ORIGINAL ARTICLE
Failure to Rescue After Proximal Femur Fracture Surgery Mariano E. Menendez, MD and David Ring, MD, PhD
Objectives: Failure to rescue (FTR)––death after a major adverse event––has recently been identified as an important determinant of variation in surgical mortality. We sought to identify patient and hospital characteristics associated with FTR after proximal femur fracture surgery, and to determine whether they are different from the predictors of the occurrence of adverse events. We also identified which adverse events are most highly associated with FTR.
Methods: Among an estimated 287,959 patients with a surgically treated proximal femur fracture identified in the 2011 Nationwide Inpatient Sample, the overall adverse event rate was 22% and the FTR rate was 6.4%. Multivariable logistic regression modeling was used to identify independent predictors of FTR and adverse events. Results: Patient-specific variables influenced adverse event occurrence but exerted little or no influence on FTR. Hospitals located in rural areas were 14% less likely than urban hospitals to have adverse events [odds ratio (OR): 0.86, 95% confidence interval (CI): 0.83– 0.89], but 30% more likely to fail to rescue patients from adverse events (OR: 1.3, 95% CI: 1.2–1.5). Compared with teaching and large hospitals, nonteaching settings and institutions of smaller size were associated with decreased risk for adverse events, but similar risk for FTR. Lower hospital volume was a risk factor for adverse events and FTR. There was a more than 3-fold increased risk for death among patients with respiratory failure (OR: 5.4, 95% CI: 5.0– 5.8), pulmonary embolism (OR: 3.6, 95% CI: 3.1–4.1), and myocardial infarction (OR: 3.0, 95% CI: 2.8–3.3). Conclusions: Because FTR is less affected by patient characteristics than morbidity, it might be a better measure of providerspecific performance in hip fracture surgery. Targeted initiatives aimed at improving the timely recognition and management of cardiorespiratory adverse events ––particularly at rural hospitals–– might be key to reducing mortality. Key Words: failure to rescue, mortality, adverse events, inpatient, Elixhauser, complications, epidemiology, nationwide inpatient sample, trauma Accepted for publication August 19, 2014. From the Department of Orthopaedic Surgery, Orthopaedic Hand and Upper Extremity Service, Yawkey Center, Massachusetts General Hospital, Boston, MA. The authors report no conflict of interest. No IRB approval is mandatory for this study. The data are deidentified and commercially available for use. This work was performed at the Orthopaedic Hand and Upper Extremity Service, Massachusetts General Hospital, Boston, MA. Reprints: David Ring, MD, PhD, Orthopaedic Hand Service, Yawkey Center, Massachusetts General Hospital, Suite 2100, 55 Fruit St, Boston, MA 02114 (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
e96
| www.jorthotrauma.com
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence. (J Orthop Trauma 2015;29:e96–e102)
INTRODUCTION Surgical mortality rates are routinely used to evaluate patient safety and provider performance.1–6 Quality improvement efforts to reduce mortality have traditionally focused on reducing rates of intraoperative and postoperative adverse events with the assumption that these are directly related to mortality.7–13 There is growing recognition that the major factor influencing perioperative mortality may not be the occurrence of an adverse event, but rather how patients with an adverse event are managed.14–17 Indeed, low-mortality hospitals are distinguished not by having fewer adverse events, but by more successfully preventing death among patients that experience an adverse event.17 The emerging concept of failure to rescue (FTR)––the inability to rescue a patient from death after an adverse event––has become an important determinant of outcome for hospitalized patients and is increasingly used as a surgical quality indicator.18–20 We sought to identify which patient and hospital characteristics merit an increased alertness for “rescue” after proximal femur fracture surgery, and to determine whether they are different from the predictors of the occurrence of adverse events. In addition, we identified which adverse events are most highly associated with FTR.
PATIENTS AND METHODS Encounter data for this retrospective cohort study were abstracted from the Nationwide Inpatient Sample (NIS), currently the largest all-payer inpatient care database in the United States.21,22 Managed by the Agency for Healthcare Research and Quality, the NIS is a nationally representative cluster sample of discharges from more than 1000 short-term and nonfederal hospitals.23 Stratification of the hospitals for sampling is based on geographic region, bed size, teaching status, and rural/urban location. Discharge weight files are provided to derive statistically valid national estimates. Beside patient demographic and hospital-related data, the NIS collects up to 25 medical diagnoses and 15 procedures with the use of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. In recognition of its utility to answer valuable clinical questions, the NIS has been adopted extensively to conduct populationbased research across diverse medical disciplines.24–27 Formal approval by our institutional review board was not necessary J Orthop Trauma Volume 29, Number 3, March 2015
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
TABLE 1. In-Hospital Mortality and FTR Rates Among Patients With $1 Adverse Event Parameter Total Age group (y) ,50 50–59 60–69 70–79 80–89 .89 Sex Female Male Elixhauser comorbidity score 0 1–2 $3 Race/ethnicity White Black Hispanic Other Insurance status Medicare Medicaid Private Uninsured Other Median household income Low Medium High Highest Hospital volume Low Intermediate High Hospital bed size Small Medium Large Hospital teaching status Nonteaching Teaching Hospital location Urban Rural Trauma type Single-trauma Polytrauma
‡1 Adverse Event, n (%)
In-Hospital Mortality, n
64,248 (22)
4105
6.4
854 2055 4865 12,355 28,843 15,276
5† 69 187 699 1908 1238
0.59† 3.4 3.8 5.7 6.6 8.1
2502 1603
6.0 7.0
FTR, %*
P ,0.001
(9.7) (14) (16) (20) (24) (29)
,0.001 41,428 (21) 22,816 (27)
,0.001
1078 (6.5) 13,492 (14) 49,678 (29)
36 655 3414
3.3 4.9 6.9
57,108 2481 2776 1884
(22) (23) (20) (21)
3702 110 170 123
6.5 4.4 6.1 6.5
56,389 1376 4879 540 890
(24) (19) (16) (12) (15)
3748 66 223 14 49
6.6 4.8 4.6 2.6 5.5
0.001
,0.001
0.001 16,361 15,606 16,815 14,361
(23) (22) (22) (23)
1094 1072 996 874
6.7 6.9 5.9 6.1
21,072 (21) 22,153 (23) 21,023 (23)
1370 1471 1263
6.5 6.6 6.0
7288 (21) 16,822 (23) 39,353 (23)
484 1054 2517
6.6 6.3 6.4
,0.001
0.39
0.028 37,214 (22) 26,250 (23)
2312 1744
6.2 6.6
55,375 (23) 8089 (20)
3430 625
6.2 7.7
56,494 (22) 7754 (27)
3510 595
6.2 7.7
,0.001
,0.001
*Calculated as number of deaths/number of with $1 adverse event. †,11 patients.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
