Introduktion: Höftfrakturer drabbar många äldre och väntas öka i framtiden. En höftfraktur kan leda till en stor förändring på lång och kort sikt och är en av de största anledningarna till att äldre personer får svårt att utföra dagliga aktiviteter eller behöver flytta in på ett särskilt boende.  Syfte: Att utforska äldre personers upplevelse av att drabbas av en höftfraktur och hur livet påverkats under det efterföljande året.  Metod: Studien är gjord med kvalitativ design. Tio personer (nio kvinnor och en man) i åldern 75–91 år deltog i semi-strukturerade intervjuer i sina hem ett år efter höftfrakturen. Intervjuerna har analyserats genom kvalitativ innehållsanalys enligt Graneheim och Lundman.  Resultat: Ett tema växte fram, Att fortsätta kämpa när livet krackelerar – med självständighet som drivkraft att återfå ett gott liv.Temat belyser deltagarna som med begränsade resurser kämpat för att försöka återfå sina förmågor. Tre kategorier formulerades utifrån nio subkategorier. Kategorierna är: Höftfrakturen medför ett liv med begränsningar, Att få stöd stärker den egna kämpaglöden och Hantera och acceptera en ny tillvaro. Slutsats: Studiens resultat stärker tidigare forskning om att en höftfraktur medför långvariga konsekvenser. Utöver den fysiska skadan påverkas även det psykiska måendet. Stöd från omgivningen är av stor betydelse i efterförloppet och för att stärka motivationen att rehabiliteras.Fortsatt arbete med förebyggande åtgärder och utveckling av vården efter en höftfraktur för de sköra äldre är angeläget, inte minst ur perspektivet jämlik hälsa. / Introduction: Hip fractures affect many older people and are expected to increase in the future. A hip fracture can lead to substantial changes in the long and short term and is an important reason for dependence in daily activities or need to move into a special accommodation. Aim: To explore older peoples experience of suffering from a hip fracture and how life has been affected in subsequent year.   Method: A qualitative design was used. Ten persons (nine women and one man) aged 75-91 years participated in semi-structured interviews one year after the fracture. The interviews were analyzed using qualitative content analysis inspired by Graneheim and Lundman.  Result: One theme emerged, Continuing to fight when life cracks – with independence as a driving force to regain a good life.The theme highlights that the participants, despite limited resources, struggled to regain their abilities. Three categories were formulated from nine subcategories. The categories are: The hip fracture leads to a life of limitations, Getting support strengthens one's own fighting spirit andManage and accept a new existence.  Conclusion: The results further strengthens previous research concluding that hip fractures has long lasting consequences. Apart from the physical aspect, the injury also has an impact on mental health. Social support is of great importance as it increased the participants´ motivation towards rehabilitation. To achieve equal healthcare, the care after a hip fracture needs to be further developed and preventive measures seems urgent for the frail older population.

This dissertation aims to develop statistical methodologies for estimating the effects of time-fixed and time-varying factors in recurrent events modeling context. The research is motivated by the traffic safety research question of evaluating the influence of crash on driving risk and driver behavior. The methodologies developed, however, are general and can be applied to other fields. Four alternative approaches based on various data settings are elaborated and applied to 100-Car Naturalistic Driving Study in the following Chapters. Chapter 1 provides a general introduction and background of each method, with a sketch of 100-Car Naturalistic Driving Study. In Chapter 2, I assessed the impact of crash on driving behavior by comparing the frequency of distraction events in per-defined windows. A count-based approach based on mixed-effect binomial regression models was used. In Chapter 3, I introduced intensity-based recurrent event models by treating number of Safety Critical Incidents and Near Crash over time as a counting process. Recurrent event models fit the natural generation scheme of the data in this study. Four semi-parametric models are explored: Andersen-Gill model, Andersen-Gill model with stratified baseline functions, frailty model, and frailty model with stratified baseline functions. I derived model estimation procedure and and conducted model comparison via simulation and application. The recurrent event models in Chapter 3 are all based on proportional assumption, where effects are constant. However, the change of effects over time is often of primary interest. In Chapter 4, I developed time-varying coefficient model using penalized B-spline function to approximate varying coefficients. Shared frailty terms was used to incorporate correlation within subjects. Inference and statistical test are also provided. Frailty representation was proposed to link time-varying coefficient model with regular frailty model. In Chapter 5, I further extended framework to accommodate multi-type recurrent events with time-varying coefficient. Two types of recurrent-event models were developed. These models incorporate correlation among intensity functions from different type of events by correlated frailty terms. Chapter 6 gives a general review on the contributions of this dissertation and discussion of future research directions. / Ph. D.

