To: Administrative File CAG: # 00166N Cardiac Catheterization Performed In Other Than A Hospital Setting
From: Steve E. Phurrough, MD, MPA
Director, Coverage and Analysis Group (CAG)
Marcel E. Salive, MD, MPH
Director, Division of Medical and Surgical Services
JoAnna F. Baldwin
Policy Analyst, CAG
Joseph Chin, MD
Medical Officer, CAG
Subject: Coverage Decision Memorandum for Cardiac Catheterization Performed In Other Than a Hospital Setting
Date: January 12, 2006
The Centers for Medicare and Medicaid Services (CMS) is repealing §20.25 of the National Coverage Determination (NCD) Manual.
II. Background and History of Medicare Coverage
The Medicare National Coverage Policy for Cardiac Catheterization Performed In Other Than A Hospital Setting was implemented August 1, 1979 in the NCD Manual §20.25:
Cardiac catheterization performed in a hospital setting for either inpatients or outpatients is a covered service. The procedure may also be covered when performed in a freestanding clinic when the carrier, in consultation with the appropriate Peer Review Organization (PRO), determines that the procedure can be performed safely in all respects in the particular facility. Prior to approving Medicare payment for cardiac catheterizations performed in freestanding clinics, carriers must request PRO review of the clinic.1
At that time, the carriers (Medicare Part B contractors) were responsible for consulting with the PROs (subsequently renamed Quality Improvement Organizations [QIOs]) prior to approving Medicare payment. In 1979, the PROs were conducting reviews as part of their scope of work. In the early 1990’s, the approach and functions of the PROs changed from case reviews to quality improvement. The PROs stopped their review of freestanding cardiac catheterization facilities when it was no longer included in the PRO scope of work in 1995. The language in the NCD Manual was never revised to reflect the change in QIO functions. Therefore, the policy that remains in the NCD Manual contains outdated language that implies QIO activity outside their scope of work.
CMS implemented the national policy for cardiac catheterization in other than a hospital setting requiring PRO involvement.
Review of cardiac catheterization facilities was no longer included in the PRO scope of work.
CMS opened the NCD and requested public comment.
CMS extended the due date of the NCD.
CMS extended the due date of the NCD.
CMS posted a technology assessment commissioned from the Agency for Healthcare Research and Quality (AHRQ) to review the available evidence regarding cardiac catheterizations in freestanding facilities.
CMS issued the proposed decision memorandum for 30 days of public comment.
CMS corrected the proposed decision memorandum to accurately reflect quotations from the technology assessment and restarted the 30-day period for public comments.
Public comment period ended.
PubMed was searched from 1999 using the term “cardiac catheterization laboratory.” Within this time period, no controlled trials were identified comparing patient outcomes and safety of cardiac catheterizations performed in freestanding facilities with hospital-based procedures. The search did identify, however, a consensus document from the American College of Cardiology (ACC) and The Society for Cardiac Angiography and Interventions (SCA&I) and an article discussing trends in the growing number of freestanding facilities in the United States.2
B. Discussion of Evidence Reviewed
1. External Technology Assessment
In 2005, the Agency for Healthcare Research and Quality commissioned a technology assessment on cardiac catheterization in freestanding clinics to answer 5 key questions:
Do freestanding cardiac catheterization clinics and hospitals have comparable complication rates for diagnostic catheterization procedures?
- Do freestanding cardiac catheterization clinics and hospitals have comparable complication rates for interventional catheterization procedures?
- Do hospitals without cardiac surgical support and hospitals with cardiac surgical support have comparable complication rates for diagnostic and interventional catheterization procedures?
- What are the characteristics of patients who have had catheterization procedures in freestanding cardiac catheterization clinics vs. hospitals?
- What are the current state regulations, Certificate of Need (CON) requirements, and oversight procedures for freestanding cardiac catheterization clinics?
The assessment was performed by ECRI and stated (reproduced exactly from the report3):
"Key Question 1: Do freestanding cardiac catheterization clinics and hospitals have comparable complication rates for diagnostic catheterization procedures?
After searching the literature, retrieving articles, and applying the inclusion/exclusion criteria, we identified 23 publications that reported complication rates of diagnostic catheterization procedures in a freestanding or hospital outpatient setting. None of these studies directly compared complication rates in freestanding and hospital settings. Thus, the quality of the evidence is low. The studies’ generalizability to the Medicare population was fair.
