1. Sign our petition calling on Cochrane to withdraw their review of Exercise Therapy for CFS here.
    Dismiss Notice
  2. Guest, the 'News in Brief' for the week beginning 18th March 2024 is here.
    Dismiss Notice
  3. Welcome! To read the Core Purpose and Values of our forum, click here.
    Dismiss Notice

Jarred Younger Lab: New Chronic Disease Survey using machine learning (December 2022)

Discussion in ''Conditions related to ME/CFS' news and research' started by shak8, Dec 20, 2022.

  1. shak8

    shak8 Senior Member (Voting Rights)

    Messages:
    2,202
    Location:
    California
    Here is the email I received from the Younger lab with the link to the survey and nothing about your data privacy (which I then asked Younger to supply via email, haven't heard yet from them).

    (Beginning of the email, in full):

    "Here at the Neuroinflammation, Pain, and Fatigue Laboratory we are rolling out a new chronic disease survey. I am hoping you can complete it and tell us more about your unique medical history! If you do not have any major medical conditions, we also invite you to complete the survey. 20 respondents who complete the survey will be selected for a $50 e-gift card of their choosing.

    This survey takes a deep-dive into medical conditions, symptoms, and treatments. I believe it is the most comprehensive survey ever conducted on chronic pain and fatigue. We will be using machine learning tools to discover new targets for improved treatments.

    The survey is being launched world-wide, and we aim to collect data from thousands of individuals. It is open to anyone with a chronic condition, and can be completed even by individuals with no medical conditions. All the information is helpful to us.

    It is a long questionnaire by design, but you can stop and resume it any time (using the same device). There are some suggested places to take breaks in the survey. You can come back and complete it on another day if needed. From the small group that tested the survey, it looks like it will take between 1 and 2 hours (depending on how many symptoms and other medical conditions you have). Some respondents finish under an hour.

    The survey can be taken on a smartphone, but I suggest using a tablet or computer with a larger screen.

    You can use the link below to start the survey.
    https://uab.co1.qualtrics.com/jfe/form/SV_cJ8XAyPtbWHqI6i

    If you have any problems with the survey, you can let us know via email: youngerlab@uab.edu . When we finish the analyses, I will do a YouTube video to describe what we discovered. You can find our videos at: https://www.youtube.com/channel/UCoVoOvIX90IMEZCbBf_ycEA/videos

    Thanks!

    Jarred Younger

    Jarred W. Younger, Ph.D. | Professor
    Director of the Neuroinflammation, Pain, and Fatigue Laboratory
    Department of Psychology
    UAB | The University of Alabama at Birmingham
    Campbell Hall, Suite 233

    [​IMG]
    Follow this link to the Survey:
    Take the Survey

    Or copy and paste the URL below into your internet browser:
    https://uab.co1.qualtrics.com/jfe/f...8XAyPtbWHqI6i_CGC_227oMvp0PzZHIDa&Q_CHL=email

    Follow the link to opt out of future Y:
    Click here to unsubscribe

    (End of email)



    [​IMG]
     
    Ravn, sebaaa, EzzieD and 7 others like this.
  2. Sly Saint

    Sly Saint Senior Member (Voting Rights)

    Messages:
    9,574
    Location:
    UK
    Ravn and Peter Trewhitt like this.
  3. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    12,290
    Location:
    Canada
    I started filling in the questionnaire but I've a big snag on a page where they ask to "describe your pain" with a scale of mild-moderate-severe and they list types of pain like throbbing, stabbing, etc. and I really don't know how to answer those questions in a way that makes any sense without adding invalid data.

    It's very generic, not sure how useful it can be.
     
    SNT Gatchaman, Ravn, Milo and 3 others like this.
  4. shak8

    shak8 Senior Member (Voting Rights)

    Messages:
    2,202
    Location:
    California
    I started filling out the initial health conditions. I found the choices very old-fashioned and limited and therefore I lost interest in grinding out an hour of my time so that Younger can toy with machine learning.
     
