ORIGINAL ARTICLE
An analysis of geriatric recidivism in the era of accountable care organizations Katelyn Rittenhouse, Carissa Harnish, Brian Gross, Amelia Rogers, Jo Ann Miller, Roxanne Chandler, RHIA, and Frederick B. Rogers, MD, MS, Lancaster, Pennsylvania
To date, there are almost 500 accountable care organizations (ACOs) across the United States emphasizing cost-effective care. Readmission largely impacts health care cost; therefore, we sought to determine factors associated with geriatric trauma readmissions (recidivism) within our institution. METHODS: All admissions from 2000 to 2011 attributed to patients 65 years or older at our Level II trauma center, recently verified by Medicare as an ACO, were queried. Patients were classified as recidivist or nonrecidivist. The first admissions of recidivist patients were compared with the nonrecidivist admissions with respect to sex, age, race, primary insurance, admission Glasgow Coma Scale (GCS) score, Injury Severity Score (ISS), hospital length of stay, mechanism of injury (MOI), preexisting conditions, and discharge destination. Factors found to be significant predictors of recidivism in univariate analyses were subsequently incorporated into a multivariate logistic regression model. In addition, the second admission’s MOI was compared with the first admission’s MOI, and the proportion of first, second, and third admissions attributed to falls was calculated. A p G 0.05 was significant. RESULTS: Between 2000 and 2011, a total of 4,963 unique patients were admitted to the trauma center at 65 years or older. This population was composed of 287 recidivists (5.8%) and 4,676 nonrecidivists (94.2%). When placed in a multivariate logistic regression, female sex, admission GCS score of 15, history of head trauma, and preexisting pulmonary disease were identified as significant predictors of recidivism. A trend toward increasing proportion of injuries attributed to falls was found with each subsequent trauma admission (81.5% [234 of 287] of first admissions, 88.2% [253 of 287] of second admissions, and 90.5% [19 of 21] of third admissions). CONCLUSION: Our study identifies specific factors that should be targeted by social service and prevention resources to inhibit recidivism in the elderly. In the brave new world of ACOs, trauma centers must identify high-risk populations for the consumption of limited resources. (J Trauma Acute Care Surg. 2015;78: 409Y414. Copyright * 2015 Wolters Kluwer Health, Inc. All rights reserved.) LEVEL OF EVIDENCE: Care management study, level IV. Prognostic study, level III. KEY WORDS: Recidivism; geriatric; accountable care organization. BACKGROUND:
A
lthough traditionally identified as a singular random event, a growing body of research has identified trauma as a predictable occurrence in certain high-risk groups, including young males, substance abusers, and the elderly.1 In 1990, Reiner et al.2 ascribed the term trauma recidivism to this trend. As trauma recidivism has been linked to loss of function and socioeconomic challenges after initial injury,3 characterizing the geriatric recidivist population for future prevention initiatives should be a priority for trauma programs. The geriatric population is receiving increased attention within the medical community because it is the fastest expanding segment of the population in the United States.4 Representing
Submitted: July 31, 2014, Revised: October 14, 2014, Accepted: October 20, 2014. From the Trauma Services, Lancaster General Health, Lancaster, Pennsylvania. This work was presented as a poster at the 73rd Annual Meeting of the American Association for the Surgery of Trauma and Clinical Congress of Acute Care Surgery, September 10Y13, 2014, in Philadelphia, Pennsylvania. Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.jtrauma.com). Address for reprints: Frederick B. Rogers, MD, MS, 555 N Duke St, Lancaster, PA 17602, email: [email protected]. DOI: 10.1097/TA.0000000000000513
12% of all Americans in 2000, those 65 years and older are expected to increase to 19% of the population by 2030.5 As a result, health care providers, including trauma surgeons, in the United States will need to manage limited resources effectively and efficiently to maximize care for the elderly.6 This demographic shift presents a challenge to trauma systems, as caring for geriatric patients requires a significant knowledge of aging-related diseases and proper management of preexisting conditions (PECs).6 Commonly, geriatric patients experience outcomes poorer than those of their younger counterparts because of increased comorbidities and less physiologic reserve.7 A recent study by Ciesla et al.8 of the Florida Agency for Health Care Administration database revealed a 17% increase in injury rates for the geriatric population between 1996 and 2010, while rates of injury decreased by 18% in children and grew by 2% in adults during the same period. Furthermore, a study by McGwin et al.3 reported that individuals in a previously injured geriatric cohort were 3.25 times more likely to be injured than a noninjured geriatric cohort. Therefore, geriatric recidivism should be a significant concern for trauma centers. Further incentivizing the prevention of geriatric trauma recidivism is the emergence of accountable care organizations (ACOs).9 ACOs were established within the Medicare Shared Savings Program (MSSP) to improve medical care of Medicare
J Trauma Acute Care Surg Volume 78, Number 2
409
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
Fee-For-Service beneficiaries and simultaneously reduce cost of care by encouraging better collaboration between health care providers.9,10 ACOs aim to improve health care by focusing on four pillars as follows: informatics, clinical integration, financial management, and care coordination.11 The MSSP addresses the reality that many geriatric patients are dependent on Medicare and supplementary insurance to cover health care costs by financially rewarding ACOs if the costs of their Medicare beneficiaries are below the set benchmark standards.12 In 2010, the adjusted estimate of trauma inpatient cost burden for Medicare was $17.5 billion yearly, representing nearly 47% of inpatient costs for traumatic injuries.13 Trauma centers within ACOs, such as our Pennsylvania-accredited Level II trauma center, may be able to contribute to achieving MSSP benchmarks through geriatric recidivism prevention initiatives. To effectively prioritize care of our aging patients, recidivism must be characterized in the geriatric population. To date, very few studies have investigated this topic.3,14,15 The objective of the study was to identify characteristics of geriatric trauma recidivists that may be used to develop targeted recidivism prevention interventions. We sought to determine what specific demographic factors, comorbidities, PECs. We hypothesized that falls and increasing age would be significant predictors for trauma recidivism. There were specific clinical factors that would predict a higher rate of geriatric recidivism. The geriatric population would have a rate of recidivism higher than the overall trauma recidivism rate at our institution. Furthermore, we hypothesized that falls would be a predictor of geriatric recidivism.