Approximately 23% of Canadians over the age of 65 are considered frail, with that number predicted to increase up to 40% for the population over the age of 85. Frailty is a geriatric syndrome defined by the natural decline in muscle mass and function caused by the natural aging process. When developing to an excessive degree, frailty may present as a disease state, which is recognized as sarcopenia. The exact definition of sarcopenia relies on the presence of low muscle mass, strength, and/or function, but quantitative cut-off values are still a topic of debate. Understanding how biomarkers measured via diagnostic imaging such as magnetic resonance imaging (MRI) and dual-energy x-ray absorptiometry (DXA) describe skeletal muscle can allow doctors to develop a profile of sarcopenia and define predictors to aide in preventative therapy. 4 male and 9 female (mean age = 78 ± 6.5 years) participants from a frailty study underwent full-body DXA and had their dominant thigh scanned using a 3.0T MRI. DXA-derived appendicular lean mass (ALM) and MRI-derived cross-sectional area (CSA), fat fraction (FF), T2 relaxation (T2), magnetization transfer ratio (MTR), fractional anisotropy (FA) and mean diffusivity (MD) from 4 muscle groups at the mid-thigh were defined as muscle biomarkers. Pearson's correlation was calculated to identify relationships between biomarkers and a Wilcoxon rank-sum test was performed to assess the agreement of low-muscle mass characterization between ALM normalized by height (ALMI), ALM normalized by BMI (ALM/BMI), and the gold standard MRI cross-sectional area. Strong positive correlations between muscle quantity biomarkers such as ALMI and CSA were recognized within the quadriceps (p=0.0095), adductors (p=0.035), and sartorius (p=0.00065) muscles while muscle quality biomarkers such as FF and T2 showed significant positive correlation within the quadriceps (p=3.58*10^-5) and the hamstring (p=0.0042) muscles. Finally, ALM/BMI displayed a much stronger agreement in muscle mass quantification with the gold-standard of MRI-CSA over the more commonly researched ALMI from DXA. The main purpose of this study was to collate a vast array of skeletal muscle biomarkers obtained using DXA and MRI on a frail population, and show that significant correlations can be recognized from a single MRI-slice located at the mid-thigh. Additionally, this study recognized the potential of ALM/BMI as the DXA-derived biomarker of choice in muscle mass assessment of frailty. / Thesis / Master of Applied Science (MASc) / A person is diagnosed with sarcopenia when they present symptoms of low muscle mass, strength, and/or function. Defining these three criteria with objective measures has been long debated by researchers and clinicians alike. By understanding how different properties, or biomarkers, of skeletal muscle relate to one another and change as a person becomes more frail, we hope to better understand sarcopenia and identify the best measures to classify someone as sarcopenic. Being able to accurately diagnose someone as sarcopenic early allows for more effective treatment of this muscle disease. In this study, non-invasive magnetic resonance imaging (MRI) and dual energy x-ray absorptiometry (DXA) were used to measure many different biomarkers of skeletal muscle at the mid-thigh. Through characterizing these measures of muscle quality and quantity between different imaging techniques this study aimed to recognize which imaging techniques, and more specifically biomarkers, can best distinguish between a person who is sarcopenic and one who is non-sarcopenic.

