Ann Thorac Surg 2006;81:1393-1395
© 2006 The Society of Thoracic Surgeons
Original article: Cardiovascular
Comparison of Three Measurements of Cardiac Surgery Mortality for the Northern New England Cardiovascular Disease Study Group
Donald S. Likosky, PhD
a
,
*
,
William C. Nugent, MD
a
,
Robert A. Clough, MD
b
,
Paul W. Weldner, MD
c
,
Hebe B. Quinton, MS
a
,
Cathy S. Ross, MS
a
,
Gerald T. O'Connor, PhD, DSc
a
a Departments of Surgery, Medicine, and Community and Family Medicine, Dartmouth Medical School, Hanover, New Hampshire
b Department of Surgery, Eastern Maine Medical Center, Bangor, Maine
c Department of Surgery, Maine Medical Center, Portland, Maine
Accepted for publication November 28, 2005.
* Address correspondence to Dr Likosky, Department of Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756 (Email: likosky{at}dartmouth.edu).
 |
Abstract
|
|---|
BACKGROUND: There is no consensus on the optimal period during which to assess death after coronary artery bypass graft (CABG) surgery. Three measures are commonly used: in-hospital, 30-day, and procedural (either in-hospital or 30-day) mortality. We used a regional database to calculate the CABG mortality rate using each of these mortality measures.
METHODS: Data were collected prospectively on 31,592 consecutive isolated CABG surgeries in northern New England between January 1992 and December 2001. These data were linked to the National Death Index to obtain vital status through December 2001, and used to calculate 30-day and procedural mortality rates. Procedural mortality was defined as death occurring either within the hospital setting or within 30 days of the index procedure. Regional registry data were used to calculate in-hospital mortality rates.
RESULTS: Mortality rates and their 95% confidence intervals were calculated. In all but one medical center, the in-hospital mortality was the lowest rate, while in all centers the procedural mortality rate was the highest. There were 1,082 deaths captured by the procedural mortality measure. Of these, 927 were included in the in-hospital mortality measure; 956 occurred within 30 days of surgery.
CONCLUSIONS: Each of the measures studied has its advantages and may be used to assess the mortality outcomes of cardiac surgery. The more important issue other than the specific measure used is our ability to measure and validate it conveniently and accurately in actual practice.
|
Introduction
|
|---|
Coronary artery bypass graft (CABG) surgery is one of the most frequently performed major surgical procedures. Death associated with cardiac surgery is arguably one of the most carefully observed outcome measures in this setting. Information regarding mortality rates is of interest to patients, families, clinicians, and other local, regional, and national stakeholders. This information may be used for patient decision making regarding treatment options, quality improvement efforts, and marketing [1].
Currently, consensus does not exist regarding the optimal period for assessing mortality associated with CABG surgery. In-hospital mortality is easily captured and validated [2]. Short-term outcomes, such as 30-day and procedural mortality (within hospital or 30 days), are also of interest, and remain the most widely reported metrics [3]. While any decision regarding the chosen metric is likely dependent on its intended use, comparisons of these metrics have not been made to date.
The Northern New England Cardiovascular Disease Study Group (NNECDSG) is a voluntary research consortium composed of clinicians, research scientists, and hospital administrators. The goal of the group is to foster the continuous improvement in the quality of clinical care through the analysis of process and outcomes data and the timely feedback of information. The aim of this study was to use a regional database to calculate and compare the CABG mortality rates using each of these measures.
|
Material and Methods
|
|---|
Data Collection
We conducted a prospective cohort study of mortality associated with CABG surgery at all NNECDSG medical centers, using data collected on 31,592 consecutive patients undergoing isolated CABG surgery. The medical centers involved in this study were: Catholic Medical Center (Manchester, New Hampshire), Concord Hospital (Concord, New Hampshire), Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire), Eastern Maine Medical Center (Bangor, Maine), Fletcher Allen Health Care (Burlington, Vermont), Maine Medical Center (Portland, Maine), and Portsmouth Regional Hospital (Portsmouth, New Hampshire). Two medical centers contributed data only for years 1998 through 2001.