PATIENTS AND METHODS Following review and approval of this study by the institutional review board of Lancaster General Health, the registry of the Pennsylvania Trauma Systems Foundation (Digital Innovations, Forest Hill, MD) was queried for all trauma admissions of patients 65 years or older to the only Pennsylvania-accredited Level II trauma center in Lancaster County from 2000 to 2011. Variables collected included age, race, sex, Injury Severity Score (ISS), mechanism of injury (MOI), injury type, hospital length of stay, intensive care unit admission, mortality, PECs, discharge destination, and insurance status was collected for all trauma admissions. All variables were coded retrospectively from patients’ medical records by certified trauma registrars, with an accuracy rate of 98% to 99%, as determined by outside audits. The PECs evaluated in this study were based on the Pennsylvania System Trauma Foundation definitions (Supplemental File 1, at http://links.lww.com/TA/A508). PECs present in fewer than 50 patients were excluded from the analyses. Patients with names and medical record numbers associated with two or more trauma admissions during the study period were labeled recidivists (RCs). Patients with medical records associated with one trauma admission during the study period were labeled non-RCs (NRCs). Hospital readmissions for complications associated with the initial injury were excluded from the analyses. To calculate an overall recidivism rate for our geriatric population, patients with a trauma admission during the study period at age greater than or equal to 410
65 years with a previous trauma admission at age less than 65 years were also classified as recidivists. When determining predictors of geriatric recidivism, only initial recidivist admissions at which the patient was 65 years or older were compared with nonrecidivist admissions. The two-sample t test was used to compare continuous variables, and the W2 test was used to compare categorical variables between groups. Unadjusted odds ratios for recidivism were calculated for sex, primary insurance, admission GCS score, MOI, and PECs. Variables significantly associated or trending toward significance with recidivism in univariate analyses were subsequently incorporated into a multivariable binary logistic regression model to determine adjusted odds ratios for recidivism. A p G 0.05 was considered statistically significant. A p G 0.10 was considered a trend toward statistical significance.
RESULTS Between 2000 and 2011, there were a total of 4,963 unique patients admitted to the trauma center at 65 years or older. The patient population was composed of 2,071 males (41.7%) and 2,892 females (58.3%), with an age range from 65 years to 103 years. The median age was 80 years (interquartile range [IQR], 73Y85 years). The median ISS of the population was 9 (IQR, 5Y16). The most common MOI was fall (73.1%), followed by motor vehicle crash (MVC) (19.8%). A total of 246 patients (5.0%) had at least two trauma admissions at 65 years or older. In addition, 41 patients (0.8%) with initial trauma admissions when younger than 65 years during the study period also had subsequent trauma admissions when 65 years or older. Within this RC cohort of 287 individuals, 266 patients (92.7%) were admitted twice, 20 patients (7.0%) were admitted three times, and 1 patient (0.3%) was admitted for six unique traumatic events. The overall trauma recidivism rate for this geriatric population was determined to be 5.8% (287 of 4,963). The median time between trauma admissions was 1.9 years (695 days). The overall mortality rate (5.9%, 17 of 287) of recidivist patients was found to be statistically indistinguishable from the mortality rate (7.0%, 327 of 4,676) of nonrecidivist patients (p = 0.551). Within the recidivist patient population, 80.5% (231 of 287) of all second admissions were attributed to the MOI same as that of the initial admission (Fig. 1). Of the patients initially admitted for a fall, 93.2% (218 of 234) had the same MOI at second admission. Of the patients initially admitted for an MVC, 25% (9 of 36) had the same MOI at second admission. Of the 21 patients admitted for three unique traumatic injuries, 17 (81.0%) were admitted for the same MOI at all three occurrences. Of the 21 patients admitted to the trauma service three or more times, 16 (76.2%) were admitted for three falls. One patient was admitted for three unique motorcycle-related injuries. The other three patients were admitted for a fall two of their three admissions; this includes the first three admissions of the individual with six unique traumatic injuries. The last three admissions of the individual with six unique traumatic injuries were attributed to an MVC, an off-road vehicle accident, and a fall. Furthermore, within the recidivist population, * 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
TABLE 1. Demographics of Recidivist and Nonrecidivist Trauma Admissions
Figure 1. Rate of second admission due to same MOI as that of the initial injury.
the proportion of admissions attributed to falls increased with each admission (Fig. 2). When comparing the first admissions of recidivist patients at age 65 years or older with the admissions of nonrecidivist geriatric patients (Table 1), recidivist admissions were found to be more commonly attributed to females (RC, 67.5% vs. NRC, 57.8%; p = 0.003) and patients with Medicare as their primary insurance (RC, 78.0% vs. NRC, 68.8%; p = 0.003). A trend toward an increased likelihood for readmission of patients with a first GCS of 15 (RC, 83.6% vs. NRC, 78.2%; p = 0.053) was also observed. Furthermore, falls were more common (RC, 83.7% vs. NRC, 71.7%; p e 0.001) and MVCs were less common (RC, 10.16% vs. NRC, 21.1%; p G 0.001) as the MOI for recidivist initial admissions. A history of head trauma, pulmonary disease, and psychological disorders were all found to be significantly more common in the first admissions of recidivists than in the nonrecidivist admissions.
n Sex Male Female Race White Other Age, y 65Y84 Q85 Primary insurance Medicare Other Admission GCS score GCS score, 15 GCS score G 15 ISS Median (IQR) Hospital length of stay Median (IQR) Intensive care unit admission Yes No MOI Fall MVC Other Injury type Blunt Penetrating PECs Thyroid disease History of head trauma Renal disease Pulmonary disease Obesity Neurologic disorder Musculoskeletal disease Malignancy Psychological disorder Hematologic disease Gastrointestinal disease Diabetes Cardiac disease
RC First Admission
NRC Admission
246
4,717
32.5 (80) 67.5 (166)
42.2 (1,991) 57.8 (2,726)
0.003*
98.8 (243) 1.22 (3)
98.5 (4,638) 1.51 (71)
0.716
68.7 (169) 31.3 (77)
71.9 (3,393) 28.1 (1,324)
0.272
78.0 (191) 22.0 (54)
68.8 (3,237) 31.2 (1,465)
0.003*
83.6 (189) 16.4 (37)
78.2 (3,409) 21.8 (950)
0.053
9.00 (5.00Y14.0)
9.00 (5.00Y16.00)
4.00 (3.00Y6.00)
4.00 (2.00Y6.00)
p
42.3 (104) 57.7 (142)
44.1 (2,081) 55.9 (2,636)
0.571
83.7 (206) 10.16 (25) 0.41 (1)
71.7 (3,380) 21.1 (995) 0.53 (25)
G0.001* G0.001* 0.794
99.2 (244) 0.81 (2)
99.0 (4,669) 0.61 (29)
0.776 0.713
20.3 (50) 9.35 (23) 1.63 (4) 17.1 (42) 4.47 (11) 26.0 (64) 1.22 (3)
15.8 (743) 5.24 (247) 1.59 (75) 12.3 (578) 5.72 (270) 23.6 (1,115) 1.82 (86)
0.056 0.006* 0.965 0.026* 0.407 0.393 0.487
1.63 (4) 24.0 (59) 34.6 (85) 1.22 (3) 26.0 (64) 74.0 (182)
1.70 (80) 18.8 (885) 29.4 (1,385) 1.14 (54) 23.9 (1,126) 74.9 (3,534)
0.934 0.042* 0.082 0.915 0.442 0.742
*Significant (p G0.05).
