Abstract
Background:
Males have worse outcomes after hip fracture than female counterparts. Cognitive impairment (CI) also increases the risk of poor recovery from hip fracture; however, CI is under-recognized. Patient sex may contribute to this under-recognition through differential misclassification. The objective of this study was to measure under-recognition and differential misclassification of CI by patient sex.
Methods:
A cross-sectional analysis of baseline data from an observational cohort study of community-dwelling hip fracture patients aged 65 and older (n=339; females =171, males =168) recruited from eight hospitals in the greater Baltimore, MD area within 15 days of hospitalization for surgical repair with cognitive testing within 22 days of admission. Indication of Alzheimer’s Disease or related dementias and/or delirium as a post-operative complication in the medical record was considered evidence of documented CI. Observed CI was measured with the Modified Mini Mental State Examination (3MS, ≤78). Source of cognitive impairment identification (SCI) was defined as: ‘3MS Only’, ‘Hospital Record Only’, ‘Both’, ‘No CI’ was compared between males and females using logistic regression.
Results:
Males had more comorbidities and worse physical status upon admission, but otherwise had similar hospital experiences. SCI distribution was 12.7% ‘3MS Only’ (n=42), 11.5% ‘Hospital Record Only’ (n=38), 9.4% ‘Both’ (n=31), and ‘No CI’ (n=219). Males were more likely to be identified with CI using the (‘3MS Only’ and ‘Both’) and females were more likely to have no indication of CI
Conclusion:
There were sex differences in the documentation of CI versus observed impairment. Males had more CI using direct testing. This may be contributing to sex differences in recovery outcomes post hip fracture. Results support the implementation of cognitive testing in hip fracture patients to reduce the impact of differential misclassification by patient sex.
Keywords: hip fracture, sex, cognitive impairment, under-recognized, differential misclassification
INTRODUCTION
Cognitive impairment (CI), including delirium and Alzheimer’s disease and related dementias (ADRD), is severely under-diagnosed in the U.S. It has been estimated that hospital records capture only a third of existing age-related CI.1 Hip fracture patients have a higher prevalence of CI compared to the general population, much of which is likely under-diagnosed.1,2 Hip fracture patients with ADRD are more likely to suffer delirium during hospitalization than hip fracture patients without ADRD or other hospitalized older adults.3 An estimated 20% of hip fracture patients have ADRD prior to their fracture, and 35%−61% have prevalent CI after hip fracture.4,5 Cognitive history and new CI may be missed in medical records.
Our understanding of cognition in hip fracture patients is further complicated by patient sex. Currently, two-thirds of U.S. ADRD cases and two-thirds of hip fracture cases are females, who comprise 80–90% of the hip fracture research.6–9 Males currently represent 25–30% of the hip fracture population, are expected to comprise 51% by 2030, and are more likely to have CI at discharge.10,11 Patient sex is under-explored in U.S. research with regard to potential biases in CI assessment and diagnosis, which are at the clinician discretion. Standardized examination of cognition may identify additional cases.
Given that males are particularly vulnerable to poor outcomes after hip fracture, it is possible that under-recognized CI may impact recovery.12 This study proposed to 1) examine the degree to which CI is under-recognized in hip fracture patients, and 2) determine differences in CI recognition by sex; focusing on differential misclassification, a specific type of bias in which the outcomes are biased but the degree or direction of that bias is not the same across all groups. We hypothesized that cognitive screening of hip fracture patients using the Modified Mini-Mental State Examination (3MS) would detect previously undocumented CI, and that the proportion of documented CI in males is not equal to the proportion in females.
METHODS
Study Design
This study was a retrospective analysis of the Baltimore Hip Studies 7th Cohort (BHS-7), a longitudinal observational cohort study of community-dwelling hip fracture patients recruited from eight hospitals in the greater Baltimore area from 2006–2011. Participants were recruited and consented (self or proxy) by trained research nurses within 15 days of admission for hip fracture. The original study was approved by the institutional review boards at University of Maryland, Baltimore and participating hospitals.
