Risk of Not Being Discharged Home After Isolated Coronary Artery Bypass Graft Operations

Risk of Not Being Discharged Home After Isolated Coronary Artery Bypass Graft Operations James Edgerton, MD, Giovanni Filardo, PhD, MPH, William H. Ryan, MD, William T. Brinkman, MD, Robert L. Smith, MD, Robert F. Hebeler, Jr, MD, Baron Hamman, MD, Danielle M. Sass, MPH, CPH, Jessica P. Harbor, MS, and Michael J. Mack, MD The Heart Hospital Baylor Plano, Plano, Texas; Cardiopulmonary Research Science and Technology Institute, Medical City Dallas Hospital, Dallas, Texas; Institute for Health Care Research and Improvement, Baylor Health Care System, Dallas, Texas; Department of Statistical Science, Southern Methodist University, Dallas, Texas; Department of Infectious Diseases, University of Louisville, Louisville, Kentucky; and Baylor Heart and Vascular Institute, Baylor University Medical Center, Dallas, Texas Background. The age and risk profile of patients undergoing isolated coronary artery bypass grafting (CABG) is increasing, which will likely increase the proportion of CABG patients discharged to nursing homes, rehabilitation, or long-term care. Because discharge disposition can be important to a patient s treatment goals, developing and using predictive tools will improve informed treatment decision making. We examined the utility of The Society of Thoracic Surgeons (STS) risk of mortality score in predicting discharge disposition after CABG. Methods. From January 1, 2004 to October 31, 2011, 5,119 patients underwent isolated CABG at The Heart Hospital Baylor Plano or Baylor University Medical Center (Texas) and were discharged alive. The association between STS risk of mortality and discharge to nursing home, rehabilitation, or long-term care was assessed using multivariable logistic regression, adjusted for age, body surface area, marital status, site, and year of operation. Results. At discharge, 216 patients (4.21%) went to nursing homes, 153 (2.99%) to rehabilitation, and 115 (2.25%) to long-term care. The STS risk of mortality score was significantly positively associated with discharge status (p < 0.001). Patients with 1%, 2%, 3%, 4%, and 5% STS risk of mortality had 11.25%, 22.10%, 29.45%, 35.00%, and 38.50% probability, respectively, of not being discharged home. When the STS risk of mortality was 5%, the risk of not being discharged home was 47.9% for off-pump patients and 38.10% for on-pump patients. Conclusions. STS risk score is strongly associated with CABG discharge status. Patients with a risk score exceeding 2 are at high risk (>22%) of not being discharged home. This risk should be discussed when treatment decisions are being made. (Ann Thorac Surg 2013;96:1287 93) Ó 2013 by The Society of Thoracic Surgeons As the average age of patients undergoing coronary artery bypass graft (CABG) operations has increased [1], improved quality of life has come to rival or exceed increased survival as the patient s dominant goal for treatment [2 4]. Discharge to a nursing care facility is indicative of a decline in functional capacity and insufficient recovery after CABG for the individual to care for himself or herself independently [5]. As such, likelihood of this outcome is an important aspect of the patient s expected quality of life after CABG that should be part of the informed discussion of treatment options between physician and patient during the shared decision-making process [6]. Postdischarge needs for skilled nursing care are also important from a societal perspective: when looking at and predicting Accepted for publication May 13, 2013. Address correspondence to Dr Filardo, Institute for Health Care Research and Improvement, 8080 N Central Expy, Ste 900, Dallas, TX 75206; e-mail: giovanfi@baylorhealth.edu. health care resource utilization related to surgical revascularization, more than just in-hospital costs need to be considered [7]. Extensive work has been done to develop preoperative summary risk scores for perioperative mortality and morbidity in cardiac surgery [8 12], and these have shown some utility beyond their narrow intended purposes; for example, The Society of Thoracic Surgeons (STS) risk of mortality score shows high correlations