Figure 2. Recidivist admissions attributed to falls.
Univariate analysis of the geriatric trauma admissions showed that the median age of the recidivist population was slightly older than the nonrecidivist population (RC, 81 years vs NRC, 79 years; p = 0.001). There were significantly more
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
411
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
women in the recidivist group than in the nonrecidivist group (RC, 66.0% vs. NRC, 57.9%; p e 0.001). In addition, falls were a more common MOI in the recidivist population than in the nonrecidivist population (RC, 86.0% vs. NRC, 71.7%; p G 0.001). There was no significant difference in ISS. Significant unadjusted odds ratios for recidivism were found for sex, primary insurance, admission GCS score, MOI, and three PECs (Table 2). When put into a multivariable logistic regression model, primary insurance, fall as an MOI, and psychological disorder as a PEC were no longer found to be significant predictors of recidivism. Our final multivariable model indicates that being a female and having a history of head trauma and a preexisting pulmonary disease are predictors of geriatric recidivism. Admission for an MVC and an initial GCS score of less than 15 are both predictors of nonrecidivism.
DISCUSSION The high prevalence of PECs and decreased physiologic reserve in geriatric trauma patients result in outcomes significantly worse than in their younger counterparts.16 As this population is increasingly active and living longer,15,17geriatric trauma care presents a unique challenge to trauma centers. The high geriatric recidivism rate (5.8%) identified in this study supports trauma as a chronic disease correlated with lifestyle and environmental factors.18 This rate was found to be higher than the overall recidivism rate within the same institution during the same period (4.3%).19 With the rise of ACOs emphasizing quality-of-care improvements and cost reduction,9 addressing geriatric recidivism should be a priority for trauma centers. Integral to addressing trauma geriatric recidivism is characterizing predictors of recidivism in this population.
Although several studies have found males to be more likely to be trauma recidivists,20Y22 our study found that females were more likely to be recidivists in the geriatric population. This finding is supported by a study conducted by McGwin et al.3 of a cohort of patients 70 years or older, which found that females had a relative risk of 3.38 (95% confidence interval, 0.80Y14.23) for subsequent injury hospitalization when compared with men. Similarly, a study by Gubler et al.15 of a Medicare cohort found males to be less likely to require hospitalization for their injuries than females. The increased risk for females to be admitted for multiple traumatic injuries is likely related to the longer life expectancy of females23 and the increased likelihood of females to live alone.24 Furthermore, while younger recidivists have been found to be at a higher risk of experiencing violent injury and experiencing penetrating trauma,21,22 the elderly population has been found to more commonly experience recidivism due to falls.25 In our study, we found that significantly more geriatric recidivists were admitted for falls at their first admission compared with nonrecidivists. Of the geriatric recidivists initially admitted for a fall, 92.3% were admitted for a subsequent fall. In response to this high rate of recidivism due to falls, our trauma service has developed and implemented a fall prevention program in partnership with the skilled nursing facilities in the county. Through this program, trauma prevention taskforce, consisting of a nurse, trauma surgeon, and trauma prevention coordinator, has been developed, which visits skilled nursing facilities and presents a protocol to identify patients at risk for falls as well as assists staff to identify steps to reduce falls. The overall rate of repeat trauma admissions for the same MOI as that of initial admission was found to be 80.5% in our population. A 2013 study by Allan et al.14 identified a similar
TABLE 2. Multivariable Logistic Regression for Trauma Admissions Attributed to Recidivist Patients
Sex Male Female Primary insurance Medicare Other Admission GCS score GCS score, 15 GCS score G 15 MOI Fall MVC Other PECs History of head trauma Pulmonary disease Psychological disorder
Recidivism Rate
Odds Ratio (95%CI)
Adjusted Odds Ratio (95%CI)
p
3.9% 5.7%
Reference 1.53 (1.17Y2.02)
Reference 1.42 (1.07Y1.88)
0.016
5.6% 3.6%
1.63 (1.19Y2.22) Reference
V
V
5.3% 3.7%
Reference 0.68 (0.47Y0.98)
Reference 0.67 (0.47Y0.97)
0.033
5.7% 2.5% 3.8%
2.03 (1.44Y2.86) 0.42 (0.28Y0.65) Reference
V 0.42 (0.27Y0.65) Reference
V G0.001
8.5% 6.8% 6.3%
1.79 (1.14Y2.83) 1.48 (1.05Y2.09) 1.38 (1.02Y1.86)
1.87 (1.16Y2.99) 1.48 (1.04Y2.12) V
0.010 0.030 V
CI, confidence interval.
412
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
trend, as two thirds of those injured in falls, all-terrain vehicles/ motorcycle accidents, or stabbings were reinjured by the same mechanism. Furthermore, we found that geriatric trauma patients admitted for injuries related to MVCs are less likely to be recidivists than trauma patients with other MOIs. This is likely because MVCs are more frequently identified as chance events than other MOIs.26 In addition, state law requires medical personnel to report to the Pennsylvania Department of Transportation anyone deemed medically unfit to drive, which has been demonstrated to be a successful MVC prevention initative.27 An additional predictor of geriatric recidivism identified in this study was having a GCS score of 15 at admission. This finding is supported by the study by Allan et al.,14 which found that recidivists had a mean (SD) GCS score of 14.5 (1.4), whereas the nonrecidivists had a mean GCS score of 12.6 (4.1) (p G 0.001). This finding indicates that the less injured geriatric patients are more likely to be admitted for a subsequent traumatic event. This may be due to the fact that these patients are less severely injured and more likely to return to activities that expose them to injury risk. An unexpected finding of our study was that our geriatric recidivism rate (5.8%) was substantially higher than the 1.22% found by Allan et al.14 at the only Level I trauma center in Miami-Dade County, Florida, despite defining and classifying recidivism using the same methods. Significant demographic dissimilarities may contribute to the differences in geriatric recidivism rates. Lancaster County’s population is 91% white alone and with the geriatric population composing 16.1%. Miami-Dade County is only 77.8% white alone, and the geriatric population composes 14.9%.28 Furthermore, this finding emphasizes the importance of characterizing the geriatric population in unique patient populations to identify priorities for optimizing geriatric care and curtailing spending for programs such as the MSSP because they likely vary significantly from one location to another. While this study identifies unique characteristics of geriatric recidivists, it is not without limitations. This study has all of the inherent limitations of single-institution, retrospective studies. Patients treated at trauma centers outside Lancaster County, Pennsylvania, were not identified as recidivists in this project because these data were not available. Additional variables on socioeconomic status, alcohol and drug use, and criminal activity were not included in this study, despite being associated with general trauma recidivism in previous research studies.20Y22 The value of preventing recurrent injuries in the geriatric population cannot be underestimated, especially for trauma centers of ACOs. Our study found the rate of trauma recidivism within the geriatric population to be higher than the rate of trauma recidivism for our overall population. In addition, females, those with pulmonary disease, and those with history of head trauma were found to be more likely to be recidivists. Furthermore, geriatric patients frequently experience recidivism for the same mechanism as that of their initial injury. Trauma centers that actively characterize their high-risk patient populations for targeted prevention initiatives can assist ACOs in their mission to be patient centric and community focused.
