Dr. Brian Hooker of CHD Asked Me a Very Specific Question About Coding to MedDRA Codes...
This is a 2.5min clip of full Twitter Space (Dec 20): HERE
Did you hear Brian say at ~1:25min mark, “One of the things we are trying to do (regarding VSAFE)…” is dynamically code summary narrative free text field into ICD-10 or MedDRA codes?
I never got a chance to answer the question but he asked me if I have thought about attempting this project myself and with respects to VSAFE? I do want to answer this question in this article and also give some full transparency answers. Being transparent and honest is so vitally important considering how much the world is being gaslighted today.
Dynamically coding electronic medical records, physician operation reports, SOAP notes or encounter data into ICD-10 or MedDRA codes is nothing new. If we were in the corporate world you could also throw in Common Procedural Terminology (CPT) codes for billing insurances for reimbursement. Let’s leave out CPT for now and just keep it to MedDRA (symptom) codes for now. I created this little picture to get a better idea of what Brian might be talking about.
You could replace the CMS-1500 insurance claim from and superimpose a VAERS report or VSAFE report as it’s the same concept. The 1 trillion dollar question, is there such a software that uses sophisticated algorithms or AI to dynamically code from anywhere in VSAFE, VAERS, or encounter data and produce the correlating MedDRA codes?
Before you formulate an answer, I want to give you one more variable and that is to superimpose the doctor reading and interpreting the x-ray to a laboratory producing lab results. Doesn’t it seem a little easier for algorithms to calculate a structured set of data or combination of results and produce some diagnosis like myocarditis, or diabetes? Kind of like an automated pap smear tells you if there is low or high grade squamous cells? Wow, I wonder when diagnostic imaging is going to get automated like the pap smear test? When will the AI algorithms be able to read a x-ray or MRI? I wonder if physicians are seeing the writing on the wall? Maybe one day Skynet AI will tell the physicians, you’ve been weighed and you’ve been measured and now you’ve been found lacking? It feel like Daniel 5 in big book… I digress.
A little history:
Academia first attempted to dynamically extract billing data ICD codes from medical encounter data in the early 2000s. One of the earliest documented efforts was around 2002, as can be inferred from the publication date of related research on ICD-10 coding, which was first published in 2002 on Wikipedia.
Moreover, research into automated ICD coding using unsupervised methods to extract ICD-9 codes from electronic medical records (EMRs) was noted around 2012 with the University of Kentucky's study on in-patient visits. This indicates an ongoing interest and development in the field of automated ICD code extraction from medical data during this period.
These attempts focus on using computational methods to reduce the manual labor associated with ICD code assignment, which is crucial for billing, reimbursement, and medical research.
As a Certified Professional Coder myself, I’ve been seeing the writing on wall for quite a while myself. Before this time comes, most of medical billing is being farmed out to India, but I digress again.
It is only my opinion but the CDC/VAERS has their own internal coders, however I think some part or process of assigning MedDRA (symptom) codes to reports are partly automated? Even with AI, you still a couple of your best coders to supervise the AI. Maybe AI goes out and codes the low hanging fruit stuff then sends to the human coders for a final human eyeball check and assign any MedDRA that needs assigning? Did you know some reports have as many as 140 MedDRA codes (symptoms)?
Disclaimer:
I met with Dr. Brian Hooker and a very competent IT person from CHD to discuss a few topics, but organically landed on the topic VSAFE. Bitter sweet is that I introduced Brian and “CHD” to my multiple dashboards. Bottom line is they are just learning about my dashboards and how they are not just interactive dashboards, but specifically how I’ve spent much time doing the “data cleansing” like fixing typos of Lot Numbers, finding missing ages and populating the missing age field, etc.
So anyways, let’s read between the lines, what is CHD doing trying to extract and transcribe symptom codes from VSAFE? Are they planning or going through all that trouble to write a article or a peer reviewed paper? Hopefully CHD is planning on making their own VSAFE interactive dashboard? I mean come on let’s not be coy, aren’t CHD, ICAN, Bobby, Aaron, and Del connected at the hip? I know CHD is working with OpenVAERS.com on some kind VAERS database…
Maybe somebody will help OpenVAERS with her numbers because nobody seems to be double checking especially for the Dynamic stuff OpenVAERS claims to be doing and what academia started in ~2002?
For the grand majority that does not know, it was this exact topic that I fell out of favor with Liz Wilner creator of OpenVAERS a couple years ago, trying to get clarification and questioning her “myocarditis” numbers…
https://www.vaersaware.com/post/openvaers-com-the-myocarditis-stats-the-full-story
A young Indian Substacker who is into the Large Language Model (LLM) space and known to both OpenVAERS and myself has some keen insight. I told him about Univ of Kentucky’s project long ago…
Not to belabor the disclaimer, but this CHD IT person wanted me to “code” this narrative to see if his algorithm gets something similar.:
For everyone’s benefit including OpenVAERS, CHD and whoever else, it’s critical to know everything about the Correct Coding Initiative:
https://www.cms.gov/national-correct-coding-initiative-ncci
It’s wise to know about Washington Publishing Company:
I really wish people would stop sniffing around or beating around the bush, and just bring me in…
More will be revealed, but in the meantime Merry Christmas, God Bless and Praise Jesus!
Please accept with no obligation, implied or implicit, my best wishes for an environmentally conscious, socially responsible, low-stress, non-addictive, gender-neutral celebration of the winter solstice holiday, practiced within the most enjoyable traditions of the religious persuasion of your choice, or secular practices of your choice, with respect for the religious/secular persuasion and/or traditions of others, or their choice not to practice religious or secular traditions at all. I also wish you a fiscally successful, personally fulfilling and medically uncomplicated recognition of the onset of the generally accepted calendar year 2011, but not without due respect for the calendars of choice of other cultures whose contributions to society have helped make America great. Not to imply that America is necessarily greater than any other country nor the only America in the Western Hemisphere. Also, this wish is made without regard to the race, creed, color, age, physical ability, religious faith or sexual preference of the wished.
To All My Republican Friends: Have a Very Merry Christmas and a Happy, Healthy and Prosperous New Year!
https://www.vaersaware.com/donate
Need help finding your VAERS report?
https://www.vaersaware.com/findmyreportrequest
Please sign Federal Investigation Petition:
Merry Christmas, Alberto!
I’m guessing you have been in contact with Barry Young?