with costs of care and length of stay when patients are grouped into cohorts of similar mortality risk [13]. Given the pressing need, as the CABG population ages, for insight into the functional status patients can expect after CABG and the health care resource utilization that should be anticipated in this population, we assessed and described (overall and in the subgroups of patients who underwent off-pump and on-pump CABG) the association between the STS risk of mortality score and being discharged to a nursing home, rehabilitation, or long-term care in more than 5,000 consecutive patients Ó 2013 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc http://dx.doi.org/10.1016/j.athoracsur.2013.05.042

1288 EDGERTON ET AL Ann Thorac Surg STS RISK SCORE AND DISCHARGE AFTER CABG 2013;96:1287 93 Table 1. Characteristics of 5,119 Patients Who Were Discharged Alive After Coronary Artery Bypass Graft Operations a Table 1. Continued Characteristics Percentage or Mean SD Characteristics Percentage or Mean SD Patients 5,119 (100) Demographics Age, y 64.59 10.41 Gender Male 74.96 Female 25.04 Married or with partner 67.32 Body surface area, m 2 2.83 0.34 Race White 81.79 Black 8.39 Hispanic 6.63 Other/unknown 3.20 Risk factors Diabetes mellitus 40.11 Renal failure 8.44 Creatinine, mg/dl 1.22 1.03 Chronic lung disease 14.68 Hypertension 82.87 Peripheral vascular disease 15.42 Cerebrovascular disease 13.72 Atrial fibrillation 17.72 Tobacco Use Never 60.54 Previous 13.17 Current 26.29 Congestive heart failure 18.31 STS risk of mortality score 2.12 2.99 Previous interventions PCI 28.27 Pacing ICD operation 3.89 CABG 7.29 Valve operation 0.25 Preoperative cardiac status Myocardial infarction timing None 0.37 6 h 1.55 >6 and <24 h 2.62 >24 h 95.45 Angina 78.69 Hemodynamics and catheterization Ejection fraction 0.4857 0.1450 Left main disease 29.82 Operative status Elective 58.16 Nonelective 41.84 Preoperative IABP 15.29 Discharge status to Long-term care 2.25 (Continued) Rehabilitation 2.99 Nursing home 4.21 a Operations were performed at The Heart Hospital Baylor Plano (Plano, TX) or Baylor University Medical Center (Dallas, TX) between January 2004 and October 2011. CABG ¼ coronary artery bypass grafting; IABP ¼ intraaortic balloon pump; ICD ¼ implantable cardioversion defibrillator; PCI ¼ percutaneous coronary intervention; SD ¼ standard deviation; STS ¼ Society of Thoracic Surgeons. who underwent isolated CABG at our institution over the course of 8 years. Patients and Methods This study was approved by the Baylor Research Institute Institutional Review Board. Patient Data Consecutive patients who underwent isolated CABG surgery at The Heart Hospital Baylor Plano (Plano, TX) or Baylor University Medical Center (Dallas, TX) between January 1, 2004, and October 31, 2011, and were discharged alive, were considered for this study. The patient cohort was limited to those without preoperative endocarditis. Risk Factors Patients data on the risk factors considered in the present study were collected by both cardiac programs and entered into the STS Adult Cardiac Surgery Database [14], which includes more than 95% of the cardiac surgery programs of the United States. Data abstraction, collection procedures, and definitions of variables used for the STS Adult Cardiac Surgery Database are standardized and have been described elsewhere [14, 15]. The STS risk of operative mortality [11], based on all STS established risk factors, was considered for the present study as a surrogate of the patient s general health before the operation. Outcome The discharge status of the 5,119 patients was assessed using the STS Adult Cardiac Surgery Database and hospital administrative data. The present study considered the composite outcome of discharge to nursing home, rehabilitation facility, or long-term care. Statistical Analysis Means, standard deviations, medians, interquartile ranges (IQR), and frequencies were calculated to describe the patients case-mix. A multivariable logistic regression model [16] was used to assess the adjusted association between the STS risk of mortality and the outcomes of interest. The model included the STS risk of mortality, age, body

