For program analysis purposes, Mathematica Policy Research (MPR) has developed a set of integrated data on participants in the Medicaid Buy-In programs. The integrated data set includes data on all Medicaid Buy-In participants identified by state MIG staff. Development of the data set has been an iterative process, with new enrollees identified and added on an annual basis. Data are currently available for MBI participants as early as 1997, although there may be quality or completeness concerns with some early data.
The analytical data set includes the following broad types of information:
- Medicaid enrollment, claims, and expenditures from the Medicaid Statistical Information System (MSIS) Files and Medicaid Analytic eXtract (MAX) Files
- Medicare enrollment and claims from the Medicare Enrollment Database (EDB) and Medicare Claims Data
- Social Security program participation, including type of disability for recipients or beneficiaries from the Ticket Research File (TRF)
- Information on annualized earnings, but not raw earnings data, as reported from the Master Earnings File (MEF)
Due to confidentiality and data access restrictions, it is currently not possible to allow MIG state staff direct access to data from the integrated dataset. However, to facilitate state research, MIG-RATS and a committee of state researchers have developed a proposal process that will allow states to request specific analyses to be conducted by MPR staff. The result will be aggregate information on state programs (or multi-state projects) that will be provided to the requesting MIG researchers. The
proposal process, along with supporting materials, is posted in the MIG-RATS Workspace. The schedule for proposal submissions for Fall 2010 is as follows:
October 29 – Deadline for submitting questions for Q&A document
November 5 – Written response to submitted questions distributed to all states
November 19 – Proposals due to MPR
December 10 (approximate) – States notified of selection outcomes and tentative schedule for analysis
Below you will find information about research questions addressed by analysis requests in prior submission periods:
Who participates in the Medically Improved Group (MedIG) across states?
What are participants' patterns of enrollment in the MedIG?
What are patterns of health care use and earnings among MedIG participants?
Lead Contact: North Carolina (Kathleen Thomas)
The Medically Improved Group (MedIG) option of the Medicaid buy-in program is a critical tool that states can use to disentangle work and loss of benefits for people with disabilities. This coverage is particularly important for people who rely on Medicaid for benefits typically poorly covered under private insurance policies, such as personal care and psychotropic medications. Despite its importance, few states have included a MedIG in their buy-in laws, and among those that have, participation has been low. We anticipate an analysis population of just under 100 individuals, all Med IG participants to date.
Although the population is still small, this proposal is designed to take advantage of what may be our last chance to get a broad view of MedIG program participants due to the approaching sunset of the Ticket legislation. Since the population is small and clustered among a few states, the proposed analyses are limited to issues that can be interpreted and informative across states. Data covering 1999-2009 compiled from Mathematica's Integrated Data File augmented with the Social Security Ticket Research File (TRF) and Rural Urban Continuum Codes will be used to describe participants in 2009 and their history of health care use and earnings.
Fall 2009
How do Medicaid Buy-In (MBI) enrollees use work incentives offered through the Social Security Administration?
How do Minnesota MBI participants who use Social Security Administration work incentives compare to those who do not use work incentives?
How do these MBI sub-populations compare in the areas of length of enrollment, earnings, acquired work credits, and health care utilization?
Lead Contact: Minnesota (Stacy Myhre)
As of December 2008, over 98% Minnesota’s MBI (Medical Assistance for Employed Persons with Disabilities, MA-EPD) enrollees were also receiving SSDI through the Social Security Administration. As enrollees in SSDI and some who were dually enrolled in SSI, these MA-EPD enrollees are also eligible to use other work incentives offered through SSA. Findings will compare those MA-EPD enrollees who have never used SSA work incentives to those MA-EPD enrollees who have used either: (1) a Ticket or (2) BWE, IRWE, or PASS incentives. Specific analysis requested includes comparisons of the two MBI sub-populations in the areas of length of enrollment, earnings, acquired work credits, and health care utilization. This analysis would contribute to an initiative currently funded by the Minnesota MIG that is working with the Minnesota WorkForce Centers (Minnesota’s OneStops) to pilot the use of these Centers as Employment Networks. This analysis would also contribute to existing outreach efforts to Employment Networks and general MA-EPD outreach strategies. A summary of this project, including findings, is available here.
Spring 2009
What are the effects of “agency effort” on employment outcomes achieved by Medicaid Buy In participants?
Lead Contact: New Hampshire (Tobey Partch-Davies)
We request that Mathematica expand the data elements within the New Hampshire Medicaid Buy In finder file to include elements describing the types of services, types of vendors, and types of categories of assistance utilized by New Hampshire’s Medicaid Buy In participants. This requires data from New Hampshire’s eligibility system (New Heights) and claims system ( MSIS). This analysis will contribute to a larger project of our state MIG.