Sample
The BHS-7 sample comprised 339 hip fracture patients (ICD-9 codes 820.00–820.9) age 65 and older. Participants were hospitalized for non-pathological, non-traumatic hip fracture. Patients were excluded if not community-dwelling at the time of fracture, non-English speaking, bedbound for 6 months or more prior to fracture, lived >70 miles from the hospital, weighed >300 pounds, did not have surgical hip repair, or had hardware in the contralateral hip.13 BHS-7 oversampled males by frequency-matching the enrollment of females to males by hospital and time of admission yielding approximately 50% male sample (male=168; female=171). Proxies provided data on the patient’s behalf if participants scored <36 on the 3MS.13 Nine participants were excluded due to missing medical record abstraction information (1 male) or a baseline 3MS (4 females, 4 males). Thus, the analytic sample was 330 (163 males, 167 females).
Measurements
Cognitive Impairment (CI)
Cognitive status is not uniformly assessed across patients within or between hospitals in the U.S. To compare with standard of care documentation, this study used a standardized measure of cognition. Cognitive status was assessed within 22 days post-admission, using the 3MS,1,2 which is scored 0–100. A score of ≤78 indicates CI.14
Medical records were abstracted by a trained research nurse for ADRD diagnosis and delirium at admission and/or as a post-operative complication.
Source of cognitive impairment identification (SCI) was coded as mutually exclusive categories from two distinct sources: 1) medical record abstraction indication of ADRD and/or delirium and 2) 3MS score ≤78. The SCI categories were: ‘3MS Only’, ‘Hospital Record Only’, ‘Both’, ‘No CI’(referent).
Demographics and Clinical Information
Demographics and other clinical information were collected from the medical record or baseline interview. Demographic information included: age, sex, race (white/non-white), and education. Clinical characteristics included body mass index (BMI), site of fracture, type of surgery, length of surgery (minutes), minutes from admission to surgery, American Society of Anesthesiologists Physical Status Rating (ASA; 1–4, 4 is clinically worse); length of stay (days); number of physical therapy sessions, and days from admission to initiation of physical therapy. A modified Charlson Comorbidity Index (CCI) was calculated; dementia was not included in the CCI score because it was a dependent variable.15 These were used as potential covariates, as they have been related to hip fracture outcomes previously.
Analyses
Descriptive analyses of the study sample used Chi-square (χ2, Fisher’s exact tests as appropriate) to assess sex differences within categorical variables, and Student’s t-test for continuous variables.
A χ2 of 3MS and hospital record indication was used to examine the significance of the proportion of the under-recognized group (‘3MS Only’). The SCI group proportions were established via crosstabulation of 3MS and hospital record indication. Continuous sample characteristics were compared across SCI categories using ANOVA with a Tukey correction.
A multinomial logistic regression of the association between sex and nominal SCI was used to assess for differential misclassification. All covariates were included in the adjusted models (using forward selection at p≤0.2). In addition, interaction terms between sex and covariates were included.
RESULTS
Sample Characteristics and SCI
On average, participants were 81.6 years old, had 13.1 years of education, 1.9 comorbid conditions, and a BMI of 25.2. Most participants had an ASA rating of 3 (66.1%), femoral neck fracture (50.3%), and treated using surgical fixation (54.6%). Participants’ average length of surgery was 84.5 minutes, mean hospital stay was 5.3 days, and physical therapy was initiated on average 2.6 days after admission.
The distribution of the SCI categories was ‘3MS Only’ (n=42, 12.7%), ‘Hospital Record Only’ (n=38, 11.5%), ‘Both’ (n=31, 9.4%), or ‘No CI’ (n=219, 66.4%). Table 1 compares sample characteristics by SCI. Those with ‘No CI’ were younger (79.8 years), started physical therapy earlier (2.4 days), and had shorter hospitalization (4.9 days). Participants in ‘Both’ were older (85.6 years), started physical therapy later (3.1 days after admission), and had longer hospital stays (7.1 days). The ‘Hospital Record Only’ group had the most education (14.0 years). Among those with any indication of CI, the ‘3MS Only’ group had fewer years of education (11.1 years), shorter hospital stay (5.2 days), and were more non-white (20.0%).
Table 1.
