The new SEER Medicare data linkage has arrived. This is the third SEER Medicare linkage that PRC has received. The database links cancer cases diagnosed through 2007 from 16 US cancer registries to patient Medicare claims through 2009. For the first time, Medicare Part D drug claims are available (2007-2009). Currently, PRC has data for the following cancers: hepatocellular, colon and prostate. We expect to receive breast cancer data early 2012. Learn more about the SEER Medicare database.
We are pleased to highlight our newest PRC member, Corinne Woods. Corinne is a pharmacist who recently completed her Masters of Public Health in epidemiology. Along with her epidemiology background, she has a wealth of clinical experience working in many different types of pharmacy practice sites and she has more recently had some experience working with SAS software. Given her experience, she is well-equipped to assist investigators in understanding how drugs are used across multiple settings. Currently, Corinne is working with the SEER Medicare data and the Chronic Renal Insufficiency Cohort (CRIC). Please visit our Staff and Project pages to learn more about Corinne and these projects.
IMS Health provides the IMS LifeLinkTM 10% sample database to the Department of Pharmaceutical Health Services Research (PHSR). PRC will manage the data, which includes enrollment, medical and pharmacy claims for nearly 6 million enrollees. The data is representative of the national, commercially insured population and includes plans from employer-sponsored, government sponsored but commercially administered Medicare and Medicaid programs, and individuals purchasing insurance in the marketplace. Over 69% of enrollees have at least 18 months of continuous enrollment. On April 1, 2011, PRC staff, PHSR faculty and PHSR graduate students attended an all-day seminar by IMS staff. IMS staff provided data element details and insights into the utility of the database. Currently, the data is available for PHSR faculty and graduate students to use for pilot data for grant submissions, or to conduct studies. Anyone interested in using the IMS data should contact PRC through this website.
In the first quarter of 2011, PRC received the 2006-2008 Minimum Data Set (MDS) for use with the 2006-2008 Chronic Condition Warehouse 5% Medicare enrollment and claims files, already in house. The MDS files consist of comprehensive assessments of residents in nursing facilities. The assessments contain information not available through administrative claims, such as the resident’s customary routines, cognitive function, mood and behavior patterns, psychosocial well-being and physical functioning. Since most residents are assessed at regular intervals, as mandated by the Centers for Medicare and Medicaid Services, researchers can use this data to examine residents’ status over time in the nursing facility. As of June 2011, PRC has worked with two research teams using the MDS data.
PRC gains expertise with the recently released Center for Medicare and Medicaid Services (CMS) Chronic Condition Warehouse (CCW) data. In March 2010, PRC was one of the first research centers in the nation to receive the 5% CCW sample datasets including Part D Prescription Drug events, Part A and B claims, Chronic Condition Summary file and Denominator files for 2006 and 2007. This data set is one of the largest to date to be stored on PRC‘s high performance server.
Before the data arrived, PRC staff worked with investigators to secure the data use agreement from CMS. Once the data arrived, PRC ran preliminary analyses to assess the consistency and validity of reported values. PRC developed data cleaning algorithms and made recommendations for defining variables. Additionally, PRC programmers developed SAS code to create variables to accommodate common requests – outcomes, medication possession ratio (MPR), co-morbidity, health plan coverage, cost, drug and health care utilization measures that with slight modifications can be re-used to maximize efficiency when creating additional analytic files.To date PRC has constructed analytic files for cohorts of Medicare beneficiaries with COPD, depression, diabetes and congestive heart failure. UMB School of Pharmacy researchers are now conducting multiple research studies determining: (1) medication adherence and its impact on health outcomes and (2) impact of prescription drug plan design on health outcomes and medication adherence for Medicare beneficiaries with these chronic disease
PRC staff grows with the addition of Dr. Joel Culpepper and Mr. Peichang Shi. Dr. Culpepper has a MA in Experimental Psychology and PhD in Epidemiology. Joel joined the faculty of PHSR in June 2010. Within PRC, Dr. Culpepper will provide statistical consultations and assist in the development of PRC’s policy and procedures.
Mr. Shi joined PRC in August of 2010. He has master degrees in Applied Statistics, Biology and Computer Information Systems. Mr. Shi is an advanced SAS programmer with over 5 years of statistical experience including nonparametric methods, survival analysis, cluster analysis, stochastic models, tree-structured methods, neural network, decision trees, and other regression methods. In addition, he has experience working with data from a large health plan network, clinical trials, as well as, primary data from community health studies. With the addition of Dr. Culpepper and Mr. Shi, we believe PRC will be able to meet our client’s needs across a broader spectrum of skills including data acquisition and cleaning, database and analytic file development, pharmcotherapeutic consultation, statistical analysis, and interpretation.
PRC has upgraded its server to provide users with improved performance capabilities in the areas of processing speed, increased memory and increased data storage. This upgrade will accommodate current programming needs that use larger, more complex data sets that require faster processing and larger storage capacity. After a “test” period with a limited number of users, the move to the new server was successfully completed in June 2010.
The new server is a high performance Sun Microsystems Sun Fire x4540 with two six-core 2.6GHz processors server running Solaris 10 with 32 GB of memory. This new system provides PRC with high data access speeds in a compact package giving users twice the capacity in half the space. With a total capacity of 24 terabytes (TB) of usable disk space, PRC can store very large research datasets while the two six-core processors enable users to run SAS programs more quickly. This state-of-the-art Sun Fire x4540 system will handle the expanding needs of our current and future users.
After the short time that the server has been in wide usage we have received a number of comments from our users, we wish to share a couple:
“Impressive speed on this more muscular server!! Cool.”
“That thud you just heard…was the sound of my jaw hitting the ground with getting a better sense of the speed of the new machine…..”
We are happy to receive all feedback. Feel free to share your experience with the new server. (email@example.com)
Copyright © 2013 University of Maryland School of Pharmacy