n Feature Article
In-hospital Mortality Risk for Femoral Neck Fractures Among Patients Receiving Medicare Brandon J. Erickson, MD; Benedict U. Nwachukwu, MD, MBA; Emmanouil Kiriakopoulos, BS; Rachel M. Frank, MD; Brett Levine, MD; Leonardo Villarroel, BS; Frank M. McCormick, MD
abstract Increased risk of mortality with time has been established in association with femoral neck fractures. However, little is known about the in-hospital mortality risk associated with femoral neck fractures in the US population. This study was conducted to determine the in-hospital mortality rate associated with femoral neck fractures and to identify independent demographic features associated with an increased risk of this primary outcome. The authors queried the PearlDiver database (PearlDiver Technologies, Inc, West Conshohocken, Pennsylvania) of Medicare patients from 2005 to 2010 for International Classification of Diseases, 9th Revision (ICD-9), diagnostic codes for femoral neck fractures and related conditions. Stratified sampling was conducted by creating a group within the data set that included patients with a death discharge using ICD-9-D-820.0 through ICD-9-D-820.13. Age, sex, and year of injury were analyzed as specific demographic variables related to mortality. A total of 751,232 femoral neck fractures occurred during the index study period. There were 11,420 deaths during the initial hospital stay, for an overall mortality rate of 1.52%. The mortality rate in patients older than 84 years was 2.06%. Of all deaths, 89% occurred in patients who were 75 years and older. The mortality rate for femoral neck fractures was 1.22% in women and 2.32% in men (odds ratio, 0.5; 95% confidence interval, 0.25-1.04). The overall mortality rate for patients in the Medicare population who were treated at an inpatient center for femoral neck fractures from 2005 to 2010 was 1.5%. Men had a mortality rate almost twice that of women. Patients older than 84 years were the most likely to die soon after sustaining a femoral neck fracture. [Orthopedics. 2015; 38(7):e593-e596.] The authors are from Midwest Orthopaedics at Rush (BJE, RMF, BL), Rush University Medical Center, Chicago, Illinois; the Hospital for Special Surgery (BUN), New York, New York; Florida International University (EK), Miami, Florida; and the Holy Cross Hospital Orthopaedic Institute (LV, FMM), Fort Lauderdale, Florida. Dr Erickson, Dr Nwachukwu, Mr Kiriakopoulos, Dr Frank, Mr Villarroel, and Dr McCormick have no relevant financial relationships to disclose. Dr Levine is a paid consultant to Biomet, Conmed, DePuy, and Zimmer. Correspondence should be addressed to: Frank M. McCormick, MD, Holy Cross Hospital Orthopaedic Institute, 5597 N Dixie Hwy, Fort Lauderdale, FL 33334 (
[email protected]). Received: April 15, 2014; Accepted: September 2, 2014. doi: 10.3928/01477447-20150701-57
JULY 2015 | Volume 38 • Number 7
e593
n Feature Article
W
ith an increase in the US elderly population, 77.2 million individuals will be older than 65 years by 2040, and the incidence of hip fractures is expected to continue to rise.1,2 It is estimated that in the near future more than 300,000 hip fractures will occur annually, with an associated cost of $7 billion to the health care industry.3 Hip fractures have an annual mortality rate of approximately 30%, but the rate can vary based on fracture pattern, preinjury level of function, and current health status of the patient.1,4 However, no study to date has reported the acute in-hospital mortality risk associated with hip fractures, specifically, femoral neck fractures. Because more than 50% of all hip fractures involve the femoral neck, it is important to understand the risk associated with these fractures in the inpatient setting to properly counsel patients and families about options and associated risks. Recent computer technology has enabled accurate and efficient review of large medical databases, including 100% of the Medicare database from 2005 to 2010. With access to this information, surgeons can more accurately quantify in-hospital mortality rates and associated demographic risk factors. These calculations allow patients to better understand the mortality rates associated with their pathologic conditions. Previous studies looked at similar data in terms of geriatric trauma, a much smaller population of all hip fractures, and other osteoporotic fractures.5-7 These risk assessment tools can be expanded and applied to measure mortality rates as a result of specific injuries, such as femoral neck fractures. Accurate mortality data allow surgeons to provide more specific information with regard to each individual’s pathology, instead of grouping injuries into a catch-all diagnosis, such as “hip fractures,” to encompass all femoral head, neck, intertrochanteric, and subtrochanteric fractures, as was previously done. The goals of this study were to use a comprehensive Medicare registry database
e594
to quantify inpatient mortality rates for patients who had femoral neck fractures from 2005 to 2010 and to determine their associated demographic features. The authors hypothesized that the acute mortality rate from femoral neck fractures is less than 5%, with a higher incidence in women and a higher incidence and a higher mortality rate in patients older than 75 years.
