Letters
5. Blackwell M, Iacus S, King G, Porro G. CEM: coarsened exact matching in Stata. Stata J. 2009;9(4):524-546. 6. Agency for Healthcare Research and Quality. Guide to prevention quality indicators. http://www.qualityindicators.ahrq.gov/Downloads/Modules/PQI /V31/pqi_guide_v31.pdf. Accessed September 30, 2014.
2. Stöllberger C, Schneider B, Finsterer J. Elimination of the left atrial appendage to prevent stroke or embolism? anatomic, physiologic, and pathophysiologic considerations. Chest. 2003;124(6):2356-2362. 3. Lundh A, Sismondo S, Lexchin J, Busuioc OA, Bero L. Industry sponsorship and research outcome. Cochrane Database Syst Rev. 2012;12:MR000033.
COMMENT & RESPONSE
Left Atrial Appendage Closure for Atrial Fibrillation To the Editor The study by Dr Reddy and colleagues1 reported the 3.8-year follow-up of the PROTECT AF trial that randomized patients with atrial fibrillation (AF) to receive percutaneous left atrial appendage (LAA) closure with the WATCHMAN device or warfarin. We had a number of questions and concerns about the study. First, patients randomized to receive warfarin were older and more often had cardiovascular comorbidities and permanent AF than the patients randomized to receive the LAA closure. In Table 1 in the article, P values were missing, so it is unclear if these differences were statistically significant and may in part explain the higher mortality in the warfarin group. Second, why do the patient-years for the different end points in Table 2 differ? Third, after LAA closure, warfarin was discontinued for 345 of 370 patients at the 12-month transesophageal echocardiographic evaluation. What was the reason for continuation of warfarin in the remaining 25 patients, and what were their outcomes? Fourth, according to Table 3, 53 patients in the LAA closure group died. How many of these patients underwent autopsy, and what were the pathoanatomic findings of the left atrium and LAA? Were there any thrombi or leaks between the LAA wall and the device, and was the device completely endothelialized in all cases? Fifth, the LAA is known to play a hemodynamic role in pressure and volume overload of the left ventricle and is a site for release of atrial natriuretic peptides.2 Thus, it would be of interest to know whether patients in the LAA closure group developed new or worsening heart failure more frequently than patients in the warfarin group. Sixth, only 3 of the 12 authors reported no potential conflict of interest with the manufacturer of the closure device. Furthermore, the senior author has contracted rights to receive royalties from the license of the device. The sponsor of the study was the manufacturer of the device and was responsible for data collection, analysis, interpretation, and drafting of the manuscript. Thus, a bias in favor of the device cannot be excluded.3 Reanalysis of the data by independent scientists is needed. Claudia Stöllberger, MD Birke Schneider, MD Author Affiliations: Krankenanstalt Rudolfstiftung, Wien, Austria (Stöllberger); Sana Kliniken, Lübeck, Germany (Schneider). Corresponding Author: Claudia Stöllberger, MD, Steingasse 31, A-1030 Wien, Austria (
[email protected]). Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. jama.com
1. Reddy VY, Sievert H, Halperin J, et al; PROTECT AF Steering Committee and Investigators. Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: a randomized clinical trial. JAMA. 2014;312(19):1988-1998.
In Reply Drs Stöllberger and Schneider raise a number of questions about our article. First, with regard to baseline patient characteristics, as stated in the Results section, there were no significant differences in principal baseline characteristics between the groups. Second, the number of patient-years in Table 2 varies between end points because cumulative patient-years for each row were determined by the time to first event. In the case of the primary end point, the first event could be any component of the composite, whereas the individual rows would only include patients experiencing the specific outcome listed, thus allowing for small variations in the denominator. Third, the reason for warfarin continuation in 25 of 370 patients at 12 months was because of a residual shunt in 10 patients and physician discretion for 15 patients. Fourth, there was no systematic autopsy evaluation of patients who died during the course of the trial, so there are no pathology data for these patients. However, we have previously published an analysis 1 of patients receiving the WATCHMAN device in PROTECT AF comparing those with no peridevice leak (68% of the cohort) and those with at least some small degree of leak (32% of cohort). In that analysis,1 we found no significant difference in the primary efficacy outcome between groups if the leak was small. Fifth, the protocol did not include the systematic collection of left ventricular ejection fraction or heart failure status during follow-up, so we cannot comment on the possibility of differential development or worsening of heart failure. However, both cardiovascular mortality and all-cause mortality (both of which are known to track with severe congestive heart failure) occurred more frequently in the warfarin group. As shown in Table 3 in the article, there was no difference in heart failure–related deaths between groups. Sixth, Stöllberger and Schneider note that the manufacturer of the WATCHMAN device was responsible for the data collection and compilation. However, a data and safety monitoring board that was independent of both the company and the steering committee of the trial reviewed all of the data. Furthermore, as part of the US regulatory process, the study data were monitored in compliance with US Food and Drug Administration requirements for investigational device exemption protocols, and all data were independently reviewed and analyzed by the agency. Vivek Y. Reddy, MD David Holmes, MD Author Affiliations: Helmsley Electrophysiology Center, Mount Sinai School of Medicine, New York, New York (Reddy); Mayo Clinic College of Medicine, Rochester, Minnesota (Holmes). (Reprinted) JAMA March 10, 2015 Volume 313, Number 10
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Western Oregon University User on 05/30/2015
1057
Letters
Corresponding Author: Vivek Y. Reddy, MD, Helmsley Electrophysiology Center, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029 (
[email protected]).
