Aging Clinical and Experimental Research (2023) 35:2483–2490 https://doi.org/10.1007/s40520-023-02551-3 ORIGINAL ARTICLE Indoor falls and number of previous falls are independent risk factors for long‑term mortality after a hip fracture Montserrat Barceló1 · Jordi Casademont1 · Jordi Mascaró1 · Ignasi Gich2 · Olga Herminia Torres1 Received: 8 June 2023 / Accepted: 25 August 2023 / Published online: 9 September 2023 © The Author(s) 2023 Abstract Background Hip fractures are almost always the result of a fall. Causes and circumstances of falls may differ between frail and vigorous patients. Aim To describe the circumstances of falls causing hip fractures, number of falls during the previous year, and their associa- tion with long-term mortality. Patients and methods The study is a retrospective review conducted in a tertiary university hospital serving a population of 425,000 inhabitants in Barcelona. All patients admitted with hip fractures with medical records describing the circum- stances and number of previous falls were included. The number of falls in the previous 12 months was recorded, including the one causing the fracture. The circumstances of the index fall were dichotomized according to whether it was from the patient’s own height or above; day or night; indoors or outdoors, due to intrinsic or extrinsic causes. Cumulative mortality was recorded for almost 5 years after hip fracture. Results Indoor falls were strongly associated with shorter survival. Falling more than once in the previous year was also a risk factor for long-term mortality (hazard ratio 1.461, p < 0.001 and hazard ratio 1.035, p = 0.008 respectively). Conclusion Indoor falls and falling more than once in the previous year are long-term risk factors for mortality after hip fractures. It is always essential to take a careful patient history on admission to determine the number of falls and their cir- cumstances, and special care should be taken to reduce mortality in patients at high risk. Keywords Mortality · Hip fracture · Falls Introduction suggested as a possible risk factor for mortality [4] although this is difficult to measure once the hip fracture has occurred. It is well known that the short- and long-term mortality fol- Hip fractures are almost always the result of a fall [5]. A lowing a hip fracture is high [1]. The risk factors for mor- study by Norton et al. found that 96% of hip fractures were tality have been extensively studied in the literature. Some, associated with a fall [6]. Falls are also more common in such as advanced age, male sex, and comorbidity, have been frail older adults. In another study, Speechley et al. observed described in virtually all studies [2, 3]. Frailty has also been that the incidence of falling in frail patients in one year of follow-up was 52%, but they also found that falls in vigor- ous older adults resulted in more serious injuries [7]. Some studies have shown that causes and circumstances of falls * Montserrat Barceló differ between frail and vigorous patients. Vigorous patients mbarcelot@santpau.cat were more likely to fall during ambulatory activities, when 1 there was an environmental hazard, and on stairs [7, 8]. This Geriatric Unit, Departament de Medicina, Internal Medicine Department, Hospital de La Santa Creu I Sant Pau, difference in fall patterns has also been observed in falls that Universitat Autònoma de Barcelona, Mas Casanovas Street, result in hip fracture [9, 10]. Given that frail patients have no. 90, 08041 Barcelona, Spain an increased risk of mortality when an adverse event such 2 Department of Clinical Epidemiology and Public Health, as a hip fracture occurs [11], there may be some relationship CIBER Epidemiología Y Salud Pública (CIBERESP), between how the fall that caused the hip fracture occurred HSCSP Sant Pau Biomedical Research Institute (IIB Sant and the risk of mortality. Pau), Barcelona, Spain 13 Vol.:(0123456789) 2484 Aging Clinical and Experimental Research (2023) 35:2483–2490 The aims of our study were to analyze whether the char- such as dizziness, weakness or vertigo, and acute illnesses. acteristics of a fall that caused a fracture and the number of Extrinsic causes were those considered to be environmen- falls in the previous year were related to long-term mortality tal hazards (poor lighting, slippery floors, footwear, home in the first 5 years after a hip fracture. or neighborhood features), or drugs and alcohol use. Some authors regard medication as an intrinsic