Ann Thorac Surg EDGERTON ET AL 2013;96:1287 93 STS RISK SCORE AND DISCHARGE AFTER CABG 1289 Fig 1. Plot depicts the adjusted (by age, body surface area, marital status, site and year of operation) association between The Society of Thoracic Surgeons (STS) risk of mortality score and the estimated probability (and 95% confidence interval, gray area) of discharge to a nursing home, rehabilitation facility, or long-term care facility in 5,119 patients who underwent coronary artery bypass graft operations at The Heart Hospital Baylor Plano (Plano, TX) or Baylor University Medical Center (Dallas, TX) between January 2004 and October 2011 and were discharged alive. Adjusted p < 0.001. surface area, marital status, site where the operation was performed, and year of operation as independent variables; the dependent variable was discharge to nursing home, rehabilitation, or long-term care vs discharge to home. Restricted cubic splines were used for all continuous variables [17, 18]. No missing data were identified for any of the variables used in the models. The analysis was repeated for the subgroups of patients who underwent off-pump and on-pump CABG. All analyses were performed using R 2.14.1 software (The R Project for Statistical Computing, http://www.r-project.org/). Results The final study cohort included 5,119 patients. Mean age was 64.5 years, and 1,282 patients (25%) were women. The median STS risk of mortality was 1.14% (IQR, 0.61%, 2.39%) and the median length of stay was 7 days (mean, 8.2 days; IQR, 5, 9 days). In this cohort, 216 patients (4.21%) were discharged to a nursing home, 153 (2.99%) to a rehabilitation center, and 115 (2.25%) to a long-term care facility. Characteristics of the study population are presented in Table 1. After adjustment by age, body surface area, marital status, and site and year of operation, the analysis showed a strong, statistically significant positive association between the STS risk of mortality score and the outcome of interest for the present study (p < 0.001, Fig 1). Patients with 1%, 2%, 3%, 4%, and 5% STS risk of mortality had 11.25%, 22.10%, 29.45%, 35.00%, and 38.50% probability, respectively, of being discharged to a nursing home, rehabilitation, or long-term care facility. Likewise, the adjusted associations between the other patient risk factors considered in the analyses (marital status [p < 0.001], age [p < 0.001, Figure 2], body surface area [p ¼ 0.002, Figure 2] and the estimated probability of being discharged to a nursing home/rehabilitation/longterm care facility were highly significant. This strong association persisted when the off-pump and on-pump groups were analyzed separately (p < 0.001, Fig 3). The risk of not being discharged home was similar for off-pump and on-pump patients, with an STS risk of mortality of between 0.1% and 3%. Among patients with a 5% STS risk of mortality, those who received off-pump CABG had 46.90% risk of not being discharge home and on-pump CABG patients had 38.10% risk. Off-pump patients with an STS risk of mortality of between 3.1% and 10.5% had greater risk of not being discharged home than on-pump patients with the same STS risk of mortality (Fig 3). For example, among patients with a 5% STS risk of mortality, those who underwent off-pump CABG had a 46.90% risk of not being discharge home compared with a 38.10% risk for those who underwent on-pump CABG. Comment Our study provides strong evidence that the STS risk of mortality score is associated with discharge to a nursing home, rehabilitation facility, or long-term care in patients undergoing isolated CABG. Patients with a STS risk score of 2% or greater are at substantial risk (>22.5%) of being unable to return home at discharge. Moreover, the risk of not being discharged home was similar for off-pump and on-pump patients with an STS risk of mortality between 0.1% and 3%. From this it is clear that even CABG patients who are at relatively low risk for operative death are quite likely to not experience an optimal outcome. The comparison of patients who underwent on-pump vs. off-pump CABG did show a significantly greater risk of not being discharged home for the latter among patients with an STS risk of mortality between 3.1% and 10.5%.

