OR WAIT 15 SECS
Understanding the United States' regulatory pathways and clinical operations.
For medical devices, good engineering does not guarantee clinical trial or commercial success. The climate for introducing new medical devices into the market has become increasingly difficult in recent years. At times, regulators and device manufacturers seem to be on opposing sides of the playing field. Regulators often contend with pressure to fast-track approvals while at the same time ensuring product safety and efficacy. On the other hand, manufacturers risk spending time and resources to develop a device, only to not receive approval to go to market. For products that reach the market, several factors, including the reimbursement environment and coverage decisions, can influence profitability. Executing a successful market strategy involves navigating the US Food and Drug Administration (FDA) approval process, designing and conducting efficient clinical trials, and adeptly evaluating the existing marketplace.
The FDA's Center for Devices and Radiological Health is responsible for regulating firms who manufacture medical devices sold in the United States. The FDA classifies devices into one of three categories (Table 1). Class I devices have the lowest potential for harm. Examples are bandages, crutches, and examination gloves. Class II devices, such as powered wheelchairs, infusion pumps, EKG machines, and orthopedic drills, are subject to special controls that might include special labeling, performance standards, and postmarket surveillance. Class III devices are the most invasive. Examples would be pacemakers, stents, endovascular grafts, and heart valves. Device classification defines the regulatory requirements for a general device type.
Premarket notification 510(k) and premarket approval. Most Class I devices are exempt from the FDA's premarket notification 510(k).1 The 510(k) is a premarket submission to demonstrate that a device is as safe and effective as a predicate device. If this is established, clearance is given to the product. Most Class II devices require premarket notification 510(k). Most Class III devices require premarket approval. Premarket approval is the most stringent requirement for a new device. It is expensive and time consuming and requires extensive clinical data.
Investigational device exemption. If an investigational device study presents a significant risk to subjects, a sponsor must obtain FDA approval of an investigational device exemption (IDE) from the Office of Device Evaluation (ODE). An IDE allows the investigational device to be used in a clinical study of safety and effectiveness required to support a premarket approval application or a premarket notification 510(k) submission to FDA. The IDE must contain information concerning the study's investigational plan, report of prior investigations, device manufacture, institutional review board actions, investigator agreements, subject informed consent form, device labeling, market price of the device, and other matters related to the study. The FDA has 30 calendar days after receiving the application to approve or disapprove an IDE submission.
The FDA encourages sponsors to meet with the ODE reviewing division before the IDE application is submitted for review.2 The reviewing division provides advice for developing pre-clinical support data or the investigational plan. These meetings may take the form of telephone conference calls, video conferences, or face-to-face discussions. A formal guidance meeting, such as a determination meeting or an agreement meeting, may be arranged to review efficacy data and investigational plans. If an agreement is reached between FDA and the sponsor or applicant regarding the parameters of an investigational plan (including a clinical protocol), the terms of the agreement are put in writing and made part of the administrative record by the FDA.
Importance of seeking feedback. Working with the FDA early in the process of developing a medical device can save time, money, and frustration. Getting to know the individual review division and working with them to determine efficacy and safety criteria can prevent your study from having "red flags" down the road. The FDA looks favorably on carefully designed studies that include accurate observations made by competent personnel. Studies with unclear objectives, poorly defined or analyzed outcomes, improperly documented sample size, unexplained missing patients, unexplained unblinded changes in analyses, or serious adverse events that are dismissed without description as "intercurrent illness" reflect unfavorably on the investigational plan and possibly the device itself.
In developing medical products, planning and execution of clinical trials are extremely important. Poorly designed or executed clinical studies fail to convince payers. Trials should be designed to glean the information necessary for proving the primary and secondary objectives without gathering extraneous material. Economic information necessary for future reimbursement applications should also be obtained.
Initial clinical trials include a small number of patients and are designed primarily for safety. Subsequent clinical trials should demonstrate a new clinical benefit or the same benefit at a lower price. The Center for Medicare and Medicaid Services (CMS) is interested in trials that have comparisons to the best available therapy or to standard of care. Cost data may include such things as reduced hospital admissions or stays, reduced outpatient charges, reduced requirements for medication, or a decrease in the amount of required skilled nursing time.
Design intelligence. In planning a clinical trial, it is paramount to prioritize objectives and outcomes. For example, is the primary outcome a time to an event or survival? Are the objectives achievable? Consider the number of subjects and sites, and plan to collect only data that will be needed; most data collected for clinical trials is never used, and some data is collected in an inefficient manner. Another important consideration is whether the protocol can be adequately funded. The length of time required for documenting efficacy is usually much shorter than the time required for determining safety. Third parties, such as enrolling physicians, can provide estimates of how many patients are likely to enroll within a given time period.
