Factors Associated With Primary Hip Arthroplasty After Hip Fracture

, , ,
The American Journal of Managed Care, March 2013, Volume 19, Issue 3

Patient, clinical, and hospital factors were associated with receiving primary hip arthroplasty versus no surgery after hip fracture in the United States.


To determine patient, clinical, and hospital factors associated with receiving total hip arthroplasty (THA) and hemiarthroplasty (HA) in the United States.

Study Design:

Retrospective, cross-sectional study.


Hospital discharge records with a principal diagnosis of hip fracture and primary hip arthroplasty or no surgery were identified from the 2009 Nationwide Inpatient Sample data set of the Healthcare Cost and Utilization Project. Patient (age, sex, race, income, payer), clinical (comorbidities, severity, fracture type), hospital (region, location, teaching status, bed size, ownership), and outcome (receipt of THA or HA) variables were extracted and weighted for the analyses. Univariate and multivariate analysis were conducted and significance was set at P <.05.


A total of 92,861, 15,489, and 9863 discharges occurred for HA, no surgery, and THA, respectively. Compared with no surgery, THA or HA was significantly more likely in patients who were aged >50 years, white, and female; had >$39,000 income; lived in a medium-metro or noncore county; had comorbidities (anemia, hypertension); and had intracapsular fracture. THA or HA was significantly more likely in urban, privately owned hospitals with >249 beds. Compared with no surgery, THA was significantly more likely in nonteaching hospitals, the Northeast region, and in private insurance or self-pay patients with moderate to severe fractures; HA was more likely in teaching hospitals, in the South and West, and in Medicare patients with minor fractures.


Similarities and differences in patient, clinical, and hospital factors associated with surgical treatments of hip fracture warrant the attention of providers and payers.

Am J Manag Care. 2013;19(3):e74-e84Patient, clinical, and hospital factors are associated with receiving primary hip arthroplasty versus no surgery after hip fracture in the United States.

  • Total hip arthroplasty (THA) or hemiarthroplasty (HA) was significantly more likely if these characteristics were present: age >50 years, white, female, >$39,000 income, mediummetro or noncore patient county residence, comorbidities, intracapsular fracture, and urban, privately owned hospital with >249 beds.

  • THA was significantly more likely in nonteaching hospitals, the Northeast, and private insurance or self-pay patients with moderate-severe illness; HA was more likely in teaching hospitals, the South and West, and Medicare patients with minor illness.

Hip fractures are common, disabling, and expensive. According to the American Academy of Orthopedic Surgeons, hip fractures account for more than 350,000 hospital admissions in the United States each year at an annual cost of more than $5 billion.1,2 The number of hospital admissions is expected to increase to 700,000 by 2050, due to an expected increase in the aging population, life expectancy, and prevalence of osteoporosis.1 The hip fracture rates have been shown to increase exponentially with age, especially in individuals over 50 years of age.3 The risk of hip fractures is about 3 times higher in women, with white women more likely than African American or Asian women to suffer hip fractures.4 Further, the complications related to hip fracture and the long recovery period are associated with a high mortality rate, with an estimated 1 in 4 patients dying within 12 months of hip fracture.3

Hip fractures are serious fall injuries, and treatment options depend on the location and pattern of the fracture, patient characteristics, and the availability of US Food and Drug Administration—approved devices.5 The typical location of fracture includes the femoral neck and the intertrochanteric and subtrochanteric regions. Treatment of displaced, unstable fracture of femoral neck often requires surgical management such as internal fixation and hip arthroplasty.6 Although there is no preferred procedure for the management of hip fractures, recent studies have suggested that arthroplasty has better outcomes than internal fixation for femoral neck fractures.7-16 Earlier, hip arthroplasty was recommended for elderly patients, but it has now been shown to be successful in younger patients as well.17 Hip arthroplasty consists of hemiarthroplasty (HA) and total hip arthroplasty (THA). Hemiarthroplasty is a quick and standardized procedure compared with THA.6 However, THA has shown to have better outcomes than HA and is most commonly used in elderly patients with osteoarthritis and those who are physically active. Further, with advances in surgical techniques and technology, THA-related complications have been significantly reduced. According to the American Academy of Orthopedic Surgeons, about 231,000 THAs are performed each year in the United States.17 After hip fracture, most patients undergo surgical intervention. However, nonsurgical (conservative) management is also used for some patients. Conservative management includes a multimodal approach of medication, activity modification, and physical therapy that comprises bed rest, exercise, and the use of walking aids that help in the healing process by reducing pain and strengthening muscles around the hip joint. In general, the choice of surgical or nonsurgical management of hip fractures depends on clinical factors such as type of hip fracture and the patient’s comorbidities as well as social factors (eg, family, economic status, ethnicity, and location).

