Today’s Action: The Centers for Medicare & Medicaid Services (CMS) today posted a draft national coverage determination (NCD) to expand coverage of implantable cardioverter defibrillators (ICD) based on new research showing that the devices can benefit a wider population of patients than first indicated in the product’s original trials.
This expansion will increase the number of Medicare beneficiaries eligible for an ICD by one-third, to nearly 500,000. When this NCD becomes final, CMS expects to provide this therapy to an additional 25,000 patients in the first year of coverage, potentially saving up to 2,500 lives.
The expanded coverage for the implantable defibrillators is based on new data, including the results of the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) that shows for the first time that certain patients who have never had a heart attack are likely to benefit from these devices. In addition, careful analysis of the new data from this trial in combination with data from all previous ICD trials showed that patients with a certain finding on electrocardiograms of their heart, called a narrow QRS, may also derive a small but measurable benefit from having an ICD.
As part of the decision, CMS will work closely with product manufacturers and clinical experts, including the National Institutes of Health, to develop a practical registry that can track the progress of patients who receive the devices and help develop additional evidence to better identify who is most likely to benefit from them. The national data collection system for ICD implantation will meet several criteria for facility certification, assessment and data completeness and should not present a burden that would be a barrier to eligible patients receiving the device.
Because the decision announced today is a proposed decision, CMS is soliciting recommendation and input from clinicians, specialty societies, manufacturers, and research organizations to help ensure that the appropriate hypothesis are asked, minimal data necessary to answer the hypotheses are collected, and simple processes for data collection and submission are created.
Importance of the ICD registry
Currently, evidence-based guidelines that doctors use to decide who should receive an ICD are based on broad patient characteristics, meaning that some patients who receive ICDs are less likely to benefit than others. Because the risk of life-threatening arrhythmias is still relatively low in many patients, most individuals will never require cardioversion or defibrillation from their device. If more precise characteristics can be identified to better predict which patients are likely to benefit from these devices, or will not, patients and doctors could make more informed decisions and perhaps avoid unnecessary procedures and potentially inappropriate shocks without the guarantee of any benefit.
A national data registry can help produce this critical information. It will also help to gather long-term safety and effectiveness data on the device for the covered indications.
Often, clinical trials have only limited follow up times and do not provide adequate long-term data. While the completed ICD trials followed patients for up to 5 years, these devices are implanted for life, making it important to have data over longer periods of time. Data from registries are important in demonstrating that new technologies have acceptable long-term benefits and safety profiles, especially for low risk populations.
The value of a registry is clearly dependent on the quantity and quality of the data entered into the registry. Collecting essential and uniform data will help maintain the long-term usefulness of specific registries. CMS will work closely with clinicians, specialty societies, manufacturers, and research organizations to identify the appropriate hypotheses, collect minimum data necessary to address these hypotheses, and create simple processes for data collection and submission. Because ICD manufacturers have established an extensive network of technical assistance, there may be an opportunity to work directly with these individuals to ensure that the registry process is reliable and efficient.
A registry must include all patients who are receiving an ICD for primary prevention.
CMS’ previous national coverage determination provided coverage of ICDs as a primary prevention of sudden cardiac death in certain patients with ischemic cardiomyopathy. The proposed decision expands this coverage to more patients with ischemic cardiomyopathy and adds a large group of beneficiaries with nonischemic cardiomyopathy. Both of these groups benefit from ICD with relative reductions in mortality of ~20 percent and an absolute reduction in mortality of 6-7 percent. That means that both groups have a significant number of patients who will never receive active therapy from the ICD, and may have been able to avoid having the implant if there were more precise measures for predicting which patients would most likely require a cardioversion.
Data currently available from completed clinical trials do not provide more precise methods for identifying patients at relatively low or high risk of sudden cardiac death. Recent professional society guidelines have identified risk stratification studies as a high clinical research priority, and have suggested that registries may provide such information. To conduct studies that accurately identify the patients who are most likely to benefit and those who are unlikely to benefit requires collection of data on all patients. In addition:
- They should be given the same level of patient protection as patients covered under new policies.
- All patients who receive ICDs for primary prevention share common characteristics and indications for device implantation.
- The value of the registry is dependent on data entered. Selected inclusion of patients may introduce bias and may potentially influence the validity of results generated from analyses of the registry data.
