Health IT

Usman, I don't think this is a very good idea at all if you are looking for unique IDs. Obviously, if patients come back they might end up with a second number. I don't know what the serial number portion of the schema might be (except for sequential numbers) but generally this method is not good and cannot interoperate with other institutions or clinics, etc.. if they happen to have an Antenatal clinic, or perhaps an HIV clinic. In general good ID rules are that they should be random, and not encode information in the code itself. We are using an alphanumeric BASE-30 code which eliminates any letters which look like numbers, etc. It is short enough to also be used by cell phone entry personnel.

There is a good WHO document on something like the 8 best practices for ID generation. I have tried to find it online again but have not been able to. I hope someone on this list can provide a link since it is really a good place to start.

Andy

Thanks a lot Andrew. Actually they are using the FIRST visit of a patient to create MRN. The Serial number is basically a sequential number that starts from "1" every day. So the issue of subsequent visits is apparently not a problem (although I agree that this is not the best way to do it). The system is also generating a unique visit number for billing purposes.

This system has been in place for some time before this issue was raised recently. However, things are bit more complicated now because there are two other hospitals affiliated with this one, and sometimes patients may go to the other hospitals for various reasons. In that case, it is obvious that duplicates will be generated (which is currently happening). The hospitals are planning to integrate their systems in future, and are discussing how to deal with this issue. One suggestion was to continue using the same scheme and add a prefix denoting the hospital where the MRN was generated, thus eliminating duplicates.

I tried searching for the WHO document but haven't found it yet. I would appreciate any solid references or documents that I can use to convince these folks that this ...

Thanks a lot Andrew. Actually they are using the FIRST visit of a patient to create MRN. The Serial number is basically a sequential number that starts from "1" every day. So the issue of subsequent visits is apparently not a problem (although I agree that this is not the best way to do it). The system is also generating a unique visit number for billing purposes.

This system has been in place for some time before this issue was raised recently. However, things are bit more complicated now because there are two other hospitals affiliated with this one, and sometimes patients may go to the other hospitals for various reasons. In that case, it is obvious that duplicates will be generated (which is currently happening). The hospitals are planning to integrate their systems in future, and are discussing how to deal with this issue. One suggestion was to continue using the same scheme and add a prefix denoting the hospital where the MRN was generated, thus eliminating duplicates.

I tried searching for the WHO document but haven't found it yet. I would appreciate any solid references or documents that I can use to convince these folks that this is indeed an international standard.

Thanks again,
Usman


Usman Raza
http://www.linkedin.com/in/uraza

Hi Usman,
The standard that I've heard of is the Universal Health ID (HUID) by ASTM. The webpage where it can be found is http://www.astm.org/Standards/E1714.htm, but there's a cost in downloading the standard.

Here's a description of the standard:
The UHID Scheme consists of a sequential identifier, a delimiter, check digits and an encryption scheme to support data security. This Standards Guide covers a set of requirements outlining the properties of a national system of Universal Health Identifier (limited to the population of United States). It includes positive identification of patients, automated linkage of various computer-based records, mechanism to support data security of privileged clinical information and the use of technology to keep health care operating cost at a minimum.


Also, there's a group being led by Aamir Kahn, from John's Hopkins University and IRD in Pakistan, that is implementing electronic health records in 44 hospitals in Pakistan, starting with Indus Hospital. I'd be more than happy to connect you with his group if you you'd like.

Warm regards,

Joaquin
___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http ...

Hi Joaquin,

Thanks for sharing this information. This should be helpful. I have heard a lot about Indus Hospital and Aamir Khan but haven't had a chance to get connected. It would be nice if you can do that for me.

Thanks again,
Usman


Usman Raza
http://www.linkedin.com/in/uraza

Hi Usman, I work with Dr. Aamir at IRD although not specifically with the Indus Hospital HMIS team. Are you popping through Karachi any time, we could arrange a tour.

Best,

Sarah Bird

Thank you Sarah. I will keep this in mind if I happen to visit Karachi.

Also I have another related question for the people on this forum. We are hoping to implement ICD-10 coding in this system but the long list of 32,000 codes is not of much use in our setup at present. For this I am thinking of creating a shorter list of 500 or so diagnoses with help of clinicians, but want to maintain their ICD code alongside so that we can add later easily. I am wondering if someone here in Pakistan has already developed such a list, it would be great help.

Several hospitals in Peshawar have hired various IT companies to create their own systems, and I'm afraid each of them are using their own custom lists. Is there anyone working on these kinds of issues at policy level?

Regards,
Usman

Usman Raza
http://www.linkedin.com/in/uraza

I'd recommend starting a new thread to discuss ICD-10 issues, though... this is a big discussion.

I would NOT put the codes in as concepts directly into your dictionary. It means that you then have to change your "instance data" in the record whenever codes change, which they do frequently. The OpenMRS model is to keep the concepts separately and provide mappings to the other code sets. For example, we have ICD-10 and SNOMED included in our concept dictionary. It is large, though, as it in includes most everything. We are working on a subsetting arrangement. It would be better for you to work from this dictionary and pull out the codes you want. Then you get automatic updates when the codes change and you don't have to change your internal codes and forms, etc...

Hi Usman,
I asked a colleague of mine, Carol Hullin, from Australia and she pointed me to this document from the Australian eHealth Authority about how they are implementing Patient Identifiers.
http://www.ghdonline.org/tech/resource/australian-health-informatics-services...

Thanks Joaquin, the document does outline the standards (for IHI) in the way
we were discussing.

@ Andrew: I think you're right. I'll separate the ICD coding discussion.
Some of what you said is not clear to me, and would appreciate if you can
point to a resource that would help me understand this.

Best,
Usman

Usman Raza
http://www.linkedin.com/in/uraza

Hi all,

I just wanted to clarify something that Joaquin mentioned earlier in this thread "a group being led by Aamir Kahn, from John's Hopkins University and IRD in Pakistan, that is implementing electronic health records in 44 hospitals in Pakistan, starting with Indus Hospital."

Indus Hospital is a separate institution (www.indushospital.org.pk) who have a collaboration with IRD (which Dr. Aamir Khan is the head of) through the Indus Hospital Research Center.

Indus Hospital built their own HMIS that runs off oracle. They spent a lot of time developing over 85 modules that now run this completely paperless facility.

IRD (Interactive Research and Development) works closely with Interactive Group of Companies. Interactive are also working with Indus Hospital. Together Interactive Group and Indus Hospital are taking their system and re-working it into a network based system to deploy it over 44 Government veterans hospitals across Pakistan. This is a work in progress, but I believe the first 5 modules are now up and running.

Indus' HMIS has been operating at full tilt since April 2007 when the hospital opened.

I know it's a little convoluted but I just wanted to try and make it clear ...