The following essay was first printed in the June issue of the Houston Psychiatric Society Newsletter and contains perspectives on ways to unify the software architecture for mental health in Houston, Texas using free and open source software.

The Next Level of Efficient Practice 5/5/02

Ignacio Valdes, MD, MS

The Next Level of Efficient
Practice: We Can Do It

An interesting
aspect of being a Psychiatry resident is that you get the “grand tour” of
mental health delivery organizations in a given area. From this experience
one notices trends among them. One facet is the increasingly
wide range of styles, and complementary clinical
computing software available.

One
observation is the number of major clinical software programs used in different locations in Houston.
These number at least five. Each from a different vendor. Another observation
is that there are a number of patients who have been treated by one of these
clinical venues, sometimes more than once in the same day. The inherent duplication of work is manifest.

A laudable
goal for practitioners and patients would be to actively seek a unified computer
software architecture for all care entities in the city. Achievement of
this could result in a large-scale reduction in cost as well as improvement
in the quality of mental health care in Harris County through ease of coordination
and the resulting reductions in redundant work. It will also result in reductions in medication errors. Furthermore, a standard
software interface for clinicians would make everyone from technicians to
medical students to residents to attendings far more effective upon the arrival to any organization they happen to move to. It would also reduce errors caused by inexperience and avoid the chore of re-learning a new interface
at each site. This is not an exotic technology or concept. It has been done successfully for years at the Veterans Administration hospitals nationwide.

While
stating such a goal of software unity is easy, achieving such a goal is not. One of the biggest concerns is software security and confidentiality. Another is that several mental health organizations have already made substantial
investments in software infrastructure and are locked in to the vendors they have chosen. The lock-in is due to the high cost of purchasing software they already have and the high cost of converting to another system. Another is the far more mundane factor, and perhaps the most difficult is standardization, or approximation to a standard, of common forms (such as admission histories and mental status results) across organizations that have never collaborated so closely before. Formidable obstacles indeed.

However,
when one considers the amount of redundant work that is performed each day in Harris County by practically every worker on each patient, the potential savings and improvement in care are also prodigious.

Houston
mental health is unique in that patient data can be safeguarded more easily among its practitioners through the use of a private network, which is physically separate from the Internet. Therefore, it is much easier to secure from intrusion. Moreover, “the pandora’s” box of patient data security has already been opened through the use of computer systems that are already in place. We are not benefiting as much as we could simply because of incompatibilities.

There
is also a way around the expense of clinical software in the form of software that does not have proprietary licenses attached to them. This kind of software is covered under a license, which guarantees the freedom for the software to be used, studied, and extended without restrictions by a parent organization. This removal of restrictions has resulted in clinical software such as the Psychiatry oriented SQLclinic (http://www.sqlclinic.org)
program, which is currently in use in New York. It is freely available for download on the Internet in a fully functional form. There are other clinical software packages which have licenses that guarantee freedoms such as OSCAR
and FreePM. OSCAR and FreePM are similar to SQLclinic, but they are more oriented toward general medicine. This usage is quite distinct from proprietary clinical softwares, which are restrictively licensed so that use, study, or extension to fit needs without consent or a fee to the vendor is prohibited. Additionally, compatibility and interface issues either can be avoided using these types of software.

Among
all of the mental health organizations in Houston, there are more than enough information technology resources to use a system such as SQLclinic and to modify it to suit local conditions and have a unified software architecture for the city. The key ingredients are recognition of patient care and economic reasons for doing so and cooperation through the use of non-proprietary software.
Perhaps the most important factor is the political will and far-sightedness to undertake this. Houston could become a model of software collaboration for the nation if it has the desire to undertake this. Dare we begin?

If
you are interested in discussing the development of a common software architecture
for mental health in Houston, contact the author at ivaldes@hal-pc.org