OHDSI Home | Forums | Wiki | Github

Health Data Stories in 2018

Community:

As I presented on last week’s community call, one of my 2018 resolutions is to try and experiment with new ways to engage with health data and new ways to get our community engaged alongside with me.

As inspiration, I’m stealing a brilliant idea from Nadieh Bremer and Shirkey Wu, who put together http://www.datasketch.es/. Each month they picked a general topic, and each went their own way and ran in whatever direction they wanted to create a cool new data visualization related to the topic. Sometimes they also shared thoughts about the journey they took to produce their amazing pieces of ‘data art’. If you haven’t seen their site before, I highly recommend it, their work is pretty incredible!

My idea is simple analogue to our domain:

Each month I’m going to pick a health topic that I don’t know much (re: anything) about, and do something using our OHDSI tools to learn more about the topic. I know I’ll educate myself, but I’m also hopeful that the exercise will produce some useful nuggets of real-world evidence for the community also, and may even contribute to a broader dialogue in that therapeutic area. At the end of the month, I’ll post a blog that shares two things: 1) what did I learn? and 2) how did I use the OHDSI tools to learn it?

Now, Nadieh and Shirley had each other to be their accountabili-buddies and keep each other on track, and I think that (in addition to their tremendous talent) this was a good part of their success was seeing two different takes on the same topic come out each month.

I don’t yet have an accountabili-buddy for this project (sob…shameless plea for you to be my friend), so I’ll rely on the whole community to keep cheering me on, and if this seems like interesting, hopefully at some point, folks will join in the fun as well. As @Rijnbeek suggested, maybe this activity could also provide fodder for our various working groups to have fun, short-term targets to work on together; certainly I plan to be using this as an excuse to get more comfortable with the patient-level prediction tools that @Rijnbeek and @jennareps have built, as well as the population-level effect estimation library that @schuemie and @msuchard have put together .

To identify potential topics for the upcoming year, I found a site that lists activities to raise awareness of various diseases throughout the year. So, with that in mind, here’s my tentative list of topics I hope to learn about in 2018:

January: Thyroid Awareness Month
February: American Heart Month
March: National Endometriosis Awareness
April: National Autism Awareness Month
May: Arthritis Awareness Month
June: Cataract Awareness Month
July: World Hepatitis Day (July 28)
August: National Immunization Awareness Month
September: National Atrial Fibrillation Awareness Month
October: World Mental Health Day (Oct10)
November: American Diabetes Month
December: Crohn’s & Colitis Awareness Week (Dec1-7)

Happy hacking everyone! Here’s to a fun and productive 2018!

I’m not sure yo’ll include existing study protocol in OHDSI for this challenge. But The Comparison of combination treatment in hypertension based on Population-level effect estimation effect is on-going. I think we get the result from J&J and aggregate the result by Feb. Then I can make the evidence for February with your evidence.

One of the topic for SOS challenge in Korea is the comparison of anti hepatitis B viral agents (tenofovir vs.entecavir) led by @cys7like based on Population-level effect estimation. We’ve just started it but we can generate the evidence for the July. Then the evidence can be released with your evidence.

I’m really interested in the study for prediction of stroke in atrial fibrillation. I want to make a assisting tool or algorithm for using anti-coagulation medication in these patients. I want to collaborate with you for this topic.

I have no doubt that this challenge will set the stage for incredible development of OHDSI. It’s really inspring! @Patrick_Ryan. If there is anything I can help you, please let me know.

t