Data Products »
HRS Medicare Claims and Summary Data
1. Overview
One of the main goals of the Health and Retirement Study is to understand the relationship between medical history and financial status, and how use of health care may change as people age. To assist in this effort, the HRS endeavors to obtain information about health care costs and diagnoses from Medicare records maintained by the Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration). The HRS asks all respondents who are eligible for Medicare to provide their identification numbers; over 80% of them consent to do so.
In order to use the Medicare records linked to HRS respondents, there are two linked approval processes that must be completed. As a first step, users should obtain approval from HRS for the use of restricted data (see below). As a second step, users should obtain a Data Use Agreement (DUA) from the Centers for Medicare & Medicaid Services (CMS), working through the Research Data Assistance Center (ResDAC). Formally, the request for a DUA from CMS must come from the program officer of the funding agency that supports your research. Requests for HRS-linked CMS data are processed and provided by the MedRIC division of Acumen LLC. Inquiries regarding data file specifics should be sent to Acumen/MedRIC.
2. Background Information for Prospective Applicants
- Respondent Permissions: Procedures used by HRS to obtain permission from respondents for researcher access to CMS records.
- General Information
- MedRIC Documentation for HRS Data requestors: Provides a MedRIC overview and descriptions of the summary and claims fiies
- Getting and Using Medicare Data: What I Wish I Had Known Before I Started My Research (A training and professional resources presentation by Academy Health)
- File Structure Information
- Claim-Segment RIF Record Structure for SAF Files: Provides detailed information on the structure of the files.
- HRS-Medicare Data Dictionaries
- A visual overview of the application process:
3. Parallel Application Procedures...
3a. Negotiating a CMS Data Use Agreement
- User applies to HRS for restricted data approval (see below).
- While application is under review by HRS, user completes a data request packet (New Use Request) for CMS. A complete description of the required documentation can be found at the ResDAC Requesting CMS's Identifiable Data Files (RIFs) - New Use Requests page: (http://www.resdac.org/cms-data/request/research-identifiable-files)
- User receives PRELIMINARY approval from HRS.
- User finalizes CMS data request packet, incorporating data plan approved by HRS, and sends all paperwork, including HRS approval, to ResDAC.
- ResDAC verifies, completes, and returns the DUA to the researcher to obtain funding agency signoff
- Funding agency signs off on the request and forwards to CMS.
- CMS approves data request and notifies user, MedRIC/Acumen, and HRS.
- MedRIC processes request and distributes data to user. See the document MedRIC Fee Structure and Data Distribution Process for a description of the required licensing fees.
- HRS provides user with ID cross-reference file (encrypted) that links respondents to the record set created by Acumen.
3b. Negotiating an HRS Restricted Data Agreement
Eligibility rules and the procedures to be followed in negotiating a Restricted Data Agreement are described in detail on the Restricted Data section of the HRS Web site. In summary, a successful restricted data agreement requires:
- Affiliation with an institution with an DHHS-certified Human Subjects Review Process, known as a Federal Wide Assurance
- Current Receipt of Federal Research Funds
- Restricted CMS Data Products Order Form
- Written Research Proposal
- Restricted Data Protection Plan
- Human Subjects Review (either Full Board or Expedited)
- Confidentiality Agreement Restricting Disclosure and Use of Restricted Data from The Health and Retirement Study
4. Questions
Questions concerning HRS Restricted Datasets should be addressed as follows:
|
Procedural questions: |
Technical questions: |

