Publication
Article
Author(s):
Common vascular surgery procedures are associated with frequent and high-cost readmissions. Open wounds with infection, functional dependence, lengthy procedures, and transfusion are associated with 90-day readmission after vascular surgery.
ABSTRACT
Objectives
There are relatively sparse data regarding readmission after vascular surgery. The goal of our study is to analyze readmission rates and hospital cost for several common open and endovascular surgical procedures.
Methods
We accessed our local ACS NSQIP clinical database and hospital cost accounting for vascular surgery cases and their 30- and 90-day readmissions from January 1, 2010, to November 30, 2011. Direct hospital costs (DHC$) were analyzed during the index admission and for all readmissions. Risk factors were compared in the readmitted versus non-readmitted groups using parametric or non-parametric tests as appropriate. Significance was set at P <.05.
Results
We identified 170 patients who were readmitted. The 30-day all-cause readmission rate was 9.1% and at 90 days almost doubled to 17.9%. When readmissions occurred, on average they added DHC$ (000’s) 12.4 ± 12.3, comprising an additional 61.1% beyond index admission DHC$.
Preoperative risk factors associated with 90-day readmission included chronic obstructive pulmonary disease (COPD) (P = .027), open wound/infection (P = .005), and functional dependence (P = .027). Readmissions had longer index operative duration (P = .031) and more often received transfusions within 72 hours of the index case (P = .031). Wound infections were associated with a 90-day readmission (P = .012), as was treated DVT (P = .032) and cerebrovascular or cardiovascular events (P = .013).
Conclusions
Ninety-day readmissions after common vascular surgeries occurred at about twice our 30-day rate. The use of endovascular procedures is associated with significant readmission cost. COPD, open wounds with infection, functional dependence, lengthy procedures, and transfusion are associated with 90-day readmission after vascular surgery.
Am J Manag Care. 2014;20(9):e432-e438
Readmissions are costly and often preventable, and have emerged as a major target in healthcare reform as part of the 2010 Patient Protection and Affordable Care Act. There are relatively sparse data regarding readmission after vascular surgery. The 30-day allcause readmission rate in our study was 9.1%; at 90 days, it almost doubled, to 17.9%. When readmissions occurred, on average they added $12,400 in direct hospital costs (± SD $12,300), boosting the index admission cost by 61.1%.Readmissions are costly and often preventable, and have emerged as a major target in healthcare reform as part of the 2010 Patient Protection and Affordable Care Act, which set the foundation for a number of new quality measures for the healthcare system. The Hospital Readmissions Reduction Program is an effort by CMS to decrease the rate of readmission for Medicare patients. Under this program, it is estimated that approximately two-thirds of hospitals in the United States will suffer a penalty of up to 1% of their Medicare reimbursement if their readmission rate is high. The acuity of the initial admission and patient comorbidities have been shown to be predictors of readmission within 30 days of discharge.1 Diabetes is associated with frequent rehospitalization in a report by Jackson et al.2 Postoperative complications also increase readmission risk3; preventing complications tracked in the American College of Surgeons National Surgery Quality Improvement Program (ACS NSQIP) would reduce readmissions and lead to an estimatedMedicare savings of $620 million per year.
Cost containment efforts have led to analysis of readmissions and their financial impact. There are relatively sparse data regarding readmission after vascular surgery. Vascular surgery patients often have several comorbidities that can be associated with increased readmission rates. The 30-day readmission rate for Medicare patients undergoing vascular surgery is 24%, significantly higher than the 15.6% rate of readmission for other surgical procedures.4 The goal of our study is to analyze readmission rates and hospital costs for 7 common vascular surgical procedures. We hypothesize that readmissions related to the care of patients with vascular disease occur at greater than the current 30-day focus; therefore, examining a 90-day post surgery period may more appropriately inform total cost reduction efforts relative to readmission and the continuum of care for vascular disease.
METHODS
This Institutional Review Board—approved study retrospectively analyzed data from the University of Kentucky Hospital’s local ACS NSQIP clinical database and cost accounting system (Alliance for Decision Support, Avega Health Systems Inc, El Segundo, California) for vascular surgeries performed between January 1, 2010, and November 30, 2011. We were able to match the clinical and cost data for 170 vascular surgery cases. Fellows and residents were involved in all of these cases. All procedures were performed by the same team of experienced surgeons (>5 years each).