factor [15]; here, following Phelan et al. [16], it was considered an extrinsic Patients and methods factor, as medication is usually a modifiable environmental factor, and most intrinsic factors are not, or are only partially Setting modifiable. The study was carried out in the Department of Orthopaedic Hip fracture pathway Surgery of a tertiary university hospital serving a population of about 425,000 inhabitants in Barcelona, Spain. Our hospital has an established hip fracture pathway. Con- sequently, during admission, all patients are visited at least Subjects once by a physician from the Geriatrics Unit for a geriatric assessment, in accordance with the co-operation framework The clinical data of all patients admitted to the Orthopae- on hip fractures developed between the two departments. dic Surgery Unit with a diagnosis of hip fracture between All patients (or their relatives) were asked by the attending December 2009 and September 2015 were retrospectively physician about the number of falls in the previous year, and reviewed. Inclusion criteria: the circumstances of the fall the characteristics of the fall that led to admission. In some that led to hospital admission and the number of falls in the patients, it was not possible to collect complete data due to previous year had been recorded in the medical records, and cognitive impairment and lack of witnesses. patients for whom the date of death was known or whether they were still alive at least 5 years after hospital discharge. Mortality after discharge Patients who died during hospital admission were included. Patients with hip fractures due to accidents or neoplasm Retrospectively, we collected the date of death of the patient (bone metastases) were excluded. or whether they survived for at least 5 years after hospi- tal discharge, although in some patients it was possible to Data collection obtain data for up to 10 years after admission. The date of death was obtained from the patient’s medical records and The main risk factors for mortality collected were: age, sex, the National Death Index website of the Ministry of Health comorbidity, measured by the Charlson index [12], the Bar- (https://​www.​mscbs.​gob.​es/​estad​Estud​ios/​estad​istic​as/​estad​ thel Index of Activities of Daily Living (ADL) [13], num- istic​as/​estMi​niste​r io/​IND_​TipoD​ifusi​on.​htm), subject to ber of drugs before admission, place of residence, length of prior authorization. hospital stay, hemoglobin and albumin at admission, inpa- tient rehabilitation center at discharge, and traumatological Statistical analysis and medical complications, including blood transfusion and delirium. Continuous variables were described by calculating means The number of falls in the previous 12 months, including and standard deviations, then compared using the Student’s the one that caused the fracture, was recorded. The circum- t-test with the Welch correction. For categorical variables, stances of the index fall were dichotomized according to absolute numbers, relative frequencies, or proportions were whether it was from the patient’s height or above, during the calculated and compared using Fisher’s exact test. Mul- day or night, indoors or outdoors, and whether it was due tivariate Cox regression analysis (proportional hazards to intrinsic or extrinsic causes. In our environment, some model) was performed to determine independent variables authors define daytime as between 8 a.m. and 8 p.m. [14]. for mortality five years after sustaining a hip fracture. A We considered daytime as between sunrise and sunset, as Kaplan–Meier survival analysis over the follow-up period daylight hours in our country vary considerably between was calculated. Statistical analysis was performed using the winter and summer. Intrinsic causes were those due to age- IBM SPSS Statistical Package (version 26) (SPSS Inc., Chi- related changes, neurological diseases (gait impairment, cago, IL, USA) [17]. Statistical significance was p ≤ 0.05 in Parkinson’s disease, cognitive impairment, polyneuropa- all cases. thy, and stroke sequelae), diabetes, postural hypotension, osteoarthritis, vascular diseases, foot problems, sarcopenia or myopathies, visual or auditory impairment, or symptoms, 13 Aging Clinical and Experimental Research (2023) 35:2483–2490 2485 Table 1  Baseline characteristics, in-hospital results, and circum- Results stances of the fall n = 1349 Between December 2009 and September 2015, 2909 patients were