Fischer, John P. M.D.; Wes, Ari M. B.A.; Nelson, Jonas A. M.D.; Serletti, Joseph M. M.D.; Kovach, Stephen J. M.D.
Readmission following surgical procedures can be common and extremely costly for health care systems.1,2 Upward of 20 percent of patients undergoing inpatient surgery will experience readmission, accounting for $15 billion per year in health care expenditures.2 Emerging legislation and reform may adversely impact and even penalize institutions for any 90-day, unplanned readmissions at a rate of 3 percent cost.3 Unplanned 30-day readmission rates may be closely scrutinized as an area for improvement, thereby placing pressure on institutions to reduce rates.
Efforts to minimize the incidence of readmission, through a deeper understanding of associated risk factors, may permit a more efficient delivery of evidence-based interventions and may optimize cost containment in higher risk subgroups.4–7 Readmission can be unrelated to the acute surgical admission or even be planned; however, in this current work, we only evaluate unplanned readmissions related to the initial surgical procedure. In this study, we explore factors associated with readmission following plastic surgery procedures using a prospective, validated, national database.8 The morbidity and inconvenience of readmissions for patients coupled with the clear economic disincentives of readmission underscore the critical importance of delineating identifiable risk factors and deriving evidence-based interventions to reduce occurrence.
PATIENTS AND METHODS
Patients who underwent primary plastic surgery procedures (n = 10,669) were identified from the 2011 American College of Surgeons National Surgical Quality Improvement Program databases.8 Both inpatient and outpatient cases were identified by the variable “surgspec” being “plastic surgery.” Any patient younger than 18 years was excluded, but all other identified patients were included in this current study. American College of Surgeons National Surgical Quality Improvement Program data were collected by trained research nurses at each institution using a systematic sampling of general and vascular operations performed in each participating institution. Results from audits completed to date reveal a disagreement rate of 1.8 percent for program variables. Each data set contains 240 Health Insurance Portability and Accountability Act–compliant variables for each case encounter, including patient demographics, preoperative risk factors, baseline comorbidities, intraoperative variables, and 30-day postoperative morbidity and mortality. The list and definitions of variables collected in the database can be found at the American College of Surgeons National Surgical Quality Improvement Program Web site (http://site.acsnsqip.org/). Patients are contacted by letter or telephone survey after discharge to ensure a full 30-day follow-up period. Importantly, the 2011 American College of Surgeons National Surgical Quality Improvement Program data set contains a “readmission” variable, which represents unplanned surgical readmission. The reliability and accuracy of the readmission variable in the American College of Surgeons National Surgical Quality Improvement Program data set has recently been validated.9
Data were accessed on December 1, 2012. In addition to the predefined National Surgical Quality Improvement Program variables, we calculated body mass index (in kilograms per square meter) for each encounter and categorized patients by the World Health Organization obesity classification. Encounters were defined as follows: nonobese (body mass index < 30 kg/m2), class I obesity (body mass index of 30 to 34.9 kg/m2), class II obesity (body mass index of 34.9 to 39.9 kg/m2), and class III obesity (body mass index ≥ 40 kg/m2).10,11 For the purposes of this current study, we also characterized hypoalbuminemia as an albumin value less than 3.5 g/dl.12 Anemia was also defined as hemoglobin less than 12 g/dl in women and less than 13 g/dl in men.13 We also included multiple comorbidities, which we defined as the presence of two or more American College of Surgeons National Surgical Quality Improvement Program comorbidities. Procedures were categorized into several general groups for further analysis. Manual assessment of Current Procedural Terminology along with procedural descriptions was performed. The following categories were developed for procedure subtype: débridement, local wound reconstruction, head and neck, body contouring, decubitus ulcer management, elective or cosmetic breast, implant breast reconstruction, autologous breast reconstruction, revision breast procedure, free flap (nonbreast), craniofacial trauma and reconstruction, hand surgery, muscle flaps, and concurrent procedures.