Patient preoperative clinical risk variables, demographics, perioperative variables, and 22 uniformly defined post operative adverse occurrences up to 30 days after the operation were analyzed from our vascular service ACS NSQIP data. During the time period analyzed, our ACS NSQIP data included a prospective, 100% sample of all of the following surgeries, both elective and emergent: carotid endarterectomy, abdominal aortic aneurysm repair (AAAr; open and endoluminal [endo]), lower extremity revascularization (LER; open and endo), and aortoiliac bypass (AIB; open or endo). The NSQIP protocol excluded patients admitted for trauma, patients under 18 years of age, and procedures that followed within 30 days of another procedure. Risk and outcome variables were rigorously defined, and the experienced nurse coordinators who extracted the data had completed in-depth training on all study definitions. Follow-up with the patient included phone calls and/or letters at 30 days to identify readmission regardless of whether it was at our hospital or elsewhere.
Direct hospital costs (DHCs) for the index surgery admission and any readmission (to our hospital only) within 90 days of discharge from the index hospitalization were identified in the cost accounting system. DHCs include costs directly attributable to patient care such as nursing, operating room time, supplies, and pharmaceuticals. They exclude overhead costs related to equipment, facilities, and administration, because attribution to specific patients can vary significantly depending on the cost accounting system and methodology used. Average DHCs for index admissions by procedure group were calculated, as were average 90-day readmission DHCs.
We compared preoperative risks, perioperative processes, and postoperative outcomes in the readmitted versus non-readmitted groups using parametric or nonparametric methods as appropriate. Significance was set at P <.05 for all tests, which were performed using SPSS version 20 (IBM, New York, New York).
RESULTS
Table 1
Patients’ mean age was 64.7 years ± 11.5 SD; 62.9% were male. At our regional referral center, 16.5% of the cases presented emergently. The index case admission characteristics are shown by procedure in . Mean length of hospital stay was 6.5 days ± SD 7.5 and mean DHCs were $20,300 ± SD $19,500. Six patients died during the index case hospitalization (4 were emergent AAAr’s), leaving 164 patients for readmission analysis. All readmissions were unplanned.
Table 2
Figure
The overall rate of index admission survivors experiencing 30-day readmission was 9.1%; at 90 days, it almost doubled, to 17.9% (). The 90-day readmission rates varied depending on the procedure (χ2 P = .001) ranging from 0% in AIB endo patients to almost half (48.1%) of LER open patients (). When readmissions occurred, on average they added $12,400 in direct hospital costs (± SD $12,300), boosting the index admission cost by 61.1%. These additional costs varied by procedure (Kruskal-Wallis P <.001) and ranged from an additional 1.8% of index admission DHCs in AAAr open patients to more than doubling (147%) index admission DHCs for LER endo patients.
Table 3
Preoperative clinical conditions associated with 90-day readmission included chronic obstructive pulmonary disease (COPD; P = .027), open wound/infection (P = .005), and preoperative functional dependence (P = .027). American Society of Anesthesiologists class, age, smoking, body mass index, diabetes, emergency status of index case surgery, and index case length of stay were not significantly different in the readmitted versus non-readmitted groups. Readmissions had longer index operative duration (P = .031) and had more often received transfusions within 72 hours of the index case (P = .031). Postoperatively, 30-day adverse events associated with 90-day readmission () included wound infections (superficial to deep, P = .012), treated deep venous thrombosis (DVT; P = .032), and stroke or cardiac infarct/arrest (P = .045).
DISCUSSION
Vascular surgery procedures have been associated with high complication rates, extended hospital length of stay (LOS), and a significant risk for readmissions. Readmissions are “expensive adverse events for patients,”5 and CMS reports publicly risk-adjusted readmission rates after certain conditions, including myocardial infarction, heart failure, and pneumonia. The Patient Protection and Affordable Care Act has devised a program that will impose a financial penalty for facilities with high rates of 30-day readmissions; this is projected to include vascular surgery patients within a number of years.6 The program’s penalties range from 0.01% to 1% of hospital base operating diagnosis-related group payments.
The Medicare Payment Advisory Commission highlighted vascular procedures as one of 7 groups that were responsible for more than 30% of potentially preventable readmissions; vascular readmissions were the most costly on a per patient basis.7 Our study emphasizes the high readmission rate of vascular surgery patients and the associated cost. Overall, only 3% of vascular readmissions are planned.2 In a recent Medicare claims-based study by Jencks et al, vascular surgery procedures had the second highest readmission rate, at 23.9%, of all medical conditions and procedures,4 trailing behind only congestive heart failure diagnoses. Our readmission rate was slightly lower, at 17.9%.
