ICD10 and Patient Identifiers

classic Classic list List threaded Threaded
3 messages Options
John Bowers John Bowers
Reply | Threaded
Open this post in threaded view
|

ICD10 and Patient Identifiers

Two questions came up in meeting with hospital administration last week:

How are ICD10 codes most effectively integrated with OpenMRS (or ICD9,  
as we already have acquired the ICD9 database)?

How are you currently dealing with patient identifiers? Our problem is  
probably pretty similar to most OpenMRS installations, no national id  
number, unknown birthdates, and very similar names (since one name is  
the given name and the other typically has to do with the person's  
order in the family among the siblings). Another problem with names is  
that a patient may not know how to spell his/her name and two  
different nurses may spell it two different ways. The same nurse may  
even spell it two different ways given different days. The tribe  
attribute is a great idea also, but the majority of our patients come  
from the same tribe.

Thanks for any insight,


John Bowers
E [hidden email]
T +256 77 930 4659

Volunteer Technology Coordinator
Kagando Hospital
Kagando Village, Uganda

Owner
espressoSoft LLC
www.espressosoft.com
Tallahassee, FL, USA

_________________________________________

To unsubscribe from OpenMRS Implementers' mailing list, send an e-mail to [hidden email] with "SIGNOFF openmrs-implement-l" in the  body (not the subject) of your e-mail.

[mailto:[hidden email]?body=SIGNOFF%20openmrs-implement-l]
Kit Sumabat-2 Kit Sumabat-2
Reply | Threaded
Open this post in threaded view
|

Re: ICD10 and Patient Identifiers

I guess the best and easiest way around this is to use biometrics such as a fingerprint scanner. If this isn't possible due to resource restrictions then I guess you will have to find a way in introducing policy solutions. Sometimes technology alone will not solve things. A wise combination of policies, technology and management strategies will go a long way.

Kit Sumabat, RN
Health Informatics Specialist





i'm EMAILING FOR THE GREATER GOOD
Join me


> Date: Mon, 1 Dec 2008 15:19:37 +0300
> From: [hidden email]
> Subject: [OPENMRS-IMPLEMENTERS] ICD10 and Patient Identifiers
> To: [hidden email]
>
> Two questions came up in meeting with hospital administration last week:
>
> How are ICD10 codes most effectively integrated with OpenMRS (or ICD9,
> as we already have acquired the ICD9 database)?
>
> How are you currently dealing with patient identifiers? Our problem is
> probably pretty similar to most OpenMRS installations, no national id
> number, unknown birthdates, and very similar names (since one name is
> the given name and the other typically has to do with the person's
> order in the family among the siblings). Another problem with names is
> that a patient may not know how to spell his/her name and two
> different nurses may spell it two different ways. The same nurse may
> even spell it two different ways given different days. The tribe
> attribute is a great idea also, but the majority of our patients come
> from the same tribe.
>
> Thanks for any insight,
>
>
> John Bowers
> E [hidden email]
> T +256 77 930 4659
>
> Volunteer Technology Coordinator
> Kagando Hospital
> Kagando Village, Uganda
>
> Owner
> espressoSoft LLC
> www.espressosoft.com
> Tallahassee, FL, USA
>
> _________________________________________
>
> To unsubscribe from OpenMRS Implementers' mailing list, send an e-mail to [hidden email] with "SIGNOFF openmrs-implement-l" in the body (not the subject) of your e-mail.
>
> [mailto:[hidden email]?body=SIGNOFF%20openmrs-implement-l]

[hidden email] from OpenMRS Implementers' mailing list
Burke Mamlin Burke Mamlin
Reply | Threaded
Open this post in threaded view
|

Re: ICD10 and Patient Identifiers

John,

ICD codes would be integrated through the concept_map table.  The concept_map tables are underutilized in OpenMRS currently; however, there are efforts to add functionality (e.g., Andy has proposed some good data model changes and these mappings will be critical for the OpenMRS Concept Co-operative that's under construction).

We are getting by with names, age/birthdate, gender, address/location for now in Kenya, but definitely share the same issues with uniquely identifying patients.  There are a few options/tips in this space:
  • Person attribute(s) -- e.g., village chief's name, mother's given name, or first pet's name (anything that might be reliably queried over time without changing)
  • In Kenya, we are transitioning to universal identifiers as we expand beyond HIV care (into primary care).  This is basically a home-grown alternative to having a national id card.
  • Biometrics (as already suggested) -- e.g., acquired into a person attribute through a module
  • Shaun & Win are building a nice patient matching module that will open the door for systematically identifying duplicate patient records in your database. :-)

Cheers,

-Burke

On Dec 1, 2008, at 11:52 AM, kit sumabat wrote:

I guess the best and easiest way around this is to use biometrics such as a fingerprint scanner. If this isn't possible due to resource restrictions then I guess you will have to find a way in introducing policy solutions. Sometimes technology alone will not solve things. A wise combination of policies, technology and management strategies will go a long way.

Kit Sumabat, RN
Health Informatics Specialist

> Date: Mon, 1 Dec 2008 15:19:37 +0300
> From: [hidden email]
> Subject: [OPENMRS-IMPLEMENTERS] ICD10 and Patient Identifiers
> To: [hidden email]
> 
> Two questions came up in meeting with hospital administration last week:
> 
> How are ICD10 codes most effectively integrated with OpenMRS (or ICD9, 
> as we already have acquired the ICD9 database)?
> 
> How are you currently dealing with patient identifiers? Our problem is 
> probably pretty similar to most OpenMRS installations, no national id 
> number, unknown birthdates, and very similar names (since one name is 
> the given name and the other typically has to do with the person's 
> order in the family among the siblings). Another problem with names is 
> that a patient may not know how to spell his/her name and two 
> different nurses may spell it two different ways. The same nurse may 
> even spell it two different ways given different days. The tribe 
> attribute is a great idea also, but the majority of our patients come 
> from the same tribe.
> 
> Thanks for any insight,
> 
> 
> John Bowers
> E [hidden email]
> T +256 77 930 4659
> 
> Volunteer Technology Coordinator
> Kagando Hospital
> Kagando Village, Uganda
> 
> Owner
> espressoSoft LLC
> www.espressosoft.com
> Tallahassee, FL, USA
> 
> _________________________________________
> 
> To unsubscribe from OpenMRS Implementers' mailing list, send an e-mail to [hidden email] with "SIGNOFF openmrs-implement-l" in the body (not the subject) of your e-mail.
> 
> [[hidden email]]

[hidden email] from OpenMRS Implementers' mailing list


[hidden email] from OpenMRS Implementers' mailing list