Turning data into science

This forum gets many posts along the lines of "I took X and I felt better". If only, we could aggregate these findings, we would add an enormous amount of value.

As I see it, an important step is to measure everything: "I took 1mg of X and I felt 6% better". Even such a small step requires a lot: a measurement tool; a baseline result; confidence that you really did take 1mg and not 2mg, say; knowledge of how many times you've done this test before, but did not report it because the results were not exceptional; etc..

Any ideas?


16 Replies

  • Yes! The very info I look for in the posts.

    I also like reading what and how and how long others are getting along so I may get a feel for what I can expect in my future.

    Patients like me data


  • What a useful link! Thank you so much.

  • This link looks so interesting.  It might help @darkflower too.

  • I totally agree as the answers may well be found through a combined effort !

    For myself

    Exercise Yes

    Coconut oil Yes but more than a teaspoon a day has no benefit

    Borage oil Yes but more than a capsule every 2 days has no more benefit

    Resetting the Relax Datum in muscles (with or without Physiotherapist) through Massage ; To address imbalances, in ' motor control pairing (flexor v extensor)' and balanced co-ordination, so left and right side of body work in total harmony (and not against each other )

    Admin may help gather info using their Poll ?

  • I agree. I am guilty of not recording dates &/or amounts of things I try. I just proved that by a reply i just made to RoyProp concerning coconut oil & C/L.

  • You might be interested in this smartphone based research project


  • A better link to the project I mentioned just now 100forparkinsons.com/

  • A perspective, collecting data is one thing. Having data that is useful for some particular purpose requires several considerations. First the data represents a

    measurement , an observation. This leads to consideration of precision, error,. reliability, etc.

  • paddyfields, the 100forParkinson's initiative looks interesting. I would have preferred it if the data was open (after being made anonymous).

    BillDavid, you're right: there are complexities. But, I believe that something is better than nothing. I've been collecting side-to-side tap test data and this clearly identifies the dosing cycle. Unfortunately, an accelerometer based test is less clear.


  • John - 100forParkinson's is great for a few reasons but one of them is that it starts to help you to look at your patterns of behavior a little more objectively and to think of ways to measure. Simple example, I can see that if I get less than 7 hours of sleep a night that my mood is directly impacted, perhaps 90% of the time. I now have a hard goal to get 8 hours. This sounds simple but I think it speaks to your request to begin to measure things. Using 100forParkinson's I see the value and that impacts behavior.

    To take your point a step further, wouldn't it be nice if 100forParkinson's used your app so that people could measure the benefits of say getting 8 hours of sleep vs 7 hours?

  • clarification: To take your point a step further, wouldn't it be nice if 100forParkinson's used your app so that people could measure the benefits of say getting 8 hours of sleep vs 7 hours impact on some of the symptoms of PD using your app to measure?

  • worldpdcoalition.org/?page=...

    The patient perspectives podcast has a bit on self monitoring.

  • Today I used a 3X5 index card from my stack by the computer to take a few notes after catching up on reading the forum posts. I noted MedCure as the name of a curcumin supplement I may want to try after my brand is finished and Lexi walking poles as a brand to look into. I know walking poles can help you stand up straighter and have meant to try them someday. I think that taking notes of all the ways people are coping would be helpful to us all. The names of drugs others have been prescribed and their effects might help us when we are prescribed something by our own doctors in the future. One idea some may want to try is the hand gripper exercise for strengthening our hands. I got a blue one by ACF from Amazon last week and it seems to be easier to squeeze things like shampoo and conditioner bottles now. Instead of the index card I think a lined notebook might work better.

  • hopedope, Coincidently I tried Nordic (pole) walking for the first time yesterday. My initial reaction was very positive. This may be a good candidate to trial.

    What do other people think?

    Kudos goes to the first person to come up with a way of making the tests double blind!


  • johntPM, hum, an intriguing idea. I have worked on somewhat similar projects in the past. Here's what comes to mind: One person would have to be the overall project coordinator. They would contact Healthunlocked's admin to find out how many threads are on the parkinson's movement forum. Then, they would call for volunteers from among the participants and divide the number of threads by the number of volunteers obtained. Each volunteer would take their assigned group of threads and glean it for the data. (The coordinator would need to make sure that the same types of data were collected and were logged in a consistent manner.) The volunteers would send their collected data to the coordinator who would aggregate it. Whew! Sounds like a big job. Maybe someone else can come up with something simpler.

  • HeartSong are you volunteering?!

    My aim is to encourage people to use measurement in their posts. To this end my website:


    contains simple to use tools to measure some of the symptoms of PD, and a database, PDMeasure, to store people's scores over time.

    If people used this approach, they would create a personal baseline (e.g. over the last month my tap speed has got slower by 1%), a trial result (e.g. at the end of a month of therapy X my tap speed has increased by 5%), compared to a population baseline (e.g. an average monthly decline in the PD population as a whole of 0.1%). Of course, the problem is noise, but there's ways to address that: larger sample size, higher frequency of observation, better testing tools, stricter trial protocol.


You may also like...