The purpose of hip fracture—related surgical treatment is to reduce fracturerelated morbidity and complications, with subsequent improvement in activities of daily living and quality of life.5 A systematic review by the Agency for Healthcare Research and Quality in 2009 suggested that the fracture type was not independently related to patient outcomes such as fracture-related pain, functional status, quality of life, and mortality.5 However, there are no proven guidelines to manage hip fractures. Further, data are limited on the patient, clinical, and hospital factors that affect the decision to provide surgical treatment (arthroplasty) versus no surgical treatment. The present study was undertaken to help understand the factors associated with primary hip arthroplasty in real-world medical practice in the United States. The study objectives were (1) to examine patient, clinical, and hospital factors associated with primary hip arthroplasty (HA or THA) versus no surgical treatment; and (2) to evaluate costs, charges, and length of stay for primary HA, primary THA, and no surgical treatment.

METHODSStudy Design and Data Source

This is a retrospective cross-sectional study of hospital discharge records from the 2009 Nationwide Inpatient Sample (NIS) data. Part of the Healthcare Cost and Utilization Project (HCUP) database, NIS is the largest source of all payer hospital discharge information in the United States. The Nationwide Inpatient Sample is an approximately 20% stratified sample of US community hospitals (excluding veterans and military hospitals) and contains de-identified records of independent hospital discharges selected from 1050 hospitals across 44 states with stratified, single-stage cluster sampling. Each year, about 5 to 8 million hospital stays are recorded by NIS.18 This study was performed under the NIS Data User Agreement and was approved by the Duquesne University Institutional Review Board.

Selection Criteria


The NIS database contains information on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes for primary/secondary diagnosis and procedure codes for each discharge record. Included were the discharge records for individuals aged >18 years with an ICD-9-CM primary diagnosis code for hip fracture (codes 820.00-820.32, 820.8, and 820.9).19 Among hip fracture—related discharges, the discharge records with designated ICD- 9-CM primary procedure codes for HA (81.52) and THA (81.51)19 were compared with the discharge records for hip fracture with no documentation of surgery. Other procedures such as internal fixation with closed/open reduction or without reduction and revision arthroplasty were excluded from the study. The discharge records were then classified based on their procedure types: primary hip arthroplasty (HA or THA) and no surgery (comparison group). The schematic presentation of patient selection is shown in the.

Study Variables

Extracted patient-level variables were age (<50 years, 50- 64 years, 65-79 years, and >80 years), race (white, black, Hispanic, Asian-Pacific Islander, Native American, and other), sex, and primary payer (Medicare, Medicaid, self-pay, private, no charge, and other sources of payment). Patients’ median household income (defined as median household income of the patient’s zip code of residence) was classified into 4 quartiles: $1 to $38,999, $39,000 to $47,999, $48,000 to $62,999, and >$63,000.20 In HCUP a patient’s residing location was based on the National Center of Health Statistics urbanrural designation of the patient’s county of residence and was classified into 6 categories: large central (>1 million population), large fringe (>1 million population), medium metro (250,000-999,999 population), small metro (50,000-249,999 population), micropolitan, and noncore counties.20,21

We also included in the analysis clinical variables such as comorbidities (defined as either present or absent), severity of disease, and type of hip fracture, which predict a decision about primary hip replacement surgery. The severity of illness was assessed using All Patient Refined—Diagnosis Related Groups in the HCUP-NIS data set and was categorized as minor, moderate, major, and extreme. The type of hip fracture was grouped into 3 major categories based on the location of the fracture as indicated by their ICD-9-CM codes. These fracture types were intracapsular (820.00-820.19), extracapsular (820.20-820.32), and other (820.8, 820.9).

