Data Science Team Training 2021

Query to CSTE Health Department teams using who use Python.

Hello CSTE teams, currently we are attempting to leverage our IT department in our county to get on board with up-to-date software and have received a large pushback. In order to innovate and be as efficient as possible, we need to acquire permissions for use of new tech.

One of the programs we aim to use is Python. Are there any CSTE teams who already have employed Python into their software suite? If so, how do you use Python on your health data? We have provided the context to our IT department, their fear is we will create apps with it. This is far from the direction we aim to go if we are to obtain Python. We would like to get our Health surveillance to a point where Python can mine the data in a way that identifies key words on a population level that can point us to brewing pockets of crisis, in a sense, use it as a preventive measure to identify where we should aim our intervention efforts.

Does any Health department already use Python and how do you use it?

Thanks bunches for weighing in.

Comments & Events

Patric Prado
Hello! I mostly use R. But have used python for geospatial joining when geocoding addresses. This is important for specifying which case counts go to which counties and is done within ArcGIS. We are currently automating arcgis and python processes in R. Not sure if other coaches can also weigh in. Josh Levy, Bio- and health informatics consultant Josh Patrick Caneer Patrick Kristin Bergman Kristin  
Amanda Sapp
Thanks for your feedback!
Sarah Seidel
Hi Amanda Sapp Amanda  - we (Texas Department of State Health Services - Center for Health Statistics) use Python scripts on a daily basis to aggregate data, update datasets with new data (i.e. append), and create data files that both sit behind our public-facing dashboards or are available as Excel files for download on our website. We even use it to automatically update our website files each day. We like Python because of its efficiency and the ease of troubleshooting errors/making modifications.
Amanda Sapp
Thank you for this, very helpful!
Josh Levy, Bio- and health informatics consultant
Currently, I use python mainly for tasks that glue a bunch of stuff together, mainly because official packages for certain resources are in python and not R, for example: I've currently got a python script that pulls secrets (ie passwords) from AWS using their official boto3 package, and then it uses those secrets to login to a postgresql db using the psycopg2 package, and then the script sets up users, databases, schemas, and access permissions.

In the past I've used the more 'data scienc-y' packages in python like numpy and scipy, but my preference is to use R for these kind of operations - to each his own.
Amanda Sapp
Thanks Josh!!! Automation, is the take away for Python, I agree with your statement.
Analysis in R, automation in Python.
Patric Prado
Just wanted to mention that you can run python from R using reticulate package and RMD files. Just in case you wanted to throw another option in the mix.
Jason Lee
Thank you for the suggestion Patric.

The team here will come consider it.