AUTHORSHIP K.R., A.R., J.A.M., and F.B.R. designed this study. K.R. and R.C. performed data acquisition. K.R., C.H., B.G., and R.C. contributed to data analysis. J.A.M. and F.B.R. interpreted the data. K.R., C.H., and B.G. prepared the manuscript. A.R., J.A.M., and F.B.R. provided editorial oversight.
DISCLOSURE The authors declare no conflicts of interest.
REFERENCES 1. Smith R, Fry W, Morabito D, Organ C. Recidivism in an urban trauma center. Arch Surg. 1992;127(6):668Y670. 2. Reiner DS, Pastena JA, Swan KG, et al. Trauma recidivism. Am Surg. 1990;56:556Y560. 3. McGwin G, May AK, Melton SM, et al. Recurrent trauma in elderly patients. Arch Surg. 2001;136:197Y203. 4. United States Census Bureau. Statistical brief: sixty-five plus in the United States. Available at: http://www.census.gov/population/socdemo/statbriefs/ agebrief.html Accessed: July 29, 2014. 5. U.S. Department of Health and Human Services. Administration on Aging. A Profile of Older Americans. Available at: http://www.aoa.gov/Aging_ Statistics/Profile/2012/docs/2012profile.pdf. 2012. 6. Institute of Medicine of the National Academies. Retooling for Aging American: Building the Health Care Workforce. Available at: http://www. iom.edu/~/media/Files/Report%20Files/2008/Retooling-for-an-Aging-AmericaBuilding-the-Health-Care-Workforce/ReportBriefRetoolingforanAgingAmerica BuildingtheHealthCareWorkforce.pdf. April 2011 7. Taylor M, Tracy J, Meyer W, Pasquale M, Napolitano L. Trauma in the elderly: intensive care unit resource use and outcome. J Trauma. 2002;53:407Y414. 8. Ciesla D, Pracht E, Tepas J, Cha J, Langland-Orban B, Flint L. The injured elderly: a rising tide. Surgery. 2013;154:291Y298. 9. Centers for Medicare and Medicaid Services. Shared Savings Program. Baltimore, MD. Available at: http://www.cms.gov/Medicare/MedicareFee-for-Service-Payment/sharedsavingsprogram/index.html?redirect=/ sharedsavingsprogram. Accessed: July 12, 2014 10. Lancaster General Health. LG Health Establishes ACO. Lancaster, PA. Available at: http://www.lancastergeneralhealth.org/LGH/Healthcare-Professionals/ Progress-Notes/Announcements/LG-Health-establishes-ACO.aspx. Accessed: July 16, 2014 11. Culbert Health Solutions. Navigating the Era of Accountable Care Organizations (ACOs): A Strategic Planning Guide. Woburn, MA. Available at: http://www.culberthealth.com/press_white_paper_accountable_care.pdf. Accessed July 2014. 12. American Academy of Actuaries. Fact Sheet: Medicare Shared Savings Program Under the Affordable Care Act. Available at: http://www.actuary. org/files/ACO_fact_sheet_06_17_2011.pdf. 2011 13. Velopulos C, Enwerem N, Haider A, et al. National cost of trauma care by payer status. J Surg Res. 2013;184(1):444Y449. 14. Allan B, Davis J, Schulman C, et al. Exploring trauma recidivism in an elderly cohort. J Surg Res. 2013;184(1):582Y585. 15. Gubler K, Maier R, Davis R, et al. Trauma recidivism in the elderly. J Trauma. 1996;41(6):952Y956. 16. Jacobs DG. Special consideration in geriatric injury. Curr Opin Crit Care. 2003;9:535Y539. 17. Sahyoun NR, Lentzner H, Hoyert D, et al. Trends in causes of death among the elderly. Aging Trends. 2001;1:1Y10. 18. Poole GV, Griswold JA, Thaggard K, et al. Trauma as a recurrent disease. Surgery. 1993;113:608Y611. 19. Rogers A, Horst M, Rittenhouse K, et al. Urban versus rural trauma recidivism: Is there a difference? Eur J Trauma Emerg Surg. 2014;40(6): 701Y706. 20. Poole GV, Grisworld JA, Thaggard VK, Rhodes RD. Trauma is a recurrent disease. Surgery. 1993;113(6):608Y611.
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
413
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
21. Cooper C, Eslinger D, Nash D, et al. Repeat victims of violence: report of a large concurrent case-control study. Arch Surg. 2000;135(7): 837Y843. 22. Caufeild J, Singhal A, Moulton R, et al. Trauma recidivism in a large urban Canadian population. J Trauma. 2004;57(4):872Y876. 23. Ginter E, Simko V. Women live longer than men. Bratisl Lek Listy. 2013;114(2):45Y49. 24. Administration on Aging. Older Women. Available at: http://www.aoa.gov/ naic/may2000/factsheets/olderwomen.html. Accessed July 2014.
414
25. Nevitt M, Cummings S, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls. A prospective study. JAMA. 1989;261(18):2663Y2668. 26. Clark D. Motor vehicle crash fatalities in the elderly: rural versus urban. J Trauma. 2001;51(5):896Y900. 27. Pennsylvania Department of Transportation. Pennsylvania’s Medically Impaired Driver Law. Available at: http://www.dmv.state.pa.us/medicalReporting Center/medicallyImpairedDriver.shtml. Accessed July 2014. 28. State and County QuickFacts. US Census 2013 Estimate. Available at: http://quickfacts.census.gov. Accessed July 2014.