Bakgrund:  Antalet äldre med skörhet ökar i samband med stigande ålder. Muskelsvaghet, ofrivillig viktnedgång, känsla av trötthet och infektionskänslighet är hälsoproblem som uppstår vid biologiskt åldrande. Till följd av dessa symtom upplever ofta multisjuka äldre svårigheter med att hantera vardagen. Det är sjuksköterskans ansvar att utgå från holistiskt perspektiv samt tillämpa vetenskapliga kunskaper för självständig vardag. Syfte: Att beskriva vilka aspekter sköra äldre i behov av omvårdnadsinsatser upplever som betydelsefulla för en bra vardag. Metod: Arbetet var av systematisk litteraturstudie och genomförd enligt tematisk syntetisk design. Resultat: Studiens resultat är grundat i tolv artiklar med kvalitativ design för att besvara studiens syfte. Dataanalysen resulterade i tre huvudteman; psykisk påverkan i vardagen, fysisk påverkan i vardagen och vården och anhörigas påverkan. Äldre upplevde att deras sociala liv påverkades av fysiska begräsningar, hjälpmedelsbehov och andelen kommunala insatser. Att få klara av sin vardag självständigt var de äldres främsta önskan. / Background: The number of elderly people with frailty is growing, with increasing age. Muscle weakness, involuntary weight loss, a feeling of exhaustion, and susceptibility to infection are health problems that occur at a rising age. As a result of these symptoms, the elderly experience difficulties coping with everyday life. To achieve independence for the elderly, it is the nurse’s responsibility to have a holistic perspective and scientific knowledge. Purpose: Describing the important aspect of a good everyday life for frail elderly individuals in need of nursing interventions. Method: The method of this study has been systematic literature review and is implemented according to the thematic synthetic design. Result: The results of the study are based on twelve articles with a qualitative design to answer the purpose. The analysis results in three main categories: the psychological impact in everyday life, the physical impact in everyday life, and lastly, health care and the role of impact. The results also show that elderly people felt that their social lives were affected by physical limitations, the need for assistive devices, and the proportion of municipal interventions. To be able to manage and cope with their everyday lives independently was their main desire.

Survival analytic methods help to analyze lifetime data arising from medical and reliability experiments. The popular proportional hazards model, proposed by Cox (1972), is widely used in survival analysis to study the effect of risk factors on lifetimes. An important assumption in regression type analysis is that all relative risk factors should be included in the model. However, not all relative risk factors are observed due to measurement difficulty, inaccessibility, cost considerations, and so on. These unobservable risk factors can be modelled by the so-called frailty model, originally introduced by Vaupel et al. (1979). Furthermore, the frailty model is also applicable to clustered data. Cluster data possesses the feature that observations within the same cluster share similar conditions and environment, which are sometimes difficult to observe. For example, patients from the same family share similar genetics, and patients treated in the same hospital share the same group of profes- sionals and same environmental conditions. These factors are indeed hard to quantify or measure. In addition, this type of similarity introduces correlation among subjects within clusters. In this thesis, a semi-parametric frailty model is proposed to address aforementioned issues. The baseline hazards function is approximated by a piecewise constant function and the frailty distribution is assumed to be a Birnbaum-Saunders distribution. Due to the advancement in modern medical sciences, many diseases are curable, which in turn leads to the need of incorporating cure proportion in the survival model. The frailty model discussed here is further extended to a mixture cure rate frailty model by integrating the frailty model into the mixture cure rate model proposed originally by Boag (1949) and Berkson and Gage (1952). By linking the covariates to the cure proportion through logistic/logit link function and linking observable covariates and unobservable covariates to the lifetime of the uncured population through the frailty model, we obtain a flexible model to study the effect of risk factors on lifetimes. The mixture cure frailty model can be reduced to a mixture cure model if the effect of frailty term is negligible (i.e., the variance of the frailty distribution is close to 0). On the other hand, it also reduces to the usual frailty model if the cure proportion is 0. Therefore, we can use a likelihood ratio test to test whether the reduced