admitted to hospital with hip fracture. Complete data Mean age in years (SD) 83.83 (8.78) of the circumstances of the index fall and number of previ- Female sex n (%) 1050 (77.8) ous falls were obtained for 1364 (47.1%) patients, and 1349 Barthel index (SD) 82 (19.75) (98.9%) of these were located five years after hip fracture. Charlson index (SD) 1.27 (1.42) Cumulative mortality was 266 (19.7%) at one year, and 755 Dementia n (%) 174 (12.9) (56%) at five years. Number of drugs (SD) 6.18 (3.47) Baseline characteristics at admission and in-hospital out- Place of residence, n at home (%) 1159 (85.9) comes are shown in Table 1, highlighting a high mean age, Length of hospital stay, in days (SD) 12.26 (7.23) predominance of female sex, and a high number of medical Hemoglobin at admission, in g/L (SD) 122.27 (16.54) complications, the most frequent being respiratory infections Albumin at admission, in g/L (SD) 29.36 (3.64) (11.3%), heart failure (4.9%), urinary infections (13.6%), Delirium n (%) 203 (15) and acute renal failure (13.6%). Many patients have more Blood transfusion n (%) 657 (48.7) than one complication during admission. The mean num- Other medical complications n (%) 477 (35.4) ber of falls was more than two, most occurred indoors and Traumatological complications n (%) 30 (2.2) were due to intrinsic factors, and only a very small number Number of falls (previous 12 months)* (SD) 2.15 (2.39) occurred during the night or from a height. Characteristics Indoor fall n (%) 950 (70.4) of surviving and non-surviving patients at 5 years are com- Intrinsic factors n (%) 1031 (76.4) pared in Table 2, all were statistically significant except for Daytime fall n (%) 1237 (91.7) traumatological complications and discharge to an inpatient Own height n (%) 1309 (97) rehabilitation center. Non-survivors had more falls in the Discharge to inpatient rehabilitation center n (%) 790 (58.6) previous year, and fell more indoors, due to intrinsic fac- SD standard deviation tors, at night and from their own height. Multivariate Cox *Including index fall regression analysis for independent variables associated with mortality one year and five years after sustaining a hip Table 2  Comparison of Patients = 1349 Exitus n = 755 Survivors n = 594 p survivors and non-survivors at 5 years Mean age in years (SD) 86.40 (7.3) 80.57 (9.41) < 0.001 Female sex n (%) 544 (72.1) 506 (85.2) < 0.001 Barthel index (SD) 76.80 (20.7) 88.57 (16.25) < 0.001 Charlson index (SD) 1.58 (1.56) 0.88 (1.1) < 0.001 Dementia n (%) 124 (16.4) 50 (8.4) < 0.001 Number of drugs (SD) 6.52 (3.58) 5.75 (3.28) < 0.001 Place of residence, n at home (%) 618 (81.9) 541 (91.1) < 0.001 Length of hospital stay in days (SD) 13.03 (7.54) 11.28 (6.7) < 0.001 Hemoglobin at admission in g/L (SD) 119.83 (17.5) 125.38 (14.67) < 0.001 Albumin at admission in g/L (SD) 28.75 (3.62) 30.14 (3.53) < 0.001 Delirium n (%) 158 (20.9) 45 (7.5) < 0.001 Blood transfusion n (%) 413 (54.7) 244 (41.1) < 0.001 Other medical complications n (%) 336 (44.5) 141 (23.7) < 0.001 Traumatological complications n (%) 21 (2.8) 9 (1.5) 0.082 Number of falls (previous 12 months)* (SD) 2.39 (2.76) 1.86 (1.78) < 0.001 Indoor fall n (%) 599 (79.3) 351 (59.1) < 0.001 Intrinsic factors n (%) 618 (81.9) 413 (69.5) < 0.001 Daytime fall n (%) 683 (90.5) 544 (93.3) 0.039 Own height n (%) 744 (98.5) 565 (95.1) < 0.001 Discharge to inpatient rehabilitation center n (%) 450 (59.6) 340 (57.2) 0.206 SD standard deviation *Including index fall 13 2486 Aging Clinical and Experimental Research (2023) 35:2483–2490 fracture is shown in Tables 3 and 4. The risk factors with Discussion the strongest association with short- and long-term mortality were advanced age, male sex, higher comorbidity, delirium In our study, the risk factors for long-term mortality after a and other medical complications during admission, and falls hip fracture were a fall sustained indoors and more than one indoors. A higher number of falls were a risk factor only in fall in the previous year, including the index fall, together the long term. A higher Barthel index, and hemoglobin and with other risk factors for mortality previously described in albumin at admission were also statistically significant pro- the literature. These data can be easily obtained by asking tective factors, although the association was small. Living simple questions that should be asked of all patients admit- at home was a protective factor, but only in the long-term. ted to hospital