Empirical norms as a lever for on-line support of general practice

Frequencies, probabilities and sensitivities among symptoms, diagnoses and treatments can be inferred from large databases of health records. We investigate how these "empirical norms" can be leveraged to improve the efficiency and quality assurance capability of online systems in support of general practice medicine. In particular, we assess hotlists (case sensitive menus) of diagnoses to speed data entry. We also explore norm violation as an indicator of poor quality in practice or data recording. Using a survey based database of general practice records, we find: (a) useful hotlists of diagnoses can be generated based on conditional probability and sensitivity with respect to symptoms, and (b) records of hypertension treatments of the most frequent varieties are assessed to be of higher quality than less frequent varieties. The results support the hypothesis that empirical norms have a role to play in future online clinical information systems.