Innovation in medical technology generates a remarkable supply of new drugs, devices, and diagnostics that improve health, reduce risks, and extend life. But these technologies are too often used on the wrong patient, in the wrong setting, or at an unaffordable price. The only way to moderate the growth in health care costs without undermining the dynamic of medical innovation is to improve the process of assessing, pricing, prescribing, and using new technologies. Purchasing Medical
Innovation analyzes the contemporary revolution in the purchasing of health care technology, with a focus on the roles of the Food and Drug Administration (FDA), Medicare and private health insurers, physicians and hospitals, and consumers themselves. The FDA is more thoroughly assessing product performance under real-world conditions as well as in laboratory settings, accelerating the path to market for breakthroughs while imposing use controls on risky products.
Insurers are improving their criteria for coverage and designing payment methods that reward efficiency in the selection of new treatments. Hospitals are aligning adoption of complex supplies and equipment more closely with physicians’ preferences for the best treatment for their patients. Consumers are becoming more engaged and financially accountable for their health care choices. This book describes both the strengths and deficiencies of the current system of purchasing and highlights opportunities for buyers, sellers, and users to help improve the value of medical technology: better outcomes at lower cost.
|Publisher:||University of California Press|
|Product dimensions:||5.60(w) x 8.40(h) x 1.00(d)|
About the Author
James C. Robinson is Leonard D. Schaeffer Professor of Health Economics and Director of the Berkeley Center for Health Technology at the University of California, Berkeley. His articles appear in a broad range of scholarly, medical, and journalistic publications, including Health Affairs, JAMA, and the Wall Street Journal. Dr. Robinson brings real-world experience to academic and policy debates and scientific rigor to the professional and industry world.
Read an Excerpt
Purchasing Medical Innovation
The Right Technology, for the Right Patient, at the Right Price
By James C. Robinson
UNIVERSITY OF CALIFORNIA PRESSCopyright © 2015 The Regents of the University of California
All rights reserved.
Regulatory Access to the Market
Patients want therapies that are safe and effective, yet the nation has all too much experience with treatments that poison the patient and with nostrums that do nothing at all. The US Food and Drug Administration (FDA) was established originally to protect consumers from tainted foods, but has seen its scope of authority expand to encompass pharmaceuticals, biologics, implantable devices, laboratory tests, diagnostic imaging, and radiation therapy. New products must obtain authorization from the agency prior to being made available for sale. The FDA's product label specifies the clinical indications for which the product can be marketed to physicians and, in some cases, requires the manufacturer to support a program of risk surveillance after the product is launched.
The FDA is not usually considered to be a purchaser of health care, as the agency neither pays for the products it regulates nor insures patients against the costs. In practice, however, the FDA performs two key purchasing functions. Through its regulatory requirements the agency generates most of the performance data that subsequently are used by those who do pay for medical technology. Furthermore, the agency's authority over market access frames the purchasing decisions of others. Insurers typically will not reimburse, physicians will not prescribe, and patients will not accept products that fail to achieve FDA approval for at least some uses, and the downstream purchasers often are skeptical regarding product uses that fall outside the FDA-approved indications.
It is essential that the FDA mandate rigorous testing for safety and efficacy, as many once-promising drugs and devices have proved to be ineffective against their intended targets. Others are so toxic that they should not be prescribed for any purpose. Some laboratory assays and imaging tests impair rather than improve the physician's ability to diagnose a disease. But ensuring safety and efficacy is not a simple task, and the FDA is besieged by critics who accuse it of doing too little, or, on the contrary, of doing too much.
The agency is now seeking a balance between premarket testing and access restrictions, on the one hand, and postmarket surveillance and risk mitigation, on the other. Many treatments are dangerous when used inappropriately but offer important benefits when used for the right indications, in the right setting, and in the right manner. Denying access to these products does not protect patients, but hurts them. It also short-circuits the possibility that the product or its method of use could improve over time. This is especially important for technologies that undergo frequent improvements based on experience, such as implantable medical devices. Postmarket surveillance also can strengthen consumer protection, since many treatments are found over time to pose greater risks than were evident at the time of initial approval. The shift in emphasis from premarket to postmarket surveillance needs to be accompanied by stronger regulatory controls on the manner of use, if risks are identified, and a strengthened ability to rescind market access if the risks prove unacceptable.
This chapter describes the role of the FDA as the first stage in the purchasing of medical technology. It begins with an overview of the agency's requirements for safety and efficacy as a condition for market access. It then considers the regulatory framework for implantable medical devices, which differs in important ways from the framework for drugs. The claim that the FDA fails to adequately protect patients is examined using the example of metal-on-metal hip implants, an initially promising product approved under the FDA's lenient standards for incremental innovation. The claim that the FDA imposes excessive regulatory burdens is considered using the example of aortic valve replacement, a breakthrough innovation that gained market access in Europe years before being approved by the FDA. The final section analyzes the FDA's balancing act between premarket authorization and postlaunch surveillance of medical technology.
