G Model
JINJ-6210; No. of Pages 6 Injury, Int. J. Care Injured xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes Rishin J. Kadakia a,1, Ray Y. Hsu b,2, Roman Hayda b,2, Yoojin Lee c,3, Jason T. Bariteau a,4 a
Emory University Department of Orthopaedics, Atlanta, GA, USA Brown University Department of Geriatrics, Providence, RI, USA c Brown University Department of Orthopaedics, Providence, RI, USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 1 May 2015
Introduction: The incidence of geriatric ankle fractures will undoubtedly increase as the population continues to grow. Many geriatric patients struggle to function independently after such injury and often require placement into nursing homes. The morbidity and mortality associated with nursing homes is well documented within the field of orthopaedic surgery. However, there is currently no study examining the mortality associated with nursing home placement following hospitalization for an ankle fracture. Therefore, the purpose of this study was to determine if geriatric patients admitted to nursing homes following an ankle fracture experience elevated mortality rates. Methods: Patients were identified using diagnosis codes for ankle fractures from all 2008 part A Medicare claims, and those admitted to nursing homes were identified using a Minimum Data Set (MDS). The Medicare database was also analyzed for specific variables including over-all one year mortality, length of stay, age distribution, certain demographical characteristics, incidence of medical and surgical complications within 90 days, and the presence of comorbidities. Multivariate logistic regression analysis was used to determine if patients admitted to nursing homes had elevated mortality rates. Results: 19,648 patients with ankle fractures were identified, and 11,625 (59.0%) of these patients went to a nursing home after hospitalization. Patients who went to a nursing home had higher Elixhauser and Deyo–Charlson comorbidity scores (p < 0.0001). Nursing home patients also had significantly increased rates of postoperative medical and surgical complications. One year mortality was 6.9% for patients who did not go to a nursing home and 15.4% for patients who were admitted to a nursing home (p < 0.0001). However, multivariate logistic regression analysis demonstrated no significant difference in one year mortality between patients admitted to nursing homes and those who were not (OR = 1.1; 95% CI 0.99– 1.24, p > 0.05). Discussion: Although admission to nursing home was significantly associated with increased mortality in a bivariate statistical model, this significance was lost during multivariate analysis. This suggests that other patient characteristics may play a more prominent role in determining one year mortality following geriatric ankle fractures. ß 2015 Elsevier Ltd. All rights reserved.
Keywords: Geriatrics Injury Ankle fracture Morbidity Mortality Nursing homes Disposition
Introduction It is currently estimated that Americans 65 and older account for nearly 13% of our population, and this proportion is projected to steadily increase yearly. By 2050, older Americans will account for
E-mail address: [email protected]. 59 Executive Park South, Atlanta, GA 303029, USA. Tel.: +1 404 778 3350. 593 Eddy St, Providence, RI 02903, USA. Tel.: +1 401 444 4030. 3 593 Eddy St, Providence, RI 02903, USA. Tel.: +1 401 863 3240. 4 59 Executive Park South #2000, Atlanta, GA 30329, USA. Tel.: +1 404 778 3350. 1 2
one fifth of our nation’s population [1]. Geriatric patients can be challenging to manage because they have unique disease presentations, different ideologies for management, and important social considerations. Although the most common reasons for hospitalization within the elderly population are cardiac or pulmonary in origin, orthopaedic injuries are also a frequent cause of hospital admission [2]. The management of an orthopaedic geriatric patient is often complicated both preoperatively by medical comorbidities and postoperatively by challenging dispositions. Following orthopaedic surgery, elderly patients frequently require additional assistance and care at home. Unfortunately, not all patients have the
http://dx.doi.org/10.1016/j.injury.2015.05.020 0020–1383/ß 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
2
social support structure to return home and consequently require placement into nursing homes. Although nursing homes can be a safe and supportive environment, they can also carry significant mortality and morbidity risks. Studies have estimated that the rate of nursing home acquired pneumonia is ten times greater than community acquired pneumonia, and the fatality rate of influenza is greater than 50% in this patient population [3,4]. The rate of falls amongst nursing home residents is nearly twice the rate of falls in individuals living in the community, and it is estimated that a quarter of falls in nursing homes require admission to the hospital [5,6]. Given these statistics, it is reasonable to assume that patients admitted to nursing homes after ankle fractures may be at increased risk for complications. In fact, a few studies have examined the negative impact nursing home disposition can have on a patient’s recovery following orthopaedic injuries. For example, several studies have shown that patients discharged to nursing homes after hospitalization for hip fractures have higher mortality rates [7]. Bini et al. found that patients discharged to nursing homes after primary total joint arthroplasty are more likely to be readmitted to the hospital within 90 days of surgery [8]. Ankle fractures are the third most common type of fracture seen in the geriatric population, and the incidence of geriatric ankle fractures continues to increase [9–11]. There is currently no literature on the impact of discharge disposition in the geriatric population following admission for ankle fracture. Accordingly, the purpose of this study is to determine if there is increased mortality risk when patients are admitted to a nursing home following hospitalization for an ankle fracture. Methods This study was approved by the institutional review board (IRB) prior to initiation of data collection. Complete (100%) data from the US Centers for Medicare were obtained, and all claims from the Medicare Provider Analysis and Reviews (MedPar) File that involved International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) codes for ankle fractures (824.0– 824.9) were assessed. All 2008 part A Medicare claims were searched for diagnosis codes for ankle fractures. Any patients younger than 65 as of January 1, 2008 or who had sustained an ankle fracture in the preceding year were excluded from data analysis. Only patients admitted to the hospital were included in data analysis. In addition to the ICD-9-CM associated with the diagnosis, each claim included unique identifiers linked to the Medicare denominator file allowing determination of age, age distribution, length of stay, and discharge status. Patients who were admitted to nursing homes were identified using the Minimum Data Set (MDS), which is a federally mandated assessment completed for each nursing home resident. The MDS is stored in a national database at the CMS [12]. To stratify the overall health of each group, average Elixhauser and Deyo–Charlson scores were determined using co-morbidities listed in the MedPar file [14]. Whether the patient underwent operative fixation of their fracture within one month of their initial injury was also collected from the database. The primary outcome evaluated was 1-year post-injury mortality, which was determined by analysis of the Medicare denominator file analyzing for a date of death within one year of the initial ICD-9-CM diagnosis code. Secondary outcomes include the incidence of the following complications within 90 days of the injury: post-operative infection, deep venous thrombosis (DVT), pulmonary embolism (PE), congestive heart failure (CHF), pneumonia (PNA), urinary tract infection (UTI), pressure ulcers, myocardial infarction (MI), C. difficile infection, and gastrointestinal bleeds. The prevalence of these complications was determined by identifying the relevant ICD-9-CM codes in claims within 90