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Reddy reported receiving research grant support and consulting fees from Boston Scientific. Dr Holmes reported receiving research grant support from Boston Scientific. The LAA closure technology has been licensed to Boston Scientific; both Mayo Clinic and Dr Holmes have contractual rights to receive future royalties from this license, but to date no royalties have been received.
1. Mello MM, Studdert DM, Kachalia A. The medical liability climate and prospects for reform. JAMA. 2014;312(20):2146-2155.
1. Viles-Gonzalez JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: a PROTECT AF (Percutaneous Closure of the Left Atrial Appendage Versus Warfarin Therapy for Prevention of Stroke in Patients With Atrial Fibrillation) substudy. J Am Coll Cardiol. 2012;59(10):923-929.
5. US Department of Health and Human Services; Health Resources and Services Administration. Resources about guidebooks. http://www.npdb.hrsa .gov/resources/aboutGuidebooks.jsp. Accessed January 15, 2015.
Medical Liability and Reporting Malpractice Payments To the Editor In their overview of the current US medical liability climate, Dr Mello and colleagues1 noted that requirements to report medical malpractice payments to the National Practitioner Data Bank (NPDB) can pose a major deterrent for physicians to participate in communication-and-resolution programs. Although important, this barrier is more modest than many physicians may recognize. The Health Care Quality Improvement Act of 19862 does not require an NPDB report simply because someone paid money following a patient injury.3 Rather, a report is required when (1) an entity such as an insurance company pays on a physician’s behalf2(§11131[a]) and (2) in response to “a written claim or demand for payment based on a health care provider’s furnishing (or failure to furnish) health care services ….”2(§11151[7]) Thus, payments out of pocket are not reportable because they are not made by an entity on a physician’s behalf.3,4 In addition, the NPDB Guidebook expressly states that no report need be made if a claim or demand for payment is made orally rather than in writing.3 Indeed, “[e]ven if the practitioner transmits the demand in writing to the medical malpractice payer, the payment is not reportable if the patient’s only demand was oral.”3 In many cases for which a hospital or health care practitioner offers disclosure and compensation to an injured patient, that patient has never made even an oral, let alone written, demand; therefore, no NPDB report would be required. The Department of Health and Human Services’ Health Resources and Services Administration is currently undertaking a major revision of the NPDB Guidebook.5 I hope that this revision expressly clarifies that communication-andresolution payments and payments made during pre-suit notification periods generally do not require NPDB reports so long as such payments are not made pursuant to a written claim or demand for payment. Haavi Morreim, JD, PhD Author Affiliation: Department of Internal Medicine, University of Tennessee Health Science Center, Memphis. Corresponding Author: Haavi Morreim, JD, PhD, Department of Internal Medicine, University of Tennessee Health Science Center, 956 Court Ave, Memphis, TN 38163 (
[email protected]). 1058
2. Health Care Quality Improvement Act, 42 USC §11101 et seq (1986). 3. Morreim EH. Malpractice, mediation, and moral hazard: the virtues of dodging the data bank. Ohio State J Dispute Resolution. 2012;27(1): 109-178. 4. American Dental Association v E Shalala, 3 F3d 445 (DC Cir) (1993).
In Reply Dr Morreim suggests that requirements that malpractice payments be reported to the NPDB do not pose as great an obstacle to fully realizing the promise of communicationand-resolution programs as our article indicated. Unfortunately, we do not share her optimism. Morreim’s view is premised on the argument that reporting requirements are narrower than many believe because of 2 exceptions: claims paid directly by physicians themselves (rather than through liability insurers) and payments made in response to verbal claims (rather than written ones). In practice, however, it is these exceptions that are narrow. Historically, very few malpractice payments are made out of physicians’ own pockets, and this is unlikely to change in the context of communication-and-resolution programs. The verbal claim exception to NPDB reporting also lacks practical salience for communication-and-resolution programs. Our experience studying a number of these programs across the country revealed that even programs that are proactive in offering compensation often ask patients or family members to provide a letter outlining their version of events and what they are seeking by way of resolution.1 Communication-and-resolution programs find written demands helpful for clarifying the patient’s perspective, identifying potential misperceptions and communication gaps, and determining avenues of redress. In addition, some of these institutions’ liability insurers require a written demand to start a conversation about compensation. Hospitals and insurers could abandon such practices and conduct the resolution process purely orally, but this is not as easy as it seems. Some families state their demands in a letter or email before hospital representatives can intervene—or, indeed, even before the hospital learns of the incident. In any case, asking a family not to put any compensation demands in writing may appear suspicious and foster distrust. There is another practical challenge. Even if a hospital or insurer attempts to resolve a matter prior to a written demand, there is no guarantee such a demand will not appear later during negotiations. This makes it difficult for hospitals and insurers to reassure physicians that participating in early resolution will not trigger an NPDB report that would otherwise not have been made. It also seems illogical and arbitrary that similarly situated patients may or may not trigger NPDB reporting requirements based on how they initially approach a hospital. For example, NPDB reporting will be triggered if a patient sends an
JAMA March 10, 2015 Volume 313, Number 10 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Western Oregon University User on 05/30/2015
jama.com