for this study, because the data contained in the NIS are anonymous. A cohort of patients with an ICD-9-CM primary diagnosis of a proximal femur fracture (codes 820.0 and 820.1 for transcervical, 820.2 and 820.3 for pertrochanteric, and 820.8 and 820.9 for unspecified hip fractures) were generated.28 Only those patients who underwent subsequent open reduction or internal fixation (ICD-9-CM codes 78.55, 79.15, 79.25, 79.35), hemiarthroplasty (81.52), or total hip arthroplasty (81.51), were included for further analysis.29 Between January 1, 2011 and December 31, 2011, an estimated 287,959 patients met the inclusion criteria for the study. Patient-specific variables included age, sex, race/ ethnicity (white, black, Hispanic, and other), insurance status (Medicare, Medicaid, private, uninsured, and other), household income of the patient’s zip code of residence (low, medium, high, and highest), and trauma type (singletrauma, and polytrauma). Polytrauma was defined using the following ICD-9 secondary diagnosis codes: 800–819, 821–909, and 925–959. Baseline comorbidity status was quantified using the Elixhauser comorbidity score.30–32 Our study sample was predominantly female (70%) and white (88%), and the most common decade of fracture occurrence was in the 80s (42%). Using data from the American Hospital Association Annual Survey of Hospitals available in the NIS, hospitals were classified according to their location (urban and rural), bed size (small, medium, and large), and teaching status (teaching and nonteaching). Hospitals are considered to be teaching institutions if any of the following 3 criteria are met1: residency-training approval by the Accreditation Council for Graduate Medical Education,2 membership in the Council of Teaching Hospitals, or3 a resident to bed ratio of 0.25 or more.33 Bed size cutoff points vary so that approximately one-third of the hospitals in a given region, location, and teaching status combination falls within each bed size category (small, medium, or large). Using a unique identifier number assigned by the NIS to each hospital, we were also able to derive the hospital volume by counting the number of proximal femur fracture surgeries performed at each hospital during 2011.34 Hospital volume was then categorized into 3 groups by patient-based tertiles (low-volume: ,98 procedures; intermediate-volume: 98–179 procedures; highvolume: $180 procedures).17 By the use of ICD-9-CM diagnosis codes,16,17,35 we considered major in-hospital adverse events, such as myocardial infarction (410), acute renal failure (584), respiratory failure (518.4, 518.5, 514), cerebrovascular accident (433, 434, 997.02), deep venous thrombosis (451, 453), pulmonary embolism (415.1), pneumonia (481, 482, 483, 484, 485, 486, 997.31, 997.39), bacteremia/sepsis (038, 7907), surgical site infection (998.3, 998.5), and postoperative hemorrhage (998.1). Our primary study end point was FTR, defined as a case fatality among patients with 1 or more of the defined major adverse events.16,17 Pearson’s x2 test was used for analysis of categorical data and independent samples t test for continuous data. FTR rates were calculated for all demographic and www.jorthotrauma.com |
e97
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Menendez and Ring
TABLE 2. Risk Factors for Adverse Events and FTR After Hip Fracture Surgery Adverse Events
FTR
95% CI
95% CI
Predictor
OR
Lower
Upper
P
OR
Lower
Upper
P
Age, per 10-year increase Male sex (reference: female) Elixhauser comorbidity score, per 1-unit increase Race (reference: white) Black Hispanic Other Insurance status (reference: private insurance) Medicare Medicaid Uninsured Other Household income (reference: highest) Low Medium High Hospital volume (reference: high) Low Intermediate Hospital bed size (reference: large) Small Medium Nonteaching hospital (reference: teaching hospital) Rural hospital (reference: urban hospital) Polytrauma (reference: single-trauma) Complications (reference: no complication) Myocardial infarction Acute renal failure Respiratory failure Cerebrovascular accident Deep venous thrombosis Pulmonary embolism Pneumonia Bacteremia per sepsis Surgical site infection Postoperative hemorrhage
1.3 1.6 1.4
1.3 1.6 1.35
1.3 1.6 1.4
,0.001 ,0.001 ,0.001
1.4 1.1 1.1
1.3 1.1 1.05
1.4 1.2 1.1
,0.001 ,0.001 ,0.001
1.0 0.84 0.92
0.96 0.80 0.87
1.1 0.87 0.97
0.69 ,0.001 ,0.001
0.91 1.0 1.0
0.74 0.88 0.84
1.1 1.2 1.3
0.40 0.61 0.77
1.0 1.3 1.07 1.0
1.0 1.2 0.97 0.94
1.1 1.3 1.2 1.1
0.04 ,0.001 0.20 0.63
1.1 1.3 0.73 1.3
0.95 0.93 0.37 0.92
1.3 1.7 1.4 1.8
0.19 0.14 0.35 0.14
1.1 1.0 1.0
1.04 0.97 0.97
1.1 1.0 1.0
,0.001 0.87 0.85
1.0 1.0 1.0
0.94 0.95 0.91
1.2 1.2 1.1
0.42 0.37 0.99
1.1 1.04
1.05 1.02
1.1 1.1
,0.001 ,0.001
1.1 1.1
0.97 1.04
1.2 1.2
0.15 0.0040
0.83 0.95 0.89
0.81 0.93 0.87
0.86 0.98 0.90
,0.001 ,0.001 ,0.001
1.0 1.0 0.92
0.92 0.90 0.85
1.2 1.1 1.00
0.48 0.78 0.024
0.86
0.83
0.89
,0.001
1.3
1.2
1.5
,0.001
1.4
1.4
1.4
,0.001
1.2
1.1
1.3
,0.001
— — — — — — — — — —
— — — — — — — — — —
— — — — — — — — — —
— — — — — — — — — —
3.0 1.6 5.4 2.2 0.55 3.6 2.2 1.9 1.8 0.69
2.8 1.5 5.0 1.9 0.44 3.1 2.0 1.7 1.4 0.56
3.3 1.7 5.8 2.6 0.69 4.1 2.3 2.0 2.5 0.86
,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 0.001
Statistical significance set at P , 0.01.
RESULTS
clinical variables, as well as for all major adverse events. Multivariable logistic regression analysis was first performed to determine perioperative factors associated with the development of major adverse events. Second, multivariable analysis was undertaken to determine predictors of FTR in patients experiencing any of the predefined adverse events. Results were reported as odds ratios (OR) with 95% confidence intervals (CI). The statistical threshold for alpha error was set at 0.01.
Among an estimated 64,248 patients (22%) experiencing at least 1 adverse event, 4105 died during hospitalization, corresponding to an FTR rate of 6.4% (Table 1). Adverse event and FTR rates rose steadily with age and comorbidity burden. FTR rates were higher in male patients than in female patients (7.0% vs. 6.0%, P , 0.001), and in polytrauma patients compared with single-trauma patients (7.7% vs. 6.2%,
e98
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
FIGURE 1. FTR rates for each adverse event.
P , 0.001). Although high-volume hospitals had higher adverse event rates (23.1%) than low-volume (21.3%) and intermediate-volume (22.6%) hospitals, they were better at rescuing patients and thus exhibited lower FTR rates (6.0% vs. 6.5% and 6.6%, respectively; P , 0.001). Despite higher adverse event rates in urban (23%) than in rural (20%) hospitals, facilities located in rural areas (7.7%) had higher FTR rates than those located in urban settings (6.2%; P , 0.001). In multivariable logistic regression analysis, patient-specific variables, such as sex, Elixhauser comorbidity score, and polytrauma exerted more influence (higher ORs) on adverse event occurrence than on FTR (Table 2). Insurance status, race, and household income influenced adverse events, but not FTR. Compared with hospitals in urban settings, facilities located in rural areas were associated with lower odds of adverse events (OR: 0.86, 95% CI: 0.83–0.89, P , 0.001) but higher odds of FTR (OR: 1.3, 95% CI: 1.2–1.5, P , 0.001). Patients treated at nonteaching hospitals were less likely to experience adverse events (OR: 0.89, 95% CI: 0.87–0.90, P , 0.001) but were as likely as those treated at teaching hospitals to succumb after an adverse event. Compared with large-sized hospitals, institutions of small (OR: 0.83, 95% CI: 0.81–0.86, P , 0.001) and medium size (OR: 0.95, 95% CI: 0.93–0.98, P , 0.001) were associated with decreased risk for adverse events, but comparable risk for FTR. Patients treated at low-volume and intermediatevolume hospitals tended to be at increased risk for adverse events and FTR, compared with those treated at high-volume hospitals. Respiratory failure (19%) was the adverse event with the highest FTR rate, followed by pulmonary embolism (15%) and myocardial infarction (14%; Fig. 1). The FTR rate Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
was 16% in patients experiencing any of these 3 adverse events, and 6.0% among patients with adverse events other than the latter 3. In multivariable modeling, the risk of mortality was increased by 5.4 times (OR: 5.4, 95% CI: 5.0–5.8, P , 0.001) in patients with respiratory failure, 3.6 times (OR: 3.6, 95% CI: 3.1–4.1, P , 0.001) in patients affected by pulmonary embolism, and 3 times (OR: 3.0, 95% CI: 2.8– 3.3, P , 0.001) among those suffering myocardial infarction. Hospitals located in urban areas had generally higher rates of incidence of these adverse events, but lower rates of FTR (Fig. 2). For all 3 adverse events, high-volume hospitals reported the highest rates of incidence yet the lowest rates of FTR (Fig. 3).