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
An analysis of geriatric recidivism in the era of accountable care organizations Katelyn Rittenhouse, Carissa Harnish, Brian Gross, Amelia Rogers, Jo Ann Miller, Roxanne Chandler, RHIA, and Frederick B. Rogers, MD, MS, Lancaster, Pennsylvania
To date, there are almost 500 accountable care organizations (ACOs) across the United States emphasizing cost-effective care. Readmission largely impacts health care cost; therefore, we sought to determine factors associated with geriatric trauma readmissions (recidivism) within our institution. METHODS: All admissions from 2000 to 2011 attributed to patients 65 years or older at our Level II trauma center, recently verified by Medicare as an ACO, were queried. Patients were classified as recidivist or nonrecidivist. The first admissions of recidivist patients were compared with the nonrecidivist admissions with respect to sex, age, race, primary insurance, admission Glasgow Coma Scale (GCS) score, Injury Severity Score (ISS), hospital length of stay, mechanism of injury (MOI), preexisting conditions, and discharge destination. Factors found to be significant predictors of recidivism in univariate analyses were subsequently incorporated into a multivariate logistic regression model. In addition, the second admission’s MOI was compared with the first admission’s MOI, and the proportion of first, second, and third admissions attributed to falls was calculated. A p G 0.05 was significant. RESULTS: Between 2000 and 2011, a total of 4,963 unique patients were admitted to the trauma center at 65 years or older. This population was composed of 287 recidivists (5.8%) and 4,676 nonrecidivists (94.2%). When placed in a multivariate logistic regression, female sex, admission GCS score of 15, history of head trauma, and preexisting pulmonary disease were identified as significant predictors of recidivism. A trend toward increasing proportion of injuries attributed to falls was found with each subsequent trauma admission (81.5% [234 of 287] of first admissions, 88.2% [253 of 287] of second admissions, and 90.5% [19 of 21] of third admissions). CONCLUSION: Our study identifies specific factors that should be targeted by social service and prevention resources to inhibit recidivism in the elderly. In the brave new world of ACOs, trauma centers must identify high-risk populations for the consumption of limited resources. (J Trauma Acute Care Surg. 2015;78: 409Y414. Copyright * 2015 Wolters Kluwer Health, Inc. All rights reserved.) LEVEL OF EVIDENCE: Care management study, level IV. Prognostic study, level III. KEY WORDS: Recidivism; geriatric; accountable care organization. BACKGROUND:
A
lthough traditionally identified as a singular random event, a growing body of research has identified trauma as a predictable occurrence in certain high-risk groups, including young males, substance abusers, and the elderly.1 In 1990, Reiner et al.2 ascribed the term trauma recidivism to this trend. As trauma recidivism has been linked to loss of function and socioeconomic challenges after initial injury,3 characterizing the geriatric recidivist population for future prevention initiatives should be a priority for trauma programs. The geriatric population is receiving increased attention within the medical community because it is the fastest expanding segment of the population in the United States.4 Representing
Submitted: July 31, 2014, Revised: October 14, 2014, Accepted: October 20, 2014. From the Trauma Services, Lancaster General Health, Lancaster, Pennsylvania. This work was presented as a poster at the 73rd Annual Meeting of the American Association for the Surgery of Trauma and Clinical Congress of Acute Care Surgery, September 10Y13, 2014, in Philadelphia, Pennsylvania. Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.jtrauma.com). Address for reprints: Frederick B. Rogers, MD, MS, 555 N Duke St, Lancaster, PA 17602, email: [email protected]. DOI: 10.1097/TA.0000000000000513
12% of all Americans in 2000, those 65 years and older are expected to increase to 19% of the population by 2030.5 As a result, health care providers, including trauma surgeons, in the United States will need to manage limited resources effectively and efficiently to maximize care for the elderly.6 This demographic shift presents a challenge to trauma systems, as caring for geriatric patients requires a significant knowledge of aging-related diseases and proper management of preexisting conditions (PECs).6 Commonly, geriatric patients experience outcomes poorer than those of their younger counterparts because of increased comorbidities and less physiologic reserve.7 A recent study by Ciesla et al.8 of the Florida Agency for Health Care Administration database revealed a 17% increase in injury rates for the geriatric population between 1996 and 2010, while rates of injury decreased by 18% in children and grew by 2% in adults during the same period. Furthermore, a study by McGwin et al.3 reported that individuals in a previously injured geriatric cohort were 3.25 times more likely to be injured than a noninjured geriatric cohort. Therefore, geriatric recidivism should be a significant concern for trauma centers. Further incentivizing the prevention of geriatric trauma recidivism is the emergence of accountable care organizations (ACOs).9 ACOs were established within the Medicare Shared Savings Program (MSSP) to improve medical care of Medicare
J Trauma Acute Care Surg Volume 78, Number 2
409
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
Fee-For-Service beneficiaries and simultaneously reduce cost of care by encouraging better collaboration between health care providers.9,10 ACOs aim to improve health care by focusing on four pillars as follows: informatics, clinical integration, financial management, and care coordination.11 The MSSP addresses the reality that many geriatric patients are dependent on Medicare and supplementary insurance to cover health care costs by financially rewarding ACOs if the costs of their Medicare beneficiaries are below the set benchmark standards.12 In 2010, the adjusted estimate of trauma inpatient cost burden for Medicare was $17.5 billion yearly, representing nearly 47% of inpatient costs for traumatic injuries.13 Trauma centers within ACOs, such as our Pennsylvania-accredited Level II trauma center, may be able to contribute to achieving MSSP benchmarks through geriatric recidivism prevention initiatives. To effectively prioritize care of our aging patients, recidivism must be characterized in the geriatric population. To date, very few studies have investigated this topic.3,14,15 The objective of the study was to identify characteristics of geriatric trauma recidivists that may be used to develop targeted recidivism prevention interventions. We sought to determine what specific demographic factors, comorbidities, PECs. We hypothesized that falls and increasing age would be significant predictors for trauma recidivism. There were specific clinical factors that would predict a higher rate of geriatric recidivism. The geriatric population would have a rate of recidivism higher than the overall trauma recidivism rate at our institution. Furthermore, we hypothesized that falls would be a predictor of geriatric recidivism.
