Collection and retrieval of structured clinical data from electronic patient records in general practice. A first-phase study to create a health care database for research and quality assessment.

OBJECTIVE: To evaluate prerequisites, practicalities, attitudes and limitations related to the collection of structured clinical data in everyday general practice for use in the future establishment of a national registration network.
DESIGN: Prospective study.
SETTING: Primary health care centres in south-western Sweden.
SUBJECTS: Fourteen participating general practitioners in five primary health care centres. MAIN OUTCOME MEASURES: Feasibility and workload involved in structured data entry and in the retrieval of data from different record systems. The accuracy of clinical data in terms of clinical variables, correctness and representativeness.
RESULTS: All four record systems could deliver basic data on the patient population. One centre had to be excluded from further data retrieval because of limitations in the data retrieval export format. Collecting data in everyday practice was feasible with acceptable data accuracy and moderate workload.
CONCLUSION: It was feasible to collect, retrieve and store structured clinical data with respect to accuracy and extra workload. Interest in a national registration network and an increasing demand for information about primary health care in order to optimise clinical practices and support research, creates prerequisites for establishing a valid and reliable database. However, developmental work focusing on classification limitations, coding tools and routines for data retrieval is necessary.