days of definitive treatment. Readmissions within 30 days of hospitalization were also collected for analysis and included in the statistical model. Statistical analysis The mortality from any cause within 1 year from the fracture was measured and Kaplan–Meier survival curves were generated for patients admitted to a nursing home and those who were not admitted to a nursing home. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression analysis. Candidate variables were selected by the bivariate models or by their clinical implications. Covariates included in the final model for one-year mortality were age at time of the ankle fracture, operative treatment within one month of admission, complications (DVT, PE, CHF, MI and infection) within 90 days from the fracture, Elixhauser and Deyo–Charlson scores at time of the ankle fracture and readmission within 30 days from the fracture. All statistical analysis was performed using SAS version 9.3(SAS Institute, Cary, NC). Results 19,648 patients with ankle fractures were identified using the inclusion and exclusion criteria discussed above. 11,625 (59.0%) of these patients were admitted to a nursing home. Descriptive characteristics of both groups are demonstrated in Table 1. The mean age for patients who went to a nursing home and those who did not was 79.8 and 74.2 (p < 0.0001). The average length of hospital stay was 3.7 days for patients who did not go to a nursing home and 5.2 days for patients who went to nursing homes. Patients who went to a nursing home had a mean Elixhauser score of 2.5, and those that did not had a score of 2 (p < 0.0001). The mean Deyo–Charlson score was 1.2 for patients who went to a nursing home and 0.8 for those that did not (p < 0.0001). One year mortality was 6.9% for patients who were not admitted to a nursing home and 15.4% for patients who were admitted to a nursing home (p < 0.0001). (Table 2). The Kaplan– Meier one-year survival curve comparing each group is seen in Fig. 1. The comparative incidences of complications between the two study groups within ninety days of hospitalization are shown in Table 3 along with statistical analysis. Although the rate of DVT was significantly higher in the nursing home group, the rate of PE was not found to be significantly higher. The rates of postoperative infection and C. difficile infection were both higher in patients who went to nursing homes (p < 0.0001). The rate of medical complications such as PNA, MI, CHF, and GI bleeds was significantly higher in the nursing home group (all had p < 0.0001). The incidence of UTIs and pressure sores was also
Table 1 Patient and clinical characteristics of the study population (N = 19,648). Admitted to nursing home (N = 11,625; 59%) Age in years, mean (SD) Sex Male (%) Female (%) Average length of hospital stay, days (SD) Mean Elixhauser score (SD) Mean Deyo–Charlson score (SD) Died within one year of ankle fracture (%)
79.8 (7.9)
Not admitted to nursing home (N = 8023; 41%) 74.2 (7)
2062 (17.7%) 9563 (82.3%) 5.2 (4.1)
2179 (27.2%) 5844 (72.8%) 3.7 (4)
2.5 (1.3) 1.2 (1.3) 1786 (15.4%)
2 (1.3) 0.8 (1.1) 552 (6.9%)
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx Table 2 Chi-square test. Bivariate analysis for mortality within one year for patients admitted to nursing homes (N = 19,648). Death within one year of discharge Admitted to nursing home Yes No Totals
Yes
No
1786 (76.4%) 552 (23.6%) 2338 p-value
9839 (56.8%) 7471 (43.2%) 17,310 0.0001
significantly higher in the nursing home group, and nearly a quarter of patients in the nursing home group was diagnosed with a UTI within 90 days of hospitalisation. Nearly one fifth of the patients in the nursing home group were readmitted to the hospital within 30 days of hospitalization (p < 0.0001). A multivariate logistic regression was performed to determine which factors influence one-year mortality (Table 4). C statistics describe the goodness of fit for logistic regression analyses at predicting outcomes. A c statistic of 0.5 indicates that a model is no better than chance while a c statistic of 1.0 indicates that a model predicts perfectly. A value of 0.8 indicates a model that predicts with strong association. The c-statistic for this work was 0.805 indicating robust correlation. Admission to a nursing home was not significantly associated with mortality in the multivariate model, odds ratio 1.11 (95% CI 0.99–1.24). Operative intervention was shown to be protective of mortality with an odds ratio of 0.534 (95% CI 0.483–0.591). The most significant factors associated with mortality were: (1) readmission within thirty day, odds ratio 2.587 (95%CI 2.31–2.90); (2) myocardial infarction, odds ratio 2.731 (95% CI 2.26–3.31) and (3) congestive heart failure, odds ratio 1.813 (95%CI 1.62–2.03) Age was also associated with increased mortality to a lesser extent. Postoperative infection was associated with mortality as well (1.2) but did not demonstrate statistical significance (95% CI 0.967–1.50). While deep vein thrombosis demonstrated statistical significance in being associated with mortality (odds ratio 1.53, 95% CI 1.21–1.93), pulmonary emboli did not demonstrate a statistically significant association (1.41, 95% CI 0.897–2.21). Discussion As the geriatric population continues to grow, the incidence of orthopaedic injuries in this patient population will undoubtedly increase. Ankle fractures represent a significant portion of the orthopaedic pathology seen within this patient population. An appreciation of the factors impacting mortality and morbidity following a geriatric ankle is imperative in order to provide accurate information to patient and families and to guide management. Nearly sixty percent of patients required admission to nursing homes following ankle fractures in this study, which is in the range of reported nursing home admission rates following geriatric hip fracture (48%–64%) [13,14]. It is important to note that the present study only included patients who required inpatient admission. Therefore, this patient population represents a sicker and/or more severely injured group, which likely inflates the percentage of nursing home admissions. Further work should also include patients who are not admitted to the hospital for ankle fractures. Based on these results, patients admitted to nursing homes are at an increased risk for many adverse postoperative outcomes. Nursing home patients tend to be more medically debilitated at baseline than patients who return home, which is evident in this study based on the elevated Elixhauser and