PREMARKET TESTING AND AUTHORIZATION
The FDA has the statutory authority to deny market access to any product that has not proved itself safe and effective. Overcoming this regulatory hurtle does not guarantee, of course, that a new product will be reimbursed by insurers, prescribed by physicians, or embraced by patients. But failure to achieve FDA authorization for at least some purposes usually guarantees that it will not. And market access always is conditional, not absolute. Authorization can be retracted if the product is found to be significantly less effective or more dangerous than previously thought.
The primary mission of the FDA is consumer protection, but the agency cannot insist on complete safety. Every treatment poses at least some risk to some patients, at some doses, or in some settings. The acceptability of risk can only be assessed relative to the benefits offered. The FDA therefore requires proof of efficacy as well as safety. A new test must accurately identify the targeted trait, symptom, or condition, while a new treatment must alleviate the targeted disability or disease. Therapies that raise concerns may still receive authorization if no safer alternatives are available. Chemotherapies that target cancer, for example, often cause nausea, pain, and other serious side effects. They receive FDA authorization due to the life-threatening nature of the illness. The agency is less tolerant of toxic treatments for milder conditions. It was willing to withdraw market authorization for anti-inflammatory drugs that caused side effects in a small fraction of patients, for example, because safer and equally effective products were available.
The FDA's regulatory requirements add substantially to the time needed to get medical technology to market, cutting deeply into potential sales and revenues. Most tests and treatments are protected by patents and cannot legally be replicated for twenty years. But patents are filed early in the research and development process, and the product's twenty years of protection are often half over before it obtains FDA approval.
Drugs spend the first few years of the patent protection period undergoing laboratory assays and animal studies. The bigger delays come with the three phases of clinical trials in humans. A promising treatment is first used on a small number of healthy volunteers to ascertain toxicity and seek a threshold for safety. The second phase of clinical trials uses patients who suffer from the disease of interest, and seeks evidence on efficacy as well as additional insight into adverse side effects. These two phases often add four years of delay to the hoped-for product launch.
The third phase of testing compares the impact of the new treatment with that of a placebo or a traditional form of care, using patients who have been assigned randomly to treatment and control groups. Randomization eliminates the confounding effect of unobserved patient characteristics. Large numbers of patients are important to give the study the statistical power to detect subtle impacts and to examine differences across patient subgroups. The length of the trial is often determined by the choice of treatment endpoint. The third phase of clinical testing usually adds another two to three years to the time until product launch.
Demands for more studies, covering more patients and with longer follow-up, have added immensely to the cost of medical innovation. Clinical trials are costly because physicians need to be engaged, patients recruited, the product contributed without reimbursement, study findings assessed, and an evidence dossier assembled. It costs an average of $1 billion to develop a new drug and get it successfully through FDA review. The FDA has targeted regulatory delays and has made significant improvements with additional staffing from industry-supported fees on drug and device applications. Review times have decreased and approval rates increased.
Once the drug or device patent expires, follow-on products jump into the market. Generic drugs, biopharmaceutical "similars," and follow-on devices are held to weaker standards of evidence than their reference products. This allows them to charge lower prices. Generic drugs are often priced at an 80 percent discount from the original brand, and follow-on biologics will be priced at a 25 percent discount. Device prices usually decline over time. This competition is good for the purchaser but bad for the innovator, who must match the competitor's price cuts or face substantial loss of sales.
Treatments are authorized by the FDA for some conditions, as described in the product label, but not for others. Product developers can only market their products to physicians for these approved indications. But physicians have the right to prescribe the product for indications outside the FDA label if they believe it will benefit their patients. These "off-label" uses are not necessarily inappropriate, since it is not possible for manufacturers to test and for the FDA to evaluate every therapy for every condition. But the FDA label remains the foundation for efforts to ensure appropriate use. Medical management programs developed by insurers, clinical protocols developed by professional societies, and informed consent programs developed for patients build on the label.
The FDA does not review evidence on the economic impact of the treatments that come under its purview. Some critics feel the agency should interpret its mission as promoting the value of health care, meaning that it should assess price as well as performance. Others counter that inclusion of cost analyses would further politicize an already-volatile regulatory process. The American public resists suggestions that access be denied if costs exceed a defined threshold. And even nations more comfortable with prioritizing medical expenditures situate cost analyses as part of insurance coverage policy, keeping the regulation of initial market access focused on health risks and benefits. That is a sensible principle.
THE REGULATION OF IMPLANTABLE MEDICAL DEVICES
The regulatory structure at the FDA was originally developed for drugs, but was extended to implantable medical devices in 1976 as innovation brought increased risks as well as benefits from those products. The FDA developed a three-part classification scheme that links the stringency of regulation to the characteristics of each device. Devices that are life sustaining or pose significant risks, such as cardiac pacemakers and artificial valves, are required to go through the premarket approval (PMA) pathway. Intermediate-risk devices, such as imaging equipment and many orthopedic devices, do not need to prove safety and efficacy but are cleared through the more lenient 510(k) pathway after documenting "substantial equivalence" to an already-cleared predicate. Manufacturers of low-risk devices, such as hearing aids and prescription eyeglasses, do not need regulatory approval but merely notify the FDA prior to marketing their products.
