Imagine that a health care practitioner is about to begin an encounter with a patient. The practitioner accesses a list of current complaints along with the patient’s electronic health record. Much of that record was acquired and encoded by intelligent interviewing software prior to the patient’s visit.

The physician desires a cost-effective strategy to pursue diagnostic possibilities. Therefore, based on the patient’s unique set of problems, the physician selects medications from an array of treatment choices generated automatically by a therapy advisor component of the software. All of these tools currently exist; I have been particularly impressed by the rapid adoption of personal digital assistants (PDAs) and other tools throughout the ranks of our younger practitioners.
cheap antibiotics

As seductive as this futuristic-sounding scenario appears, imagine the details involved in creating the facility to store medication data in the software. For example, clinical checks need to be performed for drug interactions, contraindications, lab interferences, and allergies. The dose specified by the doctor has to be checked by the system for appropriateness based on renal function, weight, and other patient parameters. All the knowledge sources required to support these functions must be kept up to date.

Carol Broverman and her colleagues have contributed a great deal to our understanding of these issues, whereas I have only begun to appreciate the complexity of medical software. I am sure many of you would agree with the statement that “many clinical information tasks can benefit from the use of a standard terminology for representing drug information including electronic medical records, automated decision support, quality assurance, health care research, reimbursement, and mandatory reporting.” Yet the Food and Drug Administration’s National Drug Code, or NDC, is a system created by individual drug suppliers rather than a central authority, which has very few rules regarding editorial management. I was shocked to learn that the NDC lacks many of the widely recognized desirable characteristics for so-called “controlled terminologies.” Unfortunately, these terminologies are not interchangeable in their current forms. Dr. Chip Masarie, Jr., at Medscape’s headquarters in Hillsboro, Oregon, educated me about these issues.

What does all this mean? As we progress to an era in which we all carry web-enabled PDAs, and we write prescriptions using these tools, we must now reform the NDC and make it possible for firms like Medscape and others to begin the difficult process of standardizing our approach to the vocabulary embedded in these future technologies. I have had the privilege of working with medical scientists at Medscape and elsewhere as they have labored to begin such a process.
buy generic levitra

From these scientists, I have learned about the challenge of “graceful degrada-tion,” which means providing the best decision support possible within the context of available patient data. If every proprietary system and pharmacy benefit management company has its own language, how can we possibly communicate about pharmacy products in the future? Remember that each new drug entered requires adding information about its intended use, therapeutic classifications, interactions, contraindications, side effects, and age- and indication-specific dosing ranges to any intelligent web-enabled system. The development and maintenance of all required areas of expertise is a very daunting task for any individual vendor. I would argue that a centralized organization with appropriate authority should take on these kinds of challenges.

Such a centralized approach would have several characteristics. In part, it would have not only “graceful degradation” but also scalable precision; that is, the ability to apply increasingly expert knowledge components to a clinical context based on its richness.

The vocabulary that the centralized authority would develop must come to grips with the concepts of granularity, synonymy and ambiguity. The issues of synonymy and ambiguity must be considered as they apply to the names of drugs at various levels of description. For example, synonymy is desirable because different character strings—connections of numbers or letters—can have the same meaning, and the same character string could have different meanings or ambiguity. These issues can be handled by using unique concept identifiers for key elements of the information model proposed, such as drug descriptions at each level, dose forms, and routes of administration. cialis canadian pharmacy

What does all this jargon mean for the non-technical readers of P&T? We might envision a seamless web-enabled PDA-based pharmacy practice in the future, but the language and the specificity of that language is currently locked in the dark ages and is incompatible with this scenario. I hope that with global consolidation in the pharmaceutical industry, the few remaining manufacturers will use their clout and market savvy to streamline the NDC and provide us with intelligent language for the future. As usual, I am interested in your views.