PATIENTS AND METHODS Following review and approval of this study by the institutional review board of Lancaster General Health, the registry of the Pennsylvania Trauma Systems Foundation (Digital Innovations, Forest Hill, MD) was queried for all trauma admissions of patients 65 years or older to the only Pennsylvania-accredited Level II trauma center in Lancaster County from 2000 to 2011. Variables collected included age, race, sex, Injury Severity Score (ISS), mechanism of injury (MOI), injury type, hospital length of stay, intensive care unit admission, mortality, PECs, discharge destination, and insurance status was collected for all trauma admissions. All variables were coded retrospectively from patients’ medical records by certified trauma registrars, with an accuracy rate of 98% to 99%, as determined by outside audits. The PECs evaluated in this study were based on the Pennsylvania System Trauma Foundation definitions (Supplemental File 1, at http://links.lww.com/TA/A508). PECs present in fewer than 50 patients were excluded from the analyses. Patients with names and medical record numbers associated with two or more trauma admissions during the study period were labeled recidivists (RCs). Patients with medical records associated with one trauma admission during the study period were labeled non-RCs (NRCs). Hospital readmissions for complications associated with the initial injury were excluded from the analyses. To calculate an overall recidivism rate for our geriatric population, patients with a trauma admission during the study period at age greater than or equal to 410
65 years with a previous trauma admission at age less than 65 years were also classified as recidivists. When determining predictors of geriatric recidivism, only initial recidivist admissions at which the patient was 65 years or older were compared with nonrecidivist admissions. The two-sample t test was used to compare continuous variables, and the W2 test was used to compare categorical variables between groups. Unadjusted odds ratios for recidivism were calculated for sex, primary insurance, admission GCS score, MOI, and PECs. Variables significantly associated or trending toward significance with recidivism in univariate analyses were subsequently incorporated into a multivariable binary logistic regression model to determine adjusted odds ratios for recidivism. A p G 0.05 was considered statistically significant. A p G 0.10 was considered a trend toward statistical significance.
RESULTS Between 2000 and 2011, there were a total of 4,963 unique patients admitted to the trauma center at 65 years or older. The patient population was composed of 2,071 males (41.7%) and 2,892 females (58.3%), with an age range from 65 years to 103 years. The median age was 80 years (interquartile range [IQR], 73Y85 years). The median ISS of the population was 9 (IQR, 5Y16). The most common MOI was fall (73.1%), followed by motor vehicle crash (MVC) (19.8%). A total of 246 patients (5.0%) had at least two trauma admissions at 65 years or older. In addition, 41 patients (0.8%) with initial trauma admissions when younger than 65 years during the study period also had subsequent trauma admissions when 65 years or older. Within this RC cohort of 287 individuals, 266 patients (92.7%) were admitted twice, 20 patients (7.0%) were admitted three times, and 1 patient (0.3%) was admitted for six unique traumatic events. The overall trauma recidivism rate for this geriatric population was determined to be 5.8% (287 of 4,963). The median time between trauma admissions was 1.9 years (695 days). The overall mortality rate (5.9%, 17 of 287) of recidivist patients was found to be statistically indistinguishable from the mortality rate (7.0%, 327 of 4,676) of nonrecidivist patients (p = 0.551). Within the recidivist patient population, 80.5% (231 of 287) of all second admissions were attributed to the MOI same as that of the initial admission (Fig. 1). Of the patients initially admitted for a fall, 93.2% (218 of 234) had the same MOI at second admission. Of the patients initially admitted for an MVC, 25% (9 of 36) had the same MOI at second admission. Of the 21 patients admitted for three unique traumatic injuries, 17 (81.0%) were admitted for the same MOI at all three occurrences. Of the 21 patients admitted to the trauma service three or more times, 16 (76.2%) were admitted for three falls. One patient was admitted for three unique motorcycle-related injuries. The other three patients were admitted for a fall two of their three admissions; this includes the first three admissions of the individual with six unique traumatic injuries. The last three admissions of the individual with six unique traumatic injuries were attributed to an MVC, an off-road vehicle accident, and a fall. Furthermore, within the recidivist population, * 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
TABLE 1. Demographics of Recidivist and Nonrecidivist Trauma Admissions
Figure 1. Rate of second admission due to same MOI as that of the initial injury.
the proportion of admissions attributed to falls increased with each admission (Fig. 2). When comparing the first admissions of recidivist patients at age 65 years or older with the admissions of nonrecidivist geriatric patients (Table 1), recidivist admissions were found to be more commonly attributed to females (RC, 67.5% vs. NRC, 57.8%; p = 0.003) and patients with Medicare as their primary insurance (RC, 78.0% vs. NRC, 68.8%; p = 0.003). A trend toward an increased likelihood for readmission of patients with a first GCS of 15 (RC, 83.6% vs. NRC, 78.2%; p = 0.053) was also observed. Furthermore, falls were more common (RC, 83.7% vs. NRC, 71.7%; p e 0.001) and MVCs were less common (RC, 10.16% vs. NRC, 21.1%; p G 0.001) as the MOI for recidivist initial admissions. A history of head trauma, pulmonary disease, and psychological disorders were all found to be significantly more common in the first admissions of recidivists than in the nonrecidivist admissions.
n Sex Male Female Race White Other Age, y 65Y84 Q85 Primary insurance Medicare Other Admission GCS score GCS score, 15 GCS score G 15 ISS Median (IQR) Hospital length of stay Median (IQR) Intensive care unit admission Yes No MOI Fall MVC Other Injury type Blunt Penetrating PECs Thyroid disease History of head trauma Renal disease Pulmonary disease Obesity Neurologic disorder Musculoskeletal disease Malignancy Psychological disorder Hematologic disease Gastrointestinal disease Diabetes Cardiac disease
RC First Admission
NRC Admission
246
4,717
32.5 (80) 67.5 (166)
42.2 (1,991) 57.8 (2,726)
0.003*
98.8 (243) 1.22 (3)
98.5 (4,638) 1.51 (71)
0.716
68.7 (169) 31.3 (77)
71.9 (3,393) 28.1 (1,324)
0.272
78.0 (191) 22.0 (54)
68.8 (3,237) 31.2 (1,465)
0.003*
83.6 (189) 16.4 (37)
78.2 (3,409) 21.8 (950)
0.053
9.00 (5.00Y14.0)
9.00 (5.00Y16.00)
4.00 (3.00Y6.00)
4.00 (2.00Y6.00)
p
42.3 (104) 57.7 (142)
44.1 (2,081) 55.9 (2,636)
0.571
83.7 (206) 10.16 (25) 0.41 (1)
71.7 (3,380) 21.1 (995) 0.53 (25)
G0.001* G0.001* 0.794
99.2 (244) 0.81 (2)
99.0 (4,669) 0.61 (29)
0.776 0.713
20.3 (50) 9.35 (23) 1.63 (4) 17.1 (42) 4.47 (11) 26.0 (64) 1.22 (3)
15.8 (743) 5.24 (247) 1.59 (75) 12.3 (578) 5.72 (270) 23.6 (1,115) 1.82 (86)
0.056 0.006* 0.965 0.026* 0.407 0.393 0.487
1.63 (4) 24.0 (59) 34.6 (85) 1.22 (3) 26.0 (64) 74.0 (182)
1.70 (80) 18.8 (885) 29.4 (1,385) 1.14 (54) 23.9 (1,126) 74.9 (3,534)
0.934 0.042* 0.082 0.915 0.442 0.742
*Significant (p G0.05).
