Background
In France, general practitioners (GPs) perform out-of-hours home visits (OOH-HVs) after physician-led telephone triage at the emergency call centre. The quality of a systematic physician-led triage has not been determined in France and may affect the efficiency of the OOH-HV process. The objectives of this study were first, to evaluate the quality of reporting in the electronic patient’s file after such triage and second, to analyse the factors associated with altered reporting.
Methods
Cross-sectional study in a French urban emergency call centre (district of Paris area) from January to December 2015. For a random selection of 30 days, data were collected from electronic medical files that ended with an OOH-HV decision. Missing key quality criteria (medical interrogation, diagnostic hypothesis or ruled-out severity criteria) were analysed by univariate then multivariate logistic regression, adjusted on patient, temporal and organizational data.
Results
Among 10,284 OOH-HVs performed in 2015, 748 medical files were selected. Reasons for the encounter were digestive tract symptoms (22%), fever (19%), ear nose and throat symptoms, and cardiovascular and respiratory problems (6% each). Medical interrogation was not reported in 2% of files (
n
= 16/748) and a diagnostic hypothesis in 58% (
n
= 432/748); ruled-out severity criteria were not reported in 60% (
n
= 449/748). On multivariate analysis, altered reporting was related to the work overload of triage assistants (number of incoming calls, call duration, telephone occupation rate;
p
< 0.03).
Conclusion
In the electronic files of patients requiring an OOH-HV by a GP in a French urban area, quality in medical reporting appeared to depend on organizational factors only, especially the triage assistants-related work factors. Corrective measures are needed to ensure good quality of triage and care.
Electronic supplementary material
The online version of this article (10.1186/s12913-019-4350-4) contains supplementary material, which is available to authorized users.