Placebo-controlled trials can be performed if there is no existing standard of care. For comparing a new product with an active control, either a superiority trial or non-inferiority trial may be used. Superiority trials demonstrate whether the new treatment is better than the existing. Non-inferiority trials evaluate whether the product is similar to a control that has proven efficacy. In a non-inferiority trial, a known inferiority margin should be selected, and if treatment differences are within that range, then non-inferiority can be established. Selecting the statistical margin of non-inferiority should be done in the design stage of the trial. In most cases, sample size for a non-inferiority trial will be larger than for a superiority trial. In a non-inferiority trial, the treatment or product under investigation should offer some other advantage such as cost savings or lack of complications relative to the traditional form of treatment.
Protocol considerations. Clinical trial protocols should contain the following components: achievable objectives, appropriate statistical parameters, a comprehensive introduction with complete bibliography, a satisfactory study population, a complete safety analysis, an appropriate description of the product, and a logical study design. A study should include a minimum number of subjects to assess the primary objective. Calculation of sample size in a trial requires knowledge of the power, the significance level, the underlying event rate in the population being investigated, and the size of the treatment effect being evaluated.
Use of technology. While many clinical trials rely on pen-and-paper for documenting patient data, increasingly, electronic data capture is being used for efficiency. Although use of technology reduces complexity, long-term success depends upon up-front planning and having standardized processes. Software applications have been designed for many aspects of managing clinical trials, including planning, resourcing, operations, data capture, data management, recruitment, and site management.
Run a transparent, efficient study. During the course of a clinical trial, it is important to have complete visibility in managing operations and accrual of data. Visibility is key for operating the start-up, conduct, and closeout of a trial and managing project milestones, patient accrual, case report form status, monitoring activities, and queries from study sites. Technology-driven processes can ensure complete trial visibility and can be useful for managing performance metrics and planning future studies.
A company developing a compound or device may elect to administer studies or contract with a clinical research organization (CRO) to manage the investigation. In the latter situation, the sponsor is involved in developing the protocol, budget, and enrollment parameters, but the CRO is responsible for conducting the trial, site monitoring, coordination of centralized data collection and analysis, query resolution, and preparation of materials for submission to the FDA.
Addressable patient population. The target audience for a new medical device may be substantially less than the number of patients with a particular condition. Payers such as the CMS can have restrictions that limit payment to a subset of patients. For example, payers may require an initial use of a conservative form of therapy prior to approving payment for a new device. If patients heal during the initial therapy, the pool of patients suitable for receiving a device shrinks. With some devices, the FDA may approve the device for only a sub-segment of the target addressable market.
Know the marketplace. Is the device a quantum leap forward in technology? Examine existing literature on competing forms of treatment and determine whether the new device has specific advantages. Understand the strength of existing or potential competitors and the potential for competitors to rapidly develop a competing or identical product. A new medical device must provide added patient benefit, and there must be enthusiasm for the product from physicians, patients, and hospitals. Particular attention must be paid to the entities that will ultimately purchase the device, often hospitals or physician offices. In addition, it must be determined whether or not there will be a professional (generally physician) fee associated with use or application of the product. The potential for marketing directly to patients also must be considered.
Understand the reimbursement environment. Revenue generation for a new medical product is based on its ability to provide new diagnostic or therapeutic clinical benefit or to replace existing technology at a lower price. A poor reimbursement strategy can severely limit profitability. A new device manufacturer must know what the competitive and reimbursement environment is for their product and understand what the payment rates are for other various therapeutic options used to treat the same disease. With increasing financial constraints, government and private medical payers such as the CMS, the Agency for Healthcare Research and Quality, the Medicare Coverage Advisory Commission, and the Blue Cross/Blue Shield Technology Evaluation Center are increasingly demanding value before authorizing payment.
Coverage decisions. The CMS is tasked with paying for services and items that are reasonable and necessary. A majority of the time coverage decisions are made by local carriers, but decisions can be based on a National Coverage Decision issued by the CMS. Each year the CMS issues approximately 10-15 National Coverage Decisions for technologies that they feel will have a major impact on healthcare. They take into account the quality of individual studies and prefer data from randomized controlled trials. Key opinion leader recommendations may also be considered. There must be sufficient evidence to demonstrate a product or service is medically beneficial for the Medicare population. A National Coverage Decision determination is binding on all Medicare carriers, fiscal intermediaries, quality improvement organizations, health maintenance organizations, competitive medical plans, and healthcare prepayment plans.3
From 1999 to 2007, approximately 60% of National Coverage Decision applications resulted in approval.4 However, there are usually restrictions placed on the clinical condition, population, or setting in which the treatment can be administered. Increasingly the CMS has paid attention to whether use of the device leads to relevant outcomes or has applicability to the Medicare population.
HCPCS codes in new product reimbursement. Obtaining a Healthcare Common Procedure Coding System (HCPCS) code is an extremely important part of the reimbursement process for a new medical device. Level II of the HCPCS is used primarily to identify products, supplies, and services not included in the current procedural terminology codes such as ambulance services, durable medical equipment, prosthetics, orthotics, and supplies.5 The CMS has the authority to distribute HCPCS Level II codes. The codes offer a means of indicating to the CMS and private carriers what supplies they are paying for. The CMS, America's Health Insurance Plans, and the Blue Cross and Blue Shield Association maintain these codes jointly as members of the HCPCS National Panel. Decisions regarding changes to the national permanent codes are only made by unanimous consent of all three parties. Guidance to manufacturers and suppliers on the proper use of the HCPCS Level II codes is provided by The Statistical Analysis Durable Medical Equipment Regional Carrier, which is under contract to the CMS.
