Social-technical model for studying health information technologies17 Apr 2019
Today was dominanted by a series of seminars on the electronic medical records, safety, and all the forces that are interacting with them.
There were two seminars, one by Dr. Dean Sittig and another by Dr. David W. Bates. I’ll just touch on one of the seminars given by Sittig, which I’ve presumed came from this paper Sittig wrote a few years back with Dr. Hardeep Signh on, “A New Socio-technical Model for Studying Health Information Technology in Complex Adaptive Healthcare Systems.”
The paper outlines 8 interdependent dimenions involved in the, “design, development, implementation, use, and evaluation of HIT.”
The dimensions are:
- Hardware and software computing infrastructure
- Clinical content (i.e., the record itself and more)
- Human computer interface
- The people
- Workflow and communication (for and among the people)
- Internal organizational policies, procesures, and culture (i.e., internal pressures)
- External rules, regulations, and pressures
- System measurement and monitoring, e.g.,
- Availability and uptime
- How features are being used
- Effectiveness of care delivery for public health reasons
- Document unintended consequences
It’s been a while since I’ve seen the complexities of the EMR described in this way. It was really fascinating and mind boggling to things in each of these 8 dimensions at once! However, some of the speakers commented that this might be why some researchers and practitioners just focus on one aspect and think that is the way to fix the system.
Things to share
Working on remote branch
If you have a remote branch on, say, GitHub, and don’t have that branch on your computer, you can use that branch locally by running the following code:
git checkout --track origin/serverfix Branch serverfix set up to track remote branch serverfix from origin. Switched to a new branch 'serverfix'
But in looking this up to write about,
it appears there is a shortcut with just
git checkout serverfix Branch serverfix set up to track remote branch serverfix from origin. Switched to a new branch 'serverfix'
The gist of decorators are functions that extend the behavior of other functions without changing the code directly. Intuitively from the name itself, I’m slowly internalizing this as a simple “decoration” on functions that make them nicer or complement the function. And to go further, it is more of “wrapping paper” than anything else.
Here’s a quick abstraction that helped me understand what these are.
@dec2 @dec1 def func(arg1, arg2, ...): pass
def func(arg1, arg2, ...): pass func = dec2(dec1(func))
Note, that the order of which function is applied is from bottom to top.
In this case,
dec1 function is applied first and then
Just a quick review for me. Pearson’s chi-squared test is a statistical test for categorical data.
The Chi-squared statistic is calculated as
where is calculated as (row total times column total)/total samples.
For two-dimensional data (i.e, contingency tables), this is derived from
where are the marginal probability for each column and row.
Some quick links for more: