an adjudication process whereby payment is determined. After you meet your requirements as a beneficiary by paying your deductible and coinsurance, your insurance company typically pays the remainder of the claim for eligible services (perhaps an amount adjusted to account for negotiated purchasing agreements) according to the terms of your coverage. Your health insurance company is able to pay these bills because it maintains a fund of money reserved for just such purposes. This fund is derived in part from the premium payments made by you and other beneficiaries (and, in the case of for-profit insurance companies, the accumulation of profit). These premium payments are determined statistically by actuaries and take into account the projected risk associated with the level of health of pools of covered beneficiaries.
Medicare operates in many of the same ways as commercial health insurance. When a Medicare beneficiary goes to the doctor for a checkup, the provider submits a bill to Medicare Part B (similarly, if a Medicare beneficiary goes to the hospital, the provider submits a bill to Medicare Part A). Most of this billing is done electronically. The claim form11 contains details like the beneficiary and provider identifiers, dates of service, place of service, the procedures performed, and the patient’s diagnosis. The claim submission will go to the provider’s regional Medicare Administrative Contractor (MAC) for adjudication, processing, and payment. Once claims are paid, they are considered final action claims. Final action Medicare claims are stored in files available to the research community as Medicare administrative data. As you may have already inferred, the administrative files that are created from final action claims and enrollment information are derived from systems that are used to administer the Medicare program. In other words, the primary purpose of these systems is not to create data for research, but to adjudicate and pay claims. This fact has implications for using the administrative data files and means that the user must understand the Medicare program. We will explore this topic in detail throughout the remainder of this book. For now, let’s end with some well-known examples of how particulars of the administration of the Medicare program influence the content of administrative data files.
• With about 47 million Medicare beneficiaries, we can expect the administrative data files we use to be quite large. As such, we will need to consider efficient programming techniques. Many of the exercises in later chapters address efficiency topics.
• Some services that you want to study may not appear in Medicare administrative data or, at the least, may require searching multiple files. For example, Part D prescription drug data contains information for prescription drug fills. Prescription drugs administered during a hospital stay may not appear in the claims data at all. Additionally, services paid for by Medicare Part C may not appear in the administrative claims files because they are paid for by managed care providers.12
• Medicare pays for some services (e.g., home health agencies, hospice, hospital outpatient, skilled nursing, or acute inpatient hospitals) using what are called prospective payment systems (PPS). In very simple terms, a PPS reimburses providers using a fixed amount derived from a predetermined classification system.13 We will discuss payment systems more in Chapter 8.
• As a social insurance program, Medicare coverage is provided regardless of medical history. Therefore, if you are used to working with commercial healthcare claims data, you will likely notice some unique characteristics of the Medicare population, such as a higher prevalence of chronic conditions.
• The administrative data we use for research purposes are updated on a regular basis, but only with claims that have been received and adjudicated and deemed final action. As such, the files we use at any given time do not contain all final action claims submitted and paid up to the date of extraction of the data. It is common practice to wait at least three months for paid claims to appear in the claims files maintained by CMS. For example, a request for claims for the full calendar year 2014 is best made on or after April 1, 2015.
• Depending on what you are studying, care must be taken to determine the correct composition of your study population. For instance, our example research project will study only those beneficiaries continuously enrolled in fee-for-service Medicare during a defined timeframe. Other studies may wish to focus on beneficiaries entitled to Medicare based on being disabled. We will see in Chapter 6 that we can use enrollment data to determine a beneficiary’s reason for entitlement and define our study population.
Chapter Summary
In this chapter, we set a foundation for programming with SAS and Medicare administrative data by examining the following:
• Understanding the Medicare program and the particulars of Medicare coverage is absolutely essential to successfully programming with Medicare administrative data.
• Medicare is a social insurance program that provides beneficiaries with an array of health insurance coverage, regardless of income or medical history.
• The majority of Medicare beneficiaries are eligible for Medicare insurance because they are aged 65 and over. However, Medicare also insures beneficiaries who are permanently disabled (receiving Social Security Disability Insurance or SSDI), have ESRD, or have ALS.
• Medicare benefits are divided and defined in four parts: Part A (Hospital Insurance), Part B (Supplemental Medical Insurance), Part C (Medicare Advantage), and Part D (outpatient prescription drug coverage). Each Part covers a different type of care or set of services. These Parts are a way of describing coverage, but also a way of organizing the way we think about the administrative data files we will use throughout this book.
• Like other health insurance plans, Medicare does not cover every possible medical service or procedure.
• The primary purpose of Medicare payment systems is not to create data for research, but to adjudicate and pay claims. This fact has implications for using the administrative data files and means that the user must understand the Medicare program.
1 See: www.resdac.org, www.cms.gov, www.medicare.gov, and www.kff.org.
2 Medicare eligibility is more complicated than the simple presentation above. For more information, see federal resources such as http://www.medicare.gov/publications/pubs/pdf/11306.pdf.
3 See the Medicare timeline at http://kff.org/medicare/video/the-story-of-medicare-a-timeline/.
4 Information provided in this paragraph, and more, can be found throughout the KFF Medicare Primer (April 2010) (http://www.kff.org/medicare/7615.cfm).
5 Although we will not discuss Medicaid in this text, some beneficiaries are eligible for and enrolled in both Medicare and Medicaid. These beneficiaries are referred to as Medicare-Medicaid Enrollees, or MMEs.
6 Your Medicare Benefits can be found at http://www.medicare.gov/publications/pubs/pdf/10116.pdf.