Re: [Gnumed-devel] clinicians: coding use case survey - please respond !

[Jim Busser]

On 2011-05-04, at 1:54 PM, Rogerio Luz Coelho wrote:

> Give me an example of a work flow and I will try to explain my view.

- user starts GNUmed
- user activates Kirk
- user sees episode "Diabetes" with Diabetes ICD code
- user goes "WTF ?!? This episode is about hypertension !"
- user edits episode name to "Hypertension"

At this time do you want to end up with the now-Hypertension
episode still having the Diabetes ICD code ?

BIG YES ... although this is a good example 
- how about the user that goes to Kirk and says ... WTF I left that episode as DM because I was suspecting it at the time ... renames it again and we lose Hypertension.

If a user had manually entered "Diabtes", they might have associated ICD9 250 (Diabetes mellitus)

When a user later sees this patient

- they might want to fix the misspelling 'Diabetes"

- it might be less work to delete the text "Diabtes" and accept an auto-fill (from the ICD9) of "Diabetes mellitus"

- they may desire to add precision without altering the ICD9 e.g. "Diabetes mellitus II insulin-requiring"

instead of going to the trouble of

250.00

V58.672

However, whether or not code 250 was ever originally assigned, if it turns out that the episode was actually about Diabetes insipidus and the episode name or health issue was incompletely inputted, why on earth would you want to keep 250 when you can know in a widget or screen

Diabtes [ICD9 250 "Diabetes mellitus"]

totally misrepresents this episode or health issue. How can you accept to have the description [association]

Diabetes insipidus [ICD9 250 "Diabetes mellitus"]

and not fix it to

Diabetes insipidus [ICD9 253.5 "Diabetes insipidus"]

This is why I believe that when a user alters the description name, they cannot make a proper decision whether to keep or alter one or more codes unless they can see, in real-time, (without having to hover, side-by-side)

Description [CodeSystem CodeValue OfficialReferenceTerm]

or at minimum

Description [CodeSys1 CodeVal1; CodeSys2 CodeVal2; …]

with the official reference terms in a tooltip

example for an episode of diabetes with hypertension (I will explain lower down)

Diabetes, Hypertension [ICD9 250; ICD9 401]

tooltip:

ICD9 250 "Diabetes mellitus";

ICD9 401 "Essential hypertension"

Reasons why a user might accept / desire something to have this might be:

1) in a busy praxis where they do not care to go to the trouble of making granular the problems, and the problem list, and they are happy to lump and to store more than one problem in a SOAP note instead of threaded notelets, then the above design allows it

2) in a busy praxis where it is *normally* desired to maintain a true, problem-oriented medical record, maybe at a visit for diabetes mellitus it is noticed that the blood pressure is high. maybe it has remained the same as it has always been, but the diagnosis of diabetes now newly "makes" the patient "hypertensive" because the treatment goal is now lower. However, the visit is busy, and the doctor is running late, and they lack the time to move the physical exam finding of blood pressure from the SOAPlet "Diabetes mellitus" into a new SOAPlet. Presto! When saving this visit note, and when able to access the episode name, they just change it from

Diabetes [ICD9 250]

to

Diabetes, Hypertension [ICD9 250]

This is where the fact that Hypertension is being added need not require to "lose" the ICD9 250. The user may not care to add the [ICD9 401], or they may like to add it because they know they get paid more if in their rpaxis statistics they can show how many of their patients have more than one chronic disease.

PS yes, it is possible to put the word "hypertension: into the RFE. However, the unmanaged problem of hypertension is less apt to get missed if it would be in the episode name than if it is just in the RFE, and RFE does not accept codes if the user wished to add the code.

-- Jim