dhis2-devs team mailing list archive

[Bob Jolliffe <bobjolliffe@xxxxxxxxx>]

Hi Scott

On 17 February 2011 07:53, Merritt, Scott A. <MerrittS@xxxxxxxxxx> wrote:
> Thanks for this Crizelle. In particular, the vision in Botswana is to have
> DHIS serve as the repository and visualization tool of record for health
> data for the foreseeable future. But, as electronic tools roll out, it would
> be nice to eliminate the need for data re-entry from these systems, while
> allowing for manual entry from the programs and locations that are not yet
> electronic. At the same point, as the tools change, we would prefer to have
> a single common interface via Mirth to DHIS as opposed to creating
> interfaces for iHRIS, all of our EMR tools, DisaLab, ETR.net, …. Is anything
> similar happening elsewhere? If not, can someone help explain why not?

My understanding is that Mirth is essentially an HL7 switch which
transports and transforms electronic patient record data.  I guess
this is why all your EMR tools can talk to it and so they should.  But
DHIS (and for that matter iHRIS) is not an EMR tool.  We don't deal
with individual patient data.  And it would probably be a bad idea in
terms of application scope if we did.

There will be a point in all of your EMR tools where they generate
summary reports.  How many malaria cases this month etc.  It is this
summary data which is the province of dhis.  And HL7 is not the format
for representing that summary data.  In the absence of an HL7
equivalent for summary data we have been  strongly advocating that
systems use SDMX-HD for this to avoid the proliferation of interfaces
you refer to.

As I suggested (guessing really) it is probably possible to do the
aggregation at the HL7 level with some sort of HL7 aggregation filter,
but I doubt if this is the easiest way to do it.  When I think of the
logic engine and reporting framweork in a tool like OpenMRS for
example, I can imagine it would take some heavy lifting to get that
functionality into Mirth.  All of the EMR systems should have the
ability to produce summary reports and the challenge is to get them
all to use a comon standard reporting format (ie SDMX).  I think by
analogy you might think of dhis as your SDMX-HD hub in the same way as
mirth is your HL7 hub.

Regards
Bob

>
>
>
> Scott
>
> From: Crizelle Nel [mailto:crizellenel@xxxxxxxxx]
> Sent: Thursday, February 17, 2011 7:07 AM
> To: Knut Staring; Bob Jolliffe; Johan Ivar Sæbø
> Cc: Merritt, Scott A.; Kabelo Bitsang; Phumzile Khumalo; Chris Seebregts;
> dhis2-devs@xxxxxxxxxxxxxxxxxxx
> Subject: Re: [Dhis2-devs] HL7 in Botswana
>
>
>
> Thanks for your responses Bob, Knut and Johan.
>
>
>
> I have not yet had the opportunity to properly discuss this development with
> anyone (I just had a quick telephone call with Scott Merrit from BUTUSA
> earlier in the week).
>
>
>
> What I can tell you though is that the Tecnhical Work Group in Botswana
> (which comprises of key staff from MoH, Local Government, I-Tech, BUTUSA and
> more) is absolutely committed to the DHIS. The long-term plan in Botswana is
> to interface all applicable systems (i.e. IPMS, ETR.Net, iHRIS etc.) with
> the DHIS so that the DHIS acts as an integrator.
>
>
>
> I have a meeting with Scott this afternoon and just wanted to gather as much
> information as possible beforehand.
>
>
>
> I'll be sure to obtain answers to the questions you posed and to get back to
> you with Scott's responses.
>
>
>
> Regards,
>
> Crizelle
>
> On Wed, Feb 16, 2011 at 11:49 PM, Knut Staring <knutst@xxxxxxxxx> wrote:
>
> This was actually the approach taken by a team in South Africa in 2008
> and early 2009 to link OpenMRS and DHIS2, though I don't think it was
> put in production.
>
> Knut
>
> On Wed, Feb 16, 2011 at 10:30 PM, Bob Jolliffe <bobjolliffe@xxxxxxxxx>
> wrote:
>> HL7 is for patient data so we shouldn't need to interface directly
>> with it.  Possibly they are considering building an aggregation filter
>> in Mirth to produce something like SDMX from the patient data.   Don't
>> know.  I am not there to negotiate.
>>
>> But you need to negotiate hard (on your side of the fence) to ensure
>> that what comes out of this magic box will be suitable for dhis2
>> consumption.  That is the way with these kind of switches .. in the
>> end they are proxies for human negotiation :-)  Much like telephone
>> switches between phone operators for that matter.
>>
>> Regards
>> Bob
>>
>> On 16 February 2011 21:22, Crizelle Nel <crizellenel@xxxxxxxxx> wrote:
>>> Hi everyone,
>>>
>>> In Botswana they are discussing developing a software tool (MIRTH) to act
>>> as
>>> a switchboard between all health systems. For example if you want to
>>> export
>>> data from their patient-based system (IPMS) to the DHIS 2 for example,
>>> you
>>> would use the MIRTH interface to do it. It essentially acts as a
>>> post-office, simply passing the data between different systems.
>>>
>>> They would like to use HL7 for MIRTH.
>>>
>>> Have any of you done something similar in other countries?
>>>
>>> Has anyone used HL7 before?
>>> --
>>> Regards,
>>> Crizelle Nel
>>> Health Information Systems Programme (HISP) Software Development Team
>>>
>>> Email:  crizellenel@xxxxxxxxx
>>> Cell:     084 580 3342
>>> Fax:     0866 509 502
>>> Web:    http://www.hisp.org/
>>>
>>>
>>>
>>> _______________________________________________
>>> Mailing list: https://launchpad.net/~dhis2-devs
>>> Post to     : dhis2-devs@xxxxxxxxxxxxxxxxxxx
>>> Unsubscribe : https://launchpad.net/~dhis2-devs
>>> More help   : https://help.launchpad.net/ListHelp
>>>
>>>
>>
>> _______________________________________________
>> Mailing list: https://launchpad.net/~dhis2-devs
>> Post to     : dhis2-devs@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~dhis2-devs
>> More help   : https://help.launchpad.net/ListHelp
>>
>
>
> --
> Cheers,
> Knut Staring
>
>