In-hospital mortality was available in the registry, and was defined as death occurring before discharge from the index hospitalization. Thirty-day mortality was defined as death occurring within 30 days of the index procedure, regardless of whether the patient has or has not been discharged from the hospital. Procedural mortality was defined as death occurring either within the index hospitalization or within 30 days of the index procedure.
Patient Follow-Up
The National Death Index (US Department of Health and Human Services) was used to assess long-term mortality and death-certificate–coded cause of death for the patient population through December 2001. Probabilistic matching was performed based on patient name, Social Security number, date of birth, sex, date last known alive, and last known state of residence. The sensitivity is between 92% and 99% depending on which identifiers are available [4, 5].
Statistical Analysis
All statistics were performed using the STATA 8.0 program [6]. Logistic regression was used to calculate predicted mortality and adjusted mortality rates. Variables included in the model were age, sex, left ventricular ejection fraction, priority of surgery, history of CABG, vascular disease, diabetes mellitus, and renal failure requiring dialysis or creatinine of 2 mg/dL or greater. Complete definitions of these variables have been published previously [7]. Differences in definitions of mortality rates were tested using the McNemar test for repeated dichotomous data [8].
|
Results
|
|---|
In these data, there were a total of 1,082 procedural deaths (mortality rate 3.42%) occurring during the index admission subsequent to isolated CABG surgery or during the 30 days after the procedure. There were 927 deaths (mortality rate, 2.94%) occurring within the index hospitalization, and 956 deaths (mortality rate, 3.03%) within 30 days of the procedure (Table 1).
As expected, procedural mortality rates were higher than both in-hospital and 30-day mortality rates. This result was seen across all of our medical centers, and persisted throughout the study period (Table 2, Fig 1), even after adjusting for patient and disease characteristics.
View larger version (13K):
[in this window]
[in a new window]
|
Fig 1. Comparison of three mortality metrics by year of operation. (In-hospital = light gray line; 30-day = medium gray line; procedural = black line.)
|
|
In-hospital mortality rates were not statistically different than 30-day rates (
2 = 2.99, p = 0.836). Procedural mortality rates were statistically different than both in-hospital (2 = 155.00, p < 0.001) and 30-day rates (2 = 126.00, p < 0.001).
|
Comment
|
|---|
We calculated CABG mortality rates using three commonly used measures and found significant differences between two of the three metrics examined. Overall, in-hospital mortality will capture the fewest deaths and procedural mortality will capture the most deaths that are reasonably related to cardiac surgery.
It might initially seem surprising that in some cases a hospital's in-hospital mortality rates exceed another hospital's 30-day mortality rates. In-hospital mortality might exceed 30-day in those instances when the patient dies beyond the 30 days while still in the hospital index admission. For instance, in our own registry, 1 patient died 49 days after the index operation while still residing in the hospital. In this instance, the 30-day mortality definition identified the patient as being alive, whereas the in-hospital and procedural mortality definitions identified the patient as being dead.
We had information from four of our medical centers regarding the location in which patients surviving the index admission were discharged. Among the 3,668 patients (98%) discharged alive, 1,904 (51%) were transferred to home health services, 1,096 (29%) to home, and 378 (10%) to a skilled nursing facility. This distribution was persistent across nearly all centers. Differences in mortality measurements do not appear to be driven by variation in the location in which patients were discharged.