model is adequate to model the given data. We assume the baseline hazard to be that of Weibull distribution since Weibull distribution possesses increasing, constant or decreasing hazard rate, and the frailty distribution to be Birnbaum-Saunders distribution. D ́ıaz-Garc ́ıa and Leiva-Sa ́nchez (2005) proposed a new family of life distributions, called generalized Birnbaum-Saunders distribution, including Birnbaum-Saunders distribution as a special case. It allows for various degrees of kurtosis and skewness, and also permits unimodality as well as bimodality. Therefore, integration of a generalized Birnbaum-Saunders distribution as the frailty distribution in the mixture cure frailty model results in a very flexible model. For this general model, parameter estimation is carried out using a marginal likelihood approach. One of the difficulties in the parameter estimation is that the likelihood function is intractable. The current technology in computation enables us to develop a numerical method through Monte Carlo simulation, and in this approach, the likelihood function is approximated by the Monte Carlo method and the maximum likelihood estimates and standard errors of the model parameters are then obtained numerically by maximizing this approximate likelihood function. An EM algorithm is also developed for the Birnbaum-Saunders mixture cure frailty model. The performance of this estimation method is then assessed by simulation studies for each proposed model. Model discriminations is also performed between the Birnbaum-Saunders frailty model and the generalized Birnbaum-Saunders mixture cure frailty model. Some illustrative real life examples are presented to illustrate the models and inferential methods developed here. / Thesis / Doctor of Science (PhD)

Nous nous sommes intéressés à l’analyse et à la mise à jour d’une typologie de l’aide reçue par les personnes âgées de 65 ans et plus vivant à domicile. Cette étude secondaire s’est basée sur les données recueuillies dans deux milieux francophones, Hochelaga-Maisonneuve (HM) et Moncton (MCT). La collecte de données avait été faite par l’entremise d’un questionnaire administré par entrevue face à face. Les deux objectifs, de cette thèse sont : 1) Établir une typologie des réseaux d’aide, résultant de la combinaison des sources d’aide et des tâches accomplies ; 2) Identifier les principaux déterminants d’appartenance aux réseaux. La typologie obtenue met en relation les ressources, formelles ou informelles, utilisées par les personnes âgées et l’aide instrumentale reçue. La capacité ou l’incapacité à effectuer neuf activités de la vie quotidienne et huit de la vie domestique ont servi à évaluer l’aide reçue. Six ressources formelles et dix informelles ont été examinées selon qu’elles étaient les 1ères, 2ièmes ou 3ièmes sources d’aide utilisées par les personnes âgées. L’approche privilégiée s’est inspirée de celle des réseaux sociaux et du modèle de Pescosolido. C’est l’influence des caractéristiques sociodémographiques des personnes âgées, de leurs états de santé, de leurs habitudes de vie sur leurs réseaux qui nous ont intéressés. Les résultats sont présentés à chaque fois pour nos deux milieux séparément. Nous commençons par un descriptif des sources d’aide utilisées et des aides reçues. Puis les profils des sources d’aide utilisées et des activités accomplies sont exposés pour l’ensemble des personnes âgées. Ces profils servent de base pour obtenir notre typologie. Elle comprend cinq catégories. Ces catégories sont toutes composées de personnes âgées faisant appel à de l’aide formelle, informelle ou mixte pour accomplir des tâches uniques ou multiples. La première catégorie « Transitoire », comprend 39% (HM) et 46% (MCT) des personnes âgées qui débute un processus d’incapacité. Elles font appel à des ressources informelles pour accomplir une tâche unique. La deuxième catégorie « Personnes âgées seules » en rassemble 14% (HM et MCT), majoritairement des femmes, avec peu d’incapacités. Ces dernières utilisent de l’aide formelle pour une tâche unique. La troisième catégorie « Familiale » regroupe 12% (HM et MCT) des personnes âgées bien entourées qui ont plusieurs incapacités. Ces gens font appel à des sources d’aide informelles pour réaliser des tâches multiples. La quatrième catégorie « Très fragile » rassemble 30% (HM) et 25% (MCT) des personnes âgées peu entourées ayant beaucoup d’incapacités. Elles utilisent des ressources d’aide mixtes pour effectuer des tâches multiples. La cinquième catégorie « Pré institutionnel » comprend 4% (HM et MCT) des personnes âgées qui ont le plus d’incapacités et qui sont seules. Ces gens font appel à de l’aide formelle pour des tâches multiples. Les déterminants d’appartenance à ces catégories proviennent des blocs sociodémographiques, état de santé et réseaux sociaux de notre modèle