| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiothoracic Surgery, University of Washington, Seattle, Washington
* Address correspondence to Dr Wood, Division of Cardiothoracic Surgery, University of Washington, 1959 NE Pacific, Rm AA-115, Box 356310, Seattle, WA 98195-6310 (Email: dewood{at}u.washington.edu).
Any editorial written by a thoracic surgeon regarding the benefits of surgeon specialty on patient outcomes should be immediately suspect due to obvious self-interest, but enough data now exist to allow the facts to speak for themselves. In this issue of the Annals, the study by Schipper and associates [1] notes that general thoracic surgeons performed only 5% to 10% of four index thoracic operations, with more than 50% performed by general surgeons [1]. However, postoperative mortality in the four thoracic surgery procedures was 34% to 51% lower and prolonged postoperative hospitalization was 36% to 46% lower when performed by general thoracic surgeons, suggesting that 17,849 deaths and 140,754 hospital stays greater than 14 days could have been avoided during a 10-year period if thoracic surgical care were standardized in the United States (U.S.).
The authors conducted a retrospective cohort investigation using data from the National Inpatient Sample evaluating the relationship between surgeon specialty and early morbidity and mortality [1]. They evaluated four index general thoracic surgical procedures—lobectomy, sublobar resection, pneumonectomy, and decortication—and three groups of surgeons—general thoracic, cardiac, and general surgeon. Outcomes were in-hospital mortality and prolonged length of stay, a surrogate for morbidity. They found that patients under the care of general thoracic surgeons had a lower adjusted risk of early death and morbidity for each index procedure compared with general surgeons. Schipper and associates' investigation adds to a growing body of evidence that suggests that the type of surgeon caring for patients with thoracic disease has an important effect on outcome.
This study is noteworthy in that, unlike prior studies using Medicare or Surveillance, Epidemiology, and End-Results-Medicare data, it examines all adult patients, regardless of payer status. As a result, this study is more likely to be generalizable to the U.S. population and thoracic surgical procedures than prior studies.
The relationship between specialty care and better short-term (safety) and long-term (efficacy) outcomes is well documented. Lung cancer patients undergoing pulmonary resection at high-volume hospitals and teaching hospitals have lower risks of operative mortality [2–4] and higher 5-year survival rates [5, 6] than those at low-volume hospitals and nonteaching hospitals. Patients under the care of higher-volume surgeons have lower risks of operative mortality independent of hospital volume [7]. More recent studies have established that lung cancer patients under the care of board-certified thoracic surgeons have lower risks of operative mortality [8] and higher-long term survival [9] independent of surgeon volume, hospital volume, and teaching status. These data, and this additional work from Schipper and colleagues [1], confirm the benefits of specialty care in thoracic surgery in terms of safety (decreased morbidity and mortality), cost (decreased complications and fewer prolonged hospitalizations), and efficacy (improved long-term survival).
From a health policy perspective, this study by Schipper and colleagues [1], and the other studies on this subject, provides strong evidence that standardized care would improve outcomes for thoracic surgical patients. A major goal of US public health policy is to save lives as well as to decrease the burden of illness. Specialized care of thoracic surgical patients accomplishes both of these goals, and one can compare the estimated magnitude of lives saved to policy regarding the influenza vaccine.
The Centers for Disease Control and Prevention promote a policy of influenza immunization that is strongly advocated at the local, regional, and national level and is extensively publicized in the media. This policy is widely accepted and is felt to be cost-effective despite the substantial expenses of vaccine development, production, distribution, and public relations. Approximately 36,000 people die each year in the United States of influenza and its complications [10], and the influenza vaccine results in an estimated mortality reduction of 4.6% [11] resulting in approximately 1656 lives saved each year from the influenza vaccine. This is virtually the same estimate of annual lives that would be saved if the four thoracic surgical procedures evaluated by Schipper and colleagues were performed by thoracic surgeons, yet there is no such policy adopted by hospitals, payers, or government agencies, and no attention to this dilemma by the media.
Thoracic surgeons are well aware of the apparent moral hazard that occurs in a community when a patient is referred to the local general surgeon for lung cancer resection but to the general thoracic surgeon if the patient is higher risk, is a "VIP" (health professional or relative, community or business leader), or if the patient demands specialist care. If high-risk or "important" patients benefit from operations done by thoracic surgeons, it seems likely that other patients will as well. This tacit understanding of the benefits of specialty care is obvious and is supported by research from Schipper and others, yet appears to be undermined by local factors that have yet to be confronted by hospitals, payers, patient advocacy groups, or policy makers.
Physicians referring patients requiring thoracic operations may prefer to direct a patient to a nonspecialist due to local politics and economics, potentially benefiting directly or indirectly if the patient is cared for within the same hospital or same medical group. Although many hospital credentials committees require specialty board certification to provide specialty care, this is often overlooked because of local traditions, reluctance to restrict or offend current medical staff, and concern about potential financial implications if lack of hospital "specialists" results in redirection of certain patients to a competing hospital. National specialty societies representing surgeons are generally silent on the issue in an effort to avoid disenfranchising one or more of their constituencies. These well-intended but incongruous local incentives could be overcome by policy decisions by health care systems, payers, agencies evaluating quality, and government policy makers.