3
Deyo–Charlson scores seen in patients admitted to nursing homes. Therefore, it is no surprise that these patients had higher rates of postoperative medical complications such as pneumonia, congestive heart failure, myocardial infarctions, and gastrointestinal bleeds. These findings are more likely attributed to the poor baseline health status of the nursing home patient as opposed to the fact that they are residing in a nursing home. Patients sent to nursing homes had increased rates of infections, and several studies have identified infections as a serious and common problem in nursing homes. Infections are responsible for over fifty percent of nursing home deaths, and each nursing home resident experiences on average at least one serious infection annually [15,16]. Nursing home patients were also more likely to develop a pressure ulcer. It is estimated that one in five current nursing home patients have a pressure ulcer, and patients admitted to hospitals directly from nursing homes were five times more likely to have pressure ulcers than those admitted from home [17,18]. A recent study found that stage four pressure ulcer hospitalisations can cost over $120,000, which highlights the severity of this condition and its dire medical and financial consequences [19]. The 30 day readmission rate amongst the patients discharged to nursing homes was also higher than those who did not go to nursing homes. Hospital readmissions are often viewed as indicator of poor healthcare quality, and they can burden healthcare systems both logistically and financially. Orthopaedic surgeons must be vigilant to ensure that nursing home patients are followed closely by a team of specialists including geriatric and medicine specialists in order to help prevent many of these adverse complications. The general public opinion of nursing homes is largely negative and there is significant data to support this view as discussed earlier. A recent study examining articles from four major American newspapers over a ten year span found that most articles published about nursing homes were largely negative in nature [20]. In addition, only one third of Americans believe that nursing homes are doing a ‘‘good job’’ of caring for patients [20]. Although admission to a nursing home was found to be significantly associated with mortality during bivariate analysis, this significance was lost during multivariate analysis. This suggests that other factors likely contribute more to the mortality risk of the patient. As expected, the patient’s general health state is important as patients with heart attacks, congestive heart failure, and elevated Deyo–Charlson comorbidity index scores all had increased mortality risks on multivariable analysis. When discussing outcomes with patients and families following geriatric ankle fracture, these results suggest that admission to nursing home may not carry a significant mortality risk, and the baseline health status of the patient is likely much more important. A recent study on patients following surgical fixation of ankle fractures in a general patient population found that older patients and increased comorbidities was associated with increased mortality, which supports the findings of this study and highlights the importance of patient health status in predicting mortality risk [21]. Interestingly, the Elixhauser comorbidity score was not found to be significantly associated with mortality while the Deyo– Charlson comorbidity index was found to be significantly associated. The Deyo Charlson utilizes 17 co-morbidities, while the Elixhauser score was developed utilising 30 comorbidities with the hopes of more precisely measuring co-morbidity as a whole [22]. In addition, the Elixhauser score was developed looking at comorbidities associated with in hospital mortality, while the Charlson comorbidity index was developed by identifying comorbidities associated with overall one year mortality [23,24]. Although there will be similarities between the scoring systems, the ‘‘point value’’ of certain comorbidities will vary depending on
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 4
R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
Fig. 1. Kaplan–Meier curve for mortality in one year following ankle fracture (A) patients admitted to nursing homes (B) patients not admitted to nursing homes.
the measured outcome. While these comorbidity scoring systems have valuable clinical utility, it is important to understand the differences between them when interpreting any statistical analysis. Multivariable analysis discovered several other variables associated with mortality. Patients who underwent operative fixation within one month exhibited a decrease mortality risk,
which is in concordance with the current literature on management of geriatric ankle fractures [25]. Patients readmitted to the hospital within 30 days of discharge were also at increased mortality risk. This is an expected finding as these patients likely have several other health issues present which place them at an increased risk for mortality within one year.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
5
Table 3 Complications within 90 days of discharge between two study groups (N = 19,648).
Deep vein thrombosis Yes No Pulmonary embolism Yes No Post-operative infection Yes No Pneumonia Yes No Congestive heart failure Yes No Urinary tract infection Yes No Pressure ulcers Yes No Myocardial infarction Yes No C. difficile infection Yes No Gastrointestinal bleeds Yes No 30 day readmission Yes No
Not admitted to nursing home (N = 8023)
Admitted to nursing home (N = 11,625)
p-value
1.5% (124/8023) 98.5% (7899/8023)
3.3% (384/11,625) 96.7% (11,241/11,625)
p < 0.0001
0.9% (71/8023) 99.1% (7952/8023)
1.0% (115/11,625) 99.0% (11,510/11,625)
p = 0.46
1.9% (156/8023) 98.1% (7867/8023)
4.8% (559/11,625) 95.2% (11,066/11,625)
p < 0.0001
0.6% (45/8023) 99.4% (7978/8023)
1.6% (188/11,625) 98.4% (11,437/11,625)
p < 0.0001
8.1% (646/8023) 91.9% (7377/8023)
20.6% (2397/11,625) 79.4% (9228/11,625)
p < 0.0001
8.5% (682/8023) 91.5% (7341/8023)
22.0% (2561/11,625) 78.0% (9064/11,625)
p < 0.0001
1.3% (106/8023) 98.7% (7917/8023)
4.9% (569/11,625) 95.1% (11,056/11,625)
p < 0.0001
1.7% (136/8023) 98.3% (7887/8023)
4.0% (460/11,625) 96.0% (11,165/11,625)
p < 0.0001
0.5% (42/8023) 99.5% (7981/8023)
2.3% (271/11,625) 97.7% (11,354/11,625)
p < 0.0001
1.2% (100/8023) 98.8% (7923/8023)
3.0% (350/11,625) 97.0% (11,275/11,625)
p < 0.0001
7.0% (558/8023) 93.0% (7465/8023)
17.6% (2050/11,625) 82.4% (9575/11,625)
p < 0.0001
Table 4 Multivariable logistic regression analysis of factors associated with one-year mortality following geriatric ankle fracture (N = 19,648).
*
Risk factor
OR (95% CI)
Operative fixation within one month Age Elixhauser score Deyo–Charlson comorbidity index Admission to nursing home Readmission within 30 days: yes vs. no (reference) Complications within 90 days of discharge: yes vs. no (reference) Deep vein thrombosis Pulmonary embolism Congestive heart failure Myocardial infarction Postoperative infection
0.534 (0.483–0.591)* 1.08 (1.07–1.09)* 1.0 (0.959–1.04) 1.39 (1.34–1.44)* 1.11 (0.99–1.24) 2.59 (2.31–2.90)*
1.53 1.41 1.81 2.73 1.20
(1.21–1.93)* (0.897–2.21) (1.62–2.03)* (2.26–3.31)* (0.967–1.50)
p < 0.05.