DISCUSSION Ideally, decreases in mortality after surgery would reflect minimization of errors and adverse events. Accumulating evidence suggests that timely recognition and effective management of adverse events may also be important in explaining rates of death, particularly among infirm patients or patients having high-risk procedures.15,17,36,37 The Agency for Healthcare Research and Quality has recently adopted FTR as a patient safety indicator,38,39 and the American College of Surgeons National Surgical Quality Improvement Program has begun to report FTR rates as part of their annual summary report.36 Using nationally representative data, we identified factors associated with FTR after proximal femur fracture surgery and determined whether they were different from those associated with the development of adverse events. Although our analysis benefits from access to large numbers and associated power, we acknowledge a number of www.jorthotrauma.com |
e99
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Menendez and Ring
J Orthop Trauma Volume 29, Number 3, March 2015
FIGURE 2. Rates of adverse events and FTR according to hospital location.
important shortcomings.40,41 First, the primary purpose of administrative claims data is for billing, and, as such, there is likely an underreporting of adverse events. To minimize this bias, we limited our analysis to major postoperative adverse events that are likely to generate a claim.16 Second, the possibility of errors in coding of the diagnoses and procedures cannot be avoided; nonetheless, coding mistakes tend to be evenly distributed in large-scale studies.42 Third, the NIS does not include relevant data such as the number of healthcare personnel and specialties involved in patient care, and the chronological occurrence and severity of adverse events.43 Fourth, another limitation was our inability to adjust for surgeon volume and energy of injury, factors known to influence inpatient outcomes after proximal femur fracture surgery.29 However, we controlled for polytrauma, which may act as a surrogate for energy of injury. Fifth, we limited our analysis to a single year of data. However, restricting the population to a single time period could also be seen as strength, because the diagnostic evaluation and operative techniques are unlikely to have changed and are more likely to be consistent across the population.44 Sixth, the retrospective nature of the NIS does not allow ascertainment of the exact reason for death. Finally, the reader should be aware that findings in large-scale studies can be statistically
significant yet clinically insignificant. The reader should consider whether small but significant differences are clinically important. Older age and higher comorbidity burden were associated with greater risk for FTR after proximal femur fracture surgery. Our finding is consistent with recent studies in patients undergoing cytoreductive nephrectomy and radical cystectomy.45,46 Male patients were 60% more likely than women to experience an adverse event, but only 10% more likely to die of an adverse event. Other patient characteristics, such as insurance status, race, and household income influenced adverse events, but not FTR. Confirming previous research,20 FTR was not as affected by patient characteristics as were adverse events. Because FTR seems to be less sensitive to conditions outside of the control of hospital staff such as patient demographics and comorbidity, it might be a better measure of provider-specific performance in proximal femur fracture surgery. Compared with urban facilities, hospitals located in rural areas were less likely to have adverse events, but more likely to fail from preventing death once they occurred. It is possible that sicker––more prone to morbidity––patients were admitted to urban facilities, which would explain the decreased risk for adverse events in rural hospitals.47 The
e100
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
FIGURE 3. Rates of adverse events and FTR according to hospital volume.
higher risk for dying after an adverse event in rural hospitals might indicate that these facilities are less prepared to manage adverse events, perhaps because of limited expertise,48 restricted access to high-technology equipment,49 or less coordinated care.50 Consistent with research in other surgical disciplines,16,17,45,46,51 lower hospital volume was an independent predictor of adverse events and FTR after proximal femur fracture surgery. Hospital teaching status and bed size were determinants of adverse events, but not of FTR: compared with teaching and large-sized hospitals, nonteaching hospitals and institutions of smaller size were associated with decreased risk for adverse events, but similar risk for mortality after an adverse event. These findings add to the current body of knowledge suggesting that an increased risk for morbidity does not always correlate with a higher mortality risk (eg, FTR). Furthermore, they suggest that FTR might be incorporated in pay-for-performance initiatives, because it seems to be a more discriminative measure of hospital performance than morbidity,15,17,36,37 at least for infirm patients and high-risk procedures. Hospital staff should be particularly mindful of the more than 3-fold increased risk for death among patients with respiratory failure, pulmonary embolism, and myocardial infarction. In view of these findings, quality improvement Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
efforts should be prioritarily aimed at both reducing and effectively managing cardiorespiratory complications in hospitalized patients with a proximal femur fracture. Testing FTR as a quality-of-care metric in orthopaedic surgery is worthy of future work, as much of the outcomebased orthopaedic literature has primarily focused on traditional morbidity and mortality rates. For instance, the analysis of temporal trends in FTR rates for common elective orthopaedic procedures would be of value. Furthermore, it would be interesting to evaluate the impact of surgical safety checklists on FTR rates.52 Quality initiatives to reduce intraoperative and postoperative adverse events (such as the World Health Organization protocol) are important ways to reduce surgical morbidity and mortality, particularly for surgeries with low rates of adverse events and death.52–54 In a low-risk setting, the proportion of adverse events related to error is expected to be higher and the impact of checklists and other measures greater. In contrast, with high-risk patients or procedures, most adverse events relate more to patient infirmity and misfortune than to quality of care. Improved surgical safety should therefore aim to both reduce adverse events related to errors or inadequate preparation or communication, and to optimize the ability to rescue patients from adverse events. www.jorthotrauma.com |
e101
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Menendez and Ring
Additional research is needed to determine if rescue of infirm patients leads to an acceptable quality extension of life.
1. Dimick JB, Ghaferi AA, Osborne NH, et al. Reliability adjustment for reporting hospital outcomes with surgery. Ann Surg. 2012;255:703–707. 2. Dimick JB, Staiger DO, Hall BL, et al. Composite measures for profiling hospitals on surgical morbidity. Ann Surg. 2013;257:67–72. 3. Dimick JB, Welch HG, Birkmeyer JD. Surgical mortality as an indicator of hospital quality: the problem with small sample size. JAMA. 2004;292: 847–851. 4. Hall BL, Hamilton BH, Richards K, et al. Does surgical quality improve in the American College of Surgeons National Surgical Quality Improvement Program: an evaluation of all participating hospitals. Ann Surg. 2009;250:363–376. 5. Teng CI, Dai YT, Shyu YI, et al. Professional commitment, patient safety, and patient-perceived care quality. J Nurs Scholarsh. 2009;41: 301–309. 6. Hashmi ZG, Schneider EB, Castillo R, et al. Benchmarking trauma centers on mortality alone does not reflect quality of care: Implications for pay-for-performance. J Trauma Acute Care Surg. 2014;76:1184–1191. 7. Hemmila MR, Nathens AB, Shafi S, et al. The Trauma Quality Improvement Program: pilot study and initial demonstration of feasibility. J Trauma. 2010;68:253–262. 8. Jha AK, Joynt KE, Orav EJ, et al. The long-term effect of premier pay for performance on patient outcomes. N Engl J Med. 2012;366:1606–1615. 9. Moore L, Stelfox HT, Turgeon AF. Complication rates as a trauma care performance indicator: a systematic review. Crit Care. 2012;16:R195. 10. Rosenthal MB, Landon BE, Normand SL, et al. Pay for performance in commercial HMOs. N Engl J Med. 2006;355:1895–1902. 11. Shih T, Nicholas LH, Thumma JR, et al. Does pay-for-performance improve surgical outcomes? An evaluation of phase 2 of the premier hospital quality incentive demonstration. Ann Surg. 2014;259:677–681. 12. Sutton M, Nikolova S, Boaden R, et al. Reduced mortality with hospital pay for performance in England. N Engl J Med. 2012;367:1821–1828. 13. Brailer DJ, Kroch E, Pauly MV, et al. Comorbidity-adjusted complication risk: a new outcome quality measure. Med Care. 1996;34:490–505. 14. Ghaferi AA, Dimick JB. Variation in mortality after high-risk cancer surgery: failure to rescue. Surg Oncol Clin N Am. 2012;21:389–395, vii. 15. Sheetz KH, Waits SA, Krell RW, et al. Improving mortality following emergent surgery in older patients requires focus on complication rescue. Ann Surg. 2013;258:614–617; discussion 617–618. 16. Wright JD, Ananth CV, Ojalvo L, et al. Failure to rescue after major gynecologic surgery. Am J Obstet Gynecol. 2013;209:420e1–428e1. 17. Wright JD, Herzog TJ, Siddiq Z, et al. Failure to rescue as a source of variation in hospital mortality for ovarian cancer. J Clin Oncol. 2012;30: 3976–3982. 18. Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in hospital mortality associated with inpatient surgery. N Engl J Med. 2009;361:1368–1375. 19. Sheetz KH, Guy K, Allison JH, et al. Improving the care of elderly adults undergoing surgery in Michigan. J Am Geriatr Soc. 2014;62:352–357. 20. Silber JH, Williams SV, Krakauer H, et al. Hospital and patient characteristics associated with death after surgery. A study of adverse occurrence and failure to rescue. Med Care. 1992;30:615–629. 21. Bohl DD, Basques BA, Golinvaux NS, et al. Nationwide inpatient sample and national surgical quality improvement program give different results in hip fracture studies. Clin Orthop Relat Res. 2014;472:1672–1680. 22. Pallin DJ, Espinola JA, Camargo CA Jr. US population aging and demand for inpatient services. J Hosp Med. 2014;9:193–196. 23. Wakeam E, Hevelone ND, Maine R, et al. Failure to rescue in safety-net hospitals: availability of hospital resources and differences in performance. JAMA Surg. 2014;149:229–235. 24. D’Apuzzo MR, Keller TC, Novicoff WM, et al. CT pulmonary angiography after total joint arthroplasty: overdiagnosis and iatrogenic harm? Clin Orthop Relat Res. 2013;471:2737–2742. 25. Maltenfort MG, Rasouli MR, Morrison TA, et al. Clostridium difficile colitis in patients undergoing lower-extremity arthroplasty: rare infection with major impact. Clin Orthop Relat Res. 2013;471:3178–3185. 26. Saeed F, Adil MM, Khursheed F, et al. Acute renal failure is associated with higher death and disability in patients with acute ischemic stroke: analysis of nationwide inpatient sample. Stroke. 2014;45:1478–1480.
27. Walkey AJ, Wiener RS. Use of noninvasive ventilation in patients with acute respiratory failure, 2000–2009: a population-based study. Ann Am Thorac Soc. 2013;10:10–17. 28. Neuhaus V, King J, Hageman MG, et al. Charlson comorbidity indices and in-hospital deaths in patients with hip fractures. Clin Orthop Relat Res. 2013;471:1712–1719. 29. Browne JA, Pietrobon R, Olson SA. Hip fracture outcomes: does surgeon or hospital volume really matter? J Trauma. 2009;66:809–814. 30. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27. 31. Rasouli MR, Maltenfort MG, Purtill JJ, et al. Has the rate of in-hospital infections after total joint arthroplasty decreased? Clin Orthop Relat Res. 2013;471:3102–3111. 32. Menendez ME, Neuhaus V, van Dijk CN, et al. The Elixhauser comorbidity method outperforms the Charlson index in predicting inpatient death after orthopaedic surgery. Clin Orthop Relat Res. 2014;472:2878–2886. 33. Nandyala SV, Marquez-Lara A, Fineberg SJ, et al. Perioperative characteristics and outcomes of patients undergoing anterior cervical fusion in July: analysis of the “July effect”. Spine (Phila Pa 1976). 2014;39:612–617. 34. Jain NB, Pietrobon R, Guller U, et al. Influence of provider volume on length of stay, operating room time, and discharge status for rotator cuff repair. J Shoulder Elbow Surg. 2005;14:407–413. 35. Reames BN, Birkmeyer NJ, Dimick JB, et al. Socioeconomic disparities in mortality after cancer surgery: failure to rescue. JAMA Surg. 2014;149: 475–481. 36. Ghaferi AA, Osborne NH, Birkmeyer JD, et al. Hospital characteristics associated with failure to rescue from complications after pancreatectomy. J Am Coll Surg. 2010;211:325–330. 37. Waits SA, Sheetz KH, Campbell DA, et al. Failure to rescue and mortality following repair of abdominal aortic aneurysm. J Vasc Surg. 2014; 59:909–914e1. 38. Reddy HG, Shih T, Englesbe MJ, et al. Analyzing “failure to rescue”: is this an opportunity for outcome improvement in cardiac surgery? Ann Thorac Surg. 2013;95:1976–1981; discussion 1981. 39. Silber JH, Romano PS, Rosen AK, et al. Failure-to-rescue: comparing definitions to measure quality of care. Med Care. 2007;45:918–925. 40. Fleischut PM, Mazumdar M, Memtsoudis SG. Perioperative database research: possibilities and pitfalls. Br J Anaesth. 2013;111:532–534. 41. Memtsoudis SG. Limitations associated with the analysis of data from administrative databases. Anesthesiology. 2009;111:449; author reply 450–441. 42. Tseng VL, Yu F, Lum F, et al. Risk of fractures following cataract surgery in Medicare beneficiaries. JAMA. 2012;308:493–501. 43. Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery. 1992;111:518–526. 44. Nikkel LE, Fox EJ, Black KP, et al. Impact of comorbidities on hospitalization costs following hip fracture. J Bone Joint Surg Am. 2012;94:9–17. 45. Trinh QD, Bianchi M, Hansen J, et al. In-hospital mortality and failure to rescue after cytoreductive nephrectomy. Eur Urol. 2013;63:1107–1114. 46. Trinh VQ, Trinh QD, Tian Z, et al. In-hospital mortality and failure-torescue rates after radical cystectomy. BJU Int. 2013;112:E20–E27. 47. Glenn LL, Jijon CR. Risk-adjusted in-hospital death rates for peer hospitals in rural and urban regions. J Rural Health. 1999;15:94–107. 48. Sasichay-Akkadechanunt T, Scalzi CC, Jawad AF. The relationship between nurse staffing and patient outcomes. J Nurs Adm. 2003;33:478–485. 49. James PA, Li P, Ward MM. Myocardial infarction mortality in rural and urban hospitals: rethinking measures of quality of care. Ann Fam Med. 2007;5:105–111. 50. Ghaferi AA, Birkmeyer JD, Dimick JB. Complications, failure to rescue, and mortality with major inpatient surgery in Medicare patients. Ann Surg. 2009;250:1029–1034. 51. Henneman D, van Leersum NJ, Ten Berge M, et al. Failure-to-rescue after colorectal cancer surgery and the association with three structural hospital factors. Ann Surg Oncol. 2013;20:3370–3376. 52. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491–499. 53. Sparkes D, Rylah B. The World Health Organization surgical safety checklist. Br J Hosp Med (Lond). 2010;71:276–280. 54. Weiser TG, Haynes AB, Dziekan G, et al. Effect of a 19-item surgical safety checklist during urgent operations in a global patient population. Ann Surg. 2010;251:976–980.
e102
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
REFERENCES
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Failure to Rescue After Proximal Femur Fracture Surgery Mariano E. Menendez, MD and David Ring, MD, PhD
Objectives: Failure to rescue (FTR)––death after a major adverse event––has recently been identified as an important determinant of variation in surgical mortality. We sought to identify patient and hospital characteristics associated with FTR after proximal femur fracture surgery, and to determine whether they are different from the predictors of the occurrence of adverse events. We also identified which adverse events are most highly associated with FTR.
