|
|
||||||||
Ann Thorac Surg 2006;81:835-842
© 2006 The Society of Thoracic Surgeons
a Taipei Medical University, School of Health Care Administration, Tri-Service General Hospital, Taipei, Taiwan
b Division of Cardiovascular Surgery, Tri-Service General Hospital, Taipei, Taiwan
c Department of Anesthesiology, Tri-Service General Hospital, Taipei, Taiwan
d Department of Economics, National Taipei University, Taipei, Taiwan
Accepted for publication September 15, 2005.
* Address correspondence to Dr Lin, School of Health Care Administration, Taipei Medical University, 250 Wu-Hsing St, Taipei 110, Taiwan (Email: henry11111{at}tmu.edu.tw).
| Abstract |
|---|
|
|
|---|
METHODS: This study uses the Taiwan National Health Insurance Research Database covering the period 2000 to 2002, with the study sample comprising 9,895 first-time coronary artery bypass graft admissions, treated by 316 surgeons in 46 hospitals.
RESULTS: Of the sampled patients, 356 (3.6%) were discharged after death. Those patients treated by low-volume (150 cases) surgeons had significantly higher mortality rates than those treated by medium-volume (51100 cases) surgeons (7.0% vs 3.8%), high-volume (101150 cases) surgeons (7.0% vs 2.7%), or very-high-volume (
151 cases) surgeons (7.0% vs 3.2%). However, hospital coronary artery bypass graft volume alone is an insufficient predictor of hospital in-patient deaths (p = 0.078). The adjusted odds ratio of hospital in-patient deaths declined with increasing surgeon volume, with the odds of in-patient death for those patients treated by low-volume surgeons being 1.52 times those of medium-volume surgeons, 1.89 times those of high-volume surgeons, and 2.04 times those of very-high-volume surgeons.
CONCLUSIONS: We conclude that for all coronary artery bypass graft surgeries taking place in Taiwan, the skill and experience of individual surgeons is a more critical factor for patient outcome than either hospital equipment or surgical teams.
| Introduction |
|---|
|
|
|---|
Most of the prior studies on the association between CABG volume by healthcare providers and subsequent patient outcomes have been heavily reliant upon statewide samples or subpopulations of patients, and as such, have failed to present unequivocal conclusions. Furthermore, most of these studies were conducted on hospital-level volume alone, with very few seeking to examine the simultaneous contribution to patient outcomes from both hospital and surgeon volumes. The majority of the studies on CABG volume and subsequent patient outcomes have also lacked case-specific measures of either surgeon or hospital volumes within their dataset. All of these issues have hampered the efforts of both clinicians and policymakers alike, to optimize CABG patient outcomes through the simultaneous development of hospital-level and surgeon-level strategies.
Using three-year population-based data on Taiwan, this study sets out to examine the association between surgeon and hospital CABG surgery volume, and subsequent patient outcomes. We have found no other study on CABG volume, and subsequent patient outcomes, to be conducted within any Asian country, and indeed, we also have found no other CABG volume-outcome study outside of the United States.
| Material and Methods |
|---|
|
|
|---|
Study Sample
The study sample was identified from the database by a principal performed operational procedure International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) code 36.1036.20 (broadly defined as bypass anastomosis for heart revascularization). Of the six million in-patient records within the dataset covering the period of this study, 10,844 hospitalized patients had undergone CABG surgery. In order to limit our study sample to the adult population, all patients aged below 18 years (n = 15) were excluded from the dataset. In addition, our study sample was limited to those patients who had undergone first-time CABG surgeries only; ultimately, our study sample comprised 9,895 admissions.
Surgeon and Hospital CABG Volume Groups
Since unique physician and hospital identifiers are available within the NHIRD for each medical claim submitted, this enabled us to identify the same physician, or the same hospital, carrying out one or more CABG surgeries during our three-year study period. Surgeon volume was calculated by counting all claims for principal performed operational procedure ICD-9-CM code 36.10-36.20 submitted in 2000, 2001, or 2002.
In order to permit the comparison of our finding to the experiences of the US, we have used the same volume thresholds as those adopted in the US studies [6, 12]. The sample of 9,895 patients was divided into four surgeon volume groups: 50 or less cases (hereafter referred to as low volume), 51 to 100 cases (medium volume), 101 to 150 cases (high volume), and 151 or greater cases (very high volume), while the three hospital volume groups were 249 or less cases (low volume), 250 to 499 cases (medium volume), and 500 or greater cases (high volume).