The reduction of rehospitalization rates has been identified as a way of achieving improved care and reduced cost, and has also been evaluated as a pay-for-performance criterion.8,9 These healthcare policy developments may prolong LOS since patients may be kept in the hospital longer to diminish the risk of readmissions. It is therefore important to assess the correlation of readmission risks and LOS. We found that the index case length of stay did not influence readmission risk. It is important to identify risk factors that are responsible for readmission after vascular interventions and to also access the financial impact of readmission to institutions. Our report quantifies readmission costs for the most common surgical procedures and identifies perioperative risk factors that are associated with higher readmission rates.
A readmission rate of 24% has been reported after lower extremity bypass10; factors related to readmission were smoking, dialysis, and tissue loss. Female gender was associated with high readmission rate, which could be related to the higher wound complication rate in females after bypass surgery.11 The readmission rate for our lower extremity revascularization patients was 11.1% over a 30-day period and 48.1% over a 90-day period. Readmission added 67% of DHCs to the overall index admission cost.
In another recent study, Belkin et al reported that patients undergoing open lower extremity revascularization and critical limb ischemia patients had the highest rates of unplanned 30-day rehospitalization. These findings are in agreement with the high readmission rates in our population, which were 33% after endoluminal revascularization and 46% after open revascularization. Interestingly, readmissions after endoluminal lower extremity revascularization were associated with the highest costs.
It has been reported that patients with diabetes are more likely to undergo unplanned 30-day readmission to the hospital,12 but this finding was not substantiated by our analysis. However, history of COPD and functional dependence were identified as significant readmission risk factors. A recent report also indicates that endovascular procedures for lower extremity revascularization, which are traditionally deemed less invasive, were not associated with decreased readmission. After multivariable adjustment, endovascular procedures in our study tended to be associated with increased readmission.13
Several investigators have measured readmission over time periods of 6 to 12 months14 and included readmission in outcomes of abdominal aortic aneurysm repair.15 One study reported that complications leading to prolonged length of stay and discharge destination other than home had a great effect on the likelihood of readmission.16 These investigators observed that not all complications influenced readmission equally. Their data suggested that some complications if recognized and treated did not lead to prolonged length of stay or readmission, but other complications resulted in the need for readmission. Wound complication was the most common reason for readmission after both endovascular aortic aneurysm repair (EVAR) and open repair. Graft problems were more common after EVAR, but bowel obstruction was common after open repair. An analysis of national Medicare data indicated that readmission after abdominal aortic aneurysm (AAA) repair is common, with more than 1 in 8 individuals readmitted within 30 days of discharge after nonruptured AAA repair. In one prior study, Vogel and Kruse17 reported on a Washington state administrative database: 30-day readmission rates were 11.6% for EVAR and 13.1% for open repair. In our study, the 30-day readmission rate was 11.7% for EVAR and 11.1% for open repair. With extension to 90 days, readmission increased to 17.65% for EVAR but remained the same for open repair.
We evaluated risk factors that were associated with higher readmission rates. In our population, patients who had open wounds preoperatively had a readmission rate almost 3 times higher than those without such wounds. Perioperative transfusions and longer operative duration also increased readmission risk. Postoperative complications that were associated with higher 90-day readmission risk included wound infection; DVT; and renal, cardiac, or cerebrovascular complications.
One limitation of our study is the relatively small number of patients, which restricts the ability to perform risk factor analysis. Also, the rate of readmission after open aneurysm repair is affected by the fact that 4 of the 13 patients who underwent open repair did not survive (all presented with ruptured AAAs). Nevertheless, we have been able to report actual cost, and a significant strength of our analysis is the high quality of the ACS NSQIP database. This database is a clinical database constructed by clinical personnel, and it does not suffer from the known faults of administrative databases constructed for billing purposes.18
CONCLUSIONS
Our findings offer supporting evidence that vascular surgery patients have an overall high readmission rate, 17.9% at 90 days. We draw attention to patients undergoing lower extremity revascularization, who have the highest readmission rate as well as the highest costs. When averaged over all patients, including non-readmitted, the total cost of 90-day care rises by 75% in LER endo patients as a result of readmission cost. COPD, open wounds with infection, functional dependence, lengthy procedures, and transfusion are all associated with 90-day readmission after vascular surgery. Interventions that target comorbidities and perioperative risk factors and complications may be of value in reducing readmission rates and decreasing costs. Further research is needed in this area.Author Affiliations: Department of Vascular Surgery Service (ESX), Department of Surgery (DLD), College of Medicine, University of Kentucky, Lexington; UK Healthcare, Lexington, KY (RLK); and Duke University School of Medicine, Durham, NC (JAL).