Hospital-related variables included geographic region (Northeast, Midwest, West, and South), location (urban and rural), teaching status (teaching and nonteaching), ownership (government, nonfederal; private, nonprofit; and private, investor owned), and hospital bed size (small 1-249 beds, medium 250-449 beds, and large >450 beds).18 The outcome variables were the receipt of either HA or THA versus no surgery. The study also included charges and length of stay attributable to primary hip arthroplasty. Because charges include operating, capital, and other costs, these are usually higher than the true economic costs of medical services. Costto- charge ratios provided in the NIS-HCUP data set were applied to medical service charges to estimate the true costs.

Statistical Analyses

For analyses, we used IBM SPSS Statistics for Windows, version 20 (IBM Corp, Armonk, NY). The differences between the hip arthroplasty (HA or THA) and no-surgery groups were assessed using x2 and independent sample t tests. Multivariate analysis using logistic regression was also conducted, in which all the predictor variables were entered into the model in order to evaluate the association of each predictor variable with the likelihood of undergoing THA or HA while controlling for other variables. Population sampling weights from NIS were applied to all statistical computations to calculate the national estimates for community hospitals across the United States. Significance for all statistical tests was set a priori at P <.05, with all P values being 2 tailed. Missing data were considered to be missing at random, were generally below 5% of the data, and were excluded from the analyses.


A weighted sample of 9863 discharges for THA, 92,861 discharges for HA, and 15,489 discharges for no surgery were identified in the 2009 NIS data set. The mean ages for the patients undergoing THA, HA, and no surgery were 75.62 ± 0.12 years, 81.19 ± 0.03 years, and 80.85 ± 0.10 years, respectively (P <.001).

Table 1

reports several significant factors for THA and HA versus no surgery. Both THA and HA were more likely in whites, patients aged 50 to 79 years (THA) and 50 to 64 years (HA), patients with >$39,000 income, patients residing in the large central and metro counties, patients with comorbidities (rheumatoid arthritis [RA] and chronic blood loss/ anemia), and patients with intracapsular or other fractures. The hospital-related factors (urban location, private/nonprofit ownership, teaching status, and >450 bed size) were more likely to be associated with receiving both THA and HA.

Table 2

presents the multivariate logistic regression analysis results with THA as the outcome variable versus no surgery. Total hip arthroplasty was most likely in patients aged 50 to 64 years (odds ratio [OR] = 2.410; P <.001), followed by patients aged 65 to 79 years (OR = 1.720; P <.001) versus patients younger than 50 years. Women were more likely to undergo THA (OR = 1.147; P <.05). Patients with private insurance (OR = 1.250; P <.001) or self-pay (OR = 1.393; P <.05) versus those with Medicare were significantly more likely to undergo THA; that was also true for patients residing in medium metro (OR = 1.386; P <.001) or noncore (OR = 1.292; P = 0.006) counties versus those residing in large central counties.

Patients with comorbidities including RA (OR = 1.637; P <.001), chronic blood loss/anemia (OR = 5.214; P <.001), hypertension (OR = 1.229; P <.001), and obesity (OR = 1.348; P <.05) were more likely to undergo THA. Compared with patients with minor illness, patients with increasing severity of illness were more likely to undergo THA: moderate (OR = 1.279; P <.001), major (OR = 1.378; P <.001), and extreme (OR = 1.685; P <.001). Having extracapsular fracture (OR = 0.053; P <.001) or other types of fracture (OR = 0.844; P <.001) versus intracapsular fractures decreased the likelihood of undergoing THA.

