Washington Report: "D" Is for Data

In the rush to write the rules for the new Medicare prescription drug benefit, one aspect of the program has been overlooked: the vast amount of data it will create. Medicare expects to be covering more than 40 percent of all prescriptions in the United States, noted Mark McClellan, administrator of the Centers for Medicare and Medicaid Services (CMS), in a recent speech. As a result, McClellan believes it is critical for Medicare to address important questions on drug safety and utilization, and about how prescribing decisions affect health outcomes and costs.

Jill Wechsler

In pursuit of that goal, he plans to link reimbursement data from Medicare prescription drug plans (PDPs) with existing Medicare hospitalization and physician-care databases. The result will be an unprecedented collection of data on the prescription drug use of millions of elderly patients. Both as a tool and a challenge for pharma, Medicare's new knowledge base promises to have a major impact on the industry.

Links to Outcomes

In order to be reimbursed under the prescription drug program, PDPs have to submit electronically to CMS extensive information (36 data elements) for each prescription filled under Part D, including the particular drug form, dosage, quantity and refill schedule.

The FDA Connection

"This will be the largest scale implementation ever of such electronic data on prescription drugs, by far," McClellan predicted, explaining that it will have an even greater impact through links to existing Medicare databases on hospital and physician services. The program, he said, will "provide insights about the association of drug use with overall outcomes and costs of care." This will address a number of specific purposes:

Identifying drug adverse events Medicare already provides FDA with information related to safety issues that emerge for those drugs covered by Medicare Part B, such as data on what kinds of patients use a certain treatment and possible long-term complications. Integrated outpatient drug use data will help identify drug safety issues more broadly, which may involve rare but potentially important adverse events. This could enhance FDA post-market surveillance activities by reducing reliance on spontaneous reporting of adverse events by healthcare providers. McClellan anticipates that FDA will be able to determine much more quickly if certain drugs are associated with important complications in particular kinds of patients. Drug manufacturers and prescribers ideally will use such information to target drugs to patients who don't have such side effects, and may truly benefit.

Augmenting post-market studies By monitoring patients for longer periods—and on a much larger scale than occurs in pre-approval clinical trials—Medicare can help pharma companies gain information more quickly and cheaply on many issues now addressed through costly Phase IV studies.

Information, Please

Assessing off-label risks and benefits Medicare drug information can provide evidence for marketers seeking to expand indications. Utilization and safety data may better answer questions involving subpopulations with different demographics, co-morbidities, risk factors, and baseline therapeutic regimens than those studied in clinical trials. The flip side, of course, is that Medicare data may reveal negative side effects and added risks from off-label use.

Comparing treatment costs and effectiveness Part D data combined with other health information will help determine whether certain medication management techniques or treatment approaches lead to best results and lower costs for certain types of patients. McClellan envisions studies to show » which drugs keep patients with complex diseases out of hospitals or how certain treatments can reduce side effects. Such analysis, which would be conducted by outside experts and research organizations, would support decisions on best practices in using medications to improve outcomes and lower overall costs of care.

McClellan is seeking input from experts from FDA advisory committees, the Institute of Medicine, and the Agency for Healthcare Research and Quality to help set research priorities. Clinical researchers who focus on particular diseases also could identify key study issues. For example, CMS is collaborating with the American College of Cardiology and the Heart Rhythm Society to identify ways to use Medicare claims data in evaluating treatment of heart failure; similar analyses could be done by the National Cancer Institute's Clinical Trials Cooperative Program, which supports thousands of interconnected cancer research entities.

Access to Medicare data also would reduce the cost of comparative and outcomes studies by the Centers for Education and Research on Therapeutics and other research networks and individual experts. A CMS "open-door forum" this month seeks comments from stakeholders on the technical, regulatory and practical issues involved in such research activities.

Setting Data Standards

CMS expects to expand its information resources even more by linking to electronic data on drug use, health outcomes, and overall costs of care assembled by PBMs and health plans serving both Medicare and under-65 patients. While McClellan is not looking to actually pool all these data, he supports the idea of developing standards and frameworks for all segments of the US healthcare system for health utilization and payment data to be collected in a "consistent" manner.

This fits with government-wide efforts to establish a national health information system. Such a system would use electronic medical records, electronic prescribing, and health information technology (HIT) in order to allow all parts of the nation's healthcare system to communicate. Health and Human Services (HHS) secretary Michael Leavitt recently added his support for such initiatives as members of Congress began to roll out bipartisan legislation to accelerate HIT development.

Medicare is doing its part to encourage hospitals and physicians to invest in HIT. In exchange for receiving slightly higher Medicare payments, more than 4,000 hospitals have established electronic systems for reporting specific quality-of-care measures. Several pilot programs will offer physicians incentives to use electronic health records to meet quality goals.

CMS is working hard to meet a 2009 deadline for implementing an electronic prescribing system. The agency has published a proposed rule for establishing basic e-prescribing standards, which it hopes to finalize by this fall. CMS also plans to launch several e-prescribing pilot projects in January 2006.

PDP prescribing and payment data will mesh with a critical Medicare Coordination of Benefits information system that will track drug purchasing information at point of sale. This will enable pharmacists and payers to determine eligibility, coverage, and out-of-pocket costs when an elderly patient comes into a pharmacy. In tracking each beneficiary's true out-of-pocket spending and coordinating payments with states and employers offering supplemental coverage, the system will create a complete but confidential claims history for every senior. PDP claims data, electronic prescribing information, quality reporting initiatives, and beneficiary hospital and medical records together will produce much richer and more extensive information on drug utilization, efficacy, safety, and costs. The challenge for pharma marketers is to use this data to document the value of many important therapies, and to recognize the limitations inherent in some products.

Jill Wechsler is Pharmaceutical Executive's Washington correspondent.