Characteristics of BHS-7 Analytic Sample by Source of Cognitive Impairment Identification (SCI)
| SCI | |||||
|---|---|---|---|---|---|
| ‘3MS Only’ N=42 |
‘Hospital Record Only’ N=38 |
‘Both’ N=31 |
‘No CI’ N=219 |
P-value | |
| Age (years, 65+) | 81.6 ± 8.2 | 83.3 ± 6.8 | 85.6 ± 5.8 | 79.8 ± 7.8 | <0.01a |
| Education (years, n=319) | 11.1 ± 4.1 | 14.0 ± 3.2 | 12.4 ± 3.6 | 13.4 ± 3.1 | <0.01b, c |
| Charlson Comorbidity Index | 2.0 ± 1.7 | 2.4 ± 1.9 | 2.0 ± 1.4 | 1.7 ± 1.7 | 0.14 |
| Body Mass Index | 25.6 ± 5.3 | 24.8 ± 4.8 | 24.6 ± 5.0 | 25.3 ± 5.0 | 0.81 |
| Length of Surgery (minutes) | 88.8 ± 44.1 | 98.0 ± 45.4 | 73.7 ± 45.4 | 82.8 ± 43.9 | 0.11 |
| Admission to PT (days) | 2.9 ± 1.3 | 2.7 ± 1.2 | 3.2 ± 1.7 | 2.4 ± 1.1 | <0.01d |
| Length of Hospital Stay (days) | 5.3 ± 2.1 | 6.4 ± 2.8 | 7.2 ± 5.2 | 4.8 ± 1.9 | <0.01d |
| Admission to 3MS (days) | 15.4 ± 4.8 | 16.5 ± 4.4 | 16.7 ± 4.1 | 15.5 ± 5.3 | 0.45 |
| Number of PT Sessions | 3.0 ± 1.6 | 3.8 ± 2.1 | 3.7 ± 3.0 | 3.4 ± 1.5 | 0.15 |
| Site of Fracture, n (%) | 0.68 | ||||
| Intertrochanteric | 17 (40.5) | 13 (34.2) | 15 (48.4) | 84 (38.4) | |
| Femoral Neck | 23 (54.8) | 21 (55.3) | 14 (45.2) | 108 (49.3) | |
| Other | 2 (4.8) | 4 (10.5) | 2 (6.5) | 27 (12.3) | |
| Surgery Type, n (%) | 0.34 | ||||
| Fixation | 21 (50.0) | 16 (42.1) | 20 (64.5) | 123 (56.2) | |
| Arthroplasty | 19 (45.2) | 19 (50.0) | 11 (35.5) | 90 (41.1) | |
| Other | 2 (4.8) | 3 (7.9) | 0 (0) | 6 (2.7) | |
| ASA Physical Status Rating | 0.01 | ||||
| 2 | 4 (9.5) | 4 (10.5) | 3 (9.7) | 54 (24.7) | |
| 3 | 30 (71.4) | 24 (63.2) | 21 (67.7) | 143 (65.3) | |
| 4 | 8 (19.0) | 10 (26.3) | 7 (22.6) | 22 (10.0) | |
| Sex | 0.02 | ||||
| Male | 27 (64.3) | 20 (52.6) | 20 (64.5) | 96 (43.8) | |
| Female | 15 (35.7) | 18 (47.4) | 11 (35.5) | 123 (56.2) | |
| Race | <0.01 | ||||
| White | 30 (75.0) | 37 (97.4) | 28 (90.3) | 196 (92.5) | |
| Non-white, mixed | 10 (20.0) | 1 (2.6) | 3 (9.7) | 16 (7.6) | |
Note: Physical Therapy (PT), Modified Mini-Mental State Examination (3MS), p-value is based on F-test and chi-square (Fisher’s exact as needed), Charlson Comorbidity Index −1 if ADRD indicated in Medical Chart, total sample values available in Table 2.; Notation: N (%), Mean ± SD, difference significant at p ≤ 0.05:
‘No CI’ vs ‘Hospital Record Only’,
‘Hospital Record Only’ vs ‘3MS Only’,
‘No CI’ vs ‘3MS Only’,
‘No CI’ vs ‘Both’.
In ASA rating significant difference in proportions of ASA 2 vs ASA 4 (Bonferroni p=0.0027), other comparisons in ASA not significantly different.