Regarding hospital characteristics, THA was less common in the Midwest (OR = 0.801; P <.001) and the West (OR = 0.684; P <.001) versus the Northeast. The odds of undergoing THA were higher among hospitals in urban versus rural settings (OR = 2.64; P <.001), and with bed size of 450 or more (OR = 2.536; P <.001) and bed size of 250 to 449 (OR = 2.260; P <.001) versus bed size of fewer than 250. After adjusting for other factors, patients were less likely to undergo THA in hospitals with teaching status (OR = 0.798; P <.001) versus nonteaching status and more likely to undergo THA in hospitals with private/nonprofit status (OR = 1.346; P <.001) or private/investor-owned status (OR = 1.151; P <.05) versus nonfederal government hospitals.

Table 3

presents the multivariate logistic regression analysis results with HA as the outcome variable versus no surgery. Compared with patients younger than 50 years, patients aged 50 to 64 years (OR = 4.263; P <.001), 65 to 79 years (OR = 6.084; P <.001), and 80 years or older (OR = 5.196; P <.001) were more likely to undergo HA. Patients residing in medium metro (OR = 1.121; P = 0.009) and noncore (OR = 1.476; P <.001) counties were more likely to undergo HA than those residing in large central counties.

Patients with comorbidities including chronic blood loss/ anemia (OR = 9.333; P <.001), congestive heart failure (OR = 2.587; P <.001), depression (OR = 1.149; P <.05), uncomplicated diabetes (OR = 2.338; P <.001), diabetes with complications (OR = 2.45; P <.001), hypertension (OR = 1.450; P <.001), neurologic disorders (OR = 1.146; P <.001), and paralysis (OR = 1.289; P <.05) were more likely to undergo HA. Patients with moderate (OR = 0.153; P <.001), major (OR = 0.007; P <.001) or extreme (OR = 0.281; P <.001) severity of illness and those with extracapsular fractures (OR = 0.037; P <.001) or other types of fractures (OR = 0.847; P <.001) were less likely to undergo HA.

Receiving HA was more likely in the South (OR = 1.411; P <.001) and West (OR = 1.146; P <.05) versus the Northeast; in an urban (OR = 2.159; P <.001) versus a rural setting; in a hospital with private ownership (nonprofit: OR = 1.839, P <.001; investor owned: OR = 1.545; P <.001) versus nonfederal government ownership; in a teaching hospital (OR = 1.197; P <.001) versus a nonteaching hospital; and in a hospital with a bed size of 250 to 449 (OR = 2.577; P <.001) or a bed size of 450 or more (OR = 2.927; P <.001) versus a bed size of fewer than 250.

Table 4

reports significant (P <.001) differences in the outcomes associated with arthroplasty (THA and HA) versus no surgery. Patients who underwent THA or HA versus no surgery had higher costs ($21,003 and $18,507 vs $5832), higher charges ($59,876 and $52,610 vs $14,811), and longer length of stay (6 days and 6 days vs 4 days).


Hip fractures occur frequently, especially in older people, and are associated with substantial morbidity and mortality. The decision for surgical versus nonsurgical treatment of hip fracture is multifactorial and involves patient, clinical, and hospital factors. The present study was conducted to understand the factors associated with primary hip arthroplasty in a real-world practice using the 2009 NIS database. Data from NIS can be extrapolated for the entire country and provide valuable information on current practices in the United States. To our knowledge, this is the first study focusing on patient, clinical, and hospital factors associated with receipt of primary hip arthroplasty versus no surgery in patients with all types of hip fractures.

Factors Associated With Receiving Hemiarthroplasty or Total Hip Arthroplasty Versus No Surgery

Patient-Related Factors. Age is an important factor in deciding whether to receive HA or THA. Hemiarthroplasty was more common among patients 50 years and older, whereas THA was seen more in patients aged 50 to 79 years than in patients younger than 50 years. This may be due to the fact that arthroplasty in patients younger than 50 years who are physically active may result in extra stress on the artificial hip, which may require frequent replacements due to the wear and tear.22 Total hip arthroplasty has been shown to be safe and durable in the elderly, but is sometimes associated with perioperative mental confusion and cognitive function deterioration in patients 80 years and older. That may result in a lower likelihood of THA in this patient population.23

Whites were more likely to undergo HA or THA, consistent with the results from the studies by Liu et al24 and Jain et al.20 The lower rates of arthroplasty in non-whites can be attributed to the increased risk of complications and mortality in this group.25 Women were more likely than men to undergo THA because osteoarthritis (a common cause of hip fractures) is a prevalent condition among women,21,26 and treatment requires arthroplasty, especially THA.

