Skip to content

An electronic health record on your USB? CHITS-on-a-stick

In Asia Source 3, Bobby S demonstrated FrontlineSMS and described how it may be possible to hack Huawei USB modem and make it into an Ubuntu liveUSB. This is mind-boggling for several reasons:

– it means there will come a time when all you need to do is stick your USB in any machine and you’ll be able to run your own personal health record AND transmit your data to a centralized backup system

– it means you can run an automated SMS service from any PC

– it means you can run CHITS on any machine, on any operating system, and submit FHSIS data to the National Epidemiology Center (or any other agency you wish) in real-time.

Next year will be a great year for health informatics. Because there will be a new president…

Mike has more info on how to do this here:

http://www.vdomck.org/2009/11/ubuntu-web-appliance.html

Asia Source 3 Peppered with Health Apps (and health profs!)

The largest source camp in Asia was held at Silang, Cavite last November 7 to 12, 2009. Amazingly, this camp has quite a number of health professionals in attendance. Out of the one hundred and fifty participants, five were doctors and three were nurses.

Topics on OpenMRS, CHITS, and other eHealth-relevant applications (like FrontlineSMS) were discussed.

It might actually be a good idea to hold another source camp just for health…watch out..

Book on Telehealth Out (and it’s free!)

International Development Research Centre of Canada has just released a book on telehealth in developing countries. Download it now from http://www.idrc.ca/en/ev-136734-201-1-DO_TOPIC.html

(Full disclosure: I wrote chapter 3 🙂

What happens when regulators become implementers?

A lot of people have asked me why there are no standards for health information in the Philippines*. This is a tricky question and one that does not have a right or wrong answer. But allow me to re-phrase it as follows: what happens when regulators (those who impose standards) are also implementers of those standards?

Let’s look at the positive side. We can almost be sure the implementations will be according to the regulations/standards they have formulated. There will be no question about compliance. But this sunny option presumes the standards were in place before they were implemented.

On the downside, the problem occurs when the (regulator’s) implementation comes _before_ the standards are defined. More often than not, the implementation is not compliant (as it preceded the standard). The implementer-regulator then finds itself in a tight fix. The implementer-regulator has three possible options:

1) Keep the status quo. Don’t publish anything. This way the implementation is off radar, and can continue to exist. It’s not standard, but who cares? There is no standard anyway. (Anticipate health informatics chaos here to the same degree America finds itself now.)

2) Make the implementation the standard. This option is too strong, and it won’t hold scrutiny in public. The standards definition process has to be participatory and coming out with the standard (through the implementation) at the outset diminishes its potency. This option is too dictatorial it only serves to highlight the imperfection of the implementation rather than the benefits of having standards.

3) Make pronouncements about the ideal standard (based on evidence), and bite the bullet that the existing implementation may need to be overhauled or even decommissioned. It is also entirely possible to forego the business of implementation and shift to testing and certification (it might actually be more lucrative for the regulator).

Whatever the regulator chooses, there is a huge lesson for all of us here: regulators should not be implementers. Regulators should define the standards and provide tests/certification for implementations. Once regulators become implementers, the check and balance is lost.

And the country will not have the standards it direly needs. In the year 2009.

* There was an attempt in 1999 to publish standards for health information in the Philippines. The UP Manila-NIH facilitated that effort together with the DOH, PHIC, PHA, PMA, and other major professional organizations. In retrospect, a standard has to be implemented to become part of civil society. A theoretical standard (like this 1999 document) is useless without reference implementations.

Does the Philippines need a eHealth Council?

In the last Med-e-tel conference in Luxembourg, it became apparent that a national eHealth masterplan was crucial if any country were to progress definitively into the integration of ICT in health. In most countries, an eHealth council leads this process.

Surprisingly, the Philippines was mentioned as a model for the crafting a national council. Other countries took the guidelines for the Philippine National Council on HIV/AIDS and re-purposed it for eHealth.

When will our country have its eHealth masterplan?

The PHIN, the PHIS

Not many know that there is a thing called the Philippine Health Information Network (PHIN) which is spearheading the Philippine Health Information System. The effort is led by DOH with support from the National Statistics Office, the National Statistical Coordinating Board (NSCB), PhilHealth, and other government agencies. I will post more about the PHIN as we go along and about the PHIS as more details come in.

EHR for Hacking

The International Open Source Network ASEAN+3 together with DabaweGNU has started the FFEHR project — a Firefox extension project that functions as an electronic health record. The FFEHR project aims to instruct new open source developers on the use of the Cruxade and Firefox 3.x environment for application development. It also aims to engage medical students in the creation of the software.

The project page is here.

Protege: Ontologies for relationships

Modeling the real-world has always been a challenge to any health informatics professional. The fact is that the way humans perceive their world is vastly beyond the capabilities of technology to document. In essence, the best way to model the world is to live in it and be part of it.

But for lesser mortals, who are given the responsibility to use modeling tools to create software, there is hope. The Protégé Ontology Editor and Knowledge Acquisition System
at Stanford University aims to assist modelers organize and manage entities and their relationships. The relationships are conceptual (as opposed to the programmatic relationships in databases).

So how do we use Protege? There is the left-to-right approach and (of course!) the right-to-left. The right-to-left means there is an existing form already being used and the data elements from that form are extracted to form the concepts (facets) in the ontology. The left-to-right requires the modeler to think of concepts, to place them on the ontology and then later, on the form. L-R is more difficult but it could be a more comprehensive and coherent way of constructing the ontology.

If you’re working on some ontologies, let us know!

Open Source for Health

Much has been said about open source and there are better sites that give more information (www.fsf.org, www.opensource.org). But let me write about free and/or open source software (FOSS) for health.

I am biased towards enforcing FOSS fundamentals for health applications especially those which are mission-critical. The reason is that a full audit of the software should be possible *before* the application is used in a production environment and especially when a mishap occurs causing injury.

Our experience at NThC reveals that there is a need for an auditing system for software to be applied in clinical environments. Tests should be created to guide developers, and a monitoring system to document application performance should be made available to the regulators.

What do you mean? The role of data dictionaries in eHealth

What is an admission? You ask an ER physician and they will say it is when the patient with a stab wound enters the emergency room. Ask the trauma surgeon and they think it’s when they write “Admit to Trauma” on the chart (even if the patient is still physically at the ER). And when you ask the trauma ward nurse, they’ll say it’s when the patient is actually on the bed in the ward.

To add to the confusion, a neurosurgical patient (in PGH at least) who is placed in a borrowed bed in Ophthalmology will be counted by the NSS as an admission, but also by the Ophthalmology ward nurses.

What is happening here?

This is the result of the lack of a data dictionary. Essentially, a formalized method for defining concepts in a given milieu.

The Australian Institute for Health and Welfare has their METeOR, the Canadians have this.

The Philippines has started work on it. The lead agency is the Department of Health.

How about your organization? Do you have a data dictionary?