Implementing a low-cost computer-based patient record: a controlled vocabulary reduces data base design complexity.

In order to build a computer-based patient record (CPR) system suitable for use in solo and small group practice settings it is necessary to use development methods that minimize cost. Design complexity is a major source of high cost. Reducing complexity should result in lower development, deployment and maintenance costs as well as higher reliability. We have developed a simplified relational model and have used that model, in conjunction with a controlled vocabulary, to implement a CPR that can capture and store patient examinations and other forms of clinical notes as well as laboratory and other test results. The information can be viewed in a familiar document format and it can accessed for other types of processing using standard Structured Query Language (SQL) techniques. The database, as implemented, uses inexpensive components resulting in a system that is not prohibitively expensive for solo practitioners and small groups. In addition the architecture is scaleable and can accommodate very large numbers of patients and practitioners.