Structured data entry in ORCA: the strengths of two models combined.

The capture of patient data in a structured format receives increasing attention. Data can be extracted from free text using natural language processing techniques, but it can also be collected in a structured fashion at the time of data entry. The latter has the advantage that completeness and unambiguity can be promoted by offering predefined terms and options for description of findings. The paper discusses two models for supporting structured data entry. In the direct model, there is an immediate relationship between the terms and options for data entry and the structure of the underlying database. In the indirect model, terms and options for data entry are based on a controlled vocabulary and not directly related to the structure in which actual data is represented. Both models have been utilized by ORCA (Open Record for CAre). We discuss the pros and cons of these two models in relation to the type of patient data and the task involved. It is concluded that a strategic combination of both models has more strengths and less weaknesses than the use of each model only.