Data Mining

The CTSC houses a number of databases which are available to HSC investigators. Each of these provides access to different kinds of patient data that can be mined for research.

Some of our databases contain de-identified data and can be explored by the investigator or members of the research team on a self-serve basis. These databases can help identify potential cohorts to determine study feasibility.

Access to identified patient data requires IRB approval, and usually requires coordination with a CTSC honest broker.

Requests for data or access to CTSC databases can be made through the CTSC Data Request Form.

Clinical Data Warehouse

HSC researchers may request data from the UNMH Electronic Health Record. Approved requests are assigned to a CTSC honest broker who extracts the data and releases it in accordance with all applicable approvals and regulations. All requests must be approved by the CTSC Data Request Review Committee and be associated with a protocol submitted to the HSC Human Research Protections Office.

Health Facts

HealthFacts contains de-identified electronic health record (EHR) data from over 600 participating Cerner client hospitals and clinics in the United States, representing over 106 million unique patients. With this longitudinal, relational database, UNM researchers can analyze detailed sets of de-identified clinical data at the patient level. Available data include demographics, encounters, diagnoses, procedures, lab results, medication orders, medication administration, vital signs, microbiology, surgical cases, other clinical observations, and health systems attributes. Currently available to all UNM HSC faculty.

i2b2

i2b2 (Informatics Integrating Biology & the Bedside) provides a self-service means of exploring de- identified data from UNM’s electronic medical record and from billing diagnoses. Its easy-to-use web interface offers researchers the ability to explore aggregate, de-identified data in order to determine study feasibility. Currently available to all UNM HSC faculty. Training for i2b2 is available in Learning Central (Course CLT 144). After successful completion of the course, users will have an i2b2 account created and they can access the i2b2 system.

Truven MarketScan

Truven MarketScan is the most recent database to be installed on CTSC servers. It contains complete information on inpatient prescriptions, as well as links to outpatient prescription data. MarketScan data also contain diagnoses which can be used to determine the influence of clinical, demographic, and provider characteristics upon prescribing patterns. Individual patients’ prescription fills are recorded so concurrently prescribed drugs can also be identified. MarketScan is expected to be available to all UNM HSC faculty in January, 2018.