[Rapport] AUTOMATE Project stage’s report First year of the project

, by Myriam Le Goff-Pronost

The AUTOMATE project aims to propose a collection of indicators, evaluation methods on the uses made of the information systems developped within health informatics. The modification of the cooperation practises among professionals has been focused on through the partnership with the Uni-Medecine company and follows up the implementation of an Electronic Patient Record (EPR) within two particular networks: the first one takes care of addicted people (ADDICA network, Champagne Ardennes) and the second one monitors diabetic people (REDIAB network, Boulogne-sur-Mer).

The Uni-Medecine company aims to extract data from the EPR logs and implements a data processing device. The other members of the project follow a research-action approach to study these statistical data and meet the EPR users in order to understand how the health informatics globally function, what the place of the EPR is, how it is used, what image the EPR users make of this tool as far as its role in the coordination is concerned. The multidisciplinary evaluation method suggested mixes the statistical analysis of the logs (analysis of the real uses of the EPR) and a socio-economic evaluation (analysis of the uses observed, of the EPR economic model and of the levers promoting the uses in terms of financing and remuneration).

Today, both of the networks have been contacted and the sociological study has been made for the ADDICA network. The first stage of the sociological study of the ADDICA network includes 13 semi-structured interviews. All of them last about one hour and were carried out with two general pratictioners, a smoking cessation specialist, a nurse, a dietician, two project managers, two administrative coordinators, three social workers. Two sessions of observation also took place at a training and coordinating meeting and an administrative council. The history of the network, the way it functions (integration of the members, organization of the trainings, data sharing,...) and its size could have been written with the first analysis of the materials collected. The more thematic analysis of the interviews questions the place of the Electronic Patient Record within the network and the modifications in the professional practises related to networked work. A second wave of interviews will be done in this network, once the statistical analysis has permitted to identify typical routes from the information system. A common evaluation protocol has been created between the participants of the AUTOMATE project, the ORS Champagne-Ardennes, the ENDEL company and the ADDICA network to coordinate the actions and the plannings so that the different parts of the network evaluation can be coherent. As the REDIAB network has just recently been contacted, a such thorough evaluation of this network can’t have been made yet. As for the statistical analysis, it can’t have been launched yet as far as the CNIL has not given yet the authorization to analyze the secured network data (the 4-month-period has already come to terms). Modules of data analysis (factorial methods and classification models producing growing classes) have been prepared by the statisticians from a base of test files (Uni-Médecine Virtual ERP). The data which will be available are the patient characteristics (ages, sexes, origins, cities of residence), the list of the possible events (checkup, examination), the characteristics of the medical assistant (CPS, geographical data), the number of enclosures, the key-boarding time. The aim is to identify new indicators and to integrate them into a sharper analysis which will include how many times the files, the enclosures, the patient records, their backgrounds have been consulted.

An informational watch is being carried out on the electronic patient record and the development of health informatics.

The same evaluation methodology has been applied on both of the networks. The final goal is to compare the results which have been obtained in terms of use and concerning the role of the EPR for two different diseases: addicion and diabetes. This would make us able to give recommendations as far as the use of information systems within health informatics is concerned.