following a fall; indeed, the recent World The Kaplan–Meier curve shows the estimated probability Guidelines for falls prevention and management for older of survival after hip fracture during a mean follow-up of adults has emphasized, among other things, the importance 4.24 years, as some patients died before 5 years shown in of asking for details of the event and previous falls [18]. Fig. 1a, b, which compares the post-discharge survival over In our patients, we found an association between mor- time between patients who fell indoors versus those who fell tality and a fall sustained indoors, from the height of the outdoors, while Fig. 1c compares long-term survival after patient, at night, and due to intrinsic factors; after multi- discharge between patients who sustained one fall versus variate analysis, only the place of the index fall (indoors), those who sustained two or more falls in the previous year. remained an independent risk factor of mortality. Falling indoors is strongly correlated with lower survival, since the trend continues throughout the entire period, and even ten years after discharge. To our knowledge, the only previously reported study that directly relates the causes and circum- stances of a fall to mortality after hip fracture is a paper Table 3  Risk factors for Hazard ratio 95% confidence interval p mortality at 1 year based on multivariate Cox regression Age 1.038 1.019–1.058 < 0.001 analysis Male sex 2.094 1.595–2.749 < 0.001 Charlson index 1.131 1.045–1.225 0.002 Barthel index 0.992 0.986–0.997 0.004 Hemoglobin at admission 0.990 0.983–0.997 0.006 Albumin at admission 0.941 0.908–0.975 0.001 Delirium 1.546 1.154–2.071 0.004 Other medical complications 1.417 1.097–1.831 0.008 Indoor falls* 1.784 1.256–2.534 0.001 *Including index fall Table 4  Risk factors for Hazard ratio 95% confidence interval p mortality at 5 years based on multivariate Cox regression Age 1.055 1.044–1.067 < 0.001 analysis Male sex 2.051 1.746–2.410 < 0.001 Charlson index 1.165 1.111–1.223 < 0.001 Barthel index 0.992 0.989–0.996 < 0.001 Place of residence at home 0.794 0.661–0.954 0.014 Hemoglobin at admission 0.991 0.987–0.995 < 0.001 Albumin at admission 0.958 0.938–0.978 < 0.001 Delirium 1.386 1.161–1.656 < 0.001 Other medical complications 1.246 1.080–1.437 0.003 Indoor falls* 1.461 1.233–1.732 < 0.001 Number of falls (previous 12 months)* 1.035 1.009–1.062 0.008 *Including index fall 13 Aging Clinical and Experimental Research (2023) 35:2483–2490 2487 Fig. 1  Kaplan–Meier curves for cumulative survival after hip frac- survival in relation to more than one fall in the previous 12 months, ture. a Cumulative survival in relation to all parameters. b Cumula- including the index fall tive survival in relation to an indoor/outdoor index fall. c Cumulative by Burm et al. with an average follow-up of 1.5 years; the We have not found any studies that associated other cir- same authors also found that indoor falls had higher mortal- cumstances of the index fall (falls from the patient’s height ity than outdoor falls [19]. Other studies link indoor falls to or above, during the day or night, and due to intrinsic or poor physical function although they do not directly corre- extrinsic causes) to mortality. We were surprised that falls late with mortality after hip fracture [10, 11]. Some papers due to intrinsic factors were not risk factors for mortality, have pointed out that even if an indoor fall does not cause as those due to extrinsic factors have been related to bet- a fracture, it may signal functional decline in older adults ter levels of physical function than those with physiological [20] and indoor fallers show a significant excess of mortality causes [9]. In our study, the number of patients who fell due [21]. Although indoors are where most falls occur [5], it is to intrinsic factors was very high (more than three-quarters also true that frail older patients spend more hours at home of the index falls) and very few patients fell from a height, than more active ones [22]. such as a ladder. This may be one of the reasons why this Our study also showed that more than one fall in the pre- circumstance was not significant in multivariate analysis. vious year (including the index fall) was a risk factor for Frailty has been associated with increased hospitali- long-term mortality, but not for short-term mortality, and zations, higher rates of falls, hip fractures, admission to was associated with increased mortality over more