Figure 2. Recidivist admissions attributed to falls.
Univariate analysis of the geriatric trauma admissions showed that the median age of the recidivist population was slightly older than the nonrecidivist population (RC, 81 years vs NRC, 79 years; p = 0.001). There were significantly more
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
411
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
women in the recidivist group than in the nonrecidivist group (RC, 66.0% vs. NRC, 57.9%; p e 0.001). In addition, falls were a more common MOI in the recidivist population than in the nonrecidivist population (RC, 86.0% vs. NRC, 71.7%; p G 0.001). There was no significant difference in ISS. Significant unadjusted odds ratios for recidivism were found for sex, primary insurance, admission GCS score, MOI, and three PECs (Table 2). When put into a multivariable logistic regression model, primary insurance, fall as an MOI, and psychological disorder as a PEC were no longer found to be significant predictors of recidivism. Our final multivariable model indicates that being a female and having a history of head trauma and a preexisting pulmonary disease are predictors of geriatric recidivism. Admission for an MVC and an initial GCS score of less than 15 are both predictors of nonrecidivism.
DISCUSSION The high prevalence of PECs and decreased physiologic reserve in geriatric trauma patients result in outcomes significantly worse than in their younger counterparts.16 As this population is increasingly active and living longer,15,17geriatric trauma care presents a unique challenge to trauma centers. The high geriatric recidivism rate (5.8%) identified in this study supports trauma as a chronic disease correlated with lifestyle and environmental factors.18 This rate was found to be higher than the overall recidivism rate within the same institution during the same period (4.3%).19 With the rise of ACOs emphasizing quality-of-care improvements and cost reduction,9 addressing geriatric recidivism should be a priority for trauma centers. Integral to addressing trauma geriatric recidivism is characterizing predictors of recidivism in this population.
Although several studies have found males to be more likely to be trauma recidivists,20Y22 our study found that females were more likely to be recidivists in the geriatric population. This finding is supported by a study conducted by McGwin et al.3 of a cohort of patients 70 years or older, which found that females had a relative risk of 3.38 (95% confidence interval, 0.80Y14.23) for subsequent injury hospitalization when compared with men. Similarly, a study by Gubler et al.15 of a Medicare cohort found males to be less likely to require hospitalization for their injuries than females. The increased risk for females to be admitted for multiple traumatic injuries is likely related to the longer life expectancy of females23 and the increased likelihood of females to live alone.24 Furthermore, while younger recidivists have been found to be at a higher risk of experiencing violent injury and experiencing penetrating trauma,21,22 the elderly population has been found to more commonly experience recidivism due to falls.25 In our study, we found that significantly more geriatric recidivists were admitted for falls at their first admission compared with nonrecidivists. Of the geriatric recidivists initially admitted for a fall, 92.3% were admitted for a subsequent fall. In response to this high rate of recidivism due to falls, our trauma service has developed and implemented a fall prevention program in partnership with the skilled nursing facilities in the county. Through this program, trauma prevention taskforce, consisting of a nurse, trauma surgeon, and trauma prevention coordinator, has been developed, which visits skilled nursing facilities and presents a protocol to identify patients at risk for falls as well as assists staff to identify steps to reduce falls. The overall rate of repeat trauma admissions for the same MOI as that of initial admission was found to be 80.5% in our population. A 2013 study by Allan et al.14 identified a similar
TABLE 2. Multivariable Logistic Regression for Trauma Admissions Attributed to Recidivist Patients
Sex Male Female Primary insurance Medicare Other Admission GCS score GCS score, 15 GCS score G 15 MOI Fall MVC Other PECs History of head trauma Pulmonary disease Psychological disorder
Recidivism Rate
Odds Ratio (95%CI)
Adjusted Odds Ratio (95%CI)
p
3.9% 5.7%
Reference 1.53 (1.17Y2.02)
Reference 1.42 (1.07Y1.88)
0.016
5.6% 3.6%
1.63 (1.19Y2.22) Reference
V
V
5.3% 3.7%
Reference 0.68 (0.47Y0.98)
Reference 0.67 (0.47Y0.97)
0.033
5.7% 2.5% 3.8%
2.03 (1.44Y2.86) 0.42 (0.28Y0.65) Reference
V 0.42 (0.27Y0.65) Reference
V G0.001
8.5% 6.8% 6.3%
1.79 (1.14Y2.83) 1.48 (1.05Y2.09) 1.38 (1.02Y1.86)
1.87 (1.16Y2.99) 1.48 (1.04Y2.12) V
0.010 0.030 V
CI, confidence interval.
412
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
trend, as two thirds of those injured in falls, all-terrain vehicles/ motorcycle accidents, or stabbings were reinjured by the same mechanism. Furthermore, we found that geriatric trauma patients admitted for injuries related to MVCs are less likely to be recidivists than trauma patients with other MOIs. This is likely because MVCs are more frequently identified as chance events than other MOIs.26 In addition, state law requires medical personnel to report to the Pennsylvania Department of Transportation anyone deemed medically unfit to drive, which has been demonstrated to be a successful MVC prevention initative.27 An additional predictor of geriatric recidivism identified in this study was having a GCS score of 15 at admission. This finding is supported by the study by Allan et al.,14 which found that recidivists had a mean (SD) GCS score of 14.5 (1.4), whereas the nonrecidivists had a mean GCS score of 12.6 (4.1) (p G 0.001). This finding indicates that the less injured geriatric patients are more likely to be admitted for a subsequent traumatic event. This may be due to the fact that these patients are less severely injured and more likely to return to activities that expose them to injury risk. An unexpected finding of our study was that our geriatric recidivism rate (5.8%) was substantially higher than the 1.22% found by Allan et al.14 at the only Level I trauma center in Miami-Dade County, Florida, despite defining and classifying recidivism using the same methods. Significant demographic dissimilarities may contribute to the differences in geriatric recidivism rates. Lancaster County’s population is 91% white alone and with the geriatric population composing 16.1%. Miami-Dade County is only 77.8% white alone, and the geriatric population composes 14.9%.28 Furthermore, this finding emphasizes the importance of characterizing the geriatric population in unique patient populations to identify priorities for optimizing geriatric care and curtailing spending for programs such as the MSSP because they likely vary significantly from one location to another. While this study identifies unique characteristics of geriatric recidivists, it is not without limitations. This study has all of the inherent limitations of single-institution, retrospective studies. Patients treated at trauma centers outside Lancaster County, Pennsylvania, were not identified as recidivists in this project because these data were not available. Additional variables on socioeconomic status, alcohol and drug use, and criminal activity were not included in this study, despite being associated with general trauma recidivism in previous research studies.20Y22 The value of preventing recurrent injuries in the geriatric population cannot be underestimated, especially for trauma centers of ACOs. Our study found the rate of trauma recidivism within the geriatric population to be higher than the rate of trauma recidivism for our overall population. In addition, females, those with pulmonary disease, and those with history of head trauma were found to be more likely to be recidivists. Furthermore, geriatric patients frequently experience recidivism for the same mechanism as that of their initial injury. Trauma centers that actively characterize their high-risk patient populations for targeted prevention initiatives can assist ACOs in their mission to be patient centric and community focused.