There is substantial precedent for such intervention, with payers requiring certain volume and specialty training for specific cancers [12], the Leapfrog Group setting volume standards for certain procedures [13], and the National Quality Forum establishing important processes of care and outcomes for thoracic surgery [14]. Yet none of these organizations has acted on the mounting evidence of the benefits of specialist care in thoracic surgery. Likewise, Medicare has set specific rigorous standards for the relatively rare procedure of lung volume reduction for emphysema [15, 16], yet has avoided creating such criteria for common procedures such as pulmonary resection for lung cancer, where the potential for beneficial patient impact is highest.
There are two disparate views on how to achieve standardization—selective referral vs "raising the tide." Selective referral involves the regionalization of care to specialty surgeons or centers of excellence, or both. Raising the tide may involve outcomes assessment and/or process compliance. An example of outcomes assessment would be surgeon or institutional enrollment in a national quality database, such as the Society of Thoracic Surgeons General Thoracic Database or the American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP). Process compliance might involve a pay-for-performance model for reimbursing surgeons based on adherence to practice guidelines. Both approaches to standardization share the common goal of providing equal, safe, and effective care to all patients.
Selective referral requires the health care infrastructure (ie, insurance coverage, patient transportation) and an adequate workforce of specialists to ensure adequate access to care. Selective referral runs the risk of an unintended consequence whereby some patients who may have once received suboptimal care might not receive any care at all. A 2002 survey of board-certified thoracic surgeons suggested that a sufficient specialty workforce exists to provide care for patients through 2012 [17], although the survey did not necessarily consider an increase in case-volume secondary to a policy of selective referral.
Risks notwithstanding, a key benefit of a selective referral policy is that it does not require an understanding of the modifiable processes and structures of care that presumably underlie the relationship between specialty care and better outcomes. As a result, improvements in outcome could be realized in a relatively short timeframe, with potentially lower costs and less resource utilization.
Raising the tide requires funding for national quality improvement database enrollment, scientific evidence (or least national consensus) to identify acceptable high-impact process measures, and access to and ability to perform important processes of care. Enrollment in a national database is expensive—the annual hospital costs for participating in ACS-NSQIP is estimated to exceed $100,000 [18]. Processes of care that might have an important effect on outcome may not be available to all surgeons, and not all surgeons may be able to safely perform potentially important processes of care—such as mediastinoscopy. Prohibitive costs and limited access to important processes of care may preclude some surgeons or hospitals from providing care.
A raise the tide approach to standardization does not implicitly specify a mechanism for patients to seek care elsewhere. Although the key benefit of a raise the tide approach is that it provides all surgeons and hospitals the opportunity to deliver high-quality care to patients, it fails to address those issues that may be inherent to specialty training, such as technical expertise and experience in a domain of patient disease.
Achieving standardized care may ultimately require a mix of selective referral, process compliance, and outcomes assessment. Among patients undergoing pulmonary resection for lung cancer by a general surgeon, 79% lived in a metropolitan area [9], where presumably a thoracic surgeon practiced. A hybrid policy would mandate selective referral to a thoracic surgeon in (metropolitan) areas where thoracic surgeons already exist. For surgeons working in areas where thoracic surgeons are less likely to practice, this policy would also subsidize enrollment in national outcomes databases or link reimbursement to compliance with practice guidelines, or both, such as those provided by the National Comprehensive Cancer Network.
Such a policy would reap the benefits of predominantly specialty surgical care (about 95% of all patients) without requiring changes in the existing surgical specialty workforce or massive overhauls in the health care infrastructure, especially those linked to transportation of patients. Patients in urban or rural areas would not be denied access to care on the basis of their closest surgeon's specialty, and generalist care in those areas would be held to the same standards as specialty care in more populated areas. Payers might have to be more inclusive of specialty providers in their payment plan, although they would have a financial incentive to do so given the reduced morbidity and, therefore, reduced cost of care by thoracic surgeons. A recent study of Medicare patients undergoing coronary artery bypass grafting observed a relationship between decreasing outlier payments to hospitals with lower risk-adjusted mortality rates, implicating a relationship between higher quality and lower costs [19].
The study by Schipper and colleagues has certain limitations [1]. After adjusting for surgeon case-volume, the favorable association between surgeon specialty and the risk of death was only apparent for decortication, and the favorable association between surgeon specialty and prolonged length of stay was observed for all procedures but lobectomy. The authors concluded that general thoracic and cardiac surgeons achieve better outcomes than general surgeons, but that differences in mortality may be explained by volume rather than surgeon specialty and that differences in morbidity are likely explained by both volume and surgeon specialty.
However, because the authors used case-load to define specialty in this report and because case-load is inextricably linked to surgical volume, we do not agree with their conclusion that volume is more likely to account for variation in outcomes than surgeon specialty. Methodologically, the problem is collinearity of variables. When two or more highly correlated variables are included in a multivariable model, none or only one may be associated with the outcome even though both are causally related to the outcome [20].
Schipper and colleagues [1] are to be congratulated for their contribution to a large body of evidence demonstrating a relationship between specialty surgeon care and optimal outcomes. No further research is needed to document this relationship because it is intuitive and well supported by the literature. Future studies might aim to evaluate the potential benefits and risks of varying approaches to standardizing surgical care, especially in terms of health care costs. Thoracic surgeons should move forward with advocacy and political action while patient groups, payers, quality agencies, and policy makers should aggressively support health care policies that standardize safety, quality, and efficacy of thoracic surgical care in the United States.
 |
References |
|---|
 Top References |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  P. H. Schipper and B. S. Diggs Reply. Ann. Thorac. Surg., September 1, 2010; 90(3): 1063 - 1063. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