than institutions and all-cause mortality [25–27]. Assessment 5 years. Unsurprisingly, mortality was higher in recurrent of frailty in a patient admitted for a hip fracture is always fallers, as falls have frequently been linked to greater frailty complicated as we cannot use physical measures such as gait and increased mortality [7, 23, 24]. speed. Some authors have used wrist strength, as muscle 13 2488 Aging Clinical and Experimental Research (2023) 35:2483–2490 weakness is considered a key element in frailty [28, 29], Some studies have described albumin as a risk factor for but there is still no standardized method. Other methods of in-hospital [41] and one-year mortality [42]. A study car- studying frailty such as the FRAIL scale [30] have the dis- ried out in patients with intra-capsular fractures, found that advantage that, when it comes to subjective questions such albumin was an independent risk factor for long-term mor- as fatigue, proxy scores are no replacement for self-reported tality (up to 9 years), and related it to previous poor health, frailty [31]. Speechley et al. found that vigorous patients fell in-hospital complications, as well as a marker of malnutri- more often during activities involving walking about, when tion [43]. However, in many studies, neither hemoglobin nor there were environmental hazards present, and on stairs, and albumin has been described as a risk factor for mortality. that their falls resulted in more serious injuries [7], while This remains a controversial issue [3, 44]. other articles point out that fall patterns in frail and non-frail Postoperative outcome prediction still needs the perfor- patients are different [9, 27]. This suggests that obtaining a mance of traditional predictive statistical models because better picture of the circumstances of the fall could provide current implementations of artificial intelligence do not an indirect measure of the person’s frailty. seem to provide substantial benefit for hip fracture mortal- The rest of the risk factors for mortality found in our ity prediction [45]. study have been described in previous studies and are very Based on this study and others carried out at our hospital, similar worldwide [1, 32]. The factors most associated with efforts have been made to detect fragile patients, to imple- mortality are gender and advanced age. In studies carried ment measures to prevent complications during admission, out in Western countries, mortality in males can be up to such as geriatric and trauma teams with shared responsibili- twice as high as in women [3]. It is not known why mortal- ties, diagnosis of oropharyngeal dysphagia and follow-up by ity is higher in men, since women are at higher risk of falls, speech therapist, an intravenous iron therapy protocol to treat osteoporosis and frailty. Some authors have linked it to the anemia to reduce transfusions, among others. These patients higher frequency of infections in men. A study conducted by also receive advice aimed at preventing further falls. Wehren et al. found that mortality in men two years after hip The main limitation of our study is that it is a retrospec- fracture was twice as likely due a higher incidence of pneu- tive, single center study. Furthermore, it was not always easy monia, influenza, and septicemia [33]. Although in most to clarify the circumstances of the fall, although it should studies, the mean age of men with hip fractures is lower also be pointed out that the data were collected prospec- than that of women, the comorbidity burden among men is tively, according to protocol, by the geriatric physician at the higher and their health status is poorer, and men are more first patient visit, and patients for whom the circumstances of likely to die due to previous chronic diseases [34]. Older the fall were not correctly recorded were excluded. Another age is associated with higher mortality regardless of where limitation linked to the retrospective nature of the study is the study was conducted, although the cut-off point at which that those patients for whom it was not possible to obtain there is an increased risk of death is unknown [34]. A study relevant information about the fall could also be the frail- carried out in Spain found higher in-hospital mortality in est, with cognitive impairment and living in nursing homes. hip fracture patients older than 90 years [35], and a paper The main strength of our study is that it includes a large by Australian