Inventory of health databases

As part of its missions, the Etalab department has carried out the most comprehensive census of existing public databases and datasets in the field of health, and publishes this inventory in open data. 172 databases of 79 managers were identified. To facilitate data discovery, a selection of the main databases available in open format on data.gouv.fr [is also available here](https://www.data.gouv.fr/fr/pages/donnees-sante). The list is not exhaustive and is open to contributions. ### Methodology The work of carrying out the inventory was done manager after manager, working primarily on senior managers, and those for whom an interview was possible. Two different methodologies were followed according to the managers: — Managers with maintenance: research work in the room prior to the interview, in order to list all the bases concerned, as well as all the missing information and questions to ask. The interview then validates the relevance of each identified database, and obtains the required information. — Interview-free managers (these managers are those who produce a small number of datasets): In-room research work only. Interviews were conducted with the following managers: — Technical Agency for Hospital Information (ATIH); — Public assistance — Paris Hospitals (AP-HP); — National Agency for the Safety of Medicines and Health Products (ANSM). ## Limits An initial limit lies in understanding the origin of the data for each game, as well as the entity or person(s) who owns the data. This information was generally obtained through interviews conducted. Databases managed by managers that have not been contacted therefore have a higher probability of error for the “data origin” and “data owner” part. Another limitation is the understanding of the notion of statistical secrecy attached to surveys covered by Law No. 51-711 of 7 June 1951 on the obligation, coordination and confidentiality of statistics. The investigations concerned by this law are investigations with personal and identifying information. As a result, none of the relevant investigations can be published as it stands. On the other hand, these surveys generally serve as the basis for the construction of synthetic indicators and time series which are publishable and published. This distinction concerns several surveys present in the search perimeter. ## Resources This dataset contains four resources: — A list of the **managers** of the identified bases; — A ** inventory** of databases; — An inventory of **publications** and studies carried out from the databases identified; — A list of **thematics** grouped by category. ### Metadata details We detail below the few notable metadata — **thematic**: We have identified four possible themes, in order to carry out a first work of categorisation within the inventory itself. — **public health data and epidemiology**:data on lifestyles, health inequalities, epidemiology, etc. — **data on offers of care**:information on infrastructure, services offered, staff, fees, etc. — **care consumption and expenditure data:** data on health facility activities, consultations, medicines and devices, etc. — **performance and operations data:** information on financial performance, operational performance, etc. — **typology**: the inventory consists of three types of data: — Data from administrative documents — Data from surveys — Data with a directory — **presence on data.gouv.fr**: in this field, we notified whether the bases were present or not on data.gouv.fr. We also notified a partial presence when, for example, some vintages are missing on data.gouv.fr. — **data origin**: in this field, we try to explain in a summary way the process that led to the production of the data, and to mention the actors involved.