Statistical Analysis
The SAS statistical package (SAS Institute, Cary, NC) was used to perform statistical analysis of the data. Global
2 analyses were conducted in order to examine the relationship between surgeon CABG volume groups and unadjusted hospital in-patient deaths. After adjusting for surgeon, patient, and hospital characteristics, multivariate logistic regression analyses were also employed to assess the independent association between surgeon CABG volume and hospital in-patient deaths.
Finally, generalized estimated equation (GEE) was also carried out in order to account for any clustering of the sampled patients among particular surgeons. Hospital in-patient deaths were denoted by "1," while live discharges were denoted by "0." We define in-patient deaths as "the death of a patient at any time after operation if the patient does not leave hospital." The primary study outcome was dichotomous, irrespective of whether or not a CABG surgery resulted in hospital in-patient death.
Surgeon characteristics included the surgeon's gender and age (as a surrogate for practice experience). Hospital characteristics included hospital ownership, hospital level, and geographic location. The hospital ownership variable was recorded as one of three types: public hospital, private not-for-profit (NFP), or private for-profit (FP). The hospital level variable classified each hospital as a medical center (with a minimum of 500 beds), a regional hospital (minimum 250 beds), or a district hospital (minimum 20 beds). Hospital level can therefore be used as a proxy for both hospital size and clinical service capabilities. Hospital teaching status was not included within the regressions since all medical centers and regional hospitals in Taiwan are teaching hospitals. In addition, given the relatively small number of cases in private FP hospitals, as well as hospitals located in eastern Taiwan, all of the private NFP and FP hospitals, and those hospitals located in central, southern, and eastern parts of Taiwan, were combined into a single category referred to as "others."
Patient characteristics comprised age, gender, and severity of illness. Since no illness severity index is currently available in Taiwan, we used the Charlson Comorbidity Index (CCI) to quantify preexisting comorbidity. The CCI was developed in 1987 by Charlson and colleagues [13] as a means of classifying comorbid conditions that might affect the risk of death from comorbidity disease, and it has been widely used for risk adjustment in administrative datasets [14].
In addition, after the method proposed by Rathore and colleagues [6], the following ICD-9-CM codes were adjusted: principal diagnosis of myocardial infarction (MI), secondary diagnosis of MI, any other non-MI coronary disease diagnosis, concomitant valve repair, and the use of an internal mammary graft. A two-sided p value of 0.05 or less was considered statistically significant.
| Results |
|---|
|
|
|---|
|
Patient, surgeon, and hospital characteristics, by surgeon CABG volume group, are summarized in Table 2, which shows that 316 surgeons performed the CABG surgical procedure between 2000 and 2002 at a mean volume per surgeon of 33 operations. Of these, 258 (81.7%) were in the low-volume group with 50 or less operations, while a further 21 (6.7%) were in the medium-volume group with 51 to 100 operations; 16 (5.1%) were in the high-volume group with 101 to 150 operations and 21 (6.7%) were in the very-high-volume group, with 151 or greater operations. The mean age of the patients was similar across all of the groups.
|
|
|
|
| Comment |
|---|
|
|
|---|
After adjusting for patient, surgeon, and hospital characteristics, we find that a significant inverse relationship exists between surgeon volume and hospital in-patient deaths; however, this study also finds that hospital volume is not a significant predictor of hospital in-patient deaths after CABG surgery. This finding suggests that in Taiwan the skill or experience of individual surgeons is a more critical factor than hospital equipment or surgical teams in determining patient outcomes after CABG surgery. This finding also comes in light of the conclusions of five earlier studies by Hannan and others [3, 5, 12, 16, 17], which took place in New York State during different time periods. The results of these studies have consistently reported that surgeon volume is a more significant factor than hospital volume in predicting patient outcomes after CABG surgery, and that hospital volume is only marginally related to patient outcomes.
The prior literature in this area proposes three possible hypotheses to explain the inverse volume-outcome relationship [1]. The first of these hypotheses, "practice makes perfect," is based upon the rationale that a larger volume of patients allows providers to develop better skills and expertise in the management of operations or treatment procedures. Therefore, high-volume providers are more likely to achieve better clinical performance due to their greater skills and experience. Under such a hypothesis, there is the likelihood that low-volume surgeons with poor outcomes can improve their clinical performance substantially by increasing their patient volume. However, we must remain cautious here because an increase in the volume of low-volume surgeons may lead to adverse effects, such as incentives being created for low-volume surgeons to lower artificially the threshold for CABG operations [18]. Furthermore, although it is difficult to refute the role that "practice makes perfect" has played in the system of healthcare delivery in Taiwan [19], we are unable to demonstrate through our cross-sectional study whether the volume-outcome relationship observed in this study can be fully explained by such a hypothesis.