authors described similar results [36]. A study number of patients with a long follow-up, and it was pos- in hip fracture patients older than 75 years found an overall sible to locate the vast majority, at least during the first five mortality of 40% at one year, which was significantly higher years after discharge. To our knowledge, there are very few in those older than 85 years. The authors concluded that age studies linking the number and circumstances of falls to was the most important determinant of long-term mortality mortality after hip fracture, and this is the first study in our [37]. environment. Other well-known risk factors confirmed in our study were higher comorbidity (with an increased risk for more than one comorbidity) [3], medical complications during Conclusions admission, including delirium [38], or living in a nursing home [4], as has also been reported previously. Falling indoors and the number of falls in the previous year A higher Barthel index, and hemoglobin and albumin are related to long-term mortality after hip fracture. Answers at admission were all protective factors in our study. Low to simple questions, such as where the patient fell and the hemoglobin at admission has been associated with higher number of falls in the previous year, are useful, not only in-hospital mortality [35], at 30 days, at one year [2, 39], to help prevent future falls, but also to provide an indirect and also in the long-term [40], this author indicates that estimate of frailty and to better understand the patient’s mor- only hemoglobin at admission, and not other controls during tality risk. admission, is an independent predictor, and links anemia to advanced age, type of fracture and previous comorbidities. 13 Aging Clinical and Experimental Research (2023) 35:2483–2490 2489 Author contributions All authors met the criteria for authorship stated Soc 45:1108–1112. https://​doi.​org/​10.​1111/j.​1532-​5415.​1997.​ in the Uniform Requirements for Manuscripts Submitted to Biomedical tb059​75.x Journals. Study concept and design: MB, JC, JM, IG, OT. Acquisition 7. Speechley M, Tinetti M (1991) Falls and injuries in frail and vig- of data: MB. Analysis and interpretation of data: MB, IG, OT. Draft- orous community elderly persons. J Am Geriatr Soc 39:46–52. ing of the manuscript: MB, OT. Critical revision of the manuscript for https://​doi.​org/​10.​1111/j.​1532-​5415.​1991.​tb059​05.x important intellectual content: MB, JC, JM, IG, OT. 8. Berg WP, Alessio HM, Mills EM et al (1997) Circumstances and consequences of falls in independent community-dwelling older Funding Open Access Funding provided by Universitat Autonoma de adults. Age Ageing 26:261–268. https://​doi.​org/​10.​1093/​ageing/​ Barcelona. This research did not receive any funding from agencies in 26.4.​261 the public, commercial, or not-for-profit sectors. 9. Leavy B, Byberg L, Michaëlsson K et al (2015) The fall descrip- tions and health characteristics of older adults with hip fracture: Declarations a mixed methods study. BMC Ger 15:40. https://​doi.​org/​10.​1186/​ s12877-​015-​0036-x Conflict of interest All authors declare no conflict of interest. 10. Aharonoff GB, Dennis MG, Elshinawy A et al (1998) Circum- stances of falls causing hip fractures in the elderly. Clin Orthop Ethical statement The study was approved by the institutional ethics Relat Res 348:10–14 committee and the ethical principles of the Declaration of Helsinki for 11. Lin HS, Watts NJ, Peel NM et al (2016) Frailty and post-operative medical research involving human subjects were followed. outcomes in older surgical patients: a systematic review. BMC Geriatr 16:157. https://​doi.​org/​10.​1186/​s12877-​016-​0329-8 Study approval This study protocol was reviewed and approved by 12. Charlson ME, Pompei P, Ales KL et al (1987) A new method of Ethics Committee of the Hospital de la Santa Creu i Sant Pau, approval classifying prognostic comorbidity in longitudinal studies: devel- number IIBSP-FEM-2020–111. opment and validation. J Chronic Dis 40:373–383 13. Mahoney F, Barthel D (1965) Functional evaluation: the Barthel Index. Md Med J 14:61–65 Informed consent For this type of study, no informed consent is 14. Formiga F, Ruiz D, Lopez-Soto A et al (2006) Circunstancias aso- required. ciada a las caídas responsables de fractura de fèmur en pacientes ancianos Diferencias por grupo de edad y genero. Rev Clin Esp Open Access This article is licensed under a Creative Commons Attri- 206:314–318. https://​doi.