    Ravn, Milo, SNT Gatchaman and 2 others like this.
  5. Milo

    Milo Senior Member (Voting Rights)

    Messages:
    2,107
    One questionnaire was clearly about catastrophization. Like people above, there were questions where I wanted to answer none of the above but there is still a problem, or assuming things that were not right for my case. It was horribly long and I gave up in the end.
    Not holding my breath. It would hardly be :"discoveries" because patients have been at it for so long, and answered the same type of questionnaires. They are the garden variety types of questions we are used to answer. Self-report survey, self-selected which presents risk of diverse bias.

    I have been reminded lately that the definition of insanity is to do the same thing over and over and expecting different results.

    P.S. I have not watched the video and I am willing to stand corrected. Pardon my annoyance, I am dealing with the flu and it's making me extra cranky.
     
    Ravn, shak8, Peter Trewhitt and 2 others like this.
  6. Ravn

    Ravn Senior Member (Voting Rights)

    Messages:
    2,042
    Location:
    Aotearoa New Zealand
    I'm interested in how to construct good questionnaires so I wanted to see what this one was like. I persisted for maybe half an hour all up. Then I bailed out. The questions became more and more impossible to answer in any sensible way. The results coming from this are going to be an uninterpretable mass - make that mess - of data. AI won't save it.

    I'm sorry to be so negative. We could do with a thorough symptom survey because there are discrepancies between the descriptions that keep getting regurgitated in published papers and the descriptions we see on patient forums. But it needs to be well-done and doing it well isn't going to be easy but it most definitely won't involve regurgitating - there is that word again - the same old questions that have missed the mark to date.

    Some good comments under the Youtube video by @Snow Leopard. Should have trusted his assessment and not bothered wasting half my cognitive energy for the day on this. If only they had come to us for comment before going live with their survey!
     
    rvallee, Trish, Sly Saint and 5 others like this.
  7. Ravn

    Ravn Senior Member (Voting Rights)

    Messages:
    2,042
    Location:
    Aotearoa New Zealand
    Forgot to mention, if anyone decides to plough through the survey, be aware that the survey saves entered data every time you move to a new page - but not before.

    So don't do what I did and take a break part way through completing a page (some pages are quite long). When I returned the survey had timed out and I had to redo the part of the page I had filled in before my break (previous pages were saved, just not the one I had been working on).
     
    Trish, Peter Trewhitt and Milo like this.
  8. Milo

    Milo Senior Member (Voting Rights)

    Messages:
    2,107
    Also @Raven you can skip questions or entire pages if you don't want to answer.
     
    Ravn, shak8, Trish and 1 other person like this.
  9. Kitty

    Kitty Senior Member (Voting Rights)

    Messages:
    5,036
    Location:
    UK
    I hit the same issue. I can only describe pain using sound metaphors, e.g. one element of my chronic pain is like flowing traffic on a motorway—ongoing noise that rarely alters in pitch, but varies in intensity. Other types include a continuous middle D played very loudly on an electric keyboard, a distant but intolerably shrill intruder alarm carried intermittently on wind gusts, etc.

    People may have metaphors that describe their pain well (why a middle D? No idea, it just is), but wouldn't describe someone else's. Even if you could capture it in a survey, providing for at least half a dozen different types and intensities of pain, some constant, some cyclical, and some erratic, I doubt it'd tell you much except "pain is a problem for these people".
     
    Peter Trewhitt, Ravn, rvallee and 2 others like this.
  10. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    12,290
    Location:
    Canada
    Huge difference with the MEAction survey, which is a breeze to answer as it doesn't make the usual assumptions that makes answering the questions awkward. What a difference it makes when the questions are asked correctly. I really don't think medicine will make any progress until they start working with patients. It's not enough to have good intentions, the quality is simply missing otherwise, if it's not entirely irrelevant.
     
    RedFox, Amw66, NelliePledge and 2 others like this.
  11. Sean

    Sean Moderator Staff Member

    Messages:
    7,041
    Location:
    Australia
    This.

    Asking the right questions is where science starts.
     

Share This Page