Associations of SCI and Patient Sex
SCI differences by sex are shown in Table 2. There were no significant sex differences in SCI for ‘Hospital Record Only’ nor ‘Both’. There were significantly (p<.05) more females with no indication of CI (74% vs 59%), and more males identified as ‘3MS Only’ (17% vs 9% in females). Males had significantly (p<.001) more comorbidities, higher ASA ratings, and started physical therapy later than females (2.8 vs. 2.3 days).
Table 2.
Characteristics of BHS-7 Analytic Sample by Sex, n=330
| Total N=330 |
Male N=163 |
Female N=167 |
p-value | |
|---|---|---|---|---|
| Age (years, 65+) | 81.0 ± 7.8 | 80.6 ± 7.7 | 81.4 ± 7.9 | 0.32 |
| Education (years, n=319) | 13.1 ± 3.4 | 13.2 ± 3.8 | 13.1 ± 3.0 | 0.09 |
| Race (White; n=321), n (%) | 291 (90.7) | 142 (48.8) | 149 (51.2) | 0.41 |
| Charlson Comorbidity Index | 1.9 ± 1.7 | 2.3 ± 1.8 | 1.5 ± 1.5 | <.001 |
| Body Mass Index | 25.2 ± 5.0 | 25.5 ± 4.4 | 24.9 ± 5.6 | 0.27 |
| Length of Surgery (minutes) | 84.5 ± 44.4 | 86.0 ± 44.9 | 82.9 ± 44.1 | 0.53 |
| Admission to Physical Therapy (days) | 2.6 ± 1.2 | 2.8 ± 1.4 | 2.3 ± 1.0 | <.001 |
| Length of Hospital Stay (days) | 5.3 ± 2.6 | 5.5 ± 2.5 | 5.2 ± 2.8 | 0.31 |
| Admission to 3MS (days) | 18.5 ± 49.6 | 15.8 ± 4.9 | 21.1 ± 10.5 | 0.33 |
| Number of PT Sessions | 3.4 ± 1.8 | 3.4 ± 1.6 | 3.5 ± 2.0 | 0.58 |
| Site of Fracture, n (%) | 0.30 | |||
| Intertrochanteric | 129 (39.1) | 69 (42.3) | 60 (35.9) | |
| Femoral Neck | 166 (50.3) | 75 (46.0) | 91 (54.5) | |
| Other | 35 (10.6) | 19 (11.7) | 16 (9.6) | |
| Surgery Type, n (%) | 0.20 | |||
| Fixation | 180 (54.6) | 95 (59.5) | 83 (49.7) | |
| Arthroplasty | 139 (42.1) | 61 (37.4) | 78 (46.7) | |
| Other | 11 (3.3) | 5 (3.1) | 6 (3.6) | |
| ASA Physical Status Rating, n (%) | <.001 | |||
| 2 | 65 (19.7) | 20 (12.3) | 45 (26.9) | |
| 3 | 218 (66.1) | 107 (65.6) | 111 (66.5) | |
| 4 | 47 (14.2) | 36 (22.1) | 11 (6.6) | |
| Baseline 3MS Score (0–100) | 84.2 ± 16.5 | 82.19 ± 16.4 | 86.2 ± 16.4 | 0.03 |
| ADRD in Medical Chart | 46 (13.9) | 28 (17.2) | 18 (10.8) | 0.09 |
| Post-operative Delirium | 30 (9.1) | 16 (9.8) | 14 (8.4) | 0.65 |
| ADRD and Post-operative Delirium | 7 (2.1) | 4 (2.5) | 3 (1.8) | 0.72 |
| Cognitive Impairment Source | 0.02 | |||
| ’3MS Only’ | 42 (12.7) | 27 (16.6) | 15 (9.0) | 0.04 |
| ’Hospital Record Only’ | 38 (11.5) | 20 (12.3) | 18 (10.8) | 0.67 |
| ’Both’ | 31 (9.4) | 20 (12.3) | 11 (6.6) | 0.08 |
| ’No CI’ | 219 (66.3) | 96 (58.9) | 123 (73.7) | 0.004 |
Note: Alzheimer’s Disease and Related Dementia (ADRD), Modified Mini-Mental State Examination (3MS),p-value is χ2*Cochran, Pooled as applicable, Charlson Comorbidity Index −1 if ADRD indicated in Medical Chart; Notation: N (%), Mean ± SD
Relative to females, males had significantly (p=0.03) greater odds of CI identified by ‘Both’ (2.33, 95% CI [1.07, 5.10]) or ‘3MS Only’ (2.31, 95% CI [1.16, 4.58]), but not ‘Hospital Record Only’ (1.42, 95% CI [0.71, 2.84]) as shown in Figure 1. There were no significant interactions between covariates and sex. After adjusting for age, education, and CCI, males still had significantly greater odds of having ‘Both’ indications than females (2.70 95% CI [1.15, 6.34]).