The second hypothesis relates to "selective-referral," which suggests that selectively referring physicians or patients leads to the referral to providers of more patients with superior outcomes; thus, these providers would be performing a high volume of CABG procedures. The study of Luft and colleagues [7] confirmed that at least part of the volume-outcome relationship was attributable to physician referral or patient self-referral.
The findings of our study suggest that under this hypothesis, patients or referring physicians will be more inclined to move their patients toward surgeons with better patient outcomes, as opposed to those hospitals with superior reputations. In Taiwan, although physicians work in the same department within one hospital, they may, nevertheless, have graduated from different medical schools or have undertaken their residencies in different hospitals. Therefore, even within the same department, physicians could be practicing a variety of skills or procedures in CABG operations, which could well lead to very different patient outcomes.
"Self-referral" may also be a major factor contributing to the inverse relationship between the patient outcomes and surgeon volumes observed in Taiwan, particular since, in the absence of a referral system, Taiwanese consumers have the freedom to choose their preferred provider. Physicians with good reputations or superior outcomes will tend to attract a greater number of patients as a result of word-of-mouth recommendations from relatives or friends [20]. However, prior to any policy decisions being derived from this hypothesis, further longitudinal studies will be required to determine whether surgeons with better outcomes in the initial time period would subsequently acquire any greater volume of patients.
The third hypothesis on the inverse relationship between surgeon volume and outcomes is the difference in patient characteristics between low-volume and high-volume providers, particularly with regard to "severity of illness." Although this study has controlled for patient comorbidities, the administrative database used by this study is extremely limited in its ability to account for the differences in severity of coronary diseases among patients. Nevertheless, the studies by Tu and colleagues [21] and Jones and colleagues [22] have demonstrated that a relatively small number of clinical variables are sufficient to enable a fair comparison across hospitals of risk-adjusted mortality rates after CABG surgery. Moreover, the "severity of illness" hypothesis gains no support from the study of Shook and colleagues [23], which found that low-risk patients were more likely be treated by low-volume providers.
In addition to the abovementioned limitations, one additional caveat to this study should be noted. While some of the European countries and some of the states in the US have regulations limiting the number of providers allowed to perform certain procedures, the situation is quite different in Taiwan, with some surgeons having only very small CABG caseloads. Such small caseloads may prohibit meaningful statistical comparisons of the individual surgeons concerned.
Despite these limitations, this study has found that after adjusting for patient, surgeon, and hospital characteristics, an inverse volume-outcome relationship does exist for surgeons in Taiwan, but not for hospitals. Many studies have proposed feasible policy implications such as a regionalized or centralized CABG program, or even selective referral of CABG procedures to low-mortality providers based upon the volume-outcome relationship; however, low volume as an overall indicator of poor quality must be used with considerable caution. It is difficult to deny the existence of low-volume surgeons who provide excellent CABG surgery outcomes and high-volume surgeons who provide poor outcomes; indeed, the casual mechanisms linking volume and outcomes remain unclear. We suggest, therefore, that volume should be used merely as a screening measure in the first instance, while initiating a more thorough, in-depth review of provider performance. We also suggest that investigations should be undertaken to identify the differences in clinical approaches and techniques between high-volume surgeons with excellent outcomes and low-volume surgeons with poor outcomes; the results of such studies could help the latter to improve the quality of their patient care.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. D. Auerbach, J. F. Hilton, J. Maselli, P. S. Pekow, M. B. Rothberg, and P. K. Lindenauer Shop for Quality or Volume? Volume, Quality, and Outcomes of Coronary Artery Bypass Surgery Ann Intern Med, May 19, 2009; 150(10): 696 - 704. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-C. Lin, S. Xirasagar, C.-H. Chen, and Y.-T. Hwang Physician's Case Volume of Intensive Care Unit Pneumonia Admissions and In-Hospital Mortality Am. J. Respir. Crit. Care Med., May 1, 2008; 177(9): 989 - 994. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-C. Lin, S. Xirasagar, N.-W. Tsao, Y.-T. Hwang, N.-W. Kuo, and H.-C. Lee Volume-outcome relationships in coronary artery bypass graft surgery patients: 5-year major cardiovascular event outcomes. J. Thorac. Cardiovasc. Surg., April 1, 2008; 135(4): 923 - 930. [Abstract] [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 |