Source of Funding: None reported.
Author Disclosures: Mr Korosec is an employee of UK Healthcare. Drs Xenos and Davenport and Ms Lyden report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (ESX, JAL, DLD); acquisition of data (ESX, RLK, DLD); analysis and interpretation of data (ESX, DLD); drafting of the manuscript (ESX, JAL, DLD); critical revision of the manuscript for important intellectual content (DLD); statistical analysis (ESX, DLD); supervision (ESX, ); and literature review (JAL).
Address correspondence to: Eleftherios S. Xenos, University of Kentucky, 800 Rose St, Lexington, KY 40536. E-mail: esxeno2@email.uky.edu.REFERENCES
1. van Walraven C, Dhalla IA, Bell C, et al. Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community. CMAJ. 2010;182(6):551-557.
2. Jackson BM, Nathan DP, Doctor L, Wang GJ, Woo EY, Fairman RM. Low rehospitalization rate for vascular surgery patients. J Vasc Surg. 2011;54(3):767-772.
3. Lawson EH, Hall BL, Louie R, et al. Association between occurrence of a postoperative complication and readmission: implications for quality improvement and cost savings. Ann Surg. 2013;258(1):10-18.
4. Jencks SF, Williams MV, Coleman EA: Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med. 2009;360(14):1418-1428.
5. Overview: readmission measures. QualityNet website. http://www.qualitynet.org/dcs/ContentServer?cid=1219069855273&pagename=QnetPublic%2FPage%2FQnetTier2&c=Page. Accessed September 5, 2012.
6. Congressional Research Service. Medicare provisions in the Patient Protection and Affordable Care Act (PPACA). National Conference of State Legislatures website. http://www.ncsl.org/documents/health/MCProv.pdf.
7. Hackbarth G, Reischauer R, Miller M. Report to the Congress: Promoting Greater Efficiency in Medicare. Published 2007. http://www.wsha.org/images/activEdit/11.28.12_FINAL_CT_Toolkit_Version_1_508.pdf.
8. National Quality Forum. Candidate hospital care additional priorities: 2007 performance measure. National Quality Forum, Washington, DC (2007).
9. Application of incentives to reduce avoidable readmissions to hospitals. Fed Regist. 2008;73(84):23673-23675.
10. McPhee JT, Nguyen LL, Ho KJ, Ozaki CK, Conte MS, Belkin M. Risk prediction of 30-day readmission after infrainguinal bypass for critical limb ischemia. J Vasc Surg. 2013;57(6):1481-1488.
11. Belkin M, Conte MS, Donaldson MC, Mannick JA, Whittemore AD. The impact of gender on the results of arterial bypass with in situ greater saphenous vein. Am J Surg. 1995;170(2):97-102.
12. Jackson BM, Nathan DP, Doctor L, Wang GJ, Woo EY, Fairman RM. Low rehospitalization rate for vascular surgery patients. J Vasc Surg. 2011;54(3):767-772.
13. Vogel TR, Kruse RL. Risk factors for readmission after lower extremity procedures for peripheral artery disease. J Vasc Surg. 2013;58(1):90-97.
14. Giles KA, Landon BE, Cotterill P, O’Malley AJ, Pomposelli FB, Schermerhorn ML. Thirty-day mortality and late survival with reinterventions and readmissions after open and endovascular aortic aneurysm repair in Medicare beneficiaries. J Vasc Surg. 2011; 53(1):6-12,13.e1.
15. Jetty P, Hebert P, van Walraven C. Long-term outcomes and resource utilization of endovascular versus open repair of abdominal aortic aneurysms in Ontario. J Vasc Surg. 2010; 51(3):577-583, 583.e1-3.
16. Greenblatt DY, Greenberg CC, Kind AJ, et al. Causes and implications of readmission after abdominal aortic aneurysm repair. Ann Surg. 2012;256(4):595-605.
17. Vogel TR, Kruse RL. Risk factors for readmission after lower extremity procedures for peripheral artery disease. J Vasc Surg. 2013;58:283-290.
18. Nguyen LL, Barshes NR. Analysis of large databases in vascular surgery. J Vasc Surg. 2010;52(3):768-774.