HUS Data Lake holds more than 100 patient registries, each registry containing dozens of data tables and hundreds of different variables. The data consists of structured and unstructured data that is imported from the electronic patient health records from all HUS Hospitals, with more than 1 million referrals and hospital visits per year. The earliest data is from the 1980’s but more comprehensive data is available from 2010 onwards as most HUS units started to use electronic patient health records then. The data is updated daily and new integrations from new information systems are continuously built.
For neuroscience research purposes the patient cohort can be defined, for example, by diagnoses or visits to a particular ward over a specific time period. Then the data can be queried from the HUS Data Lake for those patients. Typical data queries include patients’ diagnoses, procedures, medication, laboratory tests and hospital visits and the variables can be tailored to meet the research needs. Imaging data will also be available in spring 2021. In 2021 software specialist Paula Tanni from HUS data service will provide neuroscience research data queries as a part of the Helsinki Brain & Mind team.
The data is pseudonymized and all information that could lead to individual patient identification has been removed. Different databases of the HUS Data Lake can be connected with the pseudo-ID, making it possible to gather various information about the patient cohort. Before the data set is handed over to the researcher, the pseudo-IDs are re-pseudonymized, making it impossible to combine more data than is covered by the research permit.
The HUS data service provides data according to the Finnish law of secondary use of health and social data. Transferring the data sets to researchers requires a scientific research permit recognized by HUS. HUS data service also offers analytical workspace for powerful tools for data analyzing purposes.
The portal for data requests or data availability assessment can be found on the HUS website (in Finnish). If the research requires combining data with data outside of HUS, this is possible through Findata - Findata, Health and Social Data Permit Authority. Read more on the Findata-website.