​org/​10.​1157/​13099​0478 bution 4.0 International License, which permits use, sharing, adapta- 15. Vassallo M, Poynter L, Sharma JC et al (2008) Fall risk-assess- tion, distribution and reproduction in any medium or format, as long ment tools compared with clinical judgment: an evaluation in a as you give appropriate credit to the original author(s) and the source, rehabilitattion ward. Age Ageing 37:277–281. https://​doi.​org/​10.​ provide a link to the Creative Commons licence, and indicate if changes 1093/​ageing/​afn062 were made. The images or other third party material in this article are 16. Phelan EA, Mahoney JE, Voit JC et al (2015) Assessment and included in the article’s Creative Commons licence, unless indicated management of fall risk in primary care settings. Med Clin Borth otherwise in a credit line to the material. If material is not included in Am 99:281–293. https://​doi.​org/​10.​1016/j.​mcna.​2014.​11.​004 the article’s Creative Commons licence and your intended use is not 17. IBM Corp (2019) IBM SPSS statistics for windows version 26.0. permitted by statutory regulation or exceeds the permitted use, you will IBM Corp, Armonk need to obtain permission directly from the copyright holder. To view a 18. Montero Montero-Odasso M, van der Velde N, Martin FC et al copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. (2022) Task force on global guidelines for falls in older adults world guidelines for falls prevention and management for older adults: a global initiative. Age Ageing 51:205. https://​doi.​org/​10.​ 1093/​ageing/​afac2​05 References 19. Burm SW, Hong N, Lee SH et al (2021) Fall paterns predict mor- tality after hip fracture in older adults, independent of age, sex, and comorbidities. Calcif Tissue Int 109:372–382. https://d​ oi.o​ rg/​ 1. Hu F, Jiang C, Shen J et al (2011) Preoperative predictors for 10.​1007/​s00223-​021-​00846-z mortality following hip fracture surgery: a systematic review and 20. Mänty M, Heinonen A, Viljanen A et al (2009) Outdoor and meta-analysis. Injury 43:676–685. https://d​ oi.o​ rg/1​ 0.1​ 016/j.i​ njury.​ indoor falls as predictors of mobility limitation in older women. 2011.​05.​017 Age Ageing 38:757–761. https://​doi.​org/​10.​1093/​ageing/​afp178 2. Khan MA, Hossain FS, Ahmed I et al (2013) Predictors of early 21. Bath PA, Morgan K (1999) Differential risk factor profiles for mortality alter hip fracture surgery. Int Orthop 37:2119–2124. indoor and outdoor falls in older people living at home in Not- https://​doi.​org/​10.​1007/​s00264-​013-​2068-​1147 tingham. UK Eur J Epidemiol 15:65–73. https://d​ oi.o​ rg/1​ 0.1​ 023/a:​ 3. Liu Y, Wang Z, Xiao W (2017) Risk factors for mortality in 10075​31101​765 elderly patients with hip fractures: a meta-analysis of 18 stud- 22. Molés MP, Lavedán A, Botigué T et al (2020) Characteristics and ies. Aging Clin Exp Res 30:323–330. https://​doi.​org/​10.​1007/​ circumstances of falls in the community-dwelling older adult pop- s40520-​017-​0789-5 ulation. J Prim Care Community Health 11:2150132720940508. 4. Hjelholt TJ, Johnsen S, Brynningsen PK et al (2022) Development https://​doi.​org/​10.​1177/​21501​32720​940508 and validation of a model for predicting mortality in patients with 23. Grisso JA, Kelsey JL, Strom BL et al (1991) Risk factors for falls hip fracture. Age Ageing 51:1–10. https://d​ oi.o​ rg/1​ 0.1​ 093/a​ geing/​ as a cause of hip fracture in women. N Engl J Med 324:1326– afab2​33 1331. https://​doi.​org/​10.​1056/​NEJM1​99105​09324​1905 5. Berry SD, Miller R (2008) Falls: epidemiology, pathophysiol- 24. Donald IP, Bulpitt CJ (1999) The prognosis of falls in elderly ogy and relationship to fracture. Curr Osteoporos Rep 6:149–154. people living at home. Age Ageing 28:121–125. https://​doi.​org/​ https://​doi.​org/​10.​1007/​s11914-​008-​0026-4 10.​1093/​ageing/​28.2.​121 6. Norton R, Campbell AJ, Lee-Joe T et al (1997) Circumstances of 25. Dent E, Martin FC, Bergman H et al (2019) Management of falls resulting in hip fractures among older people. J Am Geriatr frailty: opportunities, challenges, and future directions. Lancet 394:1376–1386. https://​doi.​org/​10.​1016/​S0140-​6736(19)​31785-4 13 2490 Aging Clinical and Experimental Research (2023) 35:2483–2490 26. Ensrud KE, Ewing SK, Taylor BC et al (2007) Frailty and risk of prognostic factors. Int J Surg 24:101–104. https://​doi.​org/​10.​ falls, fracture, and mortality in older women: the study of osteo- 1016/j.​ijsu.​2015.​11.