Five separate studies reported complications in a freestanding laboratory (two of these studies were reported only in a systematic review but have not been otherwise published). The mortality rates ranged from 0 to 0.16%, as did the rate of MI, while the rate of stroke/transient ischemic attack (TIA) ranged from 0 to 0.03%. Vascular complications ranged from 0 to 2.0%.
Nineteen studies reported complication rates in a mobile or fixed hospital outpatient setting. No deaths occurred in any of the three mobile laboratory studies, while the mortality rate ranged from 0 to 0.3% among the 16 fixed hospital outpatient studies. Rates of MI ranged from 0 to 0.1% in mobile labs and 0 to 0.7% in fixed outpatient settings, while rates of stroke/TIA ranged from 0 to 0.3% in mobile labs and 0 to 0.4% in fixed outpatient clinics. Rates of vascular complications ranged from 0 to 0.1% in mobile labs and 0 to 2.0% in fixed outpatient settings. The available evidence did not reveal substantial differences in complication rates of diagnostic catheterization procedures among freestanding clinics and hospital outpatient settings. However, this indirect and informal comparison of low quality studies could not be risk-adjusted to compensate for differences in patient characteristics among the studies. Also, none of the freestanding clinic studies reported the length of followup; if it was shorter than the average followup in the hospital outpatient studies, this would create bias in the comparison. Furthermore, we cannot determine whether the relatively low complication rates reported in freestanding studies are generalizable to all freestanding centers, as this evidence base was susceptible to potential publication bias. Since all freestanding clinic studies and most hospital outpatient studies were published in the 1980s, the degree of relevance of the findings to current clinical practice is also unknown. These weaknesses in the evidence base mean that we cannot completely rule out the possibility of differences in complication rates between these settings.
Key Question 2: Do freestanding cardiac catheterization clinics and hospitals have comparable complication rates for interventional catheterization procedures?
Our literature searches identified no studies that addressed this question. No evidence- based conclusion was possible regarding interventional catheterization procedures in freestanding centers. An ACC/SCA&I consensus document recommended that such procedures not be performed in freestanding settings, and we found no information to suggest that percutaneous coronary intervention (PCI) procedures are currently being performed in this setting.
Key Question 3: Do hospitals without cardiac surgical support and hospitals with cardiac surgical support have comparable complication rates for diagnostic and interventional catheterization procedures? This question will only be addressed if the literature is insufficient for questions 1 and/or 2 for freestanding clinics.
Because no evidence was available to address Key Question 2, we evaluated outcomes of PCI procedures at hospitals with and without surgical support. However, hospitals without cardiac surgical support are an imperfect surrogate for freestanding clinics, because even hospitals without cardiac surgical support have support services and resources beyond what is typically found in freestanding settings. Thus, one cannot be certain to what extent, if any, the findings for interventional procedures in a hospital setting can be extrapolated to a freestanding setting.
We identified seven retrospective controlled studies (five articles and two meeting abstracts) that compared complication rates of non-primary or primary PCI in hospitals with or without cardiac surgical support. These studies ranged from low to fair in quality based on U.S. Preventive Services Task Force (USPSTF) ratings. All were vulnerable to potential selection bias from lack of randomization and lack of followup of patients transferred to other hospitals. Generalizability to the Medicare population was fair except for one study where it was high.
Three studies of non-primary PCI (PCI for reasons other than emergent acute MI) reported conflicting results. One of these studies exclusively evaluated elective PCI procedures and found no statistically significant differences in complication rates between care settings. The remaining two studies evaluated all non-primary PCI procedures (including some emergent procedures). The only study that exclusively evaluated Medicare patients showed a significantly higher mortality rate in hospitals without cardiac surgical support, while the remaining study showed no significant difference between care settings. However, the latter study was low quality because no adjustments were made to account for baseline differences in the characteristics of patients who were seen at the differing hospital settings. Because these studies were vulnerable to selection bias and differed from each other in several characteristics, the conflicting results cannot be explained with certainty.
Six studies of primary PCI showed consistent findings of no statistically significant difference in rates of mortality or serious morbidity between hospitals with and without cardiac surgical backup. Three of the studies adjusted for differences in patient risk. However, all of these studies were vulnerable to selection bias to a greater or lesser degree, and some may have lacked adequate statistical power to detect a meaningful difference in rates. These flaws in the evidence base mean that failure to demonstrate a difference does not eliminate the possibility that a difference may exist.
Key Question 4: What are the characteristics of patients who have had catheterization procedures in freestanding cardiac catheterization clinics vs. hospitals?