Methods: Among an estimated 287,959 patients with a surgically treated proximal femur fracture identified in the 2011 Nationwide Inpatient Sample, the overall adverse event rate was 22% and the FTR rate was 6.4%. Multivariable logistic regression modeling was used to identify independent predictors of FTR and adverse events. Results: Patient-specific variables influenced adverse event occurrence but exerted little or no influence on FTR. Hospitals located in rural areas were 14% less likely than urban hospitals to have adverse events [odds ratio (OR): 0.86, 95% confidence interval (CI): 0.83– 0.89], but 30% more likely to fail to rescue patients from adverse events (OR: 1.3, 95% CI: 1.2–1.5). Compared with teaching and large hospitals, nonteaching settings and institutions of smaller size were associated with decreased risk for adverse events, but similar risk for FTR. Lower hospital volume was a risk factor for adverse events and FTR. There was a more than 3-fold increased risk for death among patients with respiratory failure (OR: 5.4, 95% CI: 5.0– 5.8), pulmonary embolism (OR: 3.6, 95% CI: 3.1–4.1), and myocardial infarction (OR: 3.0, 95% CI: 2.8–3.3). Conclusions: Because FTR is less affected by patient characteristics than morbidity, it might be a better measure of providerspecific performance in hip fracture surgery. Targeted initiatives aimed at improving the timely recognition and management of cardiorespiratory adverse events ––particularly at rural hospitals–– might be key to reducing mortality. Key Words: failure to rescue, mortality, adverse events, inpatient, Elixhauser, complications, epidemiology, nationwide inpatient sample, trauma Accepted for publication August 19, 2014. From the Department of Orthopaedic Surgery, Orthopaedic Hand and Upper Extremity Service, Yawkey Center, Massachusetts General Hospital, Boston, MA. The authors report no conflict of interest. No IRB approval is mandatory for this study. The data are deidentified and commercially available for use. This work was performed at the Orthopaedic Hand and Upper Extremity Service, Massachusetts General Hospital, Boston, MA. Reprints: David Ring, MD, PhD, Orthopaedic Hand Service, Yawkey Center, Massachusetts General Hospital, Suite 2100, 55 Fruit St, Boston, MA 02114 (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
e96
| www.jorthotrauma.com
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence. (J Orthop Trauma 2015;29:e96–e102)
INTRODUCTION Surgical mortality rates are routinely used to evaluate patient safety and provider performance.1–6 Quality improvement efforts to reduce mortality have traditionally focused on reducing rates of intraoperative and postoperative adverse events with the assumption that these are directly related to mortality.7–13 There is growing recognition that the major factor influencing perioperative mortality may not be the occurrence of an adverse event, but rather how patients with an adverse event are managed.14–17 Indeed, low-mortality hospitals are distinguished not by having fewer adverse events, but by more successfully preventing death among patients that experience an adverse event.17 The emerging concept of failure to rescue (FTR)––the inability to rescue a patient from death after an adverse event––has become an important determinant of outcome for hospitalized patients and is increasingly used as a surgical quality indicator.18–20 We sought to identify which patient and hospital characteristics merit an increased alertness for “rescue” after proximal femur fracture surgery, and to determine whether they are different from the predictors of the occurrence of adverse events. In addition, we identified which adverse events are most highly associated with FTR.
PATIENTS AND METHODS Encounter data for this retrospective cohort study were abstracted from the Nationwide Inpatient Sample (NIS), currently the largest all-payer inpatient care database in the United States.21,22 Managed by the Agency for Healthcare Research and Quality, the NIS is a nationally representative cluster sample of discharges from more than 1000 short-term and nonfederal hospitals.23 Stratification of the hospitals for sampling is based on geographic region, bed size, teaching status, and rural/urban location. Discharge weight files are provided to derive statistically valid national estimates. Beside patient demographic and hospital-related data, the NIS collects up to 25 medical diagnoses and 15 procedures with the use of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. In recognition of its utility to answer valuable clinical questions, the NIS has been adopted extensively to conduct populationbased research across diverse medical disciplines.24–27 Formal approval by our institutional review board was not necessary J Orthop Trauma Volume 29, Number 3, March 2015
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
TABLE 1. In-Hospital Mortality and FTR Rates Among Patients With $1 Adverse Event Parameter Total Age group (y) ,50 50–59 60–69 70–79 80–89 .89 Sex Female Male Elixhauser comorbidity score 0 1–2 $3 Race/ethnicity White Black Hispanic Other Insurance status Medicare Medicaid Private Uninsured Other Median household income Low Medium High Highest Hospital volume Low Intermediate High Hospital bed size Small Medium Large Hospital teaching status Nonteaching Teaching Hospital location Urban Rural Trauma type Single-trauma Polytrauma
‡1 Adverse Event, n (%)
In-Hospital Mortality, n
64,248 (22)
4105
6.4
854 2055 4865 12,355 28,843 15,276
5† 69 187 699 1908 1238
0.59† 3.4 3.8 5.7 6.6 8.1
2502 1603
6.0 7.0
FTR, %*
P ,0.001
(9.7) (14) (16) (20) (24) (29)
,0.001 41,428 (21) 22,816 (27)
,0.001
1078 (6.5) 13,492 (14) 49,678 (29)
36 655 3414
3.3 4.9 6.9
57,108 2481 2776 1884
(22) (23) (20) (21)
3702 110 170 123
6.5 4.4 6.1 6.5
56,389 1376 4879 540 890
(24) (19) (16) (12) (15)
3748 66 223 14 49
6.6 4.8 4.6 2.6 5.5
0.001
,0.001
0.001 16,361 15,606 16,815 14,361
(23) (22) (22) (23)
1094 1072 996 874
6.7 6.9 5.9 6.1
21,072 (21) 22,153 (23) 21,023 (23)
1370 1471 1263
6.5 6.6 6.0
7288 (21) 16,822 (23) 39,353 (23)
484 1054 2517
6.6 6.3 6.4
,0.001
0.39
0.028 37,214 (22) 26,250 (23)
2312 1744
6.2 6.6
55,375 (23) 8089 (20)
3430 625
6.2 7.7
56,494 (22) 7754 (27)
3510 595
6.2 7.7
,0.001
,0.001
*Calculated as number of deaths/number of with $1 adverse event. †,11 patients.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
for this study, because the data contained in the NIS are anonymous. A cohort of patients with an ICD-9-CM primary diagnosis of a proximal femur fracture (codes 820.0 and 820.1 for transcervical, 820.2 and 820.3 for pertrochanteric, and 820.8 and 820.9 for unspecified hip fractures) were generated.28 Only those patients who underwent subsequent open reduction or internal fixation (ICD-9-CM codes 78.55, 79.15, 79.25, 79.35), hemiarthroplasty (81.52), or total hip arthroplasty (81.51), were included for further analysis.29 Between January 1, 2011 and December 31, 2011, an estimated 287,959 patients met the inclusion criteria for the study. Patient-specific variables included age, sex, race/ ethnicity (white, black, Hispanic, and other), insurance status (Medicare, Medicaid, private, uninsured, and other), household income of the patient’s zip code of residence (low, medium, high, and highest), and trauma type (singletrauma, and polytrauma). Polytrauma was defined using the following ICD-9 secondary diagnosis codes: 800–819, 821–909, and 925–959. Baseline comorbidity status was quantified using the Elixhauser comorbidity score.30–32 Our study sample was predominantly female (70%) and white (88%), and the most common decade