théorique. Une des contributions importantes de cette thèse a été de pouvoir identifier cinq catégories bien distinctes composant une typologie de l’aide reçue, indépendamment du milieu, par des personnes âgées vivant à domicile. MOTS CLÉS : Typologie, réseaux sociaux, personnes âgées, services de soins, formels, informels, aides reçues, sources d’aide, incapacités, déterminants d’appartenance, fragilité / We have been interested in the analysis and the elaboration of a typology concerned by elderlies’ support, 65 years and older living at home. This secondary study was based on data collected in two different francophone cities, Hochelaga-Maisonneuve (HM) and Moncton (MCT). Interviews “face to face” were conducted. The objectives were: 1) Establish a typology of supports networks putting in relation the support resources and the activities done by these resources, 2) Identify the belonging criteria for these networks. The typology puts into relation the resources, formal or informal, utilised by the elderlies and the instrumental aid received. The fact that elderlies were able or not able to perform, at least for one of the nine daily activities or of the eight domestic ones, was used to evaluate the support received. Six formal and ten informal resources utilized by the elderlies were looked at in terms of 1st, 2nd or 3rd resources. Social network approach and Pescosolido’s model were used. Sociodemographic characteristics, health status and way of life of the elderlies’ networks, were the main parameters in which we were concerned. The results are presented for the two environments separately. We start by describing the aid resources used and their different tasks done for the elderlies. Than, we show the different aid resources’ profiles associated to the activities’ profiles for the two populations. These profiles are the basis of our typology. It is composed of five categories. They refer to formal, informal or mixed help in order to do single or multiple activities. The first category “Transient” has 39% (HM) and 46% (MCT) of elderlies who start to have significant incapacity. They use informal support for a unique task. The second category “Singles” correspond to 14% of HM or MCT. They are mostly women with little incapacity. They use formal help for a unique task. The third category “Family” regroup 12% of HM or MCT with more incapacities. They are well supported by their family members. They use informal help for multiple activities. The fourth category “Fragile” has 30% (HM) and 25% (MCT) of elderlies with lots of incapacities. They have less family members to help them and they use mixed (informal – formal) support to be able to do multiple tasks. The fifth category “Pre-Institutional” is composed of 4% (HM or MCT) elderlies how are the most handicapped. They are by themselves and they use formal support to do multiple tasks. One of the most important goals of this thesis was to be able to construct a typology of the aid received by elderlies, living at home, in two different environments and to define five very distinct categories in relation with the type of resources. Key words: Typology, social networks, elderly, support, help, formal, informal, health services, frailty, incapacity

Réalisé en cotutelle avec l'Université de Paris-Sud / La thèse a pour objectif d’étudier les rapports entre les besoins typiquement hétérogènes des personnes âgées vivant dans la communauté et l’utilisation conséquente des services médicaux et sociaux. Inspirée du concept de la fragilité, l’approche a d’abord consisté à modéliser l’hétérogénéité des besoins en classant les individus – par l’analyse de classes latentes (ACL) – dans des profils-types, chacun représentant une constellation particulière de problèmes de santé. Appliquée aux données recueillies dans le cadre du projet de démonstration du SIPA (Système de services intégrés pour personnes âgées en perte d’autonomie), l’ACL a révélé 4 profils qui se distinguent sur le plan qualitatif par les dimensions physique et cognitive, et sur le plan quantitatif par la gravité des incapacités que les individus manifestent (Article 1). L’analyse des transitions entre les profils sur 1 an et 22 mois montre des changements différenciés selon les profils et le sexe, et confirme la stabilité de la classification et sa sensibilité aux changements d’état de santé. En outre, les profils sont robustes aux effets de la mortalité et de l’attrition due aux sorties d’étude. Les rapports entre les profils, les variables sociodémographiques et les indicateurs de résultats tel l’hébergement et la mortalité appuient la validité de la classification. Les profils ont ensuite servi à modéliser les besoins dans des modèles économétriques ajustés pour étudier l’utilisation et les coûts par profil d’état de santé pour une gamme de services financés