Public Health Databases

Epidemiological, health services and health policy research relies on access to health, environmental, demographic and other information. There are many sources, some are national in scope and others are limited to New Jersey.

 The Center for State Health Policy provides access to state level data for New Jersey from state and national sources. For a list of available database.

 The Center’s AHRQ-funded Center for Education and Research on Mental Health Therapeutics (CERTs) works with numerous data sets, both public, and those acquired through Data Use Agreements (DUAs) with the Centers for Medicare and Medicaid Services (CMS) and similar agreements with other data providers, including the Medicaid programs of several states.

Data sources include:

• CMS identifiable data include 2001-2010 Medicaid Analytic Extract (MAX) data for 45 states
• Medicare claims for all dual eligible Medicaid beneficiaries in 8 states (2001-2004), for dual eligible antipsychotic users in 45 states (2001-2005), and for dual eligible beneficiaries in 45 states who have been diagnosed with schizophrenia
• National Minimum Data Set (MDS) for 2001-2014 with linked Medicare Part A, B, and D files for 2007-2013
• National Death Index (NDI) data, linked to the MAX files, for the years 2001-2007 for all Medicaid enrollees, including dual-eligibles.
• More recent Medicaid data for a subset of states: CA, ME, MO, OH, OK, TX, and WA.
• Medicare Current Beneficiary Survey (MCBS) data for 1992-2013 and 1996-2014 Thomson Medstat Commercial Claims and Encounters data for privately insured persons in the U.S.
• CASPER (formerly OSCAR) data on nursing facilities
• The DataBank MedKnowledge file (formerly National Drug Data File) that efficiently utilizes and code drug information.

The Center has a considerable track record in obtaining, managing, and analyzing large datasets and confidential data, including the linkage of disparate data: claims and enrollment files, clinical data, mortality records, drug information, and contextual variables. Researchers interested in using these databases, should contact Margaret Koller.