of fracture occurrence was in the 80s (42%). Using data from the American Hospital Association Annual Survey of Hospitals available in the NIS, hospitals were classified according to their location (urban and rural), bed size (small, medium, and large), and teaching status (teaching and nonteaching). Hospitals are considered to be teaching institutions if any of the following 3 criteria are met1: residency-training approval by the Accreditation Council for Graduate Medical Education,2 membership in the Council of Teaching Hospitals, or3 a resident to bed ratio of 0.25 or more.33 Bed size cutoff points vary so that approximately one-third of the hospitals in a given region, location, and teaching status combination falls within each bed size category (small, medium, or large). Using a unique identifier number assigned by the NIS to each hospital, we were also able to derive the hospital volume by counting the number of proximal femur fracture surgeries performed at each hospital during 2011.34 Hospital volume was then categorized into 3 groups by patient-based tertiles (low-volume: ,98 procedures; intermediate-volume: 98–179 procedures; highvolume: $180 procedures).17 By the use of ICD-9-CM diagnosis codes,16,17,35 we considered major in-hospital adverse events, such as myocardial infarction (410), acute renal failure (584), respiratory failure (518.4, 518.5, 514), cerebrovascular accident (433, 434, 997.02), deep venous thrombosis (451, 453), pulmonary embolism (415.1), pneumonia (481, 482, 483, 484, 485, 486, 997.31, 997.39), bacteremia/sepsis (038, 7907), surgical site infection (998.3, 998.5), and postoperative hemorrhage (998.1). Our primary study end point was FTR, defined as a case fatality among patients with 1 or more of the defined major adverse events.16,17 Pearson’s x2 test was used for analysis of categorical data and independent samples t test for continuous data. FTR rates were calculated for all demographic and www.jorthotrauma.com |
e97
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Menendez and Ring
TABLE 2. Risk Factors for Adverse Events and FTR After Hip Fracture Surgery Adverse Events
FTR
95% CI
95% CI
Predictor
OR
Lower
Upper
P
OR
Lower
Upper
P
Age, per 10-year increase Male sex (reference: female) Elixhauser comorbidity score, per 1-unit increase Race (reference: white) Black Hispanic Other Insurance status (reference: private insurance) Medicare Medicaid Uninsured Other Household income (reference: highest) Low Medium High Hospital volume (reference: high) Low Intermediate Hospital bed size (reference: large) Small Medium Nonteaching hospital (reference: teaching hospital) Rural hospital (reference: urban hospital) Polytrauma (reference: single-trauma) Complications (reference: no complication) Myocardial infarction Acute renal failure Respiratory failure Cerebrovascular accident Deep venous thrombosis Pulmonary embolism Pneumonia Bacteremia per sepsis Surgical site infection Postoperative hemorrhage
1.3 1.6 1.4
1.3 1.6 1.35
1.3 1.6 1.4
,0.001 ,0.001 ,0.001
1.4 1.1 1.1
1.3 1.1 1.05
1.4 1.2 1.1
,0.001 ,0.001 ,0.001
1.0 0.84 0.92
0.96 0.80 0.87
1.1 0.87 0.97
0.69 ,0.001 ,0.001
0.91 1.0 1.0
0.74 0.88 0.84
1.1 1.2 1.3
0.40 0.61 0.77
1.0 1.3 1.07 1.0
1.0 1.2 0.97 0.94
1.1 1.3 1.2 1.1
0.04 ,0.001 0.20 0.63
1.1 1.3 0.73 1.3
0.95 0.93 0.37 0.92
1.3 1.7 1.4 1.8
0.19 0.14 0.35 0.14
1.1 1.0 1.0
1.04 0.97 0.97
1.1 1.0 1.0
,0.001 0.87 0.85
1.0 1.0 1.0
0.94 0.95 0.91
1.2 1.2 1.1
0.42 0.37 0.99
1.1 1.04
1.05 1.02
1.1 1.1
,0.001 ,0.001
1.1 1.1
0.97 1.04
1.2 1.2
0.15 0.0040
0.83 0.95 0.89
0.81 0.93 0.87
0.86 0.98 0.90
,0.001 ,0.001 ,0.001
1.0 1.0 0.92
0.92 0.90 0.85
1.2 1.1 1.00
0.48 0.78 0.024
0.86
0.83
0.89
,0.001
1.3
1.2
1.5
,0.001
1.4
1.4
1.4
,0.001
1.2
1.1
1.3
,0.001
— — — — — — — — — —
— — — — — — — — — —
— — — — — — — — — —
— — — — — — — — — —
3.0 1.6 5.4 2.2 0.55 3.6 2.2 1.9 1.8 0.69
2.8 1.5 5.0 1.9 0.44 3.1 2.0 1.7 1.4 0.56
3.3 1.7 5.8 2.6 0.69 4.1 2.3 2.0 2.5 0.86
,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 0.001
Statistical significance set at P , 0.01.
RESULTS
clinical variables, as well as for all major adverse events. Multivariable logistic regression analysis was first performed to determine perioperative factors associated with the development of major adverse events. Second, multivariable analysis was undertaken to determine predictors of FTR in patients experiencing any of the predefined adverse events. Results were reported as odds ratios (OR) with 95% confidence intervals (CI). The statistical threshold for alpha error was set at 0.01.
Among an estimated 64,248 patients (22%) experiencing at least 1 adverse event, 4105 died during hospitalization, corresponding to an FTR rate of 6.4% (Table 1). Adverse event and FTR rates rose steadily with age and comorbidity burden. FTR rates were higher in male patients than in female patients (7.0% vs. 6.0%, P , 0.001), and in polytrauma patients compared with single-trauma patients (7.7% vs. 6.2%,
e98
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
FIGURE 1. FTR rates for each adverse event.
P , 0.001). Although high-volume hospitals had higher adverse event rates (23.1%) than low-volume (21.3%) and intermediate-volume (22.6%) hospitals, they were better at rescuing patients and thus exhibited lower FTR rates (6.0% vs. 6.5% and 6.6%, respectively; P , 0.001). Despite higher adverse event rates in urban (23%) than in rural (20%) hospitals, facilities located in rural areas (7.7%) had higher FTR rates than those located in urban settings (6.2%; P , 0.001). In multivariable logistic regression analysis, patient-specific variables, such as sex, Elixhauser comorbidity score, and polytrauma exerted more influence (higher ORs) on adverse event occurrence than on FTR (Table 2). Insurance status, race, and household income influenced adverse events, but not FTR. Compared with hospitals in urban settings, facilities located in rural areas were associated with lower odds of adverse events (OR: 0.86, 95% CI: 0.83–0.89, P , 0.001) but higher odds of FTR (OR: 1.3, 95% CI: 1.2–1.5, P , 0.001). Patients treated at nonteaching hospitals were less likely to experience adverse events (OR: 0.89, 95% CI: 0.87–0.90, P , 0.001) but were as likely as those treated at teaching hospitals to succumb after an adverse event. Compared with large-sized hospitals, institutions of small (OR: 0.83, 95% CI: 0.81–0.86, P , 0.001) and medium size (OR: 0.95, 95% CI: 0.93–0.98, P , 0.001) were associated with decreased risk for adverse events, but comparable risk for FTR. Patients treated at low-volume and intermediatevolume hospitals tended to be at increased risk for adverse events and FTR, compared with those treated at high-volume hospitals. Respiratory failure (19%) was the adverse event with the highest FTR rate, followed by pulmonary embolism (15%) and myocardial infarction (14%; Fig. 1). The FTR rate Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
was 16% in patients experiencing any of these 3 adverse events, and 6.0% among patients with adverse events other than the latter 3. In multivariable modeling, the risk of mortality was increased by 5.4 times (OR: 5.4, 95% CI: 5.0–5.8, P , 0.001) in patients with respiratory failure, 3.6 times (OR: 3.6, 95% CI: 3.1–4.1, P , 0.001) in patients affected by pulmonary embolism, and 3 times (OR: 3.0, 95% CI: 2.8– 3.3, P , 0.001) among those suffering myocardial infarction. Hospitals located in urban areas had generally higher rates of incidence of these adverse events, but lower rates of FTR (Fig. 2). For all 3 adverse events, high-volume hospitals reported the highest rates of incidence yet the lowest rates of FTR (Fig. 3).
