Prometheus: the implementation of clinical coding schemes in French routine general practice

BACKGROUND: Clinical data are most useful, both at the individual level and collectively, if they are coded according to a standard classification system. However, clinicians often have little motivation to routinely code their consultation data. The main classification systems available in French primary care are the International Classification of Primary Care (ICPC) and the Dictionary of Consultation Results (DCR). OBJECTIVE: To assess the feasibility of using the ICPC-2 and the DCR for coding health problems managed in routine general practice in France. METHODS: Between December 2001 and June 2003, 61 volunteer general practitioners (GPs) from the Paris area prospectively recorded the health problems they managed at consultations, using either the ICPC (36 GPs) or the DCR (25 GPs), for a period of six months. They were equipped with one of three proprietary medical software applications specifically adapted for the study, or one open source utility, interfacing with five other, non-adapted, proprietary software programs. They had a two-day training session, were financially compensated, and were provided with electronic feedback. RESULTS: The mean reported coding time per consultation was 2.5 minutes, but 28 physicians (46%) judged the coding time excessive and reported a maximum acceptable time of 1.2 minutes. Coding consultation data was considered more useful at the collective level (by 95% of physicians) than at the individual practice level (by 69%). Only 34 physicians (56%) expressed willingness to carry on routine coding after the end of the study. Some results differed depending on the classification system used, especially due to confounding factors, as some physicians could have previously used the given system. CONCLUSIONS: Coding health problems on a routine basis proved to be feasible. However, this process can be used on a more widespread basis and linked to other management data only if physicians are specially trained and rewarded, and the software incorporates large terminologies mapped with classifications.

Abstract licence: CC BY

19 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.