Figure 1. Associations of Male Sex with Source of Cognitive Impairment Identification.
Crude and adjusted odds ratio estimates of SCI (reference ‘No CI’) for Males (reference Females). Adjusted odds are adjusted for age, education, CCI. There are large effect sizes for ‘3MS Only’ and ‘Both’, but not for ‘Hospital Record Only’. After adjustment, ‘3MS Only’ is no longer statistically significant.
Note: Source of Cognitive Impairment Identification (SCI); Charlson Comorbidities Index (CCI); Modified Mini-Mental State Examination (3MS).
DISCUSSION
One-third of participants had an indication of either CI in the hospital chart or by assessment within 22 days of admission. Those with CI by 3MS with no indicated CI in their medical charts accounted for a third of the cognitively impaired cases. This group may represent under-recognition of delirium, ADRD, or development of CI after discharge. Distinctions between those with CI by 3MS and those with medical chart documentation should be explored for effects on outcomes.
Results show that CI is more likely to be under-recognized in males than females. Zhu et al. (2020)16 found that discrepancies between the prevalence of Medicare dementia diagnostic codes and cognitive testing have narrowed over time through 2012 and found no sex differences. Their results may not apply to acute conditions, such as hip fracture, because ascertaining a diagnosis using claims data requires access to information not readily available in most clinical settings.16 The lack of sex difference in CI in this prior study was attributed to reduced rates of test-based prevalence coupled with increased diagnostic code-based ADRD prevalence. In our study, males were more likely to be identified with CI using a cognitive test versus diagnostic codes.
Older adults with CI may require more in-hospital attention and have been shown to require more conscientious effort to benefit from rehabilitation, although they can participate in physical therapy as intense as that provided to non-impaired hip fracture patients.17,18 This translates to a need for extra nursing care during hospitalization and suggests using additional screening for CI to better guide rehabilitation efforts. The under-recognized CI group had shorter hospital stays. Shorter lengths of stay combined with CI may impact recovery outcomes and should be assessed in future work. There should also be a focus on recognition of and screening for CI among the acute care team. Characteristics of ADRD, including severity and domain of impairment, have been shown to be predictive of rehabilitation outcomes.19–21 Rehabilitation should not rely solely on intake diagnosis for developing physical therapy plans, since up to 12% may have under-recognized CI.
SCI and Patient Sex
Odds of CI identification by both 3MS and Hospital Record were still greater for males than females after adjusting for age, education, and CCI. Males were, therefore, more likely to have a history of ADRD coupled with active presentation after hip fracture surgery. Although both sexes had similar medical record documentation for ADRD, males presented with poorer health at baseline. Underlying, yet undiagnosed, CI in males was potentially exacerbated by the hospitalization experience. Males with hip fractures have been found to be more disabled pre-fracture, more impaired after fracture, and have greater one-year mortality after fracture than their female counterparts.11,22 Under-recognition of CI could be contributing to these discrepancies and lead to less focused acute care as shown in the earlier discharge of ‘3MS Only’ participants, which could affect post-acute care recovery.
Strengths and Limitations
While 22.1% of participants had a 3MS≤78 within 3 weeks after fracture, there was no measure of cognitive function pre-fracture or during their hospitalization. The 3MS score is not a diagnosis and cannot differentiate ADRD, MCI, and delirium. Since date of ADRD diagnosis was not available, the duration of pre-existing CI is unknown. The severity of CI was not indicated in the medical chart, nor was it clear how the diagnoses of ADRD and delirium were delineated. BHS-7 characteristics reflect U.S hip fracture studies of community-dwelling older adults.23,24 Balanced recruitment of males and females with hip fracture was a novel research design in this population that allowed for comparison of SCI based on sex. BHS-7 also included patients with poor cognition and these patients are often excluded from research.