DISCUSSION Ideally, decreases in mortality after surgery would reflect minimization of errors and adverse events. Accumulating evidence suggests that timely recognition and effective management of adverse events may also be important in explaining rates of death, particularly among infirm patients or patients having high-risk procedures.15,17,36,37 The Agency for Healthcare Research and Quality has recently adopted FTR as a patient safety indicator,38,39 and the American College of Surgeons National Surgical Quality Improvement Program has begun to report FTR rates as part of their annual summary report.36 Using nationally representative data, we identified factors associated with FTR after proximal femur fracture surgery and determined whether they were different from those associated with the development of adverse events. Although our analysis benefits from access to large numbers and associated power, we acknowledge a number of www.jorthotrauma.com |
e99
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Menendez and Ring
J Orthop Trauma Volume 29, Number 3, March 2015
FIGURE 2. Rates of adverse events and FTR according to hospital location.
important shortcomings.40,41 First, the primary purpose of administrative claims data is for billing, and, as such, there is likely an underreporting of adverse events. To minimize this bias, we limited our analysis to major postoperative adverse events that are likely to generate a claim.16 Second, the possibility of errors in coding of the diagnoses and procedures cannot be avoided; nonetheless, coding mistakes tend to be evenly distributed in large-scale studies.42 Third, the NIS does not include relevant data such as the number of healthcare personnel and specialties involved in patient care, and the chronological occurrence and severity of adverse events.43 Fourth, another limitation was our inability to adjust for surgeon volume and energy of injury, factors known to influence inpatient outcomes after proximal femur fracture surgery.29 However, we controlled for polytrauma, which may act as a surrogate for energy of injury. Fifth, we limited our analysis to a single year of data. However, restricting the population to a single time period could also be seen as strength, because the diagnostic evaluation and operative techniques are unlikely to have changed and are more likely to be consistent across the population.44 Sixth, the retrospective nature of the NIS does not allow ascertainment of the exact reason for death. Finally, the reader should be aware that findings in large-scale studies can be statistically
significant yet clinically insignificant. The reader should consider whether small but significant differences are clinically important. Older age and higher comorbidity burden were associated with greater risk for FTR after proximal femur fracture surgery. Our finding is consistent with recent studies in patients undergoing cytoreductive nephrectomy and radical cystectomy.45,46 Male patients were 60% more likely than women to experience an adverse event, but only 10% more likely to die of an adverse event. Other patient characteristics, such as insurance status, race, and household income influenced adverse events, but not FTR. Confirming previous research,20 FTR was not as affected by patient characteristics as were adverse events. Because FTR seems to be less sensitive to conditions outside of the control of hospital staff such as patient demographics and comorbidity, it might be a better measure of provider-specific performance in proximal femur fracture surgery. Compared with urban facilities, hospitals located in rural areas were less likely to have adverse events, but more likely to fail from preventing death once they occurred. It is possible that sicker––more prone to morbidity––patients were admitted to urban facilities, which would explain the decreased risk for adverse events in rural hospitals.47 The
e100
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Failure to Rescue After Hip Fracture Surgery
FIGURE 3. Rates of adverse events and FTR according to hospital volume.
higher risk for dying after an adverse event in rural hospitals might indicate that these facilities are less prepared to manage adverse events, perhaps because of limited expertise,48 restricted access to high-technology equipment,49 or less coordinated care.50 Consistent with research in other surgical disciplines,16,17,45,46,51 lower hospital volume was an independent predictor of adverse events and FTR after proximal femur fracture surgery. Hospital teaching status and bed size were determinants of adverse events, but not of FTR: compared with teaching and large-sized hospitals, nonteaching hospitals and institutions of smaller size were associated with decreased risk for adverse events, but similar risk for mortality after an adverse event. These findings add to the current body of knowledge suggesting that an increased risk for morbidity does not always correlate with a higher mortality risk (eg, FTR). Furthermore, they suggest that FTR might be incorporated in pay-for-performance initiatives, because it seems to be a more discriminative measure of hospital performance than morbidity,15,17,36,37 at least for infirm patients and high-risk procedures. Hospital staff should be particularly mindful of the more than 3-fold increased risk for death among patients with respiratory failure, pulmonary embolism, and myocardial infarction. In view of these findings, quality improvement Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
efforts should be prioritarily aimed at both reducing and effectively managing cardiorespiratory complications in hospitalized patients with a proximal femur fracture. Testing FTR as a quality-of-care metric in orthopaedic surgery is worthy of future work, as much of the outcomebased orthopaedic literature has primarily focused on traditional morbidity and mortality rates. For instance, the analysis of temporal trends in FTR rates for common elective orthopaedic procedures would be of value. Furthermore, it would be interesting to evaluate the impact of surgical safety checklists on FTR rates.52 Quality initiatives to reduce intraoperative and postoperative adverse events (such as the World Health Organization protocol) are important ways to reduce surgical morbidity and mortality, particularly for surgeries with low rates of adverse events and death.52–54 In a low-risk setting, the proportion of adverse events related to error is expected to be higher and the impact of checklists and other measures greater. In contrast, with high-risk patients or procedures, most adverse events relate more to patient infirmity and misfortune than to quality of care. Improved surgical safety should therefore aim to both reduce adverse events related to errors or inadequate preparation or communication, and to optimize the ability to rescue patients from adverse events. www.jorthotrauma.com |
e101
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma Volume 29, Number 3, March 2015
Menendez and Ring
Additional research is needed to determine if rescue of infirm patients leads to an acceptable quality extension of life.