Materials and Methods The authors conducted a retrospective review of the Medicare database within the PearlDiver database (PearlDiver Technologies, Inc, West Conshohocken, Pennsylvania). The PearlDiver database is a publicly available national database that is compliant with the Health Insurance Portability and Accountability Act. The authors reviewed Medicare use and payment data captured as part of the PearlDiver database. This database captured 100% of Medicare data between 2005 and 2010. The database was queried based on hospital inpatient International Classification of Diseases, 9th Revision (ICD-9), diagnostic codes for femoral neck fractures and related conditions (ICD-9-D-820.00 to ICD-9-D-820.13). Stratified sampling was conducted through PearlDiver by creating a group within the data set that included patients with a death discharge from inpatient centers using the diagnostic codes ICD9-D-820.0 to ICD-9-D-820.13. Data were further analyzed for demographic parameters, assessing age, sex, and year of injury. The data used in this study were derived from the PearlDiver database, which currently includes more than 2 billion individual patient records. Diagnosis and procedure information is encrypted with Current Procedural Terminology and ICD-9 codes specific to orthopedic measures. Patients who had incomplete information (age, sex, year) were excluded. The primary goal of this study was to determine the number of death discharges from inpatient centers within the available data to statistically determine a quantified risk of death within the hospital setting for
patients with femoral neck fractures. The secondary goal was to assess the effect of demographic variables (age, sex, and year of injury) on overall acute inpatient mortality risk.
Results A total of 751,232 femoral neck fractures were treated in an inpatient setting from 2005 to 2010. Of the 751,232 femoral neck fractures, 601,503 (80%) occurred in patients 75 years or older. A total of 11,420 patients died during the initial hospital stay after femoral neck fractures. Therefore, the risk of death during the acute hospital stay for femoral neck fractures was 1.52%. Of the 11,420 deaths, 6435 occurred in patients who were older than 84 years (56%), although this group accounted for 41.5% of the total number of fractures. Furthermore, 8819 (77%) of deaths occurred in patients who were 80 years old or older, and 10,177 (90%) occurred in patients 75 years and older (Figure 1A and Table). Only 78,674 (10.5%) of femoral neck fractures occurred in patients younger than 70 years. The mortality rate from femoral neck fractures increased with age in a linear relationship (r=0.943), reaching a peak of 2.06% in patients older than 84 years. The rate was 0.65% in patients younger than 65 years (Figure 1B). There was a decrease each year in the absolute number of deaths. In 2005, a total of 2314 patients died of femoral neck fractures. In 2010, the number of deaths was 1542 (Figure 2A). The yearly mortality rate also decreased with each subsequent year, from 1.95% in 2005 to 1.55% in 2010 (Figure 2B). The overall number of femoral neck fractures also decreased each year from 2005 to 2010. In 2005, there were 148,686 femoral neck fractures and there were 99,347 in 2010. Women had more femoral neck fractures than men over the study period. A total of 525,864 (70%) femoral neck fractures occurred in women, whereas 225,368 (30%) occurred in men. However, the overall number of deaths was
Copyright © SLACK Incorporated
n Feature Article
only slightly higher in women compared with men, at 6623 (58%) vs 4797 (42%) (Figure 3A). Further, the mortality rate for femoral neck fractures was 1.22% in women and 2.32% in men (Figure 3B).
Discussion Femoral neck fractures carry a substantial risk of mortality within the first year.1 Although the mortality rate within the first year has been clearly defined, no studies have quantified the mortality risk associated with femoral neck fractures in the acute inpatient setting. The goals of this study were to quantify overall inpatient mortality rates for patients with femoral neck fractures and to determine whether a specific demographic variable increased this mortality risk from 2005 to 2010 based on a comprehensive Medicare registry database. The authors hypothesized that the acute in-house mortality rate from femoral neck fractures would be less than 5%, with a higher incidence in women and a higher incidence and a higher mortality rate in patients older than 75 years. The study hypotheses were confirmed. The overall acute inpatient mortality rate from femoral neck fractures was 1.52%, and a higher incidence was seen in women because 70% of femoral neck fractures occurred in women. Also, 80% of femoral neck fractures occurred in patients older than 75 years. The mortality rate increased with age, eventually reaching 2.06% in patients older than 84 years. The current study determined that the overall number of femoral neck fractures increased with patient age, as did the inpatient mortality rate. The mortality rate from femoral neck fractures was 0.65% in patients younger than 65 years and increased to 2.06% in patients older than 84 years. Decreased bone stock and increased likelihood of falling secondary to other causes are some of the reasons why older patients are more likely to sustain these fractures, and an increase in medical comorbidities in older patients is the likely cause of the increase in mortality.8,9 Therefore, it is im-
JULY 2015 | Volume 38 • Number 7
A
B
Figure 1: Total number of acute mortalities as a result of femoral neck fractures by patient age (A). Acute mortality rate as a result of femoral neck fractures by patient age (B).
Table
Mortality Rates for Femoral Neck Fractures in the Medicare Database From 2005 to 2010 Stratified by Patient Age Age, y Outcome
84
Sum
Deaths, No.
210
373
660
1358
2384
6435
11,420
Fractures, No.