Conclusion Geriatric ankle fractures are extremely common and their incidence will continue to rise annually. Orthopaedic surgeons must be familiar with the risk factors associated with mortality when discussing prognosis with patients and their families. This study highlights the importance of the patient’s general health status when determining mortality rates after geriatric ankle fractures. Although the public generally views nursing homes in a negative light, this study does not find any increased mortality risk associated with nursing home disposition following geriatric ankle fractures. However, the study findings do indicate that patients who go to nursing homes postoperatively are at increased risk for
postoperative complications and close follow-up is warranted and necessary. Conflict of interest There are no sources of potential conflict of interest for any of the authors of this manuscript. These include personal relationships, interests, and affiliations over the past three years, as well as any grants or funding, employment, affiliations, patents, inventions, honoraria, consultancies, royalties, stock options/ownership, or expert testimony. Acknowledgement We would like to thank Dr. Vincent Mor with the Brown University Department of Geriatrics for his assistance with data collection and analysis. References [1] U.S. Department of Health and Human Services: Administration for Community Living [Internet]. Washington, DC: Administration for Community Living; c2014. Administration of Aging (AoA): Projected Future Growth of the older population [cited 2014 Nov 9]. Available from: http://www.aoa.gov/ Aging_Statistics/future_growth/future_growth.aspx. [2] Russo CA, Elixhauser A. Hospitalizations in the elderly population. Statistical brief #6. May 2006. Rockville, MD: Agency for Healthcare Research and Quality; 2003, http://www.hcup-us.ahrq.gov/reports/statbriefs/sb6.pdf. [3] Juthani-Metha M, Quagliarello VJ. Infectious diseases in the nursing home setting: challenges and opportunities for clinical investigation. Clin Infect Dis 2010;51(October (8)):931–6. [4] Mathei C, Niclaes L, Suetens C, Jans B, Buntinx F. Infections in residents of nursing homes. Infect Dis Clin North Am 2007;(21):761–72. [5] Centers for Disease and Control Prevention: Home and Recreational Safety [Internet]. Atlanta: Centers for Disease and Control Prevention, c2014. Falls in
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
6
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Nursing Homes [cited 2014 Nov 9]. Available from: http://www.cdc.gov/ homeandrecreationalsafety/falls/nursing.html Vu MQ, Weintraub N, Rubsenstein LZ. Falls in the nursing home: are they preventable? J Am Med Dir Assoc 2004;5(November–December (6)):401–6. Neuman MD, Silber JH, Magaziner JS, Passarella MA, Metha S, Werner RM. Survival and functional outcomes after hip fracture among nursing home residents. JAMA Intern Med 2014;174(August (8)):1273–80. Bin SA, Fithian DC, Paxton LW, Khatod MX, Inacio MC, Namba RS. Does discharge disposition after primary total joint arthroplasty affect readmission rates? J Arthroplasty 2010;25(January (1)):114–7. Sporer SM, Weinstein JN, Koval KJ. The geographic incidence and treatment variation of common fractures of elderly patients. J Am Acad Orthop Surg 2006;14(4):246–55. Thur CK, Edgren G, Jansson K-A˚, Wretenberg P. Epidemiology of adult ankle fractures in Sweden between 1987 and 2004: a population-based study of 91,410 Swedish inpatients. Acta Orthop 2012;83(3):276–81. http://dx.doi.org/ 10.3109/17453674.2012.672091. Kannus P, Palvanen M, Niemi S, Parkkari J, Ja¨rvinen M. Increasing number and incidence of low-trauma ankle fractures in elderly people: finnish statistics during 1970–2000 and projections for the future. Bone 2002;31(3):430–3. Centers for Medicare and Medicaid Services [Internet]. Baltimore: Centers for Medicare and Medicaid Services; c2014. Minimum Data Set 3.0 Public Reports [cited 2014 Nov 9]. Available from: http://www.cms.gov/ResearchStatistics-Data-and-Systems/Computer-Data-and-Systems/MinimumData-Set-3-0-Public-Reports/index.html Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability, and nursing home use for persons with and without hip fracture: a populationbased study. J Am Geriatr Soc 2002;50(October (10)):1644–50. Fitzgerald JF, Fagan LF, Tierney WM, Dittus RS. Changing patterns of hip fracture care before and after implementation of the prospective payment system. JAMA 1987;258(July (2)):218–21.
[15] Smith PW. Infection control in the long-term healthcare facility. Geriatrics 1989;44(August, Suppl. A):11–4. [16] Smith PW, Bennett G, Bradley S, Drinka P, Lautenbach E, Marx J, et al. SHEA/ APIC guideline: infection prevention and control in the long-term care facility July 2008. Infect Control Hosp Epidemiol 2008;29(September (9)):785–814. [17] Caffrey C. Pressure ulcers among nursing home residents: United States 2004. NCHS data brief, no. 14. Hyattsville, MD: National Center for Health Statistics; 2009. [18] Keeleghan E, Margolis D, Zhan M, Baumgarten M. Prevalence of pressure ulcers on hospital admission among nursing home residents transferred to the hospital. Wound Repair Regen 2008;16(May–June (3)):331–6. [19] Brem H, Maggi J, Neirman D, Rolnitzky L, Bell D, Rennert R, et al. High cost of stage IV pressure ulcers. Am J Surg 2010;200(October (4)):473–7. [20] Miller EA, Tyler DA, Rozanova J, Mor V. National newspaper portrayal of U.S. nursing homes: periodic treatment of topic and tone. Milibank Q 2012;90(December (4)):725–61. [21] Basques BA, Miller CP, Golinvaux NS, Bohl DD, Grauer JN. Morbidity readmission after open reduction and interal fixation of ankle fractures are associated with preoperative patient characteristics. Clin Orthop Relat Res 2014 [Epud ahead of print]. [22] Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/ Deyo methods of comorbidity measurement in administrative data. Med Care 2004;42(4):355–60. [23] Walraven CV, Austin PC, Jennings A, Quan H, Forster AJ. A modication of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47(6):626–33. [24] Voskuijl T, Hageman M, Ring D. Higher Charlson comorbidity index scores are associated with readmission after orthopaedic surgery. Clin Orthop Relat Res 2014;472(May (5)):1638–44. [25] Koval KJ, Zhou W, Sparks MJ, Cantu RV, Hecht P, Lurie J. Complications after ankle fracture in elderly patients. Foot Ankle Int 2007;28(12):1249–55.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
JINJ-6210; No. of Pages 6 Injury, Int. J. Care Injured xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes Rishin J. Kadakia a,1, Ray Y. Hsu b,2, Roman Hayda b,2, Yoojin Lee c,3, Jason T. Bariteau a,4 a
Emory University Department of Orthopaedics, Atlanta, GA, USA Brown University Department of Geriatrics, Providence, RI, USA c Brown University Department of Orthopaedics, Providence, RI, USA b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 1 May 2015