The device sector is characterized by many incremental modifications of existing products, punctuated occasionally by an innovation that offers a significantly new mechanism of action, design, or risk profile. The FDA cannot require each variant of an evolving device to go through full premarket review. By the time the clinical trials were complete, the product would be outdated, and manufacturers might be discouraged from improving their product in the first place. Randomized trials are especially difficult to implement. It is difficult to sustain the confidentiality as to which patients have been assigned to the treatment and which to the control group, since devices involve surgery, catheterization, or another invasive procedure. Some patients resist randomization and insist on being in the treatment group. Others may acquiesce in their initial assignment but then insist on switching midway through the clinical trial. In the major clinical trial of spine fusion, for example, half of the patients assigned to surgery decided not to undergo the procedure while a third of the patients assigned to drug therapy decided they wanted the surgery.
Full premarket approval is required only rarely for medical devices. Between 2003 and 2007, only 1 percent of the 15,000 new medical devices reviewed by the FDA went through the PMA pathway. The agency appeared to set a lower bar for devices than for drugs, even for those going through the PMA pathway. Only 1 percent of the devices going through PMA were denied approval. And, in fact, the majority of the regulatory actions for high-risk devices were not treated as new reviews but as supplements to earlier reviews. Between 2003 and 2007, the FDA authorized 170 original PMA applications but a total of 664 PMA supplements. Supplements may include major or minor modifications in device structure as well as routine changes in labeling, materials, or manufacturing processes. Firms applying for a PMA supplement do not need to provide data on product safety and efficacy.
The importance of PMA supplements compared to original reviews is illustrated in the evolution of cardiac rhythm management devices such as pacemakers, defibrillators, and cardiac synchronization therapy. Between 1979 and 2012, the FDA approved 77 original PMA applications for these devices, but also 5,829 supplement applications. Half of the supplemental applications concerned changes in manufacturing techniques and half concerned changes in design or materials. Of the supplements reporting a significant change in design or label, 23 percent mentioned the collection of new clinical data, but none involved randomized clinical trials. The heavy reliance on supplemental applications significantly reduces the delay and cost of FDA review, compared to what is required for an original PMA. However, the supplement process permits incremental changes to be made without clinical data even if the changes cumulatively transform the device. For example, one type of defibrillator lead wire that the FDA recalled from the market in 2011 due to serious adverse effects had undergone 78 supplemental applications and modifications since its original PMA review in 1996.
For intermediate-risk devices, the FDA relies on comparisons with devices that are already in use, through what is referred to as the 510(k) premarket notification pathway. The 510(k) pathway is based on the concept that new devices equivalent to established devices in their structure and function are also equivalent in their risks and benefits. One-third of new devices are cleared for marketing through the j10(k) pathway. The FDA has the authority to demand clinical data from these products, but rarely does so. Between 2005 and 2007, only 8 percent of the products submitted for 510(k) clearance contained data on safety and effectiveness, and only 11 percent of the predicate devices that they were judged as equivalent to had submitted such data.
The principle of substantial equivalence is appropriate for devices that are subject to continual modification and have short shelf lives, but its application has often been inappropriate. A finding by the FDA that a new device is substantially equivalent to one on the market does not imply that the new device is safe and effective. Most established devices that are referenced in 510(k) submissions are new versions of devices that themselves never documented safety and efficacy. Rather, the predicate devices were cleared for marketing by the FDA as equivalent to an earlier product, which had been deemed equivalent to an even earlier one. Many predicate chains extend back to devices that were on the market prior to the original 1976 device legislation, and hence never passed any clinical review.
The principle of substantial equivalence has also been misused to permit the marketing of devices that the FDA classifies as posing significant risks. The 1976 legislation permitted use of the 510 (k) pathway as a temporary expedient for high-risk devices that were already on the market, with the intention of having the agency subsequently review them through a full PMA process. But as of 2011, twenty-six classes of high-risk devices still had not completed PMA review and authorization.
The FDA's principle of substantial equivalence is valid if it is applied judiciously, since many new devices can be expected to perform similarly to those they modify. Indeed, some variant of the principle is unavoidable. The PMA pathway cannot be used for every modification of every device, or the current generation of designs will be frozen in place. Ideally, the FDA builds on the 510(k) clearance process with a vigorous postmarket surveillance system. But the agency is limited in its ability to intervene in how products are used once they have been cleared for market access. The agency has bolstered its postmarket risk management programs in recent years, but it is unclear whether these will remedy the shortcomings of the premarket 510(k) process.
Excerpted from Purchasing Medical Innovation by James C. Robinson. Copyright © 2015 The Regents of the University of California. Excerpted by permission of UNIVERSITY OF CALIFORNIA PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
List of Illustrations vii
1 Regulatory Access to the Market 19
2 Insurance Coverage and Reimbursement 41
3 Methods of Payment for Medical Technology 59
4 The Hospital as Purchaser 79
5 Organizational Capabilities for Technology Purchasing 100
6 The Patient as Purchaser 120
7 Implications for the Medical Technology Industry 141