We found no studies that directly addressed this question, but 17 of 23 studies from Key Question 1 indirectly addressed the question through their patient inclusion/exclusion criteria. The quality of these studies was low, and their generalizability to the Medicare population was fair.
Two studies of freestanding facilities reported detailed inclusion/exclusion criteria, and these criteria were very similar to those reported in hospital outpatient studies. Both freestanding and hospital outpatient studies included clinically stable patients and excluded one or more subgroups of higher risk patients (with recent MI, Class IV cardiac disease, refractory unstable angina, and severe congestive heart failure, among others). Minor variability appeared in the specific subgroups of patients excluded among the different studies.
An ACC/SCAI expert consensus document recommended similar but slightly more stringent exclusion criteria for freestanding settings than for hospitals without cardiac surgical support. A published multivariable model for predicting complication risks during cardiac catheterization procedures is consistent with some of these recommendations.
Key Question 5: What are the current state regulations, Certificate of Need (CON) requirements, and oversight procedures for freestanding cardiac catheterization clinics? Include a table summarizing regulations from all 50 states. Include also a review of international regulations and guidelines; at a minimum include information from the U.K. and Canada.
Currently, 37 states (plus Washington, D.C.) do not prohibit diagnostic cardiac catheterization procedures in a freestanding setting. In these states, regulation usually occurs through CON programs (16 states plus D.C.). Sixteen states without CON programs have no regulations or licensure requirements for such clinics. Sources in 13 states (plus D.C.) that do not prohibit freestanding catheterization services reported that there were no such facilities (or at least they were not aware of any) currently operating in these states.
Thirteen states have regulations prohibiting cardiac catheterization in freestanding clinics. In three of these states, pilot programs or regulatory loopholes have allowed at least one freestanding facility to perform cardiac catheterization procedures. Any facility performing cardiac catheterization procedures can voluntarily seek Joint Commission on Accreditation of Healthcare Organizations (JCAHO) accreditation. The facility must meet several functional standards to gain accreditation.
Our survey of two other countries revealed that the United Kingdom does not allow any cardiac catheterization procedures to be performed outside of a hospital setting. Canada has no specific regulatory prohibitions on the national level, and four provinces did not report specific prohibitions, but all four provinces reported that no freestanding facilities were performing cardiac catheterization procedures. We cannot confirm whether any freestanding facilities perform these procedures in the remaining nine Canadian provinces and territories."
2. Guidelines, consensus panels and expert opinion
The ACC/SCA&I Clinical Expert Consensus Document: Cardiac Catheterization Laboratory Standards discusses recommended requirements for performing cardiac catheterizations in freestanding facilities. The freestanding facility should have a formal relationship with a tertiary hospital for the emergency transfer of patients, have equipment for intubation and ventilatory support, and have quality assurance and quality improvement programs in place. In addition, the physicians must be able to perform endotracheal intubations and insert an intra-aortic balloon pump. Patient selection is critically important in freestanding facilities without surgical backup, and is specifically addressed in the document. The consensus document recommends that only diagnostic cardiac catheterizations be performed in freestanding facilities and lists patient inclusion and exclusion criteria; no therapeutic procedures are recommended.
The ACC and SCA&I document stresses the importance of quality assurance outlining clinical proficiency, complication rate, diagnostic accuracy and patient outcomes as critical aspects of a cardiac catheterization lab. The document reports that the major complication rate for diagnostic cardiac catheterizations is estimated between 1% and 2%, which the experts feel is acceptably low.4
3. Public Comments
Initial Public Comments
CMS received comments in support of this change jointly from the ACC and SCA&I. The ACC and SCA&I acknowledge that due to changes in their scope of work, PROs have not provided oversight of freestanding facilities and that procedures have continued to be performed safely. They included in the comments a copy of the ACC/SCA&I Clinical Expert Consensus Document: Cardiac Catheterization Laboratory Standards.5
The American Hospital Association (AHA) submitted comments expressing concern over the lack of quality controls in outpatient facilities and suggesting that freestanding cardiac catheterization facilities be subject to quality standards and monitoring/enforcement requirements that are comparable to those applied to hospital outpatient departments for similar procedures. AHA also states that there is an inherently greater risk in performing the procedure in an outpatient setting where there are no emergency or cardiac surgery services on site.
