HITM Editorial Team
France’s Hospital 2012 reform plan seeks to double the share of spending on healthcare IT from 1.7 percent - a level at the bottom rungs of the EU league table - to 3 percent, and their modernisation constitutes one of the Plan’s over-arching objectives. The hub of the French healthcare IT modernisation programme is theDossier Médical Personnel (DMP) or ‘personal medical file’, which has sought to create a Electronic Medical Record for all French residents covered by health insurance.
The French government has not been hesitant to highlight the ambitions of the DMP project, as “a natural catalyst for the modernisation of the French healthcare system and the quest for greater efficiency within it.”
The DMP was established by law in August 2004. The first major trials on the DMP were conducted in 2006, followed by a call for projects in 2007.
As with similar initiatives elsewhere in Europe, the Web-based DMP seeks to enable access to patient data from anywhere at any time. It is meant to improve healthcare efficiency and quality by facilitating information exchange and coordination between health professionals as well as medical facilities, during consultations, diagnosis and treatment. According to the French government, the DMP will cut fraud and yield annual savings to the State of between two and three billion Euros.
The DMP will be delivered by via IT service vendors grouped into six regional consortia.
Ó Cegedim, Thalès;
Ó D3P (RSS, Microsoft, Medcost / Doctissimo);
Ó France Télécom, IBM, CapGemini, SNR;
Ó InVita, Accenture, La Poste, neuf cegetel, Intra Call Center, Jet Multimedia, Sun Microsystems;
Ó Santénergie (Siemens, Bull, EDS),and
Ó Santeos (Atos, Unimédecine, HP, Strateos, Cerner).
The DMP programme is administered by a central government agency, the Groupement d'Intérêt Public dossier médical personnel, or GIP-DMP. Oversight is provided by the National Council for Information and Liberty (CNIL), a government body concerned with civil liberties and data protection. Finally, a third agency, the Groupement d'Intérêt pour la modernisation du système d’information hospitalier (GIP MISH) coordinates at a national level the modernisation and adaptation of hospital information and patient information systems to ensure they meet the national DMP interface standards.
Early Concerns: Security, Heterogeneous Technology,
Synchronous/ Standalone Initiatives
The DMP has been the focus of significant controversy in France, with emotive attempts by the mass media to portray it as an invasive tool, most memorably in terms of one which would allow Big Doctor to become Big Brother. There was also a formal appeal to the Council of State to declare it unconstitutional. The Council rejected the claim, but the government was forced into the defensive.
This was followed by a report from French Senator Jean-Jacques Jegou in 2005, which called the DMP “unrealistic”
and described it as an airplane without a flight plan and a cockpit without a pilot.
Since then, the DMP has faced numerous other difficulties, due to the sheer complexity and heterogeneity of the healthcare IT market: core systems unable to share data (some of the most advanced hospitals in France were found to have 50-60 different IT systems), insufficient integration into hospital information systems, disparate products on the market as well as fragmented governance by an assortment of stakeholders, etc.
In 2006, a White Paper from Lesiss (the French association of healthcare IT professionals and industry) found only 10 percent of French healthcare facilities had shared patient dossiers at the hospital level – while 30 percent had partial sharing.
Another problem was to ensure that local and regional initiatives did not duplicate one another, and instead plugged seamlessly into a national healthcare IT/e-health infrastructure, which in turn would be in tune with still-emerging EU standards and regulations.
Such a challenge had cultural facets too: France has a strong tradition of technological elitism and resists being a follower. There was thus a real risk of an undue focus on technology for its own sake, rather than the increasingly user-facing requirements of the growingly cost-sensitive healthcare environment in Europe.
2010: From Prototype to Formal Rollout
Draft specifications for the DMP were formally published in October 2009, with responses from developer-vendors received by December 2009.
In March 2010, a proposal from a consortium led by ATOS Origin and La Poste was mandated to produce the first DMP prototype.
In June 2010, DMP compatibility technical specifications were released to enable vendors to develop software required for interfacing with the DMP.
In August 2010, a development kit containing some sample code and testing tools was released to allow developers to test their software for compliancy. In November 2010, the DMP compatibility procedure was implemented, to allow developers to attest that their software can be integrated with the DMP and provide the required quality of service for users.
The timeline remains tight. The first version of the DMP (DMP 1) is due to be rolled out by the end of 2010.
A Complex IT Project
DMP 1 has four principal sub-systems:
The core DMP IT architecture is the bedrock of the project. Its aim is to create, modify and consult electronic health records. Healthcare professionals and patients will be able to consult and add to the DMP.
For security and data confidentiality, the DMP must be interfaced with health professional card (CPS) systems to allow reliable authentication. If patients give their consent, their DMP will interface with external data sources (e.g their pharmaceutical file and reimbursement history).
The DMP itself has to be accessible through either an Internet browser or in the shape of web services to allow integration into the routine work environment of healthcare professionals.
A dedicated DMP communications portal enables (a relatively vast amount of) information to be made available to users – both patients and healthcare professionals.
Datasheets, graphics/videos and witness testimonies (by both patients and healthcare professionals) will provide answers to common questions (what, why, who, how), as well as an interactive platform for continuously updating such information.
A hotline will be implemented for user support team. This will allow helpdesk staff to answer questions from users, ranging from how-tos, through requests for information, to reports of technical malfunctions and other problems. The information feedback is aimed at giving a real-time overview of system rollout, use and growth.
Management Information System
The final facet of DMP 1 is a management information system. This is aimed at feeding back statistical information from the three sub-systems above, and aggregating the findings to provide a high-level overview of the project.
Included here will be real-time data on the number of active DMPs, users and views, communications and information sharing, as well as the overall load (documents, graphics, videos) in the system. Its aim is enable rapid responses to any faults and proactively anticipate evolution of the system, in order to support it.
Other deliverables will allow for testing user interfaces for accessibility and simplicity, and to ensure that technically, the system is operating as it should.
If all goes according to plan, DMP 1 should be operational by the end of 2010.
Thereafter, the full-scale project is due to be rolled out in stages, in a coordinated manner, with structured inputs from all of parties involved (patients, healthcare professionals, IT managers and institutions, software developments, and last, but not least, the regional and national authorities – including those with legal oversight.