How similar are the available demographic characteristics for participants in the WI SSDI cash benefit offset pilot (officially “The Wisconsin SSDI Employment Pilot”) who were Buy-in participants to those of other working age Buy-in participants who were either SSDI beneficiaries or putatively eligible to participate in the Pilot?
How similar are the patterns of employment rates and mean earnings observed for participants in the WI SSDI cash benefit offset pilot who were Buy-in participants to those of working age Buy-in participants who were either in SSDI or putatively eligible to participate in the Pilot?
How similar are the patterns of employment rates and mean earnings observed for key subgroups (e.g. by age categories, gender, etc.) of participants in the WI SSDI cash benefit offset pilot who were Buy-in participants to those for key subgroups of working age Buy-in participants who were in SSDI or putatively eligible to participate in the Pilot?
Lead Contact: Wisconsin (Barry S. Delin)
The Wisconsin MIG has the capacity to link information about Medicaid eligibility to employment and earnings data from Wisconsin Unemployment Insurance (UI) records. Wisconsin does not have complete information about who is a SSDI beneficiary, information that is available from the MPR integrated data base and that is particularly relevant to the proposed analysis because of the focus on SSDI beneficiaries.
Wisconsin also was a site of one of the four SSDI cash benefit offset pilots. All pilot participants were SSDI beneficiaries. Approximately half of those in the pilot were also enrolled in the Buy-in. However, there is little information about the degree of similarity between the pilot participants in the Buy-in and the SSDI components of the Buy-in population.
Our goals are to use information from the Integrated Data Set data to (1) compare selected demographic, disability, program, and experiential characteristics of offset pilot participants who used the Buy-in to those of other Buy-in participants who might have been eligible for the pilot or might be eligible for a future statutory offset, (2) isolate the employment outcome patterns of Buy-in participants who are SSDI from those of the larger working age population of Buy-in participants and (3) obtain a better estimate of the proportion of SSDI beneficiaries who are in the Wisconsin Buy-in.
Fall 2008
What are the dynamics of Buy-In participation and earnings relative to SSDI disability onset, SSDI approval and Medicare enrollment? Does Buy-In precede SSDI or vice versa? How do earnings differ with respect to Buy-In and SSDI participation?
Lead contact: Massachusetts (Marsha Ellison)
We propose to use the Integrated Data Set to examine the dynamics of Buy-In enrollment and earnings relative to SSDI participation. Individuals ever enrolled in the Massachusetts Buy-In (available in the dataset) will be grouped according to the timing of their enrollment relative to SSDI key events (disability onset, SSDI approval, enrollment in Medicare (sample size and percentages for the study sample in each group will be requested). Also, average length of time (number of days) elapsed for the groups during these time periods are also requested. Lastly earnings associated for the groups in each time period are requested. Findings will explore the relationship of SSDI participation to Buy-In enrollment and earnings for questions such as: do individuals use the Buy-In to acquire health insurance while they are in the Medicare waiting period, or do participants delay enrollment in the Buy-In until their SSDI coverage is secure?
What is the demographic profile of Medicaid Buy-In participants, and is it changing over time? What is the Medicaid enrollment history of MBI participants? What connections do MBI participants have with SSI, SSDI and Medicare? How much are MBI participants’ Medicaid and Medicare expenditures? How have the expenditure levels and types of expenditures changed pre- and post-MBI enrollment? How much are MBI enrollees earning? How have their earnings changed over time? How have their earnings changed pre- and post-MBI enrollment?
Lead contact: Arizona (Dara Johnson)
This proposal seeks a basic descriptive analysis of Medicaid Buy-In participants in approximately 20 states. These states have joined together with Arizona to pursue an analysis that would in large part replicate for each state the research conducted by Mathematica for the May 2006 Integrated Data Report. The specific analyses requested include demographic characteristics of MBI participants, enrollment patterns, participation in SSI/DI and Medicare, earnings history pre- and post-enrollment, and Medicaid expenditures pre- and post-enrollment. Annual data is requested, provided table cell counts are within proposal guidelines, so states can observe changes in enrollment, earnings and expenditures as their programs mature.
While meeting important information needs, this multi-state proposal is also a learning process for the participating states as they become more familiar with the integrated data access proposal process.
Note: Due to the nature and volume of the Arizona request, MPR responded by coordinating material from recently completed analyses and augmenting analyses in process to provide the request information.
For more information about the integrated dataset or analysis proposal process, please contact MIG-RATS.
Please submit proposals using the link below. If you have difficulty with the submission link, you may e-mail your proposal directly to areither@gmail.com.
Contact MIG-RATS and Submit Proposals