Introduction: The incidence of geriatric ankle fractures will undoubtedly increase as the population continues to grow. Many geriatric patients struggle to function independently after such injury and often require placement into nursing homes. The morbidity and mortality associated with nursing homes is well documented within the field of orthopaedic surgery. However, there is currently no study examining the mortality associated with nursing home placement following hospitalization for an ankle fracture. Therefore, the purpose of this study was to determine if geriatric patients admitted to nursing homes following an ankle fracture experience elevated mortality rates. Methods: Patients were identified using diagnosis codes for ankle fractures from all 2008 part A Medicare claims, and those admitted to nursing homes were identified using a Minimum Data Set (MDS). The Medicare database was also analyzed for specific variables including over-all one year mortality, length of stay, age distribution, certain demographical characteristics, incidence of medical and surgical complications within 90 days, and the presence of comorbidities. Multivariate logistic regression analysis was used to determine if patients admitted to nursing homes had elevated mortality rates. Results: 19,648 patients with ankle fractures were identified, and 11,625 (59.0%) of these patients went to a nursing home after hospitalization. Patients who went to a nursing home had higher Elixhauser and Deyo–Charlson comorbidity scores (p < 0.0001). Nursing home patients also had significantly increased rates of postoperative medical and surgical complications. One year mortality was 6.9% for patients who did not go to a nursing home and 15.4% for patients who were admitted to a nursing home (p < 0.0001). However, multivariate logistic regression analysis demonstrated no significant difference in one year mortality between patients admitted to nursing homes and those who were not (OR = 1.1; 95% CI 0.99– 1.24, p > 0.05). Discussion: Although admission to nursing home was significantly associated with increased mortality in a bivariate statistical model, this significance was lost during multivariate analysis. This suggests that other patient characteristics may play a more prominent role in determining one year mortality following geriatric ankle fractures. ß 2015 Elsevier Ltd. All rights reserved.
Keywords: Geriatrics Injury Ankle fracture Morbidity Mortality Nursing homes Disposition
Introduction It is currently estimated that Americans 65 and older account for nearly 13% of our population, and this proportion is projected to steadily increase yearly. By 2050, older Americans will account for
E-mail address: [email protected]. 59 Executive Park South, Atlanta, GA 303029, USA. Tel.: +1 404 778 3350. 593 Eddy St, Providence, RI 02903, USA. Tel.: +1 401 444 4030. 3 593 Eddy St, Providence, RI 02903, USA. Tel.: +1 401 863 3240. 4 59 Executive Park South #2000, Atlanta, GA 30329, USA. Tel.: +1 404 778 3350. 1 2
one fifth of our nation’s population [1]. Geriatric patients can be challenging to manage because they have unique disease presentations, different ideologies for management, and important social considerations. Although the most common reasons for hospitalization within the elderly population are cardiac or pulmonary in origin, orthopaedic injuries are also a frequent cause of hospital admission [2]. The management of an orthopaedic geriatric patient is often complicated both preoperatively by medical comorbidities and postoperatively by challenging dispositions. Following orthopaedic surgery, elderly patients frequently require additional assistance and care at home. Unfortunately, not all patients have the
http://dx.doi.org/10.1016/j.injury.2015.05.020 0020–1383/ß 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
2
social support structure to return home and consequently require placement into nursing homes. Although nursing homes can be a safe and supportive environment, they can also carry significant mortality and morbidity risks. Studies have estimated that the rate of nursing home acquired pneumonia is ten times greater than community acquired pneumonia, and the fatality rate of influenza is greater than 50% in this patient population [3,4]. The rate of falls amongst nursing home residents is nearly twice the rate of falls in individuals living in the community, and it is estimated that a quarter of falls in nursing homes require admission to the hospital [5,6]. Given these statistics, it is reasonable to assume that patients admitted to nursing homes after ankle fractures may be at increased risk for complications. In fact, a few studies have examined the negative impact nursing home disposition can have on a patient’s recovery following orthopaedic injuries. For example, several studies have shown that patients discharged to nursing homes after hospitalization for hip fractures have higher mortality rates [7]. Bini et al. found that patients discharged to nursing homes after primary total joint arthroplasty are more likely to be readmitted to the hospital within 90 days of surgery [8]. Ankle fractures are the third most common type of fracture seen in the geriatric population, and the incidence of geriatric ankle fractures continues to increase [9–11]. There is currently no literature on the impact of discharge disposition in the geriatric population following admission for ankle fracture. Accordingly, the purpose of this study is to determine if there is increased mortality risk when patients are admitted to a nursing home following hospitalization for an ankle fracture. Methods This study was approved by the institutional review board (IRB) prior to initiation of data collection. Complete (100%) data from the US Centers for Medicare were obtained, and all claims from the Medicare Provider Analysis and Reviews (MedPar) File that involved International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) codes for ankle fractures (824.0– 824.9) were assessed. All 2008 part A Medicare claims were searched for diagnosis codes for ankle fractures. Any patients younger than 65 as of January 1, 2008 or who had sustained an ankle fracture in the preceding year were excluded from data analysis. Only patients admitted to the hospital were included in data analysis. In addition to the ICD-9-CM associated with the diagnosis, each claim included unique identifiers linked to the Medicare denominator file allowing determination of age, age distribution, length of stay, and discharge status. Patients who were admitted to nursing homes were identified using the Minimum Data Set (MDS), which is a federally mandated assessment completed for each nursing home resident. The MDS is stored in a national database at the CMS [12]. To stratify the overall health of each group, average Elixhauser and Deyo–Charlson scores were determined using co-morbidities listed in the MedPar file [14]. Whether the patient underwent operative fixation of their fracture within one month of their initial injury was also collected from the database. The primary outcome evaluated was 1-year post-injury mortality, which was determined by analysis of the Medicare denominator file analyzing for a date of death within one year of the initial ICD-9-CM diagnosis code. Secondary outcomes include the incidence of the following complications within 90 days of the injury: post-operative infection, deep venous thrombosis (DVT), pulmonary embolism (PE), congestive heart failure (CHF), pneumonia (PNA), urinary tract infection (UTI), pressure ulcers, myocardial infarction (MI), C. difficile infection, and gastrointestinal bleeds. The prevalence of these complications was determined by identifying the relevant ICD-9-CM codes in claims within 90
