Background: With the rapid growth of elderly patients visiting the Emergency Department (ED), it is expected that there will be even more hospitalisations following ED visits in the future. The aim of this study was to examine the age effect on the performance criteria of the 10-item brief geriatric assessment (BGA) for the prolonged length of hospital stay (LHS) using artificial neural networks (ANNs) analysis. Methods: Based on an observational prospective cohort study, 1117 older patients (i.e., aged ≥ 65 years) ED users were admitted to acute care wards in a University Hospital (France) were recruited. The 10-items of BGA were recorded during the ED visit and prior to discharge to acute care wards. The top third of LHS (i.e., ≥ 13 days) defined the prolonged LHS. Analysis was successively performed on participants categorized in 4 age groups: aged ≥ 70, ≥ 75, ≥ 80 and ≥ 85 years. Performance criteria of 10-item BGA for the prolonged LHS were sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV], likelihood ratios [LR], area under receiver operating characteristic curve [AUROC]). The ANNs analysis method was conducted using the modified multilayer perceptron (MLP). Results: Values of criteria performance were high (sensitivity> 89%, specificity≥ 96%, PPV > 87%, NPV > 96%, LR+ > 22; LR- ≤ 0.1 and AUROC> 93), regardless of the age group. Conclusions: Age effect on the performance criteria of the 10-item BGA for the prediction of prolonged LHS using MLP was minimal with a good balance between criteria, suggesting that this tool may be used as a screening as well as a predictive tool for prolonged LHS. Keywords: Length of hospital stay, Prediction, Elderly Background wards compared to younger ED visitors [1–3]. The high A growing number of older adults (i.e., age 65 and over) morbidity burden and related-disabilities expose older visit the emergency departments (EDs) . In Europe, patients to an increased risk of non-fatal health out- they account for around 20% of all EDs visitors [1, 2]. comes like a long LHS [2, 4, 5]. With the rapid growth These older ED visitors, particularly the oldest group of the oldest segment of ED visitors, hospitalization after (i.e., age 85 and over), generally have a longer length of an ED admission is expected to be even greater in the hospital stay (LHS) after their ED discharge to acute care future and, thus, hospitals need to confront this new challenging issue [1–3, 6]. * Correspondence: firstname.lastname@example.org One way to reduce LHS is early identification of older An abstract of this manuscript has been presented at the 21st IAGG World ED visitors at greater risk of prolonged LHS after an ED Congress of Gerontology as an oral presentation on July 23, 2017. Service of Geriatric Medicine and Geriatric Rehabilitation, Department of discharge to acute care wards [1, 5, 6]. This screening is Medicine, Lausanne University Hospital, Lausanne, Switzerland a crucial step for targeting appropriate interventions to Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Launay et al. BMC Geriatrics (2018) 18:127 Page 2 of 6 prevent or decrease the occurrence of non-fatal health and/or year), presence of acute organ failure plus reason outcomes. The predictive tools designed for this purpose for admission, living situation (home versus institution), should provide a relevant stratification of risk and give and non-use of formal and/or informal home-help ser- information early; ideally before the hospital admission vices. The nature of the acute organ failure for ED visit in order to avoid or plan the admission [4, 7]. The use of was categorized in five groups: cardio-vascular diseases, clinical information collected by a physician has been respiratory diseases, digestive diseases, neuropsychiatric shown to be the best strategy to develop predictive tools of diseases, and other acute diseases (Table 1). Other acute unplanned hospital admissions compared to self-reported diseases referred to a heterogenous groups of diseases in- and administrative data collection [8, 9]Alimitednumber cluding traumatic injuries, hepatic failure, hematological of