32,150
46,524
71,055
116,239
173,070
312,194
751,232
Mortality rate
0.65%
0.80%
0.93%
1.17%
1.38%
2.06%
1.52%
A
B
Figure 2: Total number of acute mortalities as a result of femoral neck fractures by year from 2005 to 2010 (A). Acute mortality rate as a result of femoral neck fractures by year from 2005 to 2010 (B).
portant to counsel these older patients that their risk of death while in the acute hospital setting after a femoral neck fracture is significantly higher than in their younger counterparts. The mortality rate in men was significantly higher than that in women, although women had a significantly higher number of femoral neck fractures. Although 70% of femoral neck fractures occurred in women, the mortality rate was 1.22% in women compared with 2.32% in men (odds ratio, 0.5; 95% confidence interval, 0.25-1.04). Unfortunately, the database did not provide information on patients’ other medical problems. However, given the higher mor-
tality rate in older patients, it is possible that at the time of fracture men were older and more debilitated than women, who likely sustained their fractures at younger ages as a result of bone quality (osteoporosis, osteopenia). Therefore, the younger female patients may have been able to withstand the stress of surgery better than the older male patients, leading to the lower mortality rate in women. The exact cause of death in these patients could not be determined. However, when counseling men with femoral neck fractures, it is important to understand that their risk of death acutely is almost twice as high as that of women with these fractures. Previous studies showed
e595
n Feature Article
A
B
Figure 3: Total number of acute mortalities as a result of femoral neck fractures by sex from 2005 to 2010 (A). Acute mortality rate as a result of femoral neck fractures by sex from 2005 to 2010 (B).
an increase in the 1-year mortality risk for men compared with women, and the current study findings suggested that the acute mortality rate parallels this finding. An unexpected finding of this study was that the absolute number of femoral neck fractures in Medicare patients actually decreased each year from 2005 to 2010, from 118,686 in 2005 to 99,347 in 2010. This finding is intriguing, given the fact that the number of patients older than 75 years is increasing each year, and it would make sense that an increase in the patient population with the majority of femoral neck fractures would mean that the overall number of fractures would increase.1,2 However, this was not the case. The reason for this finding is unclear, but it could be secondary to gains in treatment for osteoporosis, underreporting, or incorrect use of ICD-9 codes for diagnosis. Alternatively, it is possible that patients were more likely to have other types of hip fractures, such as intertrochanteric or subtrochanteric fractures, as the years went on, thereby decreasing the number of femoral neck fractures.
alyzed and the use of the most up-to-date data. However, it also had several limitations. First, the study was retrospective and included only Medicare patients treated over a 6-year period. Next, the cause of death and length of stay for these patients could not be determined. Because these data were not available, the authors cannot comment on ways to decrease the mortality rate. Because the findings of this study depended on accurate reporting of ICD-9 codes, coding errors could affect the data. However, because of the large number of patients, it seems likely that a large error in reporting would be necessary to change the overall result. Finally, the mortality rate was based on all patients who were treated with various modalities for femoral neck fractures. Because numerous methods are used to treat femoral neck fractures, including total hip arthroplasty, hemiarthroplasty, closed reduction and percutaneous pinning, and a dynamic hip screw, these results can be applied broadly to all treatments, but are not specific to a single treatment.
Limitations The current study had several strengths, including the large number of patients an-
Conclusion
e596
In the Medicare population, from 2005 to 2010, the overall mortality rate for
patients who had femoral neck fractures treated at an inpatient center was 1.5%. Men had a mortality rate that was almost twice as high as that in women, and patients older than 84 years were most likely to die soon after sustaining a femoral neck fracture.
References 1. Bhandari M, Devereaux PJ, Tornetta P III, et al. Operative management of displaced femoral neck fractures in elderly patients: an international survey. J Bone Joint Surg Am. 2005; 87(9):2122-2130. 2. Cummings SR, Rubin SM, Black D. The future of hip fractures in the United States: numbers, costs, and potential effects of postmenopausal estrogen. Clin Orthop Relat Res. 1990; (252):163-166. 3. Haentjens P, Autier P, Barette M, Boonen S, Belgian Hip Fracture Study Group. Costs of care after hospital discharge among women with a femoral neck fracture. Clin Orthop Relat Res. 2003; (414):250-258. 4. Wilkins K. Health care consequences of falls for seniors. Health Rep. 1999; 10(4):47-55. 5. Center JR, Nguyen TV, Schneider D, Sambrook PN, Eisman JA. Mortality after all major types of osteoporotic fracture in men and women: an observational study. Lancet. 1999; 353(9156):878-882. 6. Panula J, Pihlajamäki H, Mattila VM, et al. Mortality and cause of death in hip fracture patients aged 65 or older: a population-based study. BMC Musculoskelet Disord. 2011; 12:105. 7. Keller JM, Sciadini MF, Sinclair E, O’Toole RV. Geriatric trauma: demographics, injuries, and mortality. J Orthop Trauma. 2012; 26(9):e161-e165. 8. Yeung PY, Chan W, Woo J. A communitybased Falls Management Exercise Programme (FaME) improves balance, walking speed and reduced fear of falling. Prim Health Care Res Dev. 2015; 16(2):138-146. 9. Peel N, Steinberg M, Williams G. Home safety assessment in the prevention of falls among older people. Aust N Z J Public Health. 2000; 24(5):536-539.
Copyright © SLACK Incorporated