days of definitive treatment. Readmissions within 30 days of hospitalization were also collected for analysis and included in the statistical model. Statistical analysis The mortality from any cause within 1 year from the fracture was measured and Kaplan–Meier survival curves were generated for patients admitted to a nursing home and those who were not admitted to a nursing home. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression analysis. Candidate variables were selected by the bivariate models or by their clinical implications. Covariates included in the final model for one-year mortality were age at time of the ankle fracture, operative treatment within one month of admission, complications (DVT, PE, CHF, MI and infection) within 90 days from the fracture, Elixhauser and Deyo–Charlson scores at time of the ankle fracture and readmission within 30 days from the fracture. All statistical analysis was performed using SAS version 9.3(SAS Institute, Cary, NC). Results 19,648 patients with ankle fractures were identified using the inclusion and exclusion criteria discussed above. 11,625 (59.0%) of these patients were admitted to a nursing home. Descriptive characteristics of both groups are demonstrated in Table 1. The mean age for patients who went to a nursing home and those who did not was 79.8 and 74.2 (p < 0.0001). The average length of hospital stay was 3.7 days for patients who did not go to a nursing home and 5.2 days for patients who went to nursing homes. Patients who went to a nursing home had a mean Elixhauser score of 2.5, and those that did not had a score of 2 (p < 0.0001). The mean Deyo–Charlson score was 1.2 for patients who went to a nursing home and 0.8 for those that did not (p < 0.0001). One year mortality was 6.9% for patients who were not admitted to a nursing home and 15.4% for patients who were admitted to a nursing home (p < 0.0001). (Table 2). The Kaplan– Meier one-year survival curve comparing each group is seen in Fig. 1. The comparative incidences of complications between the two study groups within ninety days of hospitalization are shown in Table 3 along with statistical analysis. Although the rate of DVT was significantly higher in the nursing home group, the rate of PE was not found to be significantly higher. The rates of postoperative infection and C. difficile infection were both higher in patients who went to nursing homes (p < 0.0001). The rate of medical complications such as PNA, MI, CHF, and GI bleeds was significantly higher in the nursing home group (all had p < 0.0001). The incidence of UTIs and pressure sores was also
Table 1 Patient and clinical characteristics of the study population (N = 19,648). Admitted to nursing home (N = 11,625; 59%) Age in years, mean (SD) Sex Male (%) Female (%) Average length of hospital stay, days (SD) Mean Elixhauser score (SD) Mean Deyo–Charlson score (SD) Died within one year of ankle fracture (%)
79.8 (7.9)
Not admitted to nursing home (N = 8023; 41%) 74.2 (7)
2062 (17.7%) 9563 (82.3%) 5.2 (4.1)
2179 (27.2%) 5844 (72.8%) 3.7 (4)
2.5 (1.3) 1.2 (1.3) 1786 (15.4%)
2 (1.3) 0.8 (1.1) 552 (6.9%)
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx Table 2 Chi-square test. Bivariate analysis for mortality within one year for patients admitted to nursing homes (N = 19,648). Death within one year of discharge Admitted to nursing home Yes No Totals
Yes
No
1786 (76.4%) 552 (23.6%) 2338 p-value
9839 (56.8%) 7471 (43.2%) 17,310 0.0001
significantly higher in the nursing home group, and nearly a quarter of patients in the nursing home group was diagnosed with a UTI within 90 days of hospitalisation. Nearly one fifth of the patients in the nursing home group were readmitted to the hospital within 30 days of hospitalization (p < 0.0001). A multivariate logistic regression was performed to determine which factors influence one-year mortality (Table 4). C statistics describe the goodness of fit for logistic regression analyses at predicting outcomes. A c statistic of 0.5 indicates that a model is no better than chance while a c statistic of 1.0 indicates that a model predicts perfectly. A value of 0.8 indicates a model that predicts with strong association. The c-statistic for this work was 0.805 indicating robust correlation. Admission to a nursing home was not significantly associated with mortality in the multivariate model, odds ratio 1.11 (95% CI 0.99–1.24). Operative intervention was shown to be protective of mortality with an odds ratio of 0.534 (95% CI 0.483–0.591). The most significant factors associated with mortality were: (1) readmission within thirty day, odds ratio 2.587 (95%CI 2.31–2.90); (2) myocardial infarction, odds ratio 2.731 (95% CI 2.26–3.31) and (3) congestive heart failure, odds ratio 1.813 (95%CI 1.62–2.03) Age was also associated with increased mortality to a lesser extent. Postoperative infection was associated with mortality as well (1.2) but did not demonstrate statistical significance (95% CI 0.967–1.50). While deep vein thrombosis demonstrated statistical significance in being associated with mortality (odds ratio 1.53, 95% CI 1.21–1.93), pulmonary emboli did not demonstrate a statistically significant association (1.41, 95% CI 0.897–2.21). Discussion As the geriatric population continues to grow, the incidence of orthopaedic injuries in this patient population will undoubtedly increase. Ankle fractures represent a significant portion of the orthopaedic pathology seen within this patient population. An appreciation of the factors impacting mortality and morbidity following a geriatric ankle is imperative in order to provide accurate information to patient and families and to guide management. Nearly sixty percent of patients required admission to nursing homes following ankle fractures in this study, which is in the range of reported nursing home admission rates following geriatric hip fracture (48%–64%) [13,14]. It is important to note that the present study only included patients who required inpatient admission. Therefore, this patient population represents a sicker and/or more severely injured group, which likely inflates the percentage of nursing home admissions. Further work should also include patients who are not admitted to the hospital for ankle fractures. Based on these results, patients admitted to nursing homes are at an increased risk for many adverse postoperative outcomes. Nursing home patients tend to be more medically debilitated at baseline than patients who return home, which is evident in this study based on the elevated Elixhauser and
3