studies have used tools aimed at identifying older pa- failure and kidney failure. tients at greater risk of prolonged LHS after an ED visit, with low predictive accuracy [2, 3, 5, 6]. Recently, the Outcome measure 10-item Brief Geriatric Assessment (BGA), was reported to The LHS was calculated using the administrative registry of have a high specificity (97%) but a lower sensitivity 63% the University Hospital and corresponded to number of . This study reported the best criteria performance to days between the first day of ED visit and the last day of date. This result was explained in part by the use of hospitalization on an acute care ward. Prolonged LHS was artificial neural networks (ANNs), and in particular the defined as being in the top third of LHS, which corre- modified multilayer perceptron (MLP) . Indeed, ANNs sponded to more than 13 days in the studied sample. The analysis is particularly adapted to predict an inherent com- main issue to identify this threshold value is that there is no plex event like prolonged LHS [11, 12]. The main limit of consensus on the definition of a prolonged length of hos- this previous study was the unbalance between sensitivity pital stay is in geriatric acute care unit. The absence of def- and specificity, which could be related to the high amount inition is due to the fact that a prolonged length of hospital of data required by ANNs [11, 12]. In addition, because the stay depends on an accumulation and complex interplay risk of hospitalization increases with age, it could be between several variables. These variables are related to the suggested that the best balance with greater values of health status of patients but alsotothe environmentwhere performance criteria could be reported specifically in the they are hospitalized (e.g., flux of patients, number of health oldest age group (i.e., age 85 and over) of ED users [1–4]. professionals, type of hospital, organization of care etc.…). The reported study aims to examine the effect of age on Thus, the unique solution to determine this threshold is to thepredictiveabilities(i.e.,sensitivity, specificity, positive use the consensus methods of tertilization [6, 10]. predictive value [PPV], negative predictive value [NPV], likelihood ratios [LR], area under receiver operating char- Table 1 The 10-item Brief Geriatric Assessment acteristic curve [AUROC]) of the 10-item BGA for the pro- Items Yes No longed LHS using MLP in geriatric ED visitors. Age ≥ 85 years Male gender Methods ≥ 5 drugs per day Participants A total of 1117 older patients (i.e., aged ≥ 65 years) were Use psychoactive drugs recruited upon their hospitalization after an ED visit in a History of falls in the past 6 months University Hospital (France) from January 2013 and Temporal disorientation December 2013. This study is an ongoing study which Acute organ failure began in 2011 and its procedure for participant’s recruit- Reason for admission: ment has been previously described in detail [6, 10]. To Cardio-vascular diseases be included, patients had to be hospitalized on acute care wards after an ED visit, age 65 years and over, and Respiratory diseases willingness to participate in research. Patients who died Digestive diseases during hospitalization were excluded. Neuropsychiatric diseases Other acute diseases Assessment Living situation The 10-item BGA was fulfilled upon admission to the ED Home and was composed of the following items: age ≥ 85 years, male gender, polypharmacy defined as ≥5 drugs per day, Institution use psychoactive drugs (i.e., benzodiazepines, antidepres- Non-use of formal and/or informal home-help services sants or neuroleptics), history of falls in previous 6 months, a hypnotics, anti-depressants or neuroleptics temporal disorientation (i.e., inability to give the month unable to give the current year and/or month Launay et al. BMC Geriatrics (2018) 18:127 Page 3 of 6 Standard protocol approvals, registrations, and groups were identified: ≥ 70, ≥ 75, ≥ 80 and ≥ 85 years old. participant consents Performance criteria were sensitivity, specificity, PPV, Patients recruited in this study provided