlations that professionals are too boxed in to see themselves.

    Plus that puts more brainpower at work onto the task - just by sheer numbers. You all might be aware, for example, that major comets have been discovered by amateurs.

    Let's say there was an pilot program with just 10 questions. The survey would be completely and entirely anonymous, forever and ever. There likely should be assigned a Master Patient Index as someone has described, known only to the patient. Pains need to be taken that a person's identity could not possibly be discovered.

    At any future point, another survey could be done, perhaps focusing on a particular topic - let's say PEM. The same MPI would hopefully be entered by respondents, so that answers about exercise could be correlated with answers from the first survey.

    At a later point, it would be easy to do surveys on various Tx and correlate to other data. Sometime later, another module is run for results of lab tests. Anything is possible. Ad hoc groups could be formed to compose surveys on topics of interest, then be responsible for getting people to take their survey. Somebody needs to always pare back the number of questions :)

    Since the data is free, respondents can be invited from any source audience, such as other forums - or any websites that don't forbid it. No one owns the data. This is not google or bill gates.

    The data is not scientifically valid for many reasons, but it sure might be hypothesis forming.

    The big question that remains is: how many people would actually participate?

    Let's go to me and sayong being un open and tired: anyway, but not like you

    Another is if people can and will store and re-use their MPI? Otherwise each survey becomes a standalone. Maybe the MPI should be a passphrase and not a number.

    I could do the first survey easily - if I don't get pulled away. No charge, of course. The actual programing/setup wouldn't take long. There would be the survey page. There would be a download page for the data. (Note that when you send a PM here on vBulletin you get links to download a CSV etc of your msgs.) There might be an associated page to let non-tech people run a query for themselves, to follow hunches.

    There would be no community type aspects. I haven't registered with PLU to check them out but I'd suspect they are going for that community thing as a way of angling for ad income, by trying to get people to hang around there like at Facebook. Everybody wants to be a Facebook these days to get rich, but that doesn't apply here.

    Question types would be radio button, checkbox, textbox (mainly for comments, not data) and maybe rating from 1 to n.

    Whether or not to do this project depends on whether respondents will co-operate. I think that'd be difficult enough even in a population with energy :)

    So, is it possible to get PR behind an initial survey to see how things go, and what would be the 10 questions to ask initially?

    P.S> I always make it a point to never try to force respondents to answer every question. When I'm ever asked to volunteer to answer a survey, it's aggravating to be told "YOU MUST ANSWER" each question that they demand. People drop out when told that. They drop out anyway, if the list of questions gets too long.
  2. illsince1977

    illsince1977 A shadow of my former self

    Hi Sherlock,
    Welcome. I like your avatar. As a member I'm having the same rough thoughts, but lacking in current tech skills. It all sounds doable on an initial small sample as long as it's scalable for the future. You want to build a patient MYSQL database with patient ID attached that is scalable to include more patients and more questions and tests and treatment response as it scales up? That would be a dream, but wonder if it has been attempted before and what lessons could be learned from past experience. One may not need to reinvent the wheel, but I haven't investigated all the options.

    Crowdsourcing will probably be the only way we'll get these data collected. And I hope other members think this is an ideal place yo do so.

    I agree with your PS., but maybe people will be eager to answer many questions if they can break up their answering sessions.

    Eager to hear others thoughts on this.

    What do you mean depends on whether respondents will co-operate? answer questions and continually answer longitudinally or give permission to analyze data due to privacy concerns or lack of concentration or time?
  3. taniaaust1


    Sth Australia
    ask for what?

    Can you make a simplier post... I tried to read your post but its far too long and my brain started messing up all the paragraphs, words and that and I couldnt read it all. 10 questions for what?
  4. Leopardtail

    Leopardtail Senior Member

    Ratings 1..n alone don't work for ME. The issue is that very often what we call 1 vs n varies greatly with current health. Strongly defined scales are needed to make results comparable and summarisable.

    So the issues is not just what questions would you ask? But what would the multi choice answers be, and how would we grade each answer (e.g. in terms of severity).

    To get an idea of what I mean. The Bell scale is a single question but defines all ten possible answers.

See more popular forum discussions.

Share This Page