Administrative Data Systems

5. Accessing Data

Administrative medical data is necessarily “identifiable,” particularly if it contains identifying information about the individual beneficiary or insurance subscriber.  As such, privacy considerations are paramount and extensive protections are built into any request by researchers to use “identifiable” data.

While there have always been protections in the use of identifiable data, under the privacy rules promulgated pursuant to HIPAA (the Health Insurance Portability and Accountability Act, passed in 1996), these were more systematized.

Companies and their contractual agents engaged in electronic data interchange (EDI) must protect the confidentiality of their customers; any use of identifiable data must be specifically consented to by that individual unless it is approved for research purposes by a specially constituted Privacy Board. Following is a summary of the means by which researchers can safely access administrative/clinical data.

Data Use Agreements.  Whether working with Medicare, Medicaid, or individual medical care insurance companies, researchers requesting to work with individually-identifying information must complete and conform with standardized data use agreements.  Such agreements have to stipulate the precise data elements required and how these are needed to achieve the goal of the research, and describe, in detail, data security arrangements that restrict whom and how identifiers are accessed and used.

“The User agrees to establish appropriate administrative, technical, and physical safeguards to protect the confidentiality of the data and to prevent unauthorized use or access to it. The safeguards shall provide a level and scope of security that is not less than the level and scope of security established by the Office of Management and Budget (OMB) in OMB Circular No. A-130, Appendix III--Security of Federal Automated Information Systems, which sets forth guidelines for security plans for automated information systems in Federal agencies.”

Working within the framework of a data use agreement, researchers can link medical care utilization events for a given patient to demographic and pre-existing clinical characteristics longitudinally as well as hierarchically.

Reporting the results of analyses of event- and person-level files based upon identifiable data are restricted by a requirement that no fewer than 10 individuals be included in any table cell published, in order to minimize the possibility of individuals’ identities being uncovered.

The Centers for Medicare and Medicaid Services (CMS) has an elaborate procedure for processing data use requests, relying upon a subcontractor, the Research Data Assistance Center (ResDAC) at the University of Minnesota, to review, modify, and clarify data use requests before they are submitted to CMS.