themselves a NPV, LR+, LR- and AUROC. All statistics were performed verbal consent or received help from their trusted per- using R 3.1.0 and Net Beans IDE 8.0. son. The consent to participate was recorded in the patients’ digital files. Ethical Committee of Angers, Results France, approved the entire procedure. There was a trend for a greater mean age (P = 0.0699), a greater prevalence of temporal disorientation (OR = 2.65, Statistical analysis P < 0.001) in participants with prolonged LHS compared Participants were split into two subgroups based on the to those with short LHS (Table 2). In addition, partici- presence or absence of a prolonged LHS. The top third pants with prolonged LHS visited the ED less often for of LHS defined the prolonged LHS (i.e., > 13 days). digestive diseases (OR = 0.48, P = 0.0189) and more often Univariate logistic regression models were used to exam- for other diseases (OR = 1.46, P =0.089) compared to those ine the association between prolonged LHS (dependent with short LHS. The mean LHS was 21.6 ± 8.8 days for variable) and 10-item BGA (independent variables). older ED users with prolonged LHS and 5.2 ± 3.7 days for Artificial neural networks (ANNs) are inspired by ani- those who had no prolonged LHS. Whatever the age group mals’ brain and provide computational processing based considered, predictive performance were high(sensitivity> on machine learning. ANNs are more appropriate to 89%, specificity≥ 96%, PPV > 87%, NPV > 96%, LR+ > 22; examine “chaotic” events, such as prolonged LHS, be- LR- ≤ 0.1 and AUROC> 93). Participants over 75 years cause they are not linear statistical models. These sys- showed the best performance (sensitivity = 89.7%, specifi- tems are interconnected and composed of multiple city = 97.8%, PPV = 93.4, NPV = 96.5, LR + =41.0; LR- = 0.1 layers. Nodes from one layer are connected to all nodes and AUROC = 93.7) (Table 3). in the following layer, but there were no lateral connec- tions within the layer (Fig. 1). The output layer com- Discussion prised one neuron, indicating the presence or absence of The findings show that effect of age was minimal on prolonged LHS.The “neuralnet: Training of neural predictive abilities of the 10-item BGA. These results networks” R package was used for Modified multilayer suggest that analysis provided by ANNs may enable to perceptron (MLP) combining with a specific algorithm use 10-item BGA as a screening tool but also as a pre- (9, 10). To perform ANNs analysis, the sample of partic- dictive tool to identify older patients at higher risk of ipants was randomized in two subgroups (i.e., a training prolonged LHS, whatever their age. group and a testing group). There was no significant The best criteria performances for prolonged LHS were difference between training and testing group (data not shown with patients aged 75 years and over. This is an un- shown). Between-group comparisons were performed expected finding because it was hypothesized that greater using unpaired t-test, Pearson’s Chi-squared test with values of criteria performance could be reported in the Yates’ continuity correction, as appropriate. Four age oldest segment of ED users. This result is discordant with Fig. 1 General structure of modified multilayer perceptron in this study Launay et al. BMC Geriatrics (2018) 18:127 Page 4 of 6 Table 2 Baseline characteristics of participants separated in training and testing groups and univariate logistic regression showing the association between prolonged length of hospital stay (dependant variable) and 10-item Brief Geriatric Assessment (independent variables). (n = 1117) Characteristics Prolonged length hospital stay (i.e., > 13 days) P-Value* No (n = 840) Yes (n = 277) OR [95% CI] Age (years) Mean ± SD 85.14 ± 5.97 85.13 ± 5.62 0.0699 ≥ 70 years 840 (100.0) 276 (99.6) –– 0.5593 ≥ 75 years 834 (99.3) 271 (97.8) 0.325 [0.100;1.074] 0.0898 ≥ 80 years 672 (80.0) 230 (83.0) 0.268 [0.267;0.292] 0.3066 ≥ 85 years 463 (55.1) 150 (54.2) 0.779 [0.779;0.781] 0.8329 Male gender, n (%) 351 (41.8) 114 (41.2) 0.975 [0.738;1.283] 0.9090 Number of drugs daily taken Mean ± SD 6.52 ± 3.22 6.33 ± 3.33 –– 0.6094 ≥ 5, n (%) 601 (71.5) 206 (74.4) 1.152 [0.849;1.577] 0.4055 Use of psychoactive drugs , n (%) 398 (47.4) 139 (50.2) 1.267 [0.965;1.664] 0.1516 History of falls during the past 6 months, n (%) 516 (61.4) 175 (63.2) 1.122 [0.848;1.491] 0.6540 Temporal disorientation , n (%) 259 (30.8) 150 (54.2) 2.652 [2.009;3.510] < 0.0001 Non-use of formal and/or informal home-help services 576 (68.6) 196 (70.8) 1.084 [0.807;1.464] 0.6828 Acute organ failure 503 (59.9) 178 (64.3) 1.204 [0.909;1.600] 0.2207 Living at home 594 (70.7) 198 (71.5) 1.037 [0.770;1.406] 0.8673 Acute organ failure as reason for admission to Emergency Department, n (%) Cardio-vascular diseases, n (%) 92 (11.0) 37 (13.4) 1.256 [0.825;1.880] 0.3282 Respiratory diseases, n (%) 98 (11.7) 23 (8.3) 0.689 [0.418;1.091] 0.1469 Digestive diseases, n (%) 79 (9.4) 13 (4.7) 0.479 [0.250;0.848] 0.0189 Neuropsychiatric diseases, n (%) 121 (14.4) 30 (10.8) 0.724 [0.466;1.095] 0.1593 Other diseases, n (%) 450 (53.6) 174 (62.8) 1.463 [1.108;1.940] 0.0089 P-value significant (i.e., P < 0.05) in bold OR Odds Ratio, CI Confidence Interval, SD standard deviation *Comparison based on unpaired t-test or Pearson’s Chi-squared, as appropriate Use of benzodiazepines or antidepressants or neuroleptics Inability to give the month and/or year Formal (i.e., health and/or social professional) or informal (i.e., family and/or friends) previous studies which reported a strong association mean age = 87.04 years) age and gender explained 21.6% between age and the risk of prolonged LHS [1–4]. Age of area under receiver operating characteristic curve value has previously been identified as an important predictor . In the same way, Campbell et al. reported in a larger for prolonged LHS [10, 13, 14]. For instance, in a similar cohort of patients admitted to ED (1626 patients, mean sized cohort of patients admitted to ED (993 patients, age = 78.7 years) that age over 85 years was strongly Table 3 Performance criteria of 10-item brief geriatric assessment for the prediction of prolonged length hospital stay using artificial neural networks (i.e.; modified multilayer perceptron) based on age categories of participants (n = 1117) Age categories Sensitivity Specificity PPV NPV AUROC Number of individuals classified (%) (%) (%) (%) LR+ LR- TP FP FN TN ≥ 70 years 91.0 96.0 87.7 97.1 22.7 0.1 93.8 242 34 24 816 ≥ 75 years 89.7 97.8 93.4 96.5 41.0 0.1 93.7 253 18 29 805 ≥ 80 years 91.2 96.5 89.6 97.0 25.7 0.1 93.2 206 24 20 652 ≥ 85 years 90.0 96.8 90.0 96.8 27.8 0.1 95.5 135 15 15 448 PPV Positive predictive value, NPV Negative predictive value, LR+ Likelihood ratio of positive test, LR- Likelihood ratio of negative test, AUROC Area under receiver operating characteristic curve, TP True positive, FP False positive, TN True negative, FN False negative Defined as being in the highest tertile of length of hospital stay (i.e., > 13 days) only combinations involving at least 10 participants were considered Launay et al. BMC Geriatrics (2018) 18:127 Page 5 of 6 associated with prolonged LHS (OR = 7.6, P <0.001) . above 95%, which implies that there is a low false positive The association between increased age and prolonged rate when applying the 10-items BGA to predict length of LHS has been explained by incident disabilities that hospitalization. Thus, the combination of both high sensi- exceed 50% in hospitalized patients aged 85 years and over tivity and specificity indicates that the 10-items BGA is [14–16]. This finding is consistent with Sourial et al. who not only a simple screening tool but also a diagnostic tool showed that age and gender had the highest contribution with excellent predictive accuracy. The ability to analyse (C statistic values from 0.51 to 0.67) in predictive accuracy our data with the use of ANNs methods analysis is the of incident disability in a cohort composed of 6657 main explanation for these findings. Indeed, ANNs are patients (mean age = 73.68 years) . data analysis tools developed to overcome the limitations A possible explanation of the discordance about age of traditional linear models as a method to predict health