Public Comments on Proposed Decision Memorandum
CMS received a total of 66 public comments during the 30-day public comment period. Much of the correspondence received was on behalf of cardiology group practices such as one letter that represented 41 cardiovascular centers. The majority of commenters were physicians in cardiology practices while others included 12 individuals at administrative levels of group practices, individuals in the financial services area of group practices, lawyers, journalists and health care consultants. The SCA&I, the JCAHO and the Accreditation Association for Ambulatory Health Care (AAAHC) also submitted comments.
The majority of commenters requested that CMS suspend repeal of the NCD. Those commenters recommended that CMS further research the evidence and work with experts to create national requirements which should then be implemented through the NCD. Some commenters suggested that they would submit a request to CMS to require that freestanding facilities become accredited by the JCAHO or AAAHC. CMS is not able to suspend the NCD process but rather is required to issue a final decision within 3 months of posting the proposed decision (§1862(1)(3) of the Social Security Act). Therefore, we are unable to suspend the current final decision. Parties interested in seeking a reconsideration of an NCD should follow the procedures set forth in 68 Federal Register 55634 (September 26, 2003).
Overall, commenters suggested that CMS is forcing patients to use hospital-based catheterization labs, allowing hospitals to monopolize the cardiac catheterization market. They also suggested that the decision will result in the closure of some freestanding facilities since half of their patients are Medicare beneficiaries and that the decision will not allow physicians to own labs. CMS wants to be clear that repealing an NCD is not the same as establishing a national policy barring all coverage. Rather, repealing this NCD allows local Medicare contractors to establish coverage policies through their own processes, as outlined below. While these policies will not necessarily be identical in all areas of the United States, this more flexible approach permits policies that better reflect local differences in the practice of medicine.
The SCA&I supports the proposal to repeal the NCD. The organization does not believe national policy is necessary as the AHRQ technology assessment found no reason to question the quality and safety of care currently delivered in freestanding facilities. They further believe this approach reduces unnecessary regulations. The SCA&I does not view this repeal as a removal of coverage, rather, they express interest in working with local carriers in developing policy.
Both the JCAHO and the AAAHC offered to provide CMS with information related to their accrediting programs. Neither organization stated their position regarding the repeal of this NCD.
Differentiating Between Freestanding Facilities
One commenter asked that CMS differentiate between freestanding facilities that have hospital partnerships and freestanding facilities without those partnerships. The commenter stated that facilities with such partnerships have the same oversight and quality standards as the hospital. CMS is not making such a differentiation at this time as the entire policy is being repealed and dealing with this issue will be at the discretion of the local contractor.
Quality of Care
The commenters that recommended CMS suspend the repeal of this NCD suggested that there is no evidence that quality of care is better in a hospital than in a freestanding facility. Some commenters provided anecdotal information in support of this statement (e.g. patient testimony of their experience, higher staff to patient ratios than hospitals and providing a lower stress environment for patients). Other commenters reiterated the findings of the AHRQ technology assessment that lower mortality and complication rates were found in freestanding facilities when compared to hospital based catheterization labs. CMS agrees with the findings of the technology assessment and believes that because incidence of adverse events is low that establishing national policy at this time is not necessary.
Many commenters stated that they follow the consensus document issued by ACC/SCA&I to ensure quality of care and patient safety. Other commenters stated their involvement in the ACC’s National Cardiovascular Data Registry (NCDR) for catheterization labs giving them the ability to benchmark their outcomes. CMS believes that following the current professional society guidelines and participating in a national registry such as the NCDR may maintain and improve patient outcomes. However, CMS will not make these a requirement at this time and will allow the local contractor to make such considerations.
Some commenters stated that repealing the NCD would put freestanding catheterization facilities financially at risk since approximately half of their patients are Medicare beneficiaries. An additional commenter stated that CMS would be penalizing freestanding facilities for saving the agency money if it were to repeal the NCD. One commenter believed that a freestanding facility saves money by only allowing diagnostic catheterizations and, therefore, avoiding the last minute decision to perform an unnecessary intervention. The second commenter believed that money would be saved if all catheterizations were performed in the hospital where a patient could receive an immediate interventional cath following the diagnostic and therefore only bill Medicare for the therapeutic intervention. CMS did not review cost as part of this decision but rather only reviewed clinical evidence.