Conclusions
This research illustrates that as many as one-third of CI cases are undocumented during hip fracture acute care. Males are more likely to be under-recognized compared to females. Research has shown that hip fracture functional recovery and mortality can be worse in individuals with CI.21,24,25 These results identify males as a vulnerable group within the hip fracture population and support standardized testing for CI in hip fracture patients during acute care period and post-hospitalization observation at orthopedic or rehabilitation care visits. Future research should examine differences in screening for CI by sex and assess its longitudinal impact to guide clinical practice and increase recognition of CI in older patients, particularly older males, in order to develop and implement multi-disciplinary rehabilitation strategies to maximize recovery outcomes.
Key Points.
There is misclassification of CI in the medical record for hip fracture patients.
Male hip fracture patients are more likely to have CI identified when adding direct cognitive testing.
Why Does This Matter?
CI in males is under-recognized, and this may contribute to their poor outcomes after hip fracture.
Funding:
This work was supported by grants from the National Institute on Aging (NIA) (R01 AG029315, R37 AG0990, T32 AG00262, P30 AG028747).
Footnotes
Disclosures
Other Presentations: Gerontological Society of America Annual Scientific Meeting 2018 Poster and is part of a doctoral dissertation.
Other Disclosures: Dr. Magaziner serves on the boards of the Fragility Fracture Network and the Own the Bone Multi-disciplinary Advisory Board of the American Orthopedics Association.
Conflict of Interest: Heather Mutchie, Denise Orwig, Jennifer Albrecht, Yi Huang, John Boscardin, Marc Hochberg, Jay Magaziner, and Ann Gruber-Baldini have no conflicts to report.
Sponsor’s Role: The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
REFERENCES
- 1.Taylor DH, Fillenbaum GG, Ezell ME. The accuracy of Medicare claims data in identifying Alzheimer’s disease. J Clin Epidemiol 2002;55(9):929–937. doi: 10.1016/s0895-4356(02)00452-3 [DOI] [PubMed] [Google Scholar]
- 2.Olofsson B, Persson M, Bellelli G, Morandi A, Gustafson Y, Stenvall M. Development of dementia in patients with femoral neck fracture who experience postoperative delirium—a three-year follow-up study. Int J Geriatr Psychiatry 2018. doi: 10.1002/gps.4832 [DOI] [PubMed]
- 3.Mosk CA, Mus M, Vroemen JP, et al. Dementia and delirium, the outcomes in elderly hip fracture patients. Clin Interv Aging 2017;12:421–430. doi: 10.2147/CIA.S115945 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Seshadri S, Wolf PA, Beiser A, et al. Lifetime risk of dementia and Alzheimer’s disease. Neurology 1997;49(6):1498. doi: 10.1212/WNL.49.6.1498 [DOI] [PubMed] [Google Scholar]
- 5.Tom SE, Hubbard RA, Crane PK, et al. Characterization of dementia and Alzheimer’s disease in an older population: Updated incidence and life expectancy with and without dementia. Am J Public Health 2014;105(2):408–413. doi: 10.2105/AJPH.2014.301935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Alzheimer’s Association. 2017 Alzheimer’s disease facts and figures. Alzheimers Dement J Alzheimers Assoc 2017;13(4):325–373. doi: 10.1016/j.jalz.2017.02.001 [DOI] [Google Scholar]
- 7.Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010–2050) estimated using the 2010 census. Neurology 2013;80(19):1778. doi: 10.1212/WNL.0b013e31828726f5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mielke MM. Sex and gender differences in Alzheimer’s disease dementia. Psychiatr Times 2018;35(11):14–17. [PMC free article] [PubMed] [Google Scholar]
- 9.Haleem S, Lutchman L, Mayahi R, Grice JE, Parker MJ. Mortality following hip fracture: Trends and geographical variations over the last 40 years. Injury 2008;39(10):1157–1163. doi: 10.1016/j.injury.2008.03.022 [DOI] [PubMed] [Google Scholar]