Traditionally, published mortality rates have focused on both in-hospital and 30-day mortality [9, 10]. The collection of in-hospital mortality is the most convenient of the mortality measures, and allows for its validation with administrative data sources. Others have favored longer intervals, such as 180-day mortality or the indexing of mortality to length of stay [11, 12]. These same authors, however, noted appropriately that the most pressing issue of all is completeness of follow-up data. While 30-day or 180-day mortality both provide a more comprehensive and clinically relevant picture regarding clinical outcomes, their potential value is hindered by the difficulty for individual practitioners to collect and subsequently validate the data elements. Validation of these out-of-hospital mortalities is uncertain without linkage with other data sources such as the National Death Index [4]. Thus, lack of resources devoted to data collection in addition to the transient nature of patients likely favors the use of in-hospital mortality as the primary outcome of interest.
The decision regarding which mortality measure to capture and report is an important one, and likely should be based in part on the purpose of the information and ease of its capture. Procedural mortality appears to provide the most comprehensive picture of the three measures. Our findings suggest that although in-hospital and 30-day mortality appear to provide nearly equivalent results, the most convenient and easily validated metric for cardiac-related mortality would be in-hospital mortality.
|
References
|
|---|
- Nugent WC, Niles NW, Schults W, et al. Increasing the value of cardiac carethe Dartmouth approach. Quality Letter for Healthcare Leaders 1994;6:53-57.[Medline]
- O'Connor GT, Plume SK, Olmstead EM, et al. The Northern New England Cardiovascular Disease Study Group A regional intervention to improve the hospital mortality associated with coronary artery bypass graft surgery[see comments] JAMA 1996;275:841-846.[Abstract/Free Full Text]
- Shroyer AL, Plomondon ME, Grover FL, Edwards FH. The 1996 coronary artery bypass risk modelthe Society of Thoracic Surgeons Adult Cardiac National Database. Ann Thorac Surg 1999;67:1205-1208.[Abstract/Free Full Text]
- Wentworth DN, Neaton JD, Rasmussen WL. An evaluation of the Social Security Administration master beneficiary record file and the National Death Index in the ascertainment of vital status Am J Public Health 1983;73:1270-1274.[Abstract/Free Full Text]
- Williams BC, Demitrack LB, Fries BE. The accuracy of the National Death Index when personal identifiers other than Social Security number are used Am J Public Health 1992;82:1145-1147.[Abstract/Free Full Text]
- STATA statistical software. release 8.0. College Station, TX: STATA Corporation; 2003.
- O'Connor GT, Plume SK, Olmstead EM, et al. The Northern New England Cardiovascular Disease Study Group A regional prospective study of in-hospital mortality associated with coronary artery bypass grafting[see comments] JAMA 1991;266:803-809.[Abstract/Free Full Text]
- Rothman KJ, Greenland S. Modern epidemiologyPhiladelphia: Lippincott-Raven; 1998. pp. xiii737.
- O'Connor GT, Plume SK, Olmstead EM, et al. Northern New England Cardiovascular Disease Study Group Multivariate prediction of in-hospital mortality associated with coronary artery bypass graft surgery[see comments] Circulation 1992;85:2110-2118.[Abstract/Free Full Text]
- Shroyer AL, Coombs LP, Peterson ED, et al. The Society of Thoracic Surgeons30-day operative mortality and morbidity risk models. Ann Thorac Surg 2003;75:1856-1865.[Abstract/Free Full Text]
- Osswald BR, Blackstone EH, Tochtermann U, Thomas G, Vahl CF, Hagl S. The meaning of early mortality after CABG Eur J Cardiothorac Surg 1999;15:401-407.[Abstract/Free Full Text]
- Osswald BR, Tochtermann U, Schweiger P, et al. Minimal early mortality in CABG—simply a question of surgical quality? Thorac Cardiovasc Surg 2002;50:276-280.[Medline]
This article has been cited by other articles:
|
|
|
|
 
R. Ricciardi, B. A. Virnig, J. W. Ogilvie Jr, P. S. Dahlberg, H. P. Selker, and N. N. Baxter
Volume-Outcome Relationship for Coronary Artery Bypass Grafting in an Era of Decreasing Volume
Arch Surg,
April 1, 2008;
143(4):
338 - 344.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