1. Dimick JB, Ghaferi AA, Osborne NH, et al. Reliability adjustment for reporting hospital outcomes with surgery. Ann Surg. 2012;255:703–707. 2. Dimick JB, Staiger DO, Hall BL, et al. Composite measures for profiling hospitals on surgical morbidity. Ann Surg. 2013;257:67–72. 3. Dimick JB, Welch HG, Birkmeyer JD. Surgical mortality as an indicator of hospital quality: the problem with small sample size. JAMA. 2004;292: 847–851. 4. Hall BL, Hamilton BH, Richards K, et al. Does surgical quality improve in the American College of Surgeons National Surgical Quality Improvement Program: an evaluation of all participating hospitals. Ann Surg. 2009;250:363–376. 5. Teng CI, Dai YT, Shyu YI, et al. Professional commitment, patient safety, and patient-perceived care quality. J Nurs Scholarsh. 2009;41: 301–309. 6. Hashmi ZG, Schneider EB, Castillo R, et al. Benchmarking trauma centers on mortality alone does not reflect quality of care: Implications for pay-for-performance. J Trauma Acute Care Surg. 2014;76:1184–1191. 7. Hemmila MR, Nathens AB, Shafi S, et al. The Trauma Quality Improvement Program: pilot study and initial demonstration of feasibility. J Trauma. 2010;68:253–262. 8. Jha AK, Joynt KE, Orav EJ, et al. The long-term effect of premier pay for performance on patient outcomes. N Engl J Med. 2012;366:1606–1615. 9. Moore L, Stelfox HT, Turgeon AF. Complication rates as a trauma care performance indicator: a systematic review. Crit Care. 2012;16:R195. 10. Rosenthal MB, Landon BE, Normand SL, et al. Pay for performance in commercial HMOs. N Engl J Med. 2006;355:1895–1902. 11. Shih T, Nicholas LH, Thumma JR, et al. Does pay-for-performance improve surgical outcomes? An evaluation of phase 2 of the premier hospital quality incentive demonstration. Ann Surg. 2014;259:677–681. 12. Sutton M, Nikolova S, Boaden R, et al. Reduced mortality with hospital pay for performance in England. N Engl J Med. 2012;367:1821–1828. 13. Brailer DJ, Kroch E, Pauly MV, et al. Comorbidity-adjusted complication risk: a new outcome quality measure. Med Care. 1996;34:490–505. 14. Ghaferi AA, Dimick JB. Variation in mortality after high-risk cancer surgery: failure to rescue. Surg Oncol Clin N Am. 2012;21:389–395, vii. 15. Sheetz KH, Waits SA, Krell RW, et al. Improving mortality following emergent surgery in older patients requires focus on complication rescue. Ann Surg. 2013;258:614–617; discussion 617–618. 16. Wright JD, Ananth CV, Ojalvo L, et al. Failure to rescue after major gynecologic surgery. Am J Obstet Gynecol. 2013;209:420e1–428e1. 17. Wright JD, Herzog TJ, Siddiq Z, et al. Failure to rescue as a source of variation in hospital mortality for ovarian cancer. J Clin Oncol. 2012;30: 3976–3982. 18. Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in hospital mortality associated with inpatient surgery. N Engl J Med. 2009;361:1368–1375. 19. Sheetz KH, Guy K, Allison JH, et al. Improving the care of elderly adults undergoing surgery in Michigan. J Am Geriatr Soc. 2014;62:352–357. 20. Silber JH, Williams SV, Krakauer H, et al. Hospital and patient characteristics associated with death after surgery. A study of adverse occurrence and failure to rescue. Med Care. 1992;30:615–629. 21. Bohl DD, Basques BA, Golinvaux NS, et al. Nationwide inpatient sample and national surgical quality improvement program give different results in hip fracture studies. Clin Orthop Relat Res. 2014;472:1672–1680. 22. Pallin DJ, Espinola JA, Camargo CA Jr. US population aging and demand for inpatient services. J Hosp Med. 2014;9:193–196. 23. Wakeam E, Hevelone ND, Maine R, et al. Failure to rescue in safety-net hospitals: availability of hospital resources and differences in performance. JAMA Surg. 2014;149:229–235. 24. D’Apuzzo MR, Keller TC, Novicoff WM, et al. CT pulmonary angiography after total joint arthroplasty: overdiagnosis and iatrogenic harm? Clin Orthop Relat Res. 2013;471:2737–2742. 25. Maltenfort MG, Rasouli MR, Morrison TA, et al. Clostridium difficile colitis in patients undergoing lower-extremity arthroplasty: rare infection with major impact. Clin Orthop Relat Res. 2013;471:3178–3185. 26. Saeed F, Adil MM, Khursheed F, et al. Acute renal failure is associated with higher death and disability in patients with acute ischemic stroke: analysis of nationwide inpatient sample. Stroke. 2014;45:1478–1480.
27. Walkey AJ, Wiener RS. Use of noninvasive ventilation in patients with acute respiratory failure, 2000–2009: a population-based study. Ann Am Thorac Soc. 2013;10:10–17. 28. Neuhaus V, King J, Hageman MG, et al. Charlson comorbidity indices and in-hospital deaths in patients with hip fractures. Clin Orthop Relat Res. 2013;471:1712–1719. 29. Browne JA, Pietrobon R, Olson SA. Hip fracture outcomes: does surgeon or hospital volume really matter? J Trauma. 2009;66:809–814. 30. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27. 31. Rasouli MR, Maltenfort MG, Purtill JJ, et al. Has the rate of in-hospital infections after total joint arthroplasty decreased? Clin Orthop Relat Res. 2013;471:3102–3111. 32. Menendez ME, Neuhaus V, van Dijk CN, et al. The Elixhauser comorbidity method outperforms the Charlson index in predicting inpatient death after orthopaedic surgery. Clin Orthop Relat Res. 2014;472:2878–2886. 33. Nandyala SV, Marquez-Lara A, Fineberg SJ, et al. Perioperative characteristics and outcomes of patients undergoing anterior cervical fusion in July: analysis of the “July effect”. Spine (Phila Pa 1976). 2014;39:612–617. 34. Jain NB, Pietrobon R, Guller U, et al. Influence of provider volume on length of stay, operating room time, and discharge status for rotator cuff repair. J Shoulder Elbow Surg. 2005;14:407–413. 35. Reames BN, Birkmeyer NJ, Dimick JB, et al. Socioeconomic disparities in mortality after cancer surgery: failure to rescue. JAMA Surg. 2014;149: 475–481. 36. Ghaferi AA, Osborne NH, Birkmeyer JD, et al. Hospital characteristics associated with failure to rescue from complications after pancreatectomy. J Am Coll Surg. 2010;211:325–330. 37. Waits SA, Sheetz KH, Campbell DA, et al. Failure to rescue and mortality following repair of abdominal aortic aneurysm. J Vasc Surg. 2014; 59:909–914e1. 38. Reddy HG, Shih T, Englesbe MJ, et al. Analyzing “failure to rescue”: is this an opportunity for outcome improvement in cardiac surgery? Ann Thorac Surg. 2013;95:1976–1981; discussion 1981. 39. Silber JH, Romano PS, Rosen AK, et al. Failure-to-rescue: comparing definitions to measure quality of care. Med Care. 2007;45:918–925. 40. Fleischut PM, Mazumdar M, Memtsoudis SG. Perioperative database research: possibilities and pitfalls. Br J Anaesth. 2013;111:532–534. 41. Memtsoudis SG. Limitations associated with the analysis of data from administrative databases. Anesthesiology. 2009;111:449; author reply 450–441. 42. Tseng VL, Yu F, Lum F, et al. Risk of fractures following cataract surgery in Medicare beneficiaries. JAMA. 2012;308:493–501. 43. Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery. 1992;111:518–526. 44. Nikkel LE, Fox EJ, Black KP, et al. Impact of comorbidities on hospitalization costs following hip fracture. J Bone Joint Surg Am. 2012;94:9–17. 45. Trinh QD, Bianchi M, Hansen J, et al. In-hospital mortality and failure to rescue after cytoreductive nephrectomy. Eur Urol. 2013;63:1107–1114. 46. Trinh VQ, Trinh QD, Tian Z, et al. In-hospital mortality and failure-torescue rates after radical cystectomy. BJU Int. 2013;112:E20–E27. 47. Glenn LL, Jijon CR. Risk-adjusted in-hospital death rates for peer hospitals in rural and urban regions. J Rural Health. 1999;15:94–107. 48. Sasichay-Akkadechanunt T, Scalzi CC, Jawad AF. The relationship between nurse staffing and patient outcomes. J Nurs Adm. 2003;33:478–485. 49. James PA, Li P, Ward MM. Myocardial infarction mortality in rural and urban hospitals: rethinking measures of quality of care. Ann Fam Med. 2007;5:105–111. 50. Ghaferi AA, Birkmeyer JD, Dimick JB. Complications, failure to rescue, and mortality with major inpatient surgery in Medicare patients. Ann Surg. 2009;250:1029–1034. 51. Henneman D, van Leersum NJ, Ten Berge M, et al. Failure-to-rescue after colorectal cancer surgery and the association with three structural hospital factors. Ann Surg Oncol. 2013;20:3370–3376. 52. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491–499. 53. Sparkes D, Rylah B. The World Health Organization surgical safety checklist. Br J Hosp Med (Lond). 2010;71:276–280. 54. Weiser TG, Haynes AB, Dziekan G, et al. Effect of a 19-item surgical safety checklist during urgent operations in a global patient population. Ann Surg. 2010;251:976–980.
e102
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
REFERENCES
| www.jorthotrauma.com
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
