OHDSI Home | Forums | Wiki | Github

Book of OHDSI Update

(David Madigan) #1

The Book of OHDSI is now well underway! To refresh your memory, the goal is to write a book that will serve as a central knowledge repository for all things OHDSI. The book’s primary form is an electronic book written using the bookdown package. We have a wiki with pointers, meeting times, etc. and the book itself is here. Thanks to the superb efforts of our early volunteer authors, three chapters now exist:

The Common Data Model

Patient-level Prediction

OHDSI Network Research

Please take a look at these to get a sense of the (evolving) style and level we are aiming for.

Now we need volunteers to lead the other chapters listed (and also propose and lead new chapters). Please think about joining this effort. Jump on one of our fortnightly calls or reach out to one of us.

Martijn and David

(Sarah Seager) #2

Hi @David_Madigan - I’d be happy to help contribute to the Book of OHDSI. Let me review the remaining chapters and see where I am best placed with the other chapters that need writing.
I’ll be on the next call :smile:


Hi @schuemie and @David_Madigan,

I’m available in early-mid April to start contributing. I’ve been following along a bit over the past couple months, but will dig in and post some thoughts and questions under the Book of OHDSI forum posts soon.

One immediate question: how should we add comments or suggested edits to text that is already written? Have ground rules been established for this?


(Martijn Schuemie) #4

I propose we use the GitHub facilities as much as possibles, especially pull request and the issue tracker.

The chapter leads will decide what changes are made in the end. If all else fails, David and I will be the (not so) benevolent dictators to cut the knot.

(Melanie Philofsky) #5

The EHR folks (myself included) struggle/d a little at first with the CDM. EHR databases generally have 100’s or 1000’s tables! It’s very overwhelming. And then there are all these customized data and a multitude of text strings, billing data, historical data, encounter data, patient reported data, integrated registries, data labeled as a diagnosis but really it is just a reason for a procedure to justify billing the insurance company… this list goes on & I haven’t found the end. It would be great to include something specific to EHR data in the “ETL” or “Where to Start” chapters. I would love to help, I just really don’t know where to begin. I joined one call (because my other standing call was canceled) and I haven’t been back since because of said standing call. I guess I’m asking for someone to hold my hand, so I can help inform the next set of EHR collaborators.

(Todd Price) #6

I would be interested in helping.

Where could I get the weekly call-in information at?

Todd Price

(David Madigan) #7

The wiki page has the call details