Payments to hospitals. Any product that offers clinical efficacy while reducing costs will be of interest to hospital administrators. In the past, physicians had a greater input into products purchased by the hospital. However, with increasing financial constraints, hospitals often place physician requests under scrutiny. Healthcare providers may have to demonstrate the value of a product before hospital administrators purchase it.
For submission to the payer, International Classification of Diseases (ICD-9) diagnosis and procedure codes are placed on the hospital ChargeMaster. There are over 13,000 ICD-9 diagnosis codes and 5,000 ICD-9 procedure codes. Also included on the ChargeMaster are complications and co-morbidities, age, gender, and discharge destination.
For medicare patients, procedures are further classified into a Medicare Severity Adjusted Diagnosis Related Group (MS-DRG). Each MS-DRG code is assigned a weight. More complex procedures and illnesses are assigned higher weights. Payment to the hospital is based on the MS-DRG weight. There are additional factors that may be taken into account with regard to payment, such as regional labor costs.
Charges for inpatient products and services are not individually reimbursed. Hospitals are given a flat amount of money to cover all expenses involved with an in-patient hospitalization, including but not limited to lab tests, medications, disposable supplies, and human resources expenses. If a product is an adjunct to treatment of an in-patient illness for which there is a MS-DRG payment, there is no mechanism of separately paying for the device. If the hospital costs are less than the money received, the hospital makes a profit. If the costs are more, then there is a loss. Therefore it is incumbent upon device manufacturers to demonstrate value, such as decreased length of hospital stay, intensive care unit stay, or need for healthcare personnel.
Physician reimbursement. Manufacturers must be cognizant of whether or not there will be a professional reimbursement associated with the use of their product. Physicians are reimbursed for professional services under the Current Procedural Terminology (CPT) system, which is also referred to as the Level I HCPCS designation. There are approximately 8,000 CPT codes in existence. A CPT editorial panel is responsible for maintaining the database. They revise, modify, and update the codes. Of the 17 members of the panel, 11 are physicians nominated by the American Medical Association, while additional members are nominated by other groups, such as CMS, Blue Cross and Blue Shield, and the American Hospital Association. There are three categories of CPT codes. If a placement of a medical device is associated with a Category I code, the physician will generally be paid for the service. A Category II code is generally considered a tracking code not associated with reimbursement, although in the future this may have a role in "increased payment for quality." Category III codes are paid at the discretion of local carriers and may be difficult to obtain.
Private carrier payment. When a company is trying to obtain approval from managed care organizations, data must be submitted that includes peer-reviewed literature, assessments by specialty organizations, and endorsements from key opinion leaders. Unlike the CMS, private payers often consider cost of the device in their decision-making analysis.
One of the most developed entities in the private payer evaluation process is the Blue Cross Blue Shield Technology Evaluation Center that uses the following five criteria for reimbursement approval:
The ultimate goal of developing a device is to have it be both useful and profitable in the clinical setting. Doing so requires keen insight and careful planning into several unique processes—clinical testing, regulatory approval, and coding and payment procedures. As a device enters the market, it is important to document the benefits relative to established alternatives, any improvements in net health outcomes, and functions realized outside of an investigational setting. Maintaining relationships with healthcare providers and institutions is important in establishing market share and evaluating future opportunities.
Blair Keagy,* MD is CEO at MPAnalytics, Inc., and Professor of Surgery, Division of Vascular Surgery, UNC Department of Surgery, 2450 Sedgefield Drive, Chapel Hill, NC, e-mail: firstname.lastname@example.org. David Levin is Vice President at Clinipace Worldwide.
*To whom all correspondence should be addressed.
1. US Food and Drug Administration, "Overview of Device Regulation," http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/Overview/default.htm.
2. US Food and Drug Administration, "IDE Approval Process," http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/HowtoMarketYourDevice/InvestigationalDeviceExemptionIDE/ucm046164.htm#who_apply.
3. American College of Radiology, "National Medicare Coverage Decisions," http://www.acr.org/Hidden/Economics/FeaturedCategories/Coverage/nmcd.aspx.
4. P. Neumann, M. S. Kamae, and J. A. Palmer, "Medicare's National Coverage Decisions for Technologies, 1999-2007," Health Affairs, 27 (6) 1620-31 (2008).
5. US Department of Health & Human Services, Centers for Medicare & Medicaid Services, "Healthcare Common Procedure Coding System (HCPCS) Level II Coding Procedures," (June 2010), https://www.cms.gov/MedHCPCSGenInfo/Downloads/HCPCSLevelIICodingProcedures7-2011.pdf.