Andrew,

I'm not looking for applications for a form generator - I'm looking for
modelling efforts (or possibly models implemented as databases). To take a
very simple example:

I have about 4,500 primary health care facilities in my database, and I want
to capture/analyse data about all services provided by these. In the case
initially brought up by Horst, he pointed out that x fields with telephone
numbers are hopelessly inflexible, both because some entities have more than
x numbers AND because you want to know more about each number. Using the
same reasoning for health services, we can then say that setting up y fields
(let us say 20) for specific services (one for immunisation, one for STIs,
one for mental) won't do, because some facilities provide more than 20
services and managers might need to know MORE than just what services are
provided in order to plan new interventions.

So the first obvious step is to normalise the table to enable an unlimited
number of services. The second step is less obvious - what additional data
is required for each service provided? aspects like
- Hours services are provided (might not be provided every day, or during
the whole day)
- Level of service (primary, secondary, diagnosis and referral only)
- Number of trained staff available to provide the service (major
constraints for specialist services - a country like Zambia has ONE, repeat
ONE, cardiologist)
- Referral rules
- Continuity of care
etc.

Otherwise, I'm glad to see you haven't stopped recommending OIO for
everything under the sun.... <evil grin>

Regards
calle


----- Original Message ----- 
From: "Andrew Ho" <[EMAIL PROTECTED]>
To: "Calle Hedberg" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Saturday, May 01, 2004 4:42 PM
Subject: Re: HXP


> On Sat, 1 May 2004, Calle Hedberg wrote:
> ...
> > So flexibility must be high, OR the models must be based on the
> > assumption that it will usually not be directly replicated but rather
> > adapted to new countries/environments.
> >
> > Is anybody aware of any existing modelling efforts that might be
relevant
> > for our work?
>
> Calle,
>   You might want to consider OIO, which supports fully custom "forms" that
> can be easily adapted to new countries/environments.
>
> Best regards,
>
> Andrew
> ---
> Andrew P. Ho, M.D.
> OIO: Open Infrastructure for Outcomes
> www.TxOutcome.Org
>
>
>

Reply via email to