Type of payer was also associated with receipt of arthroplasty. For patients undergoing THA, the primary payer was more likely to be private insurance or self-pay versus Medicare. For HA, the primary payer was less likely to be private insurance, self-pay, or other payment source versus Medicare. Arthroplasty was more likely among patients in the higher income groups. One reason could be that higher-income patients are more likely to have access to health insurance that covers these procedures. However, the reasons for this finding are not known and require further research.

Regarding patient location, HA and THA were more likely among patients residing in either medium metro (250,000- 999,999 population) or noncore (<10,000 population) counties versus large central counties (>1 million population). Other studies have shown similar results, in that patients in rural areas were more likely to undergo hip arthroplasty than patients in urban areas.27,28 This suggests that there may be some barriers to accessing surgical treatments for patients in metro areas. Also, our findings might be related to metro areas having higher populations of ethnic minorities who are less likely to seek surgical treatment. However, further analysis of differences in receipt of arthroplasty based on patient location was beyond the scope of our study. This subject needs further investigation.

Clinical Factors. Results in the United Kingdom and Japan have shown that the presence of comorbidities is a significant determinant of post-THA complications.29 Thus, the decision for surgical versus nonsurgical treatment is affected by the presence of comorbidities. Both HA and THA were more likely in the presence of comorbid conditions such as chronic blood loss/anemia and hypertension. Although HA was also more likely in the presence of comorbid conditions such as congestive heart failure, uncomplicated diabetes, neurologic diseases, and paralysis, THA was not associated with the presence of these comorbidities. However, THA was more likely among patients with RA or obesity, conditions that impose limitations in functional ability. Although postsurgical complications are associated with arthroplasty in patients with hypertension, RA, diabetes, and obesity,29,30 our study results showed a higher likelihood of arthroplasty (HA or THA) in patients with these conditions. This may be because of advanced technology, surgical training, and optimal postsurgical management that might contribute to the improved outcomes in patients with comorbid conditions. Although the presence of RA is associated with increased rates of complications and delayed healing, THA is considered a better treatment approach than HA in these patients to relieve associated pain and repair joint damage.31 When severity of illness was assessed using All Patients Refined—Diagnosis Related Groups in the HCUP-NIS data set, varying results were obtained. Rates of THA were higher in patients with increasing severity of illness, and HA rates were higher among patients with minor severity of illness. This finding warrants further research.

In general, most extracapsular fractures are treated with open reduction internal fixation and in rare cases, with arthroplasty. Our study showed similar results: both HA and THA were less likely to be associated with extracapsular fractures than with intracapsular fractures.

Hospital-Related Factors. Regional variations were observed in receipt of arthroplasty in the United States in 2009. Compared with patients in the Northeast, patients in the South and the West were more likely to have HA, whereas THA was less likely in the Midwest and West. Although it was beyond the scope of our study to support our finding with reasons, the regional variations in receipt of THA and HA could be because of lower availability of orthopedic surgeons, different training of practitioners who care for patients with hip fracture, and differences in patient preferences among these regions.

Variations in the likelihood of receiving HA or THA were also observed by hospital location, ownership, bed size, and teaching status. Urban hospitals reported higher rates of both HA and THA than rural hospitals. Unavailability of surgical treatments (THA or HA) and longer travel times for rural patients might be responsible for this difference.32,33 Hospitals with a larger number of beds and with private ownership were also observed to have higher rates of surgery, which may be attributed to better facilities for surgery in these hospitals.34 Our study results showed that HA was more likely to be associated with teaching hospitals, whereas THA was more likely to be associated with nonteaching hospitals. It has been suggested that THA has higher postsurgery complication rates than HA.6 Similarly, it has also been suggested that postsurgery complication rates are higher among teaching hospitals.35 Therefore, the reason for a lower likelihood of THA in teaching hospitals could be the higher postsurgery complication rates in teaching hospitals. This issue needs to be examined in future studies.


