Starting chronic desease project

Hi all,
I am starting this thread in newcomers, since it’s not very specific at the moment.

I came across the video of Chris Bartley from the conference.

I think he has done a great start in QS. I was diagnosed for beginning rheumatoid illness. Since not all symptoms are established I can not be called diagnosed as Sjoegrens-Syndrome-Patient.
But so far I developed systems, of course thats why I went to the doctors. As described in the video the fatigue is one of them, also there is joint pain and others.

I can also report on the constantly changing “wellness”. Some days are great and others are really bad. On those days I can’t stop yawning all day long.

Chris Bartley describes his way he went and his findings. But unfortunately the speech he held was short and not very descriptive. So I wonder how I can apply his method to my problems. Since there must be correlations or otherwise my condition would be changing so much on an day to day basis but be more like a constant (decline perhaps over the years).

So I took some time and thought about my approach. I read so much in these forums but all are not very specific in details. Now real how-to as I would like them.
As a medical professional I know that science starts where the project is repeatable by someone else. So I think in this post we should start a discussion on how to do so.

I can imagine lots of people surfing through these pages and seeking an answer for their problems as well, but where is the starting point?

So this is what I intend to do, and I hope, by your help, fellow forum users, that we could compile a “white paper” on how to start such a project.

As I am short on time as many of you I hope I will follow through, for my own sake of course :wink:

So let’s start with the project.
(oh, btw I think there should be a section for chronic decease in the forums)

So now let’s start:

Hypothesis: Sjoegren-syndrome is a autoimmune desease. And we see a world wide increase of that kind of deceases. It may be that this kind of desease may be triggered by environmental/external factors which are unknown. These factors may also change the current “wellness” of the patient. So the removal of those factors my improve the “wellness”. Additionally there my be beneficial factors that may improve “wellness” without being directly related to the desease itself.

Through thorough self observation and statistical analysis it should be possible to find those factors.

Methods: Self observation and collecting data by jawbone up, fitbit one.

The following data is collected
outcome variables:
Wellness 0-1 representing 0% TO 100%
Joint stiffness when waking up: 0 - 1
pain: 0 -1
headache: yes/no

factor variables:
sleepduration: hours
sports activities: yes/no (I am aiming on sessions done)
chocolate: yes/no (known as trigger substance)
nuts: yes/no (allergic triggers)
seaweed: yes/no (maybe high iodine is relevant)
alcohol: yes/no
wheight: yes/no ( I am overweight, so I think of hormone active organ, known to cause diabetes 2)
%body fat: % (same reason)
holiday: yes/no
coffee: yes/no
green tea: yes/no
garlic oil: yes/no ( I use this for my fat loss diet)
black cumin seed oil (I use this as general improvement substance)

plaquenil 400mg (my regular medication)
ibuprofen (I am taking this for pain and headaches)
topical cortisone (nose and skin, but may wander off in blood stream and may have effects)
xylometazoline (also for my nose)

and for weight control
calls steps from jawbone up sensor.

Since I didn’t have a good software I was looking and came across pspp (an gnu statistical software, obviously aiming to be a SPSS replacement)

Since I used SPSS for my dissertation and also did some statistics, I hope I will get this working.

Other factors like they are suggested by all these technology friends like locations etc, don’t seem to be relevant at this moment so I skip these for now.

At the moment I am experimenting with the apps from jawbone up and fitbit. They seem to differ just in the calories databases.
On the website Fitbit allows to add other factors to be noted. But I don’t now whether this is visible in the app. I also have tictrac installed. This one looks nice as well. But I think when the holiday season is over and normal work live starts there is no time to use them parallel.

So I hope we can start the discussion on materials and methods.
Is anything missing?
Did anyone else try to get the relevance of those factors? What Software and statistics did you use?
Is there anyone who has collected data already at hand and can test on PSPP software.

So the goal should be that we can set up a standard procedure for everyone to follow and implement on any machine that has internet connection.

So this was a long post and I hope I didn’t bore you. But it’s hard to find all the clues by myself. So this is what this forum is for.

Thanks for reading and let’s start the discussion



Neither Fitbit nor TicTrac are much good for tracking custom variables. There are some mobile apps like rTracker that provide a convenient interface for recording custom variables, but you have to export the data for proper analysis and figure out how to integrate it with data from services like Fitbit.

Zenobase supports custom variables, imports data from services like Fitbit, and can do basic analyses such as correlations and effect size calculations. Only, entering custom variables is somewhat tedious, and there is no support for multi-regression (a.k.a. data dredging :-).