1290 EDGERTON ET AL Ann Thorac Surg STS RISK SCORE AND DISCHARGE AFTER CABG 2013;96:1287 93 Fig 2. Plots depict the adjusted (by The Society of Thoracic Surgeons [STS] risk of mortality score, marital status, site, and year of operation) association between (A) age and (B) body surface area, and the estimated probability (and 95% confidence interval, gray area) of discharge to a nursing home, rehabilitation facility, or long-term care facility in 5,119 patients who underwent coronary artery bypass graft operations at The Heart Hospital Baylor Plano (Plano, TX) or Baylor University Medical Center (Dallas, TX) between January 2004 and October 2011 and were discharged alive. Adjusted (by age, body surface area, marital status, site, and year of operation) p < 0.001 for age and p ¼ 0.002 for body surface area. Previous research has identified individual factors that increase CABG patients risk for needing skilled nursing care or other additional health resources after discharge, including older age, female sex, more comorbid conditions, concurrent valve operation, preadmission unemployment, and less optimism about surgical outcomes [19 21]. Development of a preoperative screening tool or predictive index has also been attempted, but with limited success [19, 22]. The first of these tools considered the preoperative presence of two or more comorbidities (ie, increased age, decreased ejection fraction, and multiple vessel disease), established durable medical supply needs, previous admission to home health care, and skilled home nursing needs as defined at the time by the Health Care Financing Administration [22]. When tested in 277 patients undergoing cardiothoracic operations, this screening tool correctly identified 95 of 102 patients (93%) as having no postacute care needs, but only 7 of 12 patients (58%) identified as needing home health care or subacute care after discharge and 44 of 163 patients (27%) identified as needing home health care [22]. The second such tool was a risk index developed specifically for CABG patients and based on a summation of nine preoperative International Classification of Diseases-9 Revision-Clinical Modification diagnoses: a patient was assigned 1 point each for the presence of osteoarthritis, congestive heart failure, atrial fibrillation, myocardial infarction, anemia, and obesity, and 2 points each for presence of chronic obstructive pulmonary

Ann Thorac Surg EDGERTON ET AL 2013;96:1287 93 STS RISK SCORE AND DISCHARGE AFTER CABG 1291 Fig 3. Plot depicts the adjusted (by age, body surface area, marital status, site and year of operation) association between The Society of Thoracic Surgeons (STS) risk of mortality score and the estimated probability of discharge to a nursing home, rehabilitation facility, or long-term care facility by use of on-pump (red) or off-pump (blue) techniques in 5,119 patients who underwent coronary artery bypass graft operations at The Heart Hospital Baylor Plano (Plano, TX) or Baylor University Medical Center (Dallas, TX) between January 2004 and October 2011 and were discharged alive. Adjusted (by Society of Thoracic Surgeons risk of mortality score, age, body surface area, marital status, site, and year of operation) p < 0.001 for both offpump and on-pump patients. disease, renal failure, and female sex [19]. Patients with a total of 3 or more points were considered high risk for skilled nursing facility admission after CABG, with a demonstrated odds ratio for patients with a score exceeding this cut point to those below of 2.34 (no confidence interval reported) in a study population of 26,040 CABG patients aged 65 years or older [19]. Our findings augment the aforementioned established evidence with novel data regarding the strong association between the STS risk of mortality score and isolated CABG patients negative discharge outcomes such as being discharged to a nursing home, rehabilitation facility, or long-term care. Accordingly, as shared decisionmaking tools and processes are developed for elective CABG, researchers should consider using the STS risk of mortality score to assess patients risks of not being discharged home. Some study limitations should be noted. First, the study was conducted at 2 centers in Texas (Dallas and Plano), which may limit the generalizability of the results. In addition, we cannot eliminate the possibility that the association between exposure (STS