effect on the prediction of prolonged LHS shown in our events . ANNs are computational models which are study compared to previous studies could be related to capable of machine learning and pattern recognition the profile of population recruited, which is oldest old pa- [7–12]. Because they apply non-linear statistics to pat- tients with a mean age around 85 years. Moreover, ANNs tern recognition, ANNs are particularly adapted to provide a different statistical approach that consider the “chaotic” events like prolonged LHS. Nowadays, the complex interplay between all items [11, 12, 18]. Indeed, advances generated by ANNs combined with improve- previous results of ANNs reported that using numerous ment of computer technology affords the opportunity to variables increased predictive accuracy (area under cover explore new perspectives using ANNs as decision-making values lie between 84.1 with 9 items and 90.5 with 10 diagnostic aids for physicians. Thus, these results can be items) but also modified the contribution of demographic applied directly to clinical practice because they can be de- items in the predictive performance (from 12.8% with 9 veloped as software applications for computers and hand items to 21.3% with 10 items) . Categorizing age held devices. The 10-item BGA may provide answers to groups provide an additional variable that limits the ana- facilitate clinical decision-making process because this lysis of ANNs to a single age group and may modify the tool provides a risk stratification of patients at risk of distribution and weight of 10 items in contribution of the non-fatal health outcomes. Such information may be rele- predictive accuracy. Thus, ANNs take into account the vant to make the right decision for the patients like the variations in the contribution of all types of variables discharge to home or to a medical ward, and to continue (demographic items, acute or chronic diseases, and envir- the appropriate interventions in the right patients and at onmental items) to increase predictive performance and the right time by the right professionals (i.e., geriatric to learn to recognize patterns of prolonged LHS in each intervention versus no geriatric intervention). age group. Our results also showed that patients with digestive Our findings underscored that regardless of age, values diseases had a shorter LHS compared to patients ad- of criteria performance were high (sensitivity> 88%, spe- mitted for other diseases. One explanation could be cificity≥ 96%, PPV > 87%, NPV > 96%, LR+ > 22; LR- ≤ 0.1 that admissions for digestive diseases are more often and AUROC> 93). To the best of our knowledge, the semi-urgent or non-urgent . This low degree of current study demonstrates the best performance and urgency may explain a lower LHS. In contrast, “Other balance between criteria reported for prediction of LHS diseases” as a reason for admission was associated after an ED visit. This result is discordant from a recent with an increases LHS. This group refers in part to previous study which reported lower values and an un- traumatic injuries related to falls. Unlike young pa- balance between sensitivity and specificity . The tients, the most common mechanism for traumatic main explanation as suggested in our hypothesis could injuries in older patients is due to fall . Falls have be related to a difference in the number of participants. been identified as major cause of unintentional injury In our study, we included 1117 individuals which likely leading to prolonged LHS and death, especially over increased the accuracy of prediction. It has been shown 80 years, that could explain our results . that ANNs may provide accurate information on an The strengths of this study include the large number event only if there is a sufficient quantity of data points of participants, the prospective cohort design, the hard to be analysed [11, 12]. outcome represented by prolonged LHS, and the use of In order for a screening test to be applicable, it re- sophisticated new statistical models. However, limita- quires high level sensitivity to limit false-negative results. tions