AUTHORSHIP K.R., A.R., J.A.M., and F.B.R. designed this study. K.R. and R.C. performed data acquisition. K.R., C.H., B.G., and R.C. contributed to data analysis. J.A.M. and F.B.R. interpreted the data. K.R., C.H., and B.G. prepared the manuscript. A.R., J.A.M., and F.B.R. provided editorial oversight.
DISCLOSURE The authors declare no conflicts of interest.
REFERENCES 1. Smith R, Fry W, Morabito D, Organ C. Recidivism in an urban trauma center. Arch Surg. 1992;127(6):668Y670. 2. Reiner DS, Pastena JA, Swan KG, et al. Trauma recidivism. Am Surg. 1990;56:556Y560. 3. McGwin G, May AK, Melton SM, et al. Recurrent trauma in elderly patients. Arch Surg. 2001;136:197Y203. 4. United States Census Bureau. Statistical brief: sixty-five plus in the United States. Available at: http://www.census.gov/population/socdemo/statbriefs/ agebrief.html Accessed: July 29, 2014. 5. U.S. Department of Health and Human Services. Administration on Aging. A Profile of Older Americans. Available at: http://www.aoa.gov/Aging_ Statistics/Profile/2012/docs/2012profile.pdf. 2012. 6. Institute of Medicine of the National Academies. Retooling for Aging American: Building the Health Care Workforce. Available at: http://www. iom.edu/~/media/Files/Report%20Files/2008/Retooling-for-an-Aging-AmericaBuilding-the-Health-Care-Workforce/ReportBriefRetoolingforanAgingAmerica BuildingtheHealthCareWorkforce.pdf. April 2011 7. Taylor M, Tracy J, Meyer W, Pasquale M, Napolitano L. Trauma in the elderly: intensive care unit resource use and outcome. J Trauma. 2002;53:407Y414. 8. Ciesla D, Pracht E, Tepas J, Cha J, Langland-Orban B, Flint L. The injured elderly: a rising tide. Surgery. 2013;154:291Y298. 9. Centers for Medicare and Medicaid Services. Shared Savings Program. Baltimore, MD. Available at: http://www.cms.gov/Medicare/MedicareFee-for-Service-Payment/sharedsavingsprogram/index.html?redirect=/ sharedsavingsprogram. Accessed: July 12, 2014 10. Lancaster General Health. LG Health Establishes ACO. Lancaster, PA. Available at: http://www.lancastergeneralhealth.org/LGH/Healthcare-Professionals/ Progress-Notes/Announcements/LG-Health-establishes-ACO.aspx. Accessed: July 16, 2014 11. Culbert Health Solutions. Navigating the Era of Accountable Care Organizations (ACOs): A Strategic Planning Guide. Woburn, MA. Available at: http://www.culberthealth.com/press_white_paper_accountable_care.pdf. Accessed July 2014. 12. American Academy of Actuaries. Fact Sheet: Medicare Shared Savings Program Under the Affordable Care Act. Available at: http://www.actuary. org/files/ACO_fact_sheet_06_17_2011.pdf. 2011 13. Velopulos C, Enwerem N, Haider A, et al. National cost of trauma care by payer status. J Surg Res. 2013;184(1):444Y449. 14. Allan B, Davis J, Schulman C, et al. Exploring trauma recidivism in an elderly cohort. J Surg Res. 2013;184(1):582Y585. 15. Gubler K, Maier R, Davis R, et al. Trauma recidivism in the elderly. J Trauma. 1996;41(6):952Y956. 16. Jacobs DG. Special consideration in geriatric injury. Curr Opin Crit Care. 2003;9:535Y539. 17. Sahyoun NR, Lentzner H, Hoyert D, et al. Trends in causes of death among the elderly. Aging Trends. 2001;1:1Y10. 18. Poole GV, Griswold JA, Thaggard K, et al. Trauma as a recurrent disease. Surgery. 1993;113:608Y611. 19. Rogers A, Horst M, Rittenhouse K, et al. Urban versus rural trauma recidivism: Is there a difference? Eur J Trauma Emerg Surg. 2014;40(6): 701Y706. 20. Poole GV, Grisworld JA, Thaggard VK, Rhodes RD. Trauma is a recurrent disease. Surgery. 1993;113(6):608Y611.
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
413
J Trauma Acute Care Surg Volume 78, Number 2
Rittenhouse et al.
21. Cooper C, Eslinger D, Nash D, et al. Repeat victims of violence: report of a large concurrent case-control study. Arch Surg. 2000;135(7): 837Y843. 22. Caufeild J, Singhal A, Moulton R, et al. Trauma recidivism in a large urban Canadian population. J Trauma. 2004;57(4):872Y876. 23. Ginter E, Simko V. Women live longer than men. Bratisl Lek Listy. 2013;114(2):45Y49. 24. Administration on Aging. Older Women. Available at: http://www.aoa.gov/ naic/may2000/factsheets/olderwomen.html. Accessed July 2014.
414
25. Nevitt M, Cummings S, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls. A prospective study. JAMA. 1989;261(18):2663Y2668. 26. Clark D. Motor vehicle crash fatalities in the elderly: rural versus urban. J Trauma. 2001;51(5):896Y900. 27. Pennsylvania Department of Transportation. Pennsylvania’s Medically Impaired Driver Law. Available at: http://www.dmv.state.pa.us/medicalReporting Center/medicallyImpairedDriver.shtml. Accessed July 2014. 28. State and County QuickFacts. US Census 2013 Estimate. Available at: http://quickfacts.census.gov. Accessed July 2014.
* 2015 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