Local Coverage Determination
Some commenters were concerned that if coverage was at the discretion of local contractors that varying methods would be used to establish such local policy. Medicare contractors use a standard process in creating local coverage determinations (LCDs) which is outlined in Chapter 13 of the Medicare Program Integrity Manual (publication 100-8). Section 13.7.1 of the manual states that LCDs should be based on: 1) published authoritative evidence derived from definitive randomized clinical trials or other definitive studies; and 2) general acceptance by the medical community (standard of practice), as supported by sound medical evidence based on: a) scientific data or research studies published in peer-reviewed medical journals; b) consensus of expert medical opinion (i.e., recognized authorities in the field); or c) medical opinion derived from consultations with medical associations or other health care experts. Further, section 13.7.2 states that contractors shall provide a comment period and notice period for all new LCDs. We believe that this process will fairly determine when cardiac catheterizations in freestanding facilities are reasonable and necessary. These local determinations may consider any state or local laws governing the furnishing of these services.
Commenters were further concerned that due to the potential of varying policies across contractors that beneficiaries would not have the same level of access to services. The Medicare program was designed to allow local contractors to take local issues into account when developing policy. In the case of cardiac catheterization, many states have established rules regarding the use of such freestanding facilities whereas other states have not. Contractors have the ability to review such information as it pertains to their geographic areas and apply it in developing a policy based on §1862(a)(1)(A) of the Social Security Act.
One commenter stated that if CMS were to move forward with the NCD repeal then state entities and contractors should be prepared to follow through quickly with establishing new policies. Since the previous policy was brief and required some level of contractor interpretation and implementation, contractors have already been dealing with this issue so CMS feels they are already prepared.
One commenter stated that CMS should not assume that contractors will not act to noncover catheterizations in freestanding facilities. Since the previous policy provided no useful information for contractors, we believe that they have been covering cardiac catheterizations in freestanding facilities as a result of their own policies and therefore we do not expect contractors to establish noncoverage. Rather, we expect contractors to review their current policies regarding the procedure and to use their processes and local information to establish policies with input from the public. It is important to note that contractors will have the authority to noncover catheterizations in freestanding facilities that do not meet the requirements they set forward in their LCDs.
V. CMS Analysis
National coverage determinations (NCDs) are determinations made by the Secretary with respect to whether or not a particular item or service is covered nationally under title XVIII of the Social Security Act § 1862(l)(6)(A). In order to be covered by Medicare, an item or service must fall within one or more benefit categories contained within Part A or Part B, and must not be otherwise excluded from coverage. Moreover, with limited exceptions, the expenses incurred for items or services must be “reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member.” § 1862(a)(1)(A).
The original coverage policy for cardiac catheterization performed in freestanding centers in 1979 was developed at a time when the PROs (now known as QIOs) were conducting facility reviews as part of their scope of work. In the early 1990’s, the approach and functions of the PROs changed from case reviews to quality improvement. In 1995, the review of freestanding cardiac catheterization was removed from the PRO scope of work, effectively ending PRO review of these clinics. The language in the NCD Manual was never revised to reflect the change in current QIO functions. Thus CMS initiated this decision to update the NCD Manual based on current evidence and practice.
When considering the evidence, specifically the AHRQ technology assessment, nearly half of the states either prohibit diagnostic cardiac catheterization in freestanding settings or have regulations or licensing programs that have prevented programs from being established. The technology assessment found that 13 states have regulations specifically prohibiting cardiac catheterization in freestanding clinics, although three of these states do have such facilities. Further, it appears that ongoing review and monitoring of cardiac catheterizations performed in freestanding clinics continues in many states. CMS concludes that the NCD Manual should not reflect inaccurate QIO functions and that no national coverage determination is appropriate at the present time.
As noted in the technology assessment, there is suggestive but insufficient evidence to determine if net health outcomes of cardiac catheterizations performed in freestanding clinics is comparable to cardiac catheterizations performed in the hospital setting. Comparative evidence on coronary catheter-based interventions was insufficient as well. Until a more complete evaluation of the safety and effectiveness of cardiac catheterizations performed in freestanding clinics is conducted, it appears prudent to consider the risks and benefits on a local coverage basis. In the absence of an NCD, local Medicare contractors can determine whether a cardiac catheterization performed in a freestanding facility is reasonable and necessary under §1862(a)(1)(A). Therefore, we conclude that cardiac catheterization may be covered when performed in a freestanding clinic. CMS is repealing the existing national policy on cardiac catheterization in freestanding clinics.
The Centers for Medicare and Medicaid Services (CMS) is repealing §20.25 of the National Coverage Determination (NCD) Manual.
1 NCD Manual §20.25
2 Sheldon, WC et al. 2001
3 AHRQ, 2005. Cardiac Catheterization in Freestanding Clinics, pages 1-6.
4 Bashore, TM et al. 2001
5 Bashore, TM et al. 2001