The identified cohort was divided into two groups: those who were readmitted and those who were not. Comparisons were made with respect to perioperative American College of Surgeons National Surgical Quality Improvement Program variables between the two groups. Several variables were defined by combining defined American College of Surgeons National Surgical Quality Improvement Program variables. We characterized a major surgical complication as a deep wound infection, graft or prosthetic loss, or an unplanned return to the operating room within 30 days. Medical complications included defined American College of Surgeons National Surgical Quality Improvement Program endpoints, such as pneumonia, pulmonary embolism, postoperative renal insufficiency (creatinine > 2 mg/dl), urinary tract infection, stroke, myocardial infarction, symptomatic deep venous thrombosis, sepsis, or septic shock. Wound complications were defined as superficial surgical-site infections or wound dehiscence.
Patients experiencing readmission and those who did not were subject to univariate analysis. The dependent variable was readmission. A variety of American College of Surgeons National Surgical Quality Improvement Program perioperative risk factors and patient comorbidities were used in our analyses. These included baseline health characteristics, reconstructive modality, preoperative laboratory values, and operative characteristics.
Pearson chi-square or Fisher’s exact test were used to analyze categorical variables; unpaired t test or Mann-Whitney tests were used for continuous variables. Variables with a value of p ≤ 0.10 in univariate analysis were used as independent variables in a logistic regression. All tests were two-tailed, and statistical significance was defined as p < 0.05. All analyses were performed using Stata IC 11.0 (StataCorp, College Station, Texas).
A total of 10,669 patients were included, with a 4.5 percent readmission rate. The study cohort was on average 49.5 years of age, 32.2 percent were obese, and 15.2 percent were smokers. The majority of patients were female (81.7 percent), Caucasian (70.1 percent), and admitted or came in from home (97.4 percent). Cases were predominantly outpatient procedures (65.6 percent) and 98 percent were nonemergent. The most commonly performed procedure types included elective or cosmetic breast (23.4 percent), implant breast reconstruction (16.5 percent), revision breast procedures (14.9 percent), hand operations (9.7 percent), and body contouring (5.9 percent). The average number of relative value units per case was 14.7 ± 8.2. A summary of patient characteristics and comorbidities can be found in Tables 1 and 2, respectively.
Most patients were healthy, with only 6.0 percent having more than two defined National Surgical Quality Improvement Program comorbidities. Anemia (17.6 percent) and obesity (32.2 percent) were the two most commonly observed comorbidities (Table 2). The majority of cases were performed under general anesthesia (94.0 percent) (Table 3). Perioperative 30-day mortality was 0.3 percent. The average operative time was 2.5 hours. Wound class was most commonly defined as clean wounds (84.1 percent). The incidence of wound complications was 4.6 percent, and the incidence of medical complications was 4.9 percent. The overall incidence of any postoperative complication was 10.9 percent, of which 4.8 percent were defined as major surgical complications. The average length of stay was 2.0 days.
Univariate analysis of various perioperative risk factors was performed comparing patients who experienced perioperative readmission to those who did not (Table 4). Readmission was more common in male patients (p < 0.001), inpatient procedures (p < 0.001), higher relative value unit procedures (p < 0.001), and in patients not admitted from home (p < 0.001) (Table 4). Procedure type was found in univariate analysis to be associated with differential rates of readmission (p < 0.001) (Table 4). Specifically, in evaluating patients who experienced readmission, there was a disproportionately higher incidence of patients undergoing muscle flap (12.8 percent), local wound reconstruction (6.3 percent), and débridement (12.0 percent), in contrast to those undergoing elective or cosmetic procedures (9.1 percent), hand surgery (3.4 percent), or revision breast operations (7.2 percent). These findings underscore the impact that operative procedure type has on readmission risk.
Analysis of preoperative patient comorbidities revealed several factors associated with readmission (Table 5). Older patients (p < 0.0001), obese patients (p < 0.001), diabetics (p < 0.001), and those with dyspnea at rest (p < 0.001) or impaired functional status (p < 0.0001) were at greater risk for readmission. In addition, patients with cardiac risk factors (p < 0.001), renal failure on dialysis (p < 0.001), preoperative open wounds (p < 0.001), recent weight loss (p < 0.001), preoperative sepsis (p < 0.001), recent operations (p < 0.001), malnutrition (p < 0.001), or anemia (p < 0.001) were also at significantly higher risk for readmission. Patients with contaminated wounds (p < 0.0001), higher American Society of Anesthesiologists physical class (p < 0.0001), longer operative times (p = 0.01), and those experiencing medical (p < 0.0001) or surgical (p < 0.0001) complications were also at greater risk for readmission (Table 6).
A multivariate regression analysis demonstrated several independent risk factors associated with readmission: procedure type (odds ratio, 1.0; p = 0.029); obesity (odds ratio, 1.2; p = 0.011); anemia (odds ratio, 1.8; p = 0.003); and medical (odds ratio, 2.3; p < 0.001), major surgical (odds ratio, 6.3; p < 0.001), and wound (odds ratio, 5.3; p < 0.001) complications (Table 7).
This is the first large, prospective, population-based study evaluating risk factors for readmission following plastic surgery procedures. Identifiable preoperative patient characteristics found to be associated with readmission in this study included obesity and anemia. The data also demonstrate a procedure-specific risk for readmission following plastic surgery interventions; it appears that certain reconstructive procedures such as débridements, muscle flaps, and tissue rearrangements are associated with higher rates of readmission compared with elective breast and hand surgery procedures. The critical finding of this current study, however, is that postoperative complications were predictive of readmission. Patients experiencing postoperative surgical complications were six times more likely to experience readmission. This finding highlights the potential benefit of elucidating perioperative risk factors for complications and morbidity, as both appear to be directly linked to unplanned readmission risk. These findings can assist surgeons and health care systems to better tailor preoperative counseling, resource allocation, and postoperative discharge services in an effort to minimize readmission risk in specific cohorts of patients.
Although the data set used in this analysis is by design heterogeneous in nature, it represents a broad sampling of plastic surgery patients. This, in our opinion, enhances the generalizability of the results across procedure types. Up-front limitations to this study that must be considered are that the occurrence of a complication cannot always be temporally assessed relative to the initial procedure, thus making it difficult to determine whether a complication was not adequately managed and then the patient was readmitted or whether the patient developed a complication as an outpatient and returned to the inpatient setting. The limitations must be recognized in interpreting the results of this analysis.
This study is a review of over 10,000 plastic surgery procedures during 2011 derived from the American College of Surgeons National Surgical Quality Improvement Program data sets, which as of 2011 include a validated and reliable variable for unplanned hospital readmission.9 We performed an analysis of risk factors for patients who experienced an unplanned readmission (n = 475) with the hypothesis that perioperative complications would be associated with a higher rate of readmissions. In assessing our results, it is clear that several risk factors are independently associated with readmission and that complications both surgical and medical are key predictors. These findings merit further discussion.
Obesity has not been found to be a risk factor for readmission following ambulatory surgery14; however, our current work suggests that obesity may be a factor associated with readmission in a mix of inpatient and ambulatory procedures considered collectively. Recent work by Reinke et al. demonstrated that obese patients were at greater risk for readmission when controlling for all other variables using a large set of Medicare data.15 This new finding is significant because obesity is prevalent and is also established as an overall strong predictor of perioperative morbidity and overall cost.16
Emerging evidence suggests that postoperative complications are strongly associated with risk of readmission17–21; thus, strategies designed to mitigate significant perioperative complications may exert a beneficial downstream effect on readmissions, cost, and resource use. Postoperative complications are becoming an established risk factor for readmission following surgical procedures. As an example, readmissions following elective spine surgery are related to postoperative complications,22 and both postoperative surgical (infection) and medical (heart failure) complications are established as common reasons for readmission following cardiac surgery.23 Kassin et al. report that postoperative complications following general surgery procedures were the most significant predictor of readmission (odds ratio, 4.2) using the National Surgical Quality Improvement Program data sets.20 These studies highlight the importance of early recognition and management of perioperative complications, further emphasizing the need for complication-reducing interventions, such as improved patient selection. Our current study of a large, relatively heterogeneous population of both cosmetic and reconstructive patients demonstrated that perioperative surgical complication was the key predictor of readmission. This may seem rather intuitive, but it is a key finding that underscores the importance of risk-reduction strategies in containing downstream costs associated with readmission and providing optimal outpatient services and follow-up for higher risk cohorts.
A recent review from Shepperd et al. demonstrated that structured discharge planning in elderly patients may reduce length of stay and readmission rates.24 Postoperative, unplanned readmission cost can account for greater than 50 percent of the index admission costs following certain complicated procedures such as abdominal aortic aneurysm repairs.19 Such findings confirm the significant economic burden that can be associated with readmission. Preventable readmissions have also been determined to be associated with underlying patient severity of illness,1 suggesting that patient selection may be an important factor that relates to readmission rates. These findings highlight the importance of early recognition of complication, careful preoperative patient selection and counseling, and early intervention and close follow-up in higher risk cohorts.
Standardization of discharge documentation and computer-based electronic discharges may be an opportunity to improve interdisciplinary communication and data exchange between providers at discharge.25 Randomized trials assessing hospital use showed that among geriatric patients, those who received detailed postoperative instructions and confirmed postoperative follow-up were less likely to use more inpatient services.7 This study further emphasizes the potential benefit of early interventions in susceptible patients and the positive impact this may impart on resource utilization and cost containment.
Institutional strategies that are geared toward minimizing perioperative medical and surgical complications could reduce patient readmission. This may begin with careful patient selection and counseling. Further recognizing complications that occur during admission and establishing a treatment plan and close postoperative outpatient follow-up may serve to reduce readmission rates.
This study is not without limitations. First, the American College of Surgeons National Surgical Quality Improvement Program includes only 30-day outcomes and thus may not capture unplanned readmissions that occur after the 30-day study period; therefore, the incidence of readmissions is likely an underestimate. Another key criticism of this current study is that the impact of readmission relative to hospital length cannot be assessed directly. Specifically, neither the length of stay nor the exact reason for readmission can be assessed directly. In addition, we do not know whether the complication was noted or occurred during the hospitalization, as this would significantly affect decision-making. This is important, as it is not clear whether every admission was necessary or whether the postoperative complication was not adequately managed. The study group used is also heterogeneous and includes patients who underwent a variety of surgical procedures, which may introduce some bias into the study design. Another interesting and potentially confounding variable is that this data set does not allow for qualification of patient insurance status or socioeconomic variables, which may affect the use and access to discharge services. This limits the clinical applicability of the study. In addition, the endpoints included in the American College of Surgeons National Surgical Quality Improvement Program do not include hematoma, seroma, or fat necrosis, which may be important in the plastic surgery cohort we studied. Lastly, with this type of data set, we rely on appropriate data collection; however, observer bias is possible and, unfortunately, cannot be well controlled for in such a situation; however, the large sample size and careful study design minimize the impact such bias might impart.
In conclusion, there is a significant need for further work in the area of readmission and its effects on patient morbidity and cost containment in the field of plastic surgery. As health care reform continues, greater scrutiny of unplanned readmissions will be an area of focus. Efforts to minimize the incidence of readmission, through targeted, evidence-based prevention strategies, will optimally address resource utilization and will emerge as the future of contemporary, cost-effective medicine. This work represents the first of possibly many studies designed to delineate risk factors for readmission and potential areas for improvement.
Readmissions following surgical procedures can be common and extremely costly for healthcare systems. As healthcare reform evolves, 30-day, unplanned readmission rates may be closely scrutinized as an area of needed improvement. Efforts to minimize the incidence of readmission, through a deeper understanding of associated risk factors, may permit a more efficient delivery of care. In this work we delineate several key independent risk factors for readmission, including perioperative complications and obesity.
Deidentified patient information is freely available to all institutional members who comply with the American College of Surgeons National Surgical Quality Improvement Program Data Use Agreement. The Data Use Agreement implements the protections afforded by the Health Insurance Portability and Accountability Act of 1996. The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors of this study.
1. Goldfield NI, McCullough EC, Hughes JS, et al. Identifying potentially preventable readmissions. Health Care Financ Rev. 2008;30:75–91
2. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med. 2009;360:1418–1428
3. Naylor MD, Aiken LH, Kurtzman ET, Olds DM, Hirschman KB. The care span: The importance of transitional care in achieving health reform. Health Aff (Millwood). 2011;30:746–754
4. Brock J, Mitchell J, Irby K, et al.Care Transitions Program Team. Association between quality improvement for care transitions in communities and rehospitalizations among Medicare beneficiaries. JAMA. 2013;309:381–391
5. Brooke B, De Martino R, Girotti M, Dimick J, Goodney P. Developing strategies for predicting and preventing readmissions in vascular surgery. J Vasc Surg. 2012;56:556–562
6. Coleman EA, Min SJ, Chomiak A, Kramer AM. Posthospital care transitions: Patterns, complications, and risk identification. Health Serv Res. 2004;39:1449–1465
7. Jack BW, Chetty VK, Anthony D, et al. A reengineered hospital discharge program to decrease rehospitalization: A randomized trial. Ann Intern Med. 2009;150:178–187
9. Sellers MM, Merkow RP, Halverson A, et al. Validation of new readmission data in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2013;216:420–427
10. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: Prevalence and trends, 1960-1994. Int J Obes Relat Metab Disord. 1998;22:39–47
11. Kuczmarski RJ, Flegal KM. Criteria for definition of overweight in transition: Background and recommendations for the United States. Am J Clin Nutr. 2000;72:1074–1081
12. Don BR, Kaysen G. Serum albumin: Relationship to inflammation and nutrition. Semin Dial. 2004;17:432–437
13. McLean E, Cogswell M, Egli I, Wojdyla D, de Benoist B. Worldwide prevalence of anaemia, WHO Vitamin and Mineral Nutrition Information System, 1993–2005. Public Health Nutr. 2009;12:444–454
14. Hofer R, Kai T, Decker P, Warner D. Obesity as a risk factor for unanticipated admissions after ambulatory surgery. Mayo Clin Proc. 2008;83:908–916
15. Reinke C, Kelz R, Zubizarreta J, et al. Obesity and readmission in elderly surgical patients. Surgery. 2012;152:355–362
16. Silber JH, Rosenbaum PR, Kelz RR, et al. Medical and financial risks associated with surgery in the elderly obese. Ann Surg. 2012;256:79–86
17. Ahmad SA, Edwards MJ, Sutton JM, et al. Factors influencing readmission after pancreaticoduodenectomy: A multi-institutional study of 1302 patients. Ann Surg. 2012;256:529–537
18. Encinosa WE, Bernard DM, Chen CC, Steiner CA. Healthcare utilization and outcomes after bariatric surgery. Med Care. 2006;44:706–712
19. Gioia L, Filion K, Haider S, Pilote L, Eisenberg M. Hospital readmissions following abdominal aortic aneurysm repair. Ann Vasc Surg. 2005;19:35–41
20. Kassin M, Owen R, Perez S, et al. Risk factors for 30-day hospital readmission among general surgery patients. J Am Coll Surg. 2012;215:322–330
21. Khan M, Hossain F, Dashti Z, Muthukumar N. Causes and predictors of early re-admission after surgery for a fracture of the hip. J Bone Joint Surg Br. 2012;94:690–697
22. Wang M, Shivakoti M, Sparapani R, Guo C, Laud P, Nattinger A. Thirty-day readmissions after elective spine surgery for degenerative conditions among US Medicare beneficiaries. Spine J. 2012;12:902–911
23. Hannan E, Racz M, Walford G, et al. Predictors of readmission for complications of coronary artery bypass graft surgery. JAMA. 2003;290:773–780
24. Shepperd S, McClaran J, Phillips C, et al. Discharge planning from hospital to home. Cochrane Database Syst Rev. 2010(1):CD000313
25. Gandara E, Moniz T, Ungar J, et al. Communication and information deficits in patients discharged to rehabilitation facilities: An evaluation of five acute care hospitals. J Hosp Med. 2009;4:E28–E33