- 10.Stevens JA, Rudd RA. The impact of decreasing U.S. hip fracture rates on future hip fracture estimates. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2013;24(10):2725–2728. doi: 10.1007/s00198-013-2375-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Arinzon Z, Shabat S, Peisakh A, Gepstein R, Berner YN. Gender differences influence the outcome of geriatric rehabilitation following hip fracture. Arch Gerontol Geriatr 2010;50(1):86–91. doi: 10.1016/j.archger.2009.02.004 [DOI] [PubMed] [Google Scholar]
- 12.Oh ES, Sieber FE, Leoutsakos J-M, Inouye SK, Lee HB. Sex Differences in Hip Fracture Surgery: Preoperative Risk Factors for Delirium and Postoperative Outcomes. J Am Geriatr Soc 2016;64(8):1616–1621. doi: 10.1111/jgs.14243 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Orwig D, Hochberg MC, Gruber-Baldini AL, et al. Examining differences in recovery outcomes between male and female hip fracture patients: Design and baseline results of a prospective cohort study from the Baltimore Hip Studies. J Frailty Aging 2018;7(3):162–169. doi: 10.14283/jfa.2018.15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jones TG, Schinka JA, Vanderploeg RD, Small BJ, Graves AB, Mortimer JA. 3MS normative data for the elderly. Arch Clin Neuropsychol 2002;17(2):171–177. doi: 10.1016/S0887-6177(00)00108-6 [DOI] [PubMed] [Google Scholar]
- 15.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40(5):373–383. doi: 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 16.Zhu Y, Chen Y, Crimmins EM, Zissimopoulos JM. Sex, race, and age differences in prevalence of dementia in medicare claims and survey data. J Gerontol B Psychol Sci Soc Sci 2020. [E-pub ahead of print] doi: 10.1093/geronb/gbaa083 [DOI] [PMC free article] [PubMed]
- 17.Thomas K, Christine L. Rehabilitation care after hip fracture in older patients with cognitive impairment: Systematic review. Int J Phys Med Rehabil 2016;04(03). doi: 10.4172/2329-9096.1000336 [DOI] [Google Scholar]
- 18.Haneuse S, Lee KH. Semi-competing risks data analysis: Accounting for death as a competing risk when the outcome of interest is non-terminal. Circ Qual Outcomes 2017;9(3):322–331. doi: 10.1161/CIRCOUTCOMES.115.001841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Albrecht JS, Marcantonio ER, Roffey DM, et al. Stability of Post-Operative Delirium Psychomotor Subtypes Among Hip Fracture Patients. J Am Geriatr Soc 2015;63(5):970–976. doi: 10.1111/jgs.13334 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Beloosesky Y, Grinblat J, Epelboym B, Weiss A, Grosman B, Hendel D. Functional gain of hip fracture patients in different cognitive and functional groups. Clin Rehabil 2002;16(3):321–328. [DOI] [PubMed] [Google Scholar]
- 21.Baker NL, Cook MN, Arrighi HM. Hip fracture risk and subsequent mortality among Alzheimer’s disease patients in the United Kingdom, 1988–2007. Age Ageing 2011;40(Journal Article). doi: 10.1093/ageing/afq146 [DOI] [PubMed] [Google Scholar]
- 22.Doyle DJ, Garmon EH. American Society of Anesthesiologists Classification (ASA Class) StatPearls Publishing; 2019. Accessed August 18, 2019. https://www.ncbi.nlm.nih.gov/books/NBK441940/ [Google Scholar]
- 23.Adeyemi A, Delhougne G. Incidence and Economic Burden of Intertrochanteric Fracture: A Medicare Claims Database Analysis. JBJS Open Access 2019;4(1):e0045. doi: 10.2106/JBJS.OA.18.00045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Abraham DS, Barr E, Ostir GV, et al. Residual disability, mortality, and nursing home placement after hip fracture over 2 decades. Arch Phys Med Rehabil 2019;100(5):874–882. doi: 10.1016/j.apmr.2018.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gruber-Baldini AL, Zimmerman S, Morrison RS, et al. Cognitive impairment in hip fracture patients: timing of detection and longitudinal follow-up. J Am Geriatr Soc 2003;51(9):1227–1236. doi: 10.1046/j.1532-5415.2003.51406.x [DOI] [PubMed] [Google Scholar]