Despite being able to generalize results to the US population, this study has some limitations. First, this retrospective database has inherent limitations including data coding/recoding errors, systematic or recorder bias, and confounding factors. Second, because the unit of analysis of the NIS database is hospital discharge and not the patient, overestimates of the true incidence of fractures from the NIS data set were not possible. Third, it was not possible to estimate results at the patient level because multiple hospital discharges could have occurred for hip fracture and its treatments for 1 patient. However, discharge data cannot be linked to a particular patient within this data set due to privacy concerns. Fourth, the information on medication use and clinical variables (degree of displacement of fractures, presence of osteoporosis, extent of osteoarthritis, patient’s preferences for surgery treatment, and prior functional status of patients) might affect the decision regarding surgical treatment. However, because information is not available in the NIS database, we were unable to control for the effect of these variables, which could provide valuable information regarding variations in patients undergoing arthroplasty. Fifth, minority groups might have been underrepresented because the data were collected in broad categories. Finally, this study excluded internal fixation/reduction, another surgical method used commonly for management of hip fractures. Arthroplasty (surgical) management of hip fractures is more expensive than internal fixation and may be associated with serious postsurgical complications.36 Therefore, our study mainly focused on primary hip arthroplasty and compared it with no surgery. Future studies are needed on how patient preferences for arthroplasty influence the outcomes of functional ability, quality of life, mortality, and costs.

Extracapsular and intracapsular fractures are different clinical entities in terms of the decision-making process; intracapsular fractures are more likely than extracapsular fractures to be treated with arthroplasty. Therefore, besides evaluating and reporting the factors associated with primary hip arthroplasty in patients with all types of hip fracture, we also conducted a separate analysis (data not shown) on a population with intracapsular femoral neck fractures. Similar results were obtained for almost all the factors, except for a few differences in comorbidities. In populations with intracapsular fractures only, the presence of uncomplicated diabetes was not significantly associated with THA and presence of depression, uncomplicated diabetes, or paralysis was not significantly associated with HA.

In summary, providers, payers, and policy makers need to consider costs as well as the similarities and differences in patient, clinical, and hospital factors when evaluating the quality of and access to surgical treatments (THA or HA) after hip fractures. Understanding the variations in surgical treatment decisions, particularly by geographic region and insurance payer, may help improve treatment quality and accessibility for patients within the healthcare system.Author Affiliations: From Division of Clinical, Social and Administrative Sciences (IC, KMK), Mylan School of Pharmacy, Pittsburgh, PA; Department of Pharmacy Practice (JS), College of Pharmacy, University of Nebraska Medical Center, Omaha, NE; Department of Foundations and Leadership (GK), Duquesne University, Pittsburgh, PA.

Funding Source: None.

Author Disclosures: The authors (IC, KMK, JS, GK) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (IC, KMK, JS); acquisition of data (IC); analysis and interpretation of data (IC, KMK, JS, GK); drafting of the manuscript (IC, KMK, JS); critical revision of the manuscript for important intellectual content (IC, KMK, JS, GK); statistical analysis (IC, KMK, JS, GK); administrative, technical, or logistic support (KMK, JS); and supervision (KMK, JS).

Address correspondence to: Ishveen Chopra, MS, Graduate Student, Division of Clinical, Social and Administrative Sciences, Mylan School of Pharmacy, 418F Mellon Hall, Duquesne University, 600 Forbes Ave, Pittsburgh, PA 15282. E-mail: choprai@duq.edu.1. Nellans K. Functional outcomes in patients with displaced femoral neck fractures undergoing hemiarthroplasty vs. total hip arthroplasty: a prospective randomized controlled trial. Doris Duke Med Students J. 2005;4:35-40.

2. Auron-Gomez M, Michota F. Medical management of hip fracture. Clin Geriatr Med. 2008;24(4):701-719.

3. American Academy of Orthopaedic Surgeons. Falls and hip fractures. http://orthoinfo.aaos.org/topic.cfm?topic=A00121. Accessed December 10, 2011.