Deyo–Charlson scores seen in patients admitted to nursing homes. Therefore, it is no surprise that these patients had higher rates of postoperative medical complications such as pneumonia, congestive heart failure, myocardial infarctions, and gastrointestinal bleeds. These findings are more likely attributed to the poor baseline health status of the nursing home patient as opposed to the fact that they are residing in a nursing home. Patients sent to nursing homes had increased rates of infections, and several studies have identified infections as a serious and common problem in nursing homes. Infections are responsible for over fifty percent of nursing home deaths, and each nursing home resident experiences on average at least one serious infection annually [15,16]. Nursing home patients were also more likely to develop a pressure ulcer. It is estimated that one in five current nursing home patients have a pressure ulcer, and patients admitted to hospitals directly from nursing homes were five times more likely to have pressure ulcers than those admitted from home [17,18]. A recent study found that stage four pressure ulcer hospitalisations can cost over $120,000, which highlights the severity of this condition and its dire medical and financial consequences [19]. The 30 day readmission rate amongst the patients discharged to nursing homes was also higher than those who did not go to nursing homes. Hospital readmissions are often viewed as indicator of poor healthcare quality, and they can burden healthcare systems both logistically and financially. Orthopaedic surgeons must be vigilant to ensure that nursing home patients are followed closely by a team of specialists including geriatric and medicine specialists in order to help prevent many of these adverse complications. The general public opinion of nursing homes is largely negative and there is significant data to support this view as discussed earlier. A recent study examining articles from four major American newspapers over a ten year span found that most articles published about nursing homes were largely negative in nature [20]. In addition, only one third of Americans believe that nursing homes are doing a ‘‘good job’’ of caring for patients [20]. Although admission to a nursing home was found to be significantly associated with mortality during bivariate analysis, this significance was lost during multivariate analysis. This suggests that other factors likely contribute more to the mortality risk of the patient. As expected, the patient’s general health state is important as patients with heart attacks, congestive heart failure, and elevated Deyo–Charlson comorbidity index scores all had increased mortality risks on multivariable analysis. When discussing outcomes with patients and families following geriatric ankle fracture, these results suggest that admission to nursing home may not carry a significant mortality risk, and the baseline health status of the patient is likely much more important. A recent study on patients following surgical fixation of ankle fractures in a general patient population found that older patients and increased comorbidities was associated with increased mortality, which supports the findings of this study and highlights the importance of patient health status in predicting mortality risk [21]. Interestingly, the Elixhauser comorbidity score was not found to be significantly associated with mortality while the Deyo– Charlson comorbidity index was found to be significantly associated. The Deyo Charlson utilizes 17 co-morbidities, while the Elixhauser score was developed utilising 30 comorbidities with the hopes of more precisely measuring co-morbidity as a whole [22]. In addition, the Elixhauser score was developed looking at comorbidities associated with in hospital mortality, while the Charlson comorbidity index was developed by identifying comorbidities associated with overall one year mortality [23,24]. Although there will be similarities between the scoring systems, the ‘‘point value’’ of certain comorbidities will vary depending on
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 4
R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
Fig. 1. Kaplan–Meier curve for mortality in one year following ankle fracture (A) patients admitted to nursing homes (B) patients not admitted to nursing homes.
the measured outcome. While these comorbidity scoring systems have valuable clinical utility, it is important to understand the differences between them when interpreting any statistical analysis. Multivariable analysis discovered several other variables associated with mortality. Patients who underwent operative fixation within one month exhibited a decrease mortality risk,
which is in concordance with the current literature on management of geriatric ankle fractures [25]. Patients readmitted to the hospital within 30 days of discharge were also at increased mortality risk. This is an expected finding as these patients likely have several other health issues present which place them at an increased risk for mortality within one year.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
5
Table 3 Complications within 90 days of discharge between two study groups (N = 19,648).
Deep vein thrombosis Yes No Pulmonary embolism Yes No Post-operative infection Yes No Pneumonia Yes No Congestive heart failure Yes No Urinary tract infection Yes No Pressure ulcers Yes No Myocardial infarction Yes No C. difficile infection Yes No Gastrointestinal bleeds Yes No 30 day readmission Yes No
Not admitted to nursing home (N = 8023)
Admitted to nursing home (N = 11,625)
p-value
1.5% (124/8023) 98.5% (7899/8023)
3.3% (384/11,625) 96.7% (11,241/11,625)
p < 0.0001
0.9% (71/8023) 99.1% (7952/8023)
1.0% (115/11,625) 99.0% (11,510/11,625)
p = 0.46
1.9% (156/8023) 98.1% (7867/8023)
4.8% (559/11,625) 95.2% (11,066/11,625)
p < 0.0001
0.6% (45/8023) 99.4% (7978/8023)
1.6% (188/11,625) 98.4% (11,437/11,625)
p < 0.0001
8.1% (646/8023) 91.9% (7377/8023)
20.6% (2397/11,625) 79.4% (9228/11,625)
p < 0.0001
8.5% (682/8023) 91.5% (7341/8023)
22.0% (2561/11,625) 78.0% (9064/11,625)
p < 0.0001
1.3% (106/8023) 98.7% (7917/8023)
4.9% (569/11,625) 95.1% (11,056/11,625)
p < 0.0001
1.7% (136/8023) 98.3% (7887/8023)
4.0% (460/11,625) 96.0% (11,165/11,625)
p < 0.0001
0.5% (42/8023) 99.5% (7981/8023)
2.3% (271/11,625) 97.7% (11,354/11,625)
p < 0.0001
1.2% (100/8023) 98.8% (7923/8023)
3.0% (350/11,625) 97.0% (11,275/11,625)
p < 0.0001
7.0% (558/8023) 93.0% (7465/8023)
17.6% (2050/11,625) 82.4% (9575/11,625)
p < 0.0001
Table 4 Multivariable logistic regression analysis of factors associated with one-year mortality following geriatric ankle fracture (N = 19,648).
*
Risk factor
OR (95% CI)
Operative fixation within one month Age Elixhauser score Deyo–Charlson comorbidity index Admission to nursing home Readmission within 30 days: yes vs. no (reference) Complications within 90 days of discharge: yes vs. no (reference) Deep vein thrombosis Pulmonary embolism Congestive heart failure Myocardial infarction Postoperative infection
0.534 (0.483–0.591)* 1.08 (1.07–1.09)* 1.0 (0.959–1.04) 1.39 (1.34–1.44)* 1.11 (0.99–1.24) 2.59 (2.31–2.90)*
1.53 1.41 1.81 2.73 1.20
(1.21–1.93)* (0.897–2.21) (1.62–2.03)* (2.26–3.31)* (0.967–1.50)
p < 0.05.