need to be considered including recruitment of With sensitivity above 89% in all age group conditions, participants from a single center and the fact that im- the 10-items BGA is a useful screening tool that can be portant items related to prolonged LHS could have been applied to identify early older ED users at higher risk of forgotten. Besides, we included inpatients who died dur- prolonged LHS after their discharge to acute care wards. ing their hospitalization and those discharged in another In addition, our results demonstrated a high specificity hospital. Both date of death or transfer to another Launay et al. BMC Geriatrics (2018) 18:127 Page 6 of 6 hospital were considered as the last day of hospitalization. Received: 4 September 2017 Accepted: 21 May 2018 Thus, a bias might exist because of a suspected higher complexity of those patients. References 1. Samaras N, Chevalley T, Samaras D, Gold G. Older patients in the emergency department: a review. Ann Emerg Med sept. 2010;56(3):261–9. Conclusion 2. Aminzadeh F, Dalziel WB. Older adults in the emergency department: a Age stratification provided minimal effect on the abilities systematic review of patterns of use, adverse outcomes, and effectiveness of the 10-item BGA to predict prolonged LHS using of interventions. Ann Emerg Med mars. 2002;39(3):238–47. 3. Xu KT, Nelson BK, Berk S. The changing profile of patients who used ANNs. Indeed, ANNs provided homogeneous predictive emergency department services in the United States: 1996 to 2005. Ann performance that enable to use 10-item BGA as a Emerg Med déc. 2009;54(6):805–10. e1–7 screening tool but also as a predictive tool to identify 4. Beauchet O, Launay CP, Fantino B, Lerolle N, Maunoury F, Annweiler C. Screening for elderly patients admitted to the emergency department older patients at higher risk of prolonged LHS, whatever requiring specialized geriatric care. J Emerg Med nov. 2013;45(5):739–45. their age. 5. Hastings SN, Schmader KE, Sloane RJ, Weinberger M, Goldberg KC, Oddone EZ. Adverse health outcomes after discharge from the emergency Abbreviations department–incidence and risk factors in a veteran population. J Gen Intern ANNs: Artificial neural networks; AUROC: Area under receiver operating Med. nov 2007;22(11):1527–31. characteristic curve; ED: Emergency department; LHS: Length of hospital stay; 6. Launay CP, de Decker L, Kabeshova A, Annweiler C, Beauchet O. Screening LR: Likelihood ratio; MLP: Modified multilayer perceptron; NPV: Negative for older emergency department inpatients at risk of prolonged hospital predictive value; PPV: Positive predictive value stay: the brief geriatric assessment tool. PLoS One. 2014;9:e110135. 7. García-Pérez L, Linertová R, Lorenzo-Riera A, Vázquez-Díaz JR, Duque- Acknowledgments González B, Sarría-Santamera A. Risk factors for hospital readmissions in We are grateful to the participants for their cooperation. elderly patients: a systematic review. QJM Mon J Assoc Physicians août. 2011;104(8):639–51. Funding 8. Wallace E, Stuart E, Vaughan N, Bennett K, Fahey T, Smith SM. Risk The study was supported by Biomathics, Paris, France. The sponsor had no prediction models to predict emergency hospital admission in community- role in the design and conduct of the study, in the collection, management, dwelling adults: a systematic review. Med Care août. 2014;52(8):751–65. analysis, and interpretation of the data, or in the preparation, review, or 9. Baldonado A, Hawk O, Ormiston T, Nelson D. Transitional care management approval of the manuscript. in the outpatient setting. BMJ Qual Improv Rep. 2017;6(1):u212974. 10. Launay CP, Rivière H, Kabeshova A, Beauchet O. Predicting prolonged Availability of data and materials length of hospital stay in older emergency department users: use of a novel Data and material concerning this article may be provided by the authors analysis method, the artificial neural network. Eur J Intern Med sept. 2015; after justified request. 26(7):478–82. 