4. Centers for Disease Control and Prevention. Hip fractures among older adults. http://www.cdc.gov/homeandrecreationalsafety/falls/ adulthipfx.html. Last updated September 10, 2010. Accessed December 10, 2011.

5. Butler M, Forte M, Kane RL, et al. Treatment of Common Hip Fractures. Evidence Report/Technology Assessment No. 184 (Prepared by the Minnesota Evidence-based Practice Center under Contract No. HHSA 290 2007 10064 1.) AHRQ Publication No. 09-E013. Rockville, MD. Agency for Healthcare Research and Quality. http://www.ahrq.gov/downloads/pub/evidence/pdf/hipfracture/hipfracture.pdf. Published August 2009. Accessed December 10, 2011.

6. Hopley C, Stengel D, Ekkemkamp A, Wich M. Primary total hip arthroplasty versus hemiarthroplasty for displaced intracapsular hip fractures in older patients: systematic review. BMJ. 2010;340:c2332.

7. Blomfeldt R, Törnkvist H, Ponzer S, Söderqvist A, Tidermark J. Comparison of internal fixation with total hip replacement for displaced femoral neck fractures: randomized, controlled trial performed at four years. J Bone Joint Surg Am. 2005;87(8):1680-1688.

8. Healy WL, Iorio R. Total hip arthroplasty: optimal treatment for displaced femoral neck fractures in elderly patients. Clin Orthop Relat Res. 2004;(429):43-48.

9. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Randomized comparison of reduction and fixation, bipolar hemiarthroplasty, and total hip arthroplasty: treatment of displaced intracapsular hip fractures in healthy older patients. J Bone Joint Surg Am. 2006;88(2): 249-260.

10. Tidermark J, Ponzer S, Svensson O, Söderqvist A, Törnkvist H. Internal fixation compared with total hip replacement for displaced femoral neck fractures in the elderly: a randomised, controlled trial. J Bone Joint Surg Br. 2003;85(3):380-388.

11. Bhandari M, Devereaux PJ, Swiontkowski MF, et al. Internal fixation compared with arthroplasty for displaced fractures of the femoral neck fractures: a meta-analysis. J Bone Joint Surg Am. 2003;85-A(9):1673-1681.

12. Johansson T, Jacobsson SA, Ivarsson I, Knutsson A, Wahlström O. Internal fixation versus total hip arthroplasty in the treatment of displaced femoral neck fractures: a prospective randomized study of 100 hips. Acta Orthop Scand. 2000;71(6):597-602.

13. Jónsson B, Sernbo I, Carlsson A, Fredin H, Johnell O. Social function after cervical hip fracture: a comparison of hook-pins and total hip replacement in 47 patients. Acta Orthop Scand. 1996;67(5):431-434.

14. Puolakka TJ, Laine HJ, Tarvainen T, Aho H. Thompson hemiarthroplasty is superior to Ullevaal screws in treating displaced femoral neck fractures in patients over 75 years: a prospective randomized study with two-year follow-up. Ann Chir Gynaecol. 2001;90(3):225-228.

15. Rogmark C, Carlsson A, Johnell O, Sernbo I. A prospective randomized trial of internal fixation versus arthroplasty for displaced fractures of the neck of the femur: functional outcome for 450 patients at two years. J Bone Joint Surg Br. 2002;84(2):183-188.

16. Söreide O, Mölster A, Raugstad TS. Internal fixation versus primary prosthetic replacement in acute femoral neck fractures: a prospective, randomized clinical study. Br J Surg. 1979;66(1):56-60.

17. National Institute of Arthritis and Musculoskeletal and Skin Diseases. Questions and Answers about hip replacement: what is a hip replacement? http://www.niams.nih.gov/health_info/hip_replacement/. Accessed December 18, 2011.

18. Healthcare Cost and Utilization Project. Introduction to the HCUP nationwide Inpatient Sample (NIS) 2009. http://www.hcup-us.ahrq.gov/db/nation/nis/NIS_Introduction_2009.jsp. Issued May 2011. Accessed December 12, 2011.

