This monthly blog highlights and discuss emerging trends and challenges related to healthcare data and its ever changing life cycle.
By Barbara A. Ryznar RPh, MSHI, CHDA, CPHIMS, CAPM
Understanding the cost of prescription drugs can be challenging. The recently introduced Centers for Medicare and Medicaid Services (CMS) Medicare Spending Dashboard, an example of the principle of collecting healthcare data once and reusing it to serve the needs of multiple stakeholders, helps users to make sense of this information. The dashboard includes a data table of drugs with pop-up windows that graphically present additional information on them, which helps to bring some perspective to drug spending and utilization. The data, representative of some of the types of healthcare data that is now becoming publicly available, showcases a total of 80 drug items selected as the top contenders using the criteria of spending per user, total spending, or significant cost increases.
The dashboard certainly calls attention to drug spending with some eye-popping dollar amounts, but the value of the data for specific analysis may require a closer look before making any conclusions. A data analyst brings the skills needed to critically analyze the data. For example, a cursory examination by a data analyst of the data might bring up some issues concerning quality and usability.
The source of the data is Medicare Part D and Part B drug claims data, which conform to standards of terminology for drug identification in a standardized format for electronic claims transmission and are bound by compliance regulations to the program. These constraints ensure consistency and accuracy in data representation. The scope of the data, though, is limited to the list of 80 drugs covered by theses programs, meeting the dashboard selection criteria, and those used in this population. It does not include spending using cash or claims coverage by other commercial insurance. By limiting the scope to the top 40 in Medicare D (retail) and top 40 Medicare B (generally administered through outpatient clinics and physician offices), this represents a rather small sample of drugs utilized. Of course, these drugs are the ones recognized as having the highest impact regarding the dashboard selection criteria.
The dashboard was created using data from 2014, which poses a problem with timeliness. In fact, of the top five in the Medicare D category for high total spending, three have since become available in generic form and do not have the same impact on spending today. The increased cost criteria for selection also represent a volatile aspect.
The data show granularity across the various parameters. The cost is defined to the individual dose, making it highly granular, but interpretation is needed to bring this into scale. For example, the cost listed for Advair Discus is as $4.94 per dose, but this is an inhaler, used twice a day, and comes packaged with 60 doses for a month’s supply. So, the actual spending per patient on a monthly basis would be $296.40. The interpretation becomes more digestible with this per month amount. With another example, the drug Lantus is represented twice, once as the vial and also the pen form, probably separated due to a slightly different cost per unit. If combined, the totals for Lantus would have been at the top of the list for total spending. The average cost for the patient’s share across the board also may not be representative of the actual cost to the beneficiary due without factoring in additional methods of payment or whether the patient has fallen into the Medicare D coverage gap known as “the donut hole.”
The dashboard is worth a look to see how the data has been presented and to gain awareness of the spending and trending of Medicare drug costs. It is a start in understanding drug utilization and cost, but combining these results with other data is needed to justify spending with outcomes to provide the total picture. To point to areas for cost containment, some comparison data with lower cost therapeutic equivalent drugs would give a dashboard for decision making.
This example is just one of many complex challenges that face data analysts as they look to define, display, and evaluate healthcare data. By applying critical thinking to data, a data analyst’s skills can be critical to making sense of complex healthcare data and help turn healthcare data into useful information for decision making.
Centers for Medicare and Medicaid Services Office of Enterprise Analytics and Data. Medicare Drug Spending Dashboard 2014. December 2015. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Dashboard/Medicare-Drug-Spending/Drug_Spending_Dashboard.html.
Barbara A. Ryznar is a community pharmacist at Medicine Shoppe and a part-time course facilitator in the MSHI program at University of Cincinnati.