11. Baxt WG, Skora J. Prospective validation of artificial neural network trained Authors’ contributions to identify acute myocardial infarction. Lancet Lond Engl. 6 janv. 1996; Study concept and design CPL and OB, acquisition of data CPL and OB. 347(8993):12–5. analysis and interpretation of data CPL, AK and AL. Drafting of the 12. Baxt WG Application of artificial neural networks to clinical medicine. manuscript CPL, AK, AL and OB. critical revision of the manuscript for Lancet. 1995;346:1135–8. important intellectual content. JC, EJL statistical expertise. AK administrative, 13. McCusker J, Bellavance F, Cardin S, Trépanier S. Screening for geriatric technical, or material support. OB Study supervision: CPL, OB. All authors problems in the emergency department: reliability and validity. read and approved the final manuscript. Identification of Seniors at Risk (ISAR) Steering Committee. Acad Emerg Med. 1998;5(9):883–93. Ethics approval and consent to participate 14. Campbell SE, Seymour DG, Primrose WR, Lynch JE, Dunstan E, Espallargues The study was conducted in accordance with the ethical standards set forth in M, et al. A multi-Centre European study of factors affecting the discharge the Helsinki Declaration (1983). All participants recruited in this study provided a destination of older people admitted to hospital: analysis of in-hospital data verbal informed consent as the study did not change the usual clinical practice. from the ACMEplus project. Age Ageing sept. 2005;34(5):467–75. The verbal informed consent was obtained from the patients themselves in the 15. De Buyser SL, Petrovic M, Taes YE, Vetrano DL, Corsonello A, Volpato S, et al. presence of a reliable caregiver who was usually a family member assisting in Functional changes during hospital stay in older patients admitted to an decision-making for medical issues. The consent to participate was recorded in acute care ward: a multicenter observational study. PLoS One. 2014;9(5): patients’ digital files. The study and the consent procedure were approved by e96398. the Ethical Committee of Angers, France, N°1,408,714. 16. Covinsky KE, Palmer RM, Fortinsky RH, Counsell SR, Stewart AL, Kresevic D, et al. Loss of independence in activities of daily living in older adults Competing interests hospitalized with medical illnesses: increased vulnerability with age. J Am The authors declare that they have no competing interest. Geriatr Soc avr. 2003;51(4):451–8. 17. Sourial N, Bergman H, Karunananthan S, Wolfson C, Payette H, Gutierrez- Robledo LM, et al. Implementing frailty into clinical practice: a cautionary Publisher’sNote tale. J Gerontol A Biol Sci Med Sci déc. 2013;68(12):1505–11. Springer Nature remains neutral with regard to jurisdictional claims in 18. Akl A., Ghoneim M. Forcasting the clinical outcome: artificial neural published maps and institutional affiliations. networks or multivariate statistical models?, Artificial neural networks - methodological advances and biomedical applications, Prof. Kenji Suzuki Author details (Ed.), 2011;InTech, ISBN: 978–953-. Service of Geriatric Medicine and Geriatric Rehabilitation, Department of 19. Martinez-Fuerte R, Sierra-Martinez L, Sanz-Gonzalez N. Emergencies in Medicine, Lausanne University Hospital, Lausanne, Switzerland. Department primary care for digestive disorders. European Geriatric Medicine. 2015;6S1: of Neuroscience, Division of Geriatric Medicine Angers University Hospital, S32–S156. Angers, France. Department of Medicine, Division of Geriatric Medicine, Sir 20. Aschkenasy MT, Rothenhaus TC. Trauma and falls in the elderly. Emerg Med Mortimer B. Davis - Jewish General Hospital and Lady Davis Institute for Clin North Am. 2006;24(2):413–32. medical research, McGill University, Montreal, QC, Canada. Dr. Joseph 21. Majdan M, Mauritz W. Unintentional fall-related mortality in the elderly: Kaufmann Chair in Geriatric Medicine, Faculty of Medicine, McGill University, comparing patterns in two countries with different demographic structure. Montreal, QC, Canada. Centre of Excellence on Aging and Chronic Diseases BMJ Open. 2015;5(8):e008672. of McGill University Health Network, Montreal, QC, Canada.
BMC Geriatrics – Springer Journals
Published: May 30, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera