Medical Device Link.

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI January 1999 Column

OUTCOMES RESEARCH

Outcomes Research: Documenting the Value of a Medical Device

Evidence gathered from outcomes research conducted from the beginning of product development can help convince purchasers and end-users of a device's economic value, regardless of its cost, and give manufacturers a competitive edge.

Bryan R. Luce and Anne Elixhauser

In the increasingly competitive struggle for healthcare dollars, the market for new medical technologies is more austere, more demanding, and more scrutinized than in the past. Buyers and end-users want to know what value a new technology provides, what advantages it offers compared to what has been available, how much it costs, and what coverage of this device will do to a budget.

Medical device manufacturers should start anticipating these questions and determine the value of their products during the planning stages to help smooth the way for product development, marketing, and acceptance.

Figure 1. Treatment functions assessed in cost-benefit and cost-effectiveness analysis (CBA/CEA) studies during the past 18 years, broken into four medical-related interventions.

In the past decade, the pharmaceutical industry has grappled with many of the issues that currently face medical device manufacturers (Figure 1). Research-oriented drug companies have dominated the effort to establish the value of their products. This effort has been so potent that a new discipline—pharmacoeconomics—has developed, as exemplified by the founding of the journal Pharmacoeconomics and of the International Society of Pharmacoeconomics and Outcomes Research. Both the journal and the society are dedicated to assessing the health, economic, and quality-of-life outcomes associated with use of drugs. Outcomes is defined as the impact of the intervention beyond safety and efficacy —for example, effectiveness, health status, satisfaction, quality of life, and cost. (Efficacy is how well an intervention achieves measured improvements in clinical outcomes under the highly controlled setting of a clinical study, and effectiveness is how well the intervention works under routine clinical conditions.)

The drug industry also lobbied Congress for section 114 of the FDA Modernization Act, which requires FDA to relax promotional regulatory standards concerning healthcare economic information for decision makers. The device industry can learn from this process and build on it to customize outcomes research for its own particular circumstances.

THE NEW HEALTHCARE ENVIRONMENT

Defining the Customers. In the past, the physician or surgeon was the individual making the decisions about purchasing and using a therapeutic medical device. A hospital purchasing manager then negotiated price, and third-party insurers paid what was billed. Today, physicians and surgeons are still important end-users, and purchasing agents still negotiate costs, but they have much less influence on the purchasing decisions and must yield to entities such as technology assessment committees (TACs). In addition, the customer base includes patients, the federal government, the courts, and even public opinion.

What Are the Customers Seeking? Physicians and patients primarily seek effective therapies and good outcomes and can be frustrated by lack of access to new technologies they consider potentially valuable. They need evidence of cost-effectiveness to support their advocacy for access to and use of technologies and services. Similarly, the courts are interested in ensuring them fair access to beneficial technologies.

FDA requires evidence to support claims. In order to promote a technology as being safe and effective, a manufacturer must have evidence of that. Similarly, FDA requires scientific evidence to back up a claim of a technology's cost-effectiveness, which is defined as providing more value for the money spent when compared with the most reasonable alternative. The term is usually expressed as cost divided by good outcome, such as cost per year of life saved, cost per case detected, or cost per case cured. The term cost-effective is often confused with cost-saving, which means that a technology saves money. A cost-effective technology may actually cost more, but the improvement in outcomes is judged to be worth the extra cost.

Institutional decision makers seek to control expenditures while keeping their physicians, patients, and beneficiaries satisfied. They want—or at least should want—good value for their money, but they seldom generate their own information concerning value.

How Can Customers Improve the Situation? Healthcare customers are becoming more informed about the value of medical technologies. Managed-care organizations, indemnity insurers, and other institutional providers are establishing TACs or are subscribing to technology assessment services (Table I).

Table I. A listing of selected organizations that conduct, use, or offer subscription services that include technology assessments (TAs).

The hallmark of the organizations listed in Table I is their use of evidence-based decision making. The evidence-based decision-making process requires systematically gathering and analyzing scientifically valid evidence of the efficacy of a new product or procedure. This evidence comes primarily from published literature, although review committees will include information from other sources. Available evidence usually includes cost of the device and clinical evidence of its effect, but relatively little information on concomitant procedures and virtually no information on cost offsets.

INDUSTRY RESPONSE

Until recently, the device industry has responded to the radical changes in the healthcare environment and to purchasers' demands mainly by negotiating discounts with insurers and providers. However, this process of discounting tends to focus on price rather than value, and places manufacturers at enormous disadvantage.

To level the negotiating playing field, manufacturers need to examine the relative value of a device and its cost offsets. For example, would use of an intervention reduce other healthcare costs, decrease the length of hospitalization, or limit the number of tests and procedures, perhaps eliminating some completely? Most effective therapeutic medical technologies offer these or other cost offsets.

Directly attributable costs are only half of the equation. It is important to consider how a device or treatment protocol compares with other available alternatives. A device that costs more than its competitors will still be a better value if it results in higher detection rates, thus allowing the condition to be better controlled or increasing survival rates. Other value factors to consider include the number and severity of side effects, and if use of the device allows patients to return to normal functioning earlier or otherwise improves their physical, mental, or social well-being.

Without these kinds of information, decision makers focus on price rather than value, and decisions about use and coverage are often made with incomplete knowledge of the facts, which can be detrimental to patients, physicians, and manufacturers.

For their part, manufacturers are responding to the increasing power of TACs by investing in health and economic outcomes assessment studies to gather evidence of product benefits. Manufacturers will thus have at their disposal during negotiations not only price information but also data on cost offsets and health benefits, including improvements in quality of life.

THE VALUE EQUATION

The value equation is a key tool in determining evidence of value. This equation shows how much it costs to achieve a particular outcome. The equation numerator comprises costs, and the denominator comprises effectiveness or the impact on health (Figure 2).

Figure 2. The value equation used to determine the economic and health consequences of a specific health intervention.

The Intervention. The first and most critical input into the value equation is the intervention, or the technology under study. The most critical question being addressed about the technology is how well it works. Acquiring firm evidence of a product's effectiveness is the first step in conducting an outcomes study.

The Numerator. The numerator of the value equation measures costs or the changes in dollar values of the following variety of resources.

• Healthcare resources. Examples of these resources include hospitalization; physician visits; emergency room visits; use of drugs; use of other devices, tests, and laboratories; procedures; rehabilitation; long-term care; and home healthcare.

• Nonhealthcare resources. Special education, costs to modify the home because of disability or other limitations, travel time, and day-care costs for children while parents seek medical care are included in this category.

• Informal caregiver time.

• Patient time for seeking care and undergoing treatment. This item and the following one are assigned monetary values that reflect the next best use of the person's time.

Note that the numerator of the equation does not include only healthcare resources. Some outcomes studies consider only healthcare resource costs dictated by the perspective of the analysis and the client for whom the study was conducted. Healthcare insurers or managed-care organizations might only concern themselves with those costs relevant to health-care services. The government might consider healthcare costs as well as nonhealthcare resources for which it would be responsible. Studies examining the benefit to society should consider all cost components.

The Denominator. The denominator of a value equation measures effectiveness or the impact of the technology on health status. One way of measuring health-status changes is to examine their intrinsic value. Some of these can be disease-specific, such as cardiovascular events avoided. Others are technology-specific, such as improved images. Still others are generic, such as number of lives or years saved or quality-adjusted life-years saved.

Another way of measuring health status changes is to examine the impact of health on production output. Reductions in illness or averted mortality can be translated into productivity gains—these individuals will continue to contribute to society and that contribution is most often measured in terms of production output, or wages.

EQUIVALENT EFFECTIVENESS

When a device is as effective as its most likely alternative, it is only necessary to examine costs. The following situations illustrate the importance of studying cost offsets for a device undergoing outcomes assessment and of comparing results with those for alternative products.

• If the device in question has lower acquisition and use costs than its alternative, an excellent argument exists for replacing the alternative, as long as there are no cost offsets in the latter's favor.

• If the device has higher or equivalent acquisition costs but would result in a reduction in total medical costs, there is an excellent argument for replacing the alternative. However, this argument will be persuasive only if the decision maker involved is concerned with a global health-care budget, as opposed to a narrow budget such as an institution's capital expenditures budget.

• Some devices may have higher or equivalent acquisition costs, offer no reduction in other healthcare costs, but have a beneficial effect on nonhealthcare resources or patient time. In such a case, use of a device may still be supported from certain perspectives, such as that of an employer who is concerned about lost productivity or a patient facing significant out-of-pocket expenses with use of the traditional treatment.

STRATEGIC OUTCOMES RESEARCH PLANNING AND EXECUTION

Not all products require a strategic outcomes research program. Outcomes research should be targeted at those devices that are expected to provide a significant therapeutic impact or that will result in changes in resource use. New devices that incorporate only minor or incremental changes compared with previous generations or competitors generally do not warrant an outcomes research program. The one exception is if a competitor's product has not established itself as a cost-effective choice, and outcomes research could potentially provide evidence that the new device is cost-effective.

Diagnostic technologies are a special challenge, since by definition they provide information that will be used to determine subsequent treatment choices. As such, they are one step removed from the treatment intervention and two steps removed from the final outcome. If there is reason to believe that the information provided by a diagnostic technology is likely to contribute to measurable differences in treatment, economic, or quality-of-life consequences, then it is a candidate for outcomes research.

Figure 3. Strategic product development for market acceptance shows the progression of the regulatory, product development, and outcomes research tracks.

Outcomes research should be introduced as early as possible, and product developers should conceive, design, and test devices with the value equation in mind. Outcomes research will help to ensure that only cost-effective products will be developed, that they will be marketed with emphasis on their perceived value, that coverage decisions will be made more quickly with higher reimbursement levels, and that commercial success is maximized. Outcomes research and health economics have a role in every stage of product development and can be used to orchestrate a well-coordinated and effective product launch (Figure 3).

HEALTH ECONOMICS AND PRODUCT DEVELOPMENT

Product Design and Early Development. During the product conception phase, outcomes research can help clarify the nature of the condition for which the device is indicated. At this stage, outcomes research can establish baseline practice patterns and costs for the condition based on its epidemiology, expected patterns, and outcomes. It is important to determine whether patients will be treated in hospitals or as outpatients and whether the device can be used only by a physician or by other medical professionals or even patients—perhaps at some cost savings.

At this early phase of the product development cycle, outcomes research can also be used to make go/no-go decisions. Factors to consider include whether the product has a health, economic, or quality-of-life advantage over its alternatives or whether its clinical advantages are significant enough to convince healthcare decision makers to change from established therapies or devices. Such decisions can be especially difficult to influence when significant capital expenditures are involved.

Early Clinical Studies. Outcomes research can be used to develop early models of the disease and of practice patterns, based on current diagnosis and treatment. Such models are built using published literature and the input of clinicians.

Early models can be used to predict the cost drivers—resources that determine the overall costs of treating a condition. These are associated with both relatively high unit costs such as hospitalization, use of ICU versus regular beds, and emergency room visits, and with relatively low-cost resources that are used very frequently, such as physician visits and use of other products or medications.

In addition to cost drivers, early health outcomes research should identify the effects of current treatment and how they are measured. Every study is not necessarily going to examine the same effects but may choose among clinical outcomes, functional status, and quality of life, to name a few. In determining what factors to measure and how, an examination of past studies can reveal potential flaws in study design.

Early models of the disease and its treatment are critical in determining which key health and economic outcomes data need to be collected in later clinical studies. These models also help determine which data are not necessary, thus helping to streamline later studies.

Later Clinical Studies. Collecting outcomes data in a clinical study is one of the most powerful ways of examining the cost-effectiveness of an intervention. Adding an outcomes assessment component often means modifying only slightly the case-report forms or adding a patient questionnaire. The marginal cost of adding an outcomes assessment component can be quite small in proportion to the expense of the entire clinical study.

At this stage, it is also possible to adapt the early models—taking into account the introduction of the new technology—by measuring effectiveness rather than efficacy. Clinical studies tend to assess efficacy rather than effectiveness, but both forms of information are necessary for TACs to make the most informed decisions.

Another reason to study effectiveness is that a device can be expected to affect costs and health outcomes well beyond the short time frame of most clinical studies. By using decision analytic modeling, results from the study can be extrapolated to approximate the long-term effects of the intervention when epidemiological information is available to support the analysis.

Dissemination of Health Outcomes Information. The preparation of a dissemination plan is a critical parallel to the product development track. Information generated during each product development stage can be part of a strategic information dissemination plan. Early models depicting current treatment patterns and costs can lay the foundation for disseminating subsequent work that supports the device's health outcomes benefits.

The questions of how to go public with health outcomes information can be complicated. Many conferences and journals will be restricted to a particular medical specialty; however, managed-care conferences and journals may be an advantageous choice for some presentations and published work.

Product Launch. The collection of information demonstrating the health, economic, and quality-of-life value of a device should be available for dissemination at product launch in the form of articles and clinical reports. Given a TAC's heavy reliance on published literature, it is imperative that high-quality outcomes research reports are available when adoption and coverage decisions are to be made.

The early-stage model can now be adapted for other major audiences pertinent to the device—the customers or payers. The primary model should be conducted from the perspective of the device's most relevant audience, and subsequent analyses can then be conducted to consider other perspectives.

CONCLUSION

A manufacturer's success in the health-care marketplace will increasingly depend on its being able to provide evidence of the health and cost outcomes associated with a device. The device industry has hesitated to embrace technology assessment and outcomes evaluation, and as a result has limited internal capacity to address the needs and demands of health-care decision makers. Nonetheless, the concepts and methods of technology assessment and outcomes research are well established, and solidly trained economists and health services researchers are available to work with the industry.

Device company personnel often lament that they do not have the financial resources of pharmaceutical manufacturers and that devices should not be held to the same standards as drugs. The market nevertheless demands evidence of value before devices will be accepted at premium prices and adopted to their fullest potential.

Recent events have heralded a more active approach on the part of industry and others to demonstrating the value of healthcare devices. Two recently formed entities epitomize these changes:

• The Medical Technology Leadership Forum is comprised of leaders from medical device firms, academic and professional organizations, and consumer or patient groups and is working to evaluate coverage policies and to promote the conduct of value assessments of medical technologies. Reports on both these subjects are available upon request (202/661-3584).

• The task force for technology assessment of medical devices met in Seattle in 1997 and included representatives from device manufacturers, HIMA, academic institutions, managed-care organizations, employers, and the federal government. Its mission was to identify strategies to improve technology assessment of medical devices and to promulgate guidelines for the process. The task force developed a set of guidelines for information exchange between device manufacturers and managed-care organizations and insurers. A special issue of the American Journal of Managed Care will be devoted to papers emanating from the task force's work.

The efforts of these groups signal the beginning of a process to establish guidelines for outcomes research of devices. The device industry can contribute to this process so that outcomes research reflects the unique characteristics, challenges, and opportunities of this sector of the health-care marketplace. Individual firms and their leaders must also rise to the challenge by conducting assessments and providing information to healthcare decision makers. Integrating outcomes research into the product development process will allow device manufacturers to meet marketplace demands, while simultaneously ensuring that their products provide healthcare benefits and enjoy commercial success.

BIBLIOGRAPHY

Drummond MF et al., Methods for the Economic Evaluation of Health Care Programmes. 2nd ed. Oxford, England: Oxford University Press, 1997.

Evidence of Value: Building a New Paradigm. Washington, DC: Medical Technology Leadership Forum, March 1998.

Gold MR et al., Cost-Effectiveness in Health and Medicine. New York: Oxford University Press, 1996.

Medicare Coverage: Time for a Public Policy Dialogue. Washington, DC: Medical Technology Leadership Forum, March 1998.

Bryan R. Luce, PhD, is CEO and senior research leader at MEDTAP International (Bethesda, MD). Anne Elixhauser, PhD, is a social science analyst at the Agency for Health Care Policy and Research (AHCPR), U.S. Public Health Service. A version of this paper was previously presented at the Medical Device Executive Forum, Anaheim, CA, in January 1998.