19. icd9data.com. International Classification of Diseases, 9th Revision, Clinical Modification. ICD-9-CM diagnosis and procedure codes. http://www.icd9data.com/. Accessed December 10, 2011.

20. Jain NB, Losina E, Ward DM, Harris MB, Katz JN. Trends in surgical management of femoral fractures in the United States. Clin Orthop Relat Res. 2008;466(12):3116-3122.

21. Centers for Disease Control and Prevention. NCHS urban-rural classification scheme for counties. http://www.cdc.gov/nchs/data_access/urban_rural.htm. Accessed December 15, 2011.

22. Medline Plus. Hip joint replacement. http://www.nlm.nih.gov/medlineplus/ency/article/002975.htm. Accessed December 10, 2011.

23. McDonald JE, Huo MH. Total hip replacement: unique challenges in the obese and geriatric population. Curr Opin Orthop. 2008;19:33-36.

24. Liu SS, Della Valle AG, Besculides MC, Gaber LK, Memtsoudis SG. Trends in mortality, complications, and demographics for primary hip arthroplasty in the United States. Int Orthop. 2009;33(3):643-651.

25. Nwachukwu BU, Kenny AD, Losina E, Chibnik LB, Katz JN. Complications for racial and ethnic minority groups after total hip and knee arthroplasty: a review of the literature. J Bone Joint Surg Am. 2010;92(2):338-345.

26. Arden NK, Grifith GO, Hart DJ, Doyle DV, Spector TD. The association between osteoarthritis and osteoporotic fracture: the Chingford Study. Br J Rheumatol. 1996;35(12):1299-1304.

27. Francis ML, Scaife SL, Zahnd WE, Cook EF, Schneeweiss S. Joint replacement surgeries among medicare beneficiaries in rural compared with urban areas. Arthritis Rheum. 2009;60(12):3554-3562.

28. Fields R. Rural residents have more back, hip and knee replacement surgeries than city dwellers. Becker’s ASC Review. http://www.beckersasc.com/orthopedic-spine-driven-ascs/study-rural-residentshave-more-back-hip-and-knee-replacement-surgeries-than-city-dwellers.html. Published May 17, 2011. Accessed March 17, 2012.

29. Imamura K, Black N. Does comorbidity affect the outcome of surgery? total hip replacement in the UK and Japan. Int J Qual Health Care. 1998;10(2):113-123.

30. Jain NB, Guller U, Pietrobon R, Bond TK, Higgins LD. Comorbidities increase complication rates in patients having arthroplasty. Clin Orthop Relat Res. 2005;(435):232-238.

31. Walker KM, Eiff P. Grayzel J. Hip fracture in adults. UpToDate. http://www.uptodate.com/contents/hip-fractures-in-adults. Published 2012. Accessed January 2, 2012.

32. Finlayson SR, Birkmeyer JD, Toteson AN, Nease RF Jr. Patient preferences for localization of care: implications for regionalization. Med Care. 1999;37(2):204-209.

33. Doty B, Zuckerman R, Finlayson S, Jenkins P, Rieb N, Heneghan S. How does degree of rurality impact the provision of surgical services at rural hospitals? J Rural Health. 2008;24(3):306-310.

34. SooHoo NF, Farng E, Zingmond DS. Disparities in the utilization of high-volume hospitals for total hip replacement. J Natl Med Assoc. 2011;103(1):31-35.

35. Rau J. Experts question Medicare’s effort to rate hospitals’ patient safety records. Kaiser Health News. http://www.kaiserhealthnews.org/ Stories/2012/February/13/medicare-hospital-patient-safety-records .aspx. Published February 13, 2012. Accessed March 17, 2012.

36. Johansson T, Bachrach-Lindström M, Aspenberg P, Jonsson D, Wahlström O. The total costs of a displaced femoral neck fracture: comparison of internal fixation and total hip replacement: a randomised study of 146 hips. Int Orthop. 2006;30(1):1-6.