risk of mortality) and outcome (discharge disposition) is confounded by other unknown factors, as is the case in any observational study. However, the STS score incorporates an extensive list of established risk factors [11]. Lastly, we did not have data on patient residence at admission (eg, primary residence is a nursing home), or on patients length of stay in facilities to which they were discharged. Our estimates of the rate of discharge to nursing home may be slightly inflated if some patients discharged to nursing homes were already long-term residents of those nursing homes and were simply discharged back to their residence. However, it is also conceivable that patients residing in nursing home had higher STS risk scores, and therefore, our results regarding the inference of lower STS risk score (eg, <5%) should not be compromised. Further research is needed to determine patients length of stay in the nursing homes or rehabilitation facilities to which they are discharged after CABG and subsequent fitness and quality of life. If these facilities are merely providing short-term transitional care as a result of shortened acute care hospital stays [21, 23], with patients returning rapidly to their homes and regular activities, discharge disposition may be an inaccurate marker of the post-cabg quality of life patients can expect. In conclusion, the STS risk of mortality score is strongly associated with discharge disposition after CABG, and developers of shared decision-making tools should consider using it in this context. Because improved quality of life is frequently a top priority for patients undergoing CABG, information on the patient s risk of requiring admission to a nursing home or longterm care facility after the operation is an important part of a truly informed, patient-centered decisionmaking process. We acknowledge the use of software from Professor Frank Harrell s Hmisc and Design libraries and thank Briget da Graca for writing and editorial assistance. Grant support was provided by the Cardiovascular Research Review Committee in cooperation with the Baylor Heart and Vascular Institute, by the Discovery Foundation, and by The Bradley Family Endowment through the Baylor Health Care Foundation. References 1. Ferguson TB Jr, Hammill BG, Peterson ED, DeLong ER, Grover FL. A decade of change risk profiles and outcomes

1292 EDGERTON ET AL Ann Thorac Surg STS RISK SCORE AND DISCHARGE AFTER CABG 2013;96:1287 93 for isolated coronary artery bypass grafting procedures, 1990 1999: a report from the STS National Database Committee and the Duke Clinical Research Institute. Society of Thoracic Surgeons. Ann Thorac Surg 2002;73:480 9. discussion 489 90. 2. Herlitz J, Wiklund I, Sjoland H, et al. Impact of age on improvement in health-related quality of life 5 years after coronary artery bypass grafting. Scand J Rehabil Med 2000;32: 41 8. 3. Jokinen JJ, Hippelainen MJ, Turpeinen AK, Pitkanen O, Hartikainen JE. Health-related quality of life after coronary artery bypass grafting: a review of randomized controlled trials. J Card Surg 2010;25:309 17. 4. Fried TR, Bradley EH, Towle VR, Allore H. Understanding the treatment preferences of seriously ill patients. N Engl J Med 2002;346:1061 6. 5. Wu AW, Yasui Y, Alzola C, et al. Predicting functional status outcomes in hospitalized patients aged 80 years and older. J Am Geriatr Soc 2000;48:S6 15. 6. Rady MY, Johnson DJ. Cardiac surgery for octogenarians: is it an informed decision? Am Heart J 2004;147:347 53. 7. Nallamothu BK, Rogers MA, Saint S, et al. Skilled care requirements for elderly patients after coronary artery bypass grafting. J Am Geriatr Soc 2005;53:1133 7. 8. Hannan EL, Wu C, Bennett EV, et al. Risk stratification of inhospital mortality for coronary artery bypass graft surgery. J Am Coll Cardiol 2006;47:661 8. 9. Higgins TL, Estafanous FG, Loop FD, Beck GJ, Blum JM, Paranandi L. Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients. A clinical severity score. JAMA 1992;267:2344 8. 10. Ivanov J, Borger MA, Rao V, David TE. The Toronto Risk Score for adverse events following cardiac surgery. Can J Cardiol 2006;22:221 7. 11. Shahian DM, O Brien SM, Filardo G, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1 coronary artery bypass grafting surgery. Ann Thorac Surg 2009;88:S2 22. 12. Tu JV, Jaglal SB, Naylor CD. Multicenter validation of a risk index for mortality, intensive care unit stay, and overall hospital length of stay after cardiac surgery. Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Circulation 1995;91:677 84. 13. Riordan CJ, Engoren M, Zacharias A, et al. Resource utilization in coronary artery bypass operation: does surgical risk predict cost? Ann Thorac Surg 2000;69:1092 7. 14. Ferguson TB Jr, Dziuban SW Jr, Edwards FH, et al. The STS National Database: current changes and challenges for the new millennium. Committee to Establish a National Database in Cardiothoracic Surgery, The Society of Thoracic Surgeons. Ann Thorac Surg 2000;69:680 91. 15. Filardo G, Hamilton C, Grayburn PA, Xu H, Hebeler RF Jr, Hamman B. Established preoperative risk factors do not predict long-term survival in isolated coronary artery bypass grafting patients. Ann Thorac Surg 2012;93:1943 8. 16. Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996;15:361 87. 17. Harrell FE Jr. Regression modeling strategies: with application to linear models, logistic regression, and survival analysis. New York: Springer-Verlag; 2001. 18. Filardo G, Hamilton C, Hamman B, Ng HK, Grayburn P. Categorizing BMI may lead to biased results in studies investigating in-hospital mortality after isolated CABG. J Clin Epidemiol 2007;60:1132 9. 19. Chang DC, Joyce DL, Shoher A, Yuh DD. Simple index to predict likelihood of skilled nursing facility admission after coronary artery bypass grafting among older patients. Ann Thorac Surg 2007;84:829 34. discussion 834 5. 20. Sawatzky JA, Naimark BJ. Coronary artery bypass graft surgery: exploring a broader perspective of risks and outcomes. J Cardiovasc Nurs 2009;24:198 206. 21. Swaminathan M, Phillips-Bute BG, Patel UD, et al. Increasing healthcare resource utilization after coronary artery bypass graft surgery in the United States. Circ Cardiovasc Qual Outcomes 2009;2:305 12. 22. Boyer CL, Wade DC, Madigan EA. Prescreening cardiothoracic surgical patient population for post acute care services. Outcomes Manag Nurs Pract 2000;4:167 71. 23. Bohmer RM, Newell J, Torchiana DF. The effect of decreasing length of stay on discharge destination and readmission after coronary bypass operation. Surgery 2002;132:10 5. INVITED COMMENTARY In this observational study from two cardiac centers, Edgerton and colleagues [1] investigate in a retrospective fashion the association between The Society of Thoracic Surgeons (STS) risk score of mortality and the probability of being discharged to a skilled nursing facility among coronary artery bypass graft surgery (CABG) patients. This included nursing homes, rehabilitation, and long-term care facilities. A striking association between STS score and discharge status is reported. Patients with a risk score of 5% were associated with a 38.5% probability of not going home, as opposed to 11.25% for patients with a score of 1%. Surprisingly, for off-pump CABG patients with a risk score of 5%, the probability of not going home increased significantly to 46.9%. The findings of this study, therefore, represent an important tool for a serious preoperative discussion with patients and their families about the probability of discharge status after surgery. The low operative mortality associated with CABG surgery is no longer the main focus of discussion. Rather, the short-term and long-term quality of life assumes center stage. Consequently, the results of this and other similar studies become increasingly important. The most striking result of this study is that only less than 10% of their patients were not discharged home. This a very low percentage compared to the experience of most US centers. They are to be congratulated for achieving these results. I wonder if the results of this study would be the same had 20% of these patients not gone home. I strongly agree that data on length of stay in discharge facilities are important in understanding the continuum of care required for these patients. Unfortunately, these data were not available, as discussed in the limitations of the study. A prospective study focusing on this aspect of care is timely and may eventually help in predicting which patients would have no improvement or deterioration in their quality of life. Finally, the findings regarding off-pump CABG appear to contradict the widely held belief that those patients Ó 2013 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc http://dx.doi.org/10.1016/j.athoracsur.2013.06.032