- Since only a small proportion of patients who get ICDs for primary prevention of SCD receive therapy from their devices, maximizing the number patients by including all primary prevention patients would help develop a more robust dataset.
- The marginal increase in burden and cost to collect data on all patients is likely to be relatively small once the infrastructure and staffing is in place to collect data on some patients.
Examples of existing cardiovascular disease registries
The feasibility and value of clinical registries has been demonstrated by several very large registry projects in cardiovascular disease device and drug therapy. Active registries are producing valuable data for patients with undergoing implantation of ventricular assist devices, thrombolytic therapy for acute MI, heart failure, acute myocardial infarction, treatment for congestive heart failure, placement of coronary stents, and cardiac catheterization. Data from these registries has resulting in dozens of peer-reviewed publications, and supported several numerous specific improvements in treatment decisions for these patients.
CMS required submission to a registry as a condition of reimbursement in the Ventricular Assist Device for Destination Therapy decision in October 2003. Most hospitals now submit data to a registry managed by the International Society for Heart/Lung Transplant. NHLBI will shortly award a contract to create and manage a VAD registry for NIH in collaboration with FDA.
The National Registry of Myocardial Infarction (NRMI), sponsored by Genentech, Inc. is one of the largest observational studies of acute myocardial infarction (AMI). Since 1990, NRMI has collected data on more than 2 million AMI patients, and helped more than 1,600 participating hospitals assess their approach to AMI treatment and identify trends in patient outcome. An analysis of over 10 years of NRMI data shows substantial improvement in key measures at NRMI participating hospitals. Trend data indicate reduced time to treatment with thrombolytic therapy, reduced length of hospital stay, and increased use of percutaneous intervention and adjunctive pharmacotherapy. The NRMI has also been accepted by the (JCAHO) for accreditation purposes in connection with the ORYX initiative.
The ADVANCENT, funded by Guidant and managed by REGISTRAT, follows patients with heart failure to allow for the analysis of entire patient populations and to identify gaps in care. For example, the registry found that women with the same indications for treatment as men received a cardiac resynchronization therapy system or implantable cardioverter defibrillator one-third of the time while men received therapy half of the time. More than 100 centers and 25,000 enrolled patients are part of the ADVANCENT registry.
The American College of Cardiology-National Cardiovascular Data Registry™ (ACC-NCDR™) Cath Lab collects data from more than 500 institutions and includes over 1.6 million patient discharges. This registry is used by the cath labs tomeasure and improve the quality of patient care. Quarterly institutional reports are created, containing national and peer group statistics relating to diagnostic cath and PCI procedures performed in the cath lab.
Cordis Corporation, a Johnson & Johnson company, recently announced plans to initiate the D.E.S.cover(sm) Registry in January 2005. It will include approximately 15,000 patients at more than 200 U.S. hospitals and is intended to yield a real-world 'snapshot' of the impact of drug-eluting stents on the practice of interventional cardiology. Consecutive patients treated with drug-eluting stents, bare metal stents and balloon angioplasty will be enrolled. The registry will capture a wide range of information about current interventional cardiology practices and procedures, including extensive clinical performance and safety findings as well as health-economic and patient quality-of-life data. Patients will be followed for at least 12 months post treatment and the resulting database will provide physicians worldwide with important information to use in 'benchmarking' their own procedural strategies with patients.
The registry must be designed to be minimally burdensome
CMS’s goal is to create a registry that gathers only those data elements that provide critical information about risks and benefits, and have the potential to provide more precise information on which doctors and patients can make decisions and to improve patient outcomes. Much has been learned from other clinical registry projects that can be applied in the design and implementation of the ICD registry.
CMS plans to design the registry process to be as simple and least burdensome as possible and should integrate with other data collection systems, be web-based, and as automated as possible. CMS is soliciting recommendations that will assure that these goals are met, and will work closely with public and private stakeholders to ensure that a practical and efficient system is developed. If properly designed and implemented, the clinical benefits and savings should easily justify the expense and effort involved.
The draft coverage decision provides an opportunity for public input
This is a proposed decision and CMS is extremely interested in working with the clinical community, specialty societies, manufacturers, and research organizations to ensure that the appropriate hypothesis are asked, minimal data necessary to answer the hypotheses are collected, and simple processes for data collection and submission are created. CMS is also interested in comments on alternative approaches to providing operational support and funding for the registry.