Conclusion Geriatric ankle fractures are extremely common and their incidence will continue to rise annually. Orthopaedic surgeons must be familiar with the risk factors associated with mortality when discussing prognosis with patients and their families. This study highlights the importance of the patient’s general health status when determining mortality rates after geriatric ankle fractures. Although the public generally views nursing homes in a negative light, this study does not find any increased mortality risk associated with nursing home disposition following geriatric ankle fractures. However, the study findings do indicate that patients who go to nursing homes postoperatively are at increased risk for
postoperative complications and close follow-up is warranted and necessary. Conflict of interest There are no sources of potential conflict of interest for any of the authors of this manuscript. These include personal relationships, interests, and affiliations over the past three years, as well as any grants or funding, employment, affiliations, patents, inventions, honoraria, consultancies, royalties, stock options/ownership, or expert testimony. Acknowledgement We would like to thank Dr. Vincent Mor with the Brown University Department of Geriatrics for his assistance with data collection and analysis. References [1] U.S. Department of Health and Human Services: Administration for Community Living [Internet]. Washington, DC: Administration for Community Living; c2014. Administration of Aging (AoA): Projected Future Growth of the older population [cited 2014 Nov 9]. Available from: http://www.aoa.gov/ Aging_Statistics/future_growth/future_growth.aspx. [2] Russo CA, Elixhauser A. Hospitalizations in the elderly population. Statistical brief #6. May 2006. Rockville, MD: Agency for Healthcare Research and Quality; 2003, http://www.hcup-us.ahrq.gov/reports/statbriefs/sb6.pdf. [3] Juthani-Metha M, Quagliarello VJ. Infectious diseases in the nursing home setting: challenges and opportunities for clinical investigation. Clin Infect Dis 2010;51(October (8)):931–6. [4] Mathei C, Niclaes L, Suetens C, Jans B, Buntinx F. Infections in residents of nursing homes. Infect Dis Clin North Am 2007;(21):761–72. [5] Centers for Disease and Control Prevention: Home and Recreational Safety [Internet]. Atlanta: Centers for Disease and Control Prevention, c2014. Falls in
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
G Model
JINJ-6210; No. of Pages 6 R.J. Kadakia et al. / Injury, Int. J. Care Injured xxx (2015) xxx–xxx
6
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Nursing Homes [cited 2014 Nov 9]. Available from: http://www.cdc.gov/ homeandrecreationalsafety/falls/nursing.html Vu MQ, Weintraub N, Rubsenstein LZ. Falls in the nursing home: are they preventable? J Am Med Dir Assoc 2004;5(November–December (6)):401–6. Neuman MD, Silber JH, Magaziner JS, Passarella MA, Metha S, Werner RM. Survival and functional outcomes after hip fracture among nursing home residents. JAMA Intern Med 2014;174(August (8)):1273–80. Bin SA, Fithian DC, Paxton LW, Khatod MX, Inacio MC, Namba RS. Does discharge disposition after primary total joint arthroplasty affect readmission rates? J Arthroplasty 2010;25(January (1)):114–7. Sporer SM, Weinstein JN, Koval KJ. The geographic incidence and treatment variation of common fractures of elderly patients. J Am Acad Orthop Surg 2006;14(4):246–55. Thur CK, Edgren G, Jansson K-A˚, Wretenberg P. Epidemiology of adult ankle fractures in Sweden between 1987 and 2004: a population-based study of 91,410 Swedish inpatients. Acta Orthop 2012;83(3):276–81. http://dx.doi.org/ 10.3109/17453674.2012.672091. Kannus P, Palvanen M, Niemi S, Parkkari J, Ja¨rvinen M. Increasing number and incidence of low-trauma ankle fractures in elderly people: finnish statistics during 1970–2000 and projections for the future. Bone 2002;31(3):430–3. Centers for Medicare and Medicaid Services [Internet]. Baltimore: Centers for Medicare and Medicaid Services; c2014. Minimum Data Set 3.0 Public Reports [cited 2014 Nov 9]. Available from: http://www.cms.gov/ResearchStatistics-Data-and-Systems/Computer-Data-and-Systems/MinimumData-Set-3-0-Public-Reports/index.html Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability, and nursing home use for persons with and without hip fracture: a populationbased study. J Am Geriatr Soc 2002;50(October (10)):1644–50. Fitzgerald JF, Fagan LF, Tierney WM, Dittus RS. Changing patterns of hip fracture care before and after implementation of the prospective payment system. JAMA 1987;258(July (2)):218–21.
[15] Smith PW. Infection control in the long-term healthcare facility. Geriatrics 1989;44(August, Suppl. A):11–4. [16] Smith PW, Bennett G, Bradley S, Drinka P, Lautenbach E, Marx J, et al. SHEA/ APIC guideline: infection prevention and control in the long-term care facility July 2008. Infect Control Hosp Epidemiol 2008;29(September (9)):785–814. [17] Caffrey C. Pressure ulcers among nursing home residents: United States 2004. NCHS data brief, no. 14. Hyattsville, MD: National Center for Health Statistics; 2009. [18] Keeleghan E, Margolis D, Zhan M, Baumgarten M. Prevalence of pressure ulcers on hospital admission among nursing home residents transferred to the hospital. Wound Repair Regen 2008;16(May–June (3)):331–6. [19] Brem H, Maggi J, Neirman D, Rolnitzky L, Bell D, Rennert R, et al. High cost of stage IV pressure ulcers. Am J Surg 2010;200(October (4)):473–7. [20] Miller EA, Tyler DA, Rozanova J, Mor V. National newspaper portrayal of U.S. nursing homes: periodic treatment of topic and tone. Milibank Q 2012;90(December (4)):725–61. [21] Basques BA, Miller CP, Golinvaux NS, Bohl DD, Grauer JN. Morbidity readmission after open reduction and interal fixation of ankle fractures are associated with preoperative patient characteristics. Clin Orthop Relat Res 2014 [Epud ahead of print]. [22] Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/ Deyo methods of comorbidity measurement in administrative data. Med Care 2004;42(4):355–60. [23] Walraven CV, Austin PC, Jennings A, Quan H, Forster AJ. A modication of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47(6):626–33. [24] Voskuijl T, Hageman M, Ring D. Higher Charlson comorbidity index scores are associated with readmission after orthopaedic surgery. Clin Orthop Relat Res 2014;472(May (5)):1638–44. [25] Koval KJ, Zhou W, Sparks MJ, Cantu RV, Hecht P, Lurie J. Complications after ankle fracture in elderly patients. Foot Ankle Int 2007;28(12):1249–55.
Please cite this article in press as: Kadakia RJ, et al. Evaluation of one-year mortality after geriatric ankle fractures in patients admitted to nursing homes. Injury (2015), http://dx.doi.org/10.1016/j.injury.2015.05.020