​009 porotic fractures. J Gerontol A Biol Sci Med Sci 62:744–751. 38. Marcantonio ER, Flacker JM, Wright RJ et al (2001) Reducing https://​doi.​org/​10.​1093/​gerona/​62.7.​744 delirium after hip fracture: a randomized trial. J Am Geriatr Soc 27. Zaslavsky O, Zelber-Sagi S, Gray SL et al (2016) A comparison 49:516–522. https://​doi.​org/​10.​1046/j.​1532-​5415.​2001.​49108.x of frailty phenotypes for prediction of mortality, incident falls, and 39. Yombi JC, Putineanu DC, Cornu O et al (2019) Low haemoglobin hip fractures in older women. J Am Geriatr Soc 64:1858–1862. at admission is associated with mortality after hip fractures in https://​doi.​org/​10.​1111/​jgs.​14233 elderly patients. Bone Joint J. 101:1122–1128. https://​doi.​org/​10.​ 28. Beloosesky Y, Weiss A, Manasian M et al (2010) Handgrip 1302/​0301-​620X.​101B9.​BJJ-​2019-​0526.​R1 strength of the elderly after hip fracture repair correlates with 40. Zhang L, Yin P, Lv H et al (2016) Anemia on Admission Is an functional outcome. Disabil Rehabil 32:367–373. https://​doi.o​ rg/​ Independent Predictor of Long-Term Mortality in Hip Fracture 10.​3109/​09638​28090​31684​9997 Population: A Prospective Study With 2-Year Follow-Up. Medi- 29. Selakovic I, Dubljanin-Raspopovic E, Markovic-Denic L et al cine (Baltimore) 95:e2469. https://​doi.​org/​10.​1097/​MD.​00000​ (2019) Can early assessment of hand grip strength in older 00000​002469 hip fracture patients predict functional outcome? PLoS ONE 41. Pimlott BJ, Jones CA, Beaupre LA et al (2011) Prognostic impact 14:e0213223. https://​doi.​org/​10.​1371/​journ​al.​pone.​02132​23 of pre-operative albumin on short-term mortality and complica- 30. Abellan van Kan G, Rolland YM, Morley JE et al (2008) Frailty: tions in patients with hip fracture. Arch Gerontol Geriatr 53:90– toward a clinical definition. J Am Med Dir Assoc 9:71–72. https://​ 94. https://​doi.​org/​10.​1016/j.​archg​er.​2010.​06.​018 doi.​org/​10.​1016/j.​jamda.​2007.​11.​005 42. Kieffer WK, Rennie CS, Gandhe AJ (2013) Preoperative albu- 31. van der Vorst A, Zijlstra GAR, De Witte N et al (2018) Can proxy min as a predictor of one-year mortality in patients with fractured assessments serve as a first screener for identifying people at risk neck of femur. Ann R Coll Surg Engl 95:26–28. https://​doi.​org/​ for multidimensional frailty? Eur Geriatr Med 9:501–507. https://​ 10.​1308/​00358​8413x​13511​60995​4815 doi.​org/​10.​1007/​s41999-​018-​0067-x 43. Thomas BK, Bajada S, Williams RL (2023) Albumin is an inde- 32. Chang W, Lv H, Feng C et al (2018) Preventable risk factors of pendent predictor of up to 9-year mortality for intracapsular femo- mortality after hip fracture surgery: systematic review and meta- ral neck fractures aiding in decision-making for total hip arthro- analysis. Int J Surg 52:320–328. https://​doi.​org/​10.​1016/j.​ijsu.​ plasty or hemiarthroplasty. J Arthroplasty 38:135–140. https://d​ oi.​ 2018.​02.​061 org/​10.​1016/j.​arth.​2022.​08.​015 33. Wehren LE, Hawkes WG, Orwig DL et al (2003) Gender differ- 44. Smith T, Pelpola K, Ball M et al (2014) Pre-operative indicators ences in mortality after hip fracture: the role of infection. J Bone for mortality following hip fracture surgery: a systematic review Miner Res 18:2231–2237. https://​doi.​org/​10.​1359/​jbmr.​2003.​18.​ and meta-analysis. Age Ageing 43:464–471. https://​doi.​org/​10.​ 12.​2231 1093/​ageing/​afu065 34. Sterling RS (2011) Gender and race/ethnicity differences in 45. Lex JR, Di Michele J, Koucheki R et al (2023) Artificial intel- hip fracture incidence, morbidity, mortality, and function. Clin ligence for hip fracture detection and outcome prediction:a sys- Orthop Relat Res 469:1913–1918. https://​d oi.​o rg/​1 0.​1 007/​ tematic review and meta-analysis. JAMA Netw Open 6:e233391. s11999-​010-​1736-3 https://​doi.​org/​10.​1001/​jaman​etwor​kopen.​2023.​3391 35. Sanz-Reig J, Salvador Marín J, Pérez Alba JM et al (2017) Risk factors for in-hospital mortality following hip fracture. Rev Esp Publisher's Note Springer Nature remains neutral with regard to Cir Ortop Traumatol. 61:209–215. https://d​ oi.o​ rg/1​ 0.1​ 016/j.r​ ecot.​ jurisdictional claims in published maps and institutional affiliations. 2017.​03.​003 36. Frost SA, Nguyen ND, Black DA et al (2011) Risk factor for in- hospital mortality. Bone 49:553–558. https://​doi.​org/​10.​1016/j.​ bone.​2011.​06.​002 37. Karademir G, Bilgin Y, Erşen A et al (2015) Hip fractures in patients older than 75 years old: restrospective analysis for 13