tion and the patient’s existing medications. To do this requires data on all the patient’s medications and the ingredients of each.

A list of validation modules that could be incorporated in a computerized physician order entry system is given in Table 6-3, together with the generic data requirements. A full set of validation modules requires a wide range of data elements: medications (including ingredients, dose levels, and administration routes), allergies (including drug allergies), diagnoses, patient age, weight, laboratory results, and contrast media used in radiology.

The various approaches to adverse event detection discussed above demonstrate that it is not possible to simply identify a small set of clinical data elements specifically for adverse event detection, especially when addressing potential injuries due to errors of omission as well as injuries due to errors of commission. On the contrary, a broad range of data elements encompassing demographic information, signs and symptoms, medications, test results, diagnoses, therapies, and outcomes are required to: (1) detect adverse events through voluntary and mandatory reporting, chart review, and automated surveillance; (2) implement performance measures (e.g., DQIP measures);