par le système public (Article 2). Ces analyses montrent que les profils permettent d’identifier des configurations distinctes d’utilisation et de coûts pour des sous-groupes représentatifs de la population âgée fragile. La sensibilité des profils aux différences d’accès et aux différences d’intensité d’utilisation révèle les choix de ces groupes cibles par les autorités locales dans la prestation des services institutionnels et des services de proximité. Enfin, une analyse fine de ces relations a été menée pour la couteuse période précédant le décès dans le but d’établir si l’intensification des coûts en fin de vie concerne les personnes âgées de manière homogène quelque soit la gravité de leur état de santé (Article 3). Les analyses comparatives des survivants et des décédés montrent une tendance inverse dans les coûts des services selon l’état de santé. Les personnes qui survivent dans des états de santé compromis coûtent significativement plus que leurs homologues relativement sains. En revanche, c’est chez les individus relativement sains, indépendamment de l’âge, qu’on retrouve les coûts de fin de vie les plus élevés – liés principalement aux hospitalisations de courte durée. Combiné aux résultats qui montrent que les coûts sont significativement moindres chez les sujets de plus de 85 ans, ce travail confirme l’hypothèse d’une compression des coûts attribuables à la morbidité chez les décédés les plus âgés. L’originalité de cette thèse tient au fait que très peu de travaux ont tenté de modéliser l’hétérogénéité de l’état de santé dans le but d’étudier ses conséquences sur l’accès, l’intensité et les coûts des services sociaux et médicaux pour notre population de référence. En outre, alors que la pertinence des soins de fin de vie et le maintien à domicile sont au cœur des préoccupations, ce travail est le premier à décrire les rapports entre l’état de santé et les coûts en fin de vie pour les personnes âgées fragiles vivant dans la communauté. / Abstract This thesis aims to study the relationship between the typically heterogeneous needs of community-living elderly and their consequent utilization of health and social services. Inspired by the concept of frailty, the approach consisted in modeling the heterogeneity – by latent class analysis (LCA) – to group individuals into homogenous categories of health status, each representing a constellation of health problems. Applied to the data collected for the demonstration project of the system of integrated services for frail elderly (SIPA), LCA revealed four health state profiles that distinguish the physical and cognitive dimensions of health and capture severity along the disability dimension (Article 1). Transition analyses over 1 year and 22 months showed differentiated and gender-specific patterns of transition probabilities, confirming the sensitivity of the profiles to change in health status. The profiles are stable over time and robust to mortality and lost to follow-up attrition. Relationships between the profiles, sociodemographic characteristics and distal outcomes, such as mortality and institutionalization, confirm the classification’s validity. These profiles were then used in two-part econometric models to study access and costs of several measures of publicly funded services (Article 2). Our results show the profiles are able to identify distinct configurations of service utilization and costs in substantially meaningful subgroups of the frail elderly population. What is important here is that the health state profiles are sensitive to differences, and changes, in available patterns of care in a specific milieu. These differences reveal choices of target groups by local authorities in delivering institutional and community-based services. Finally, a fine analysis of these differences was performed for the costly period preceding death to establish whether the high end-of-life costs affect all elderly homogenously irrespective of the severity of their health state (Article 3). A comparative analysis revealed that survivors’ and decedents’ costs of care trends according to health status go in opposite directions. Severely disabled survivors cost significantly more compared to their relatively healthy counterparts. In contrast, the highest end-of-life costs – driven by acute hospitalizations – belong to the relatively healthy, independent of age. Among survivors and among decedents, the oldest old (85+) cost significantly less compared to younger age groups. Combined, these results support the notion of a compression of costs due to morbidity in the oldest old decedents. The originality of this thesis rests on the fact that few very few studies attempted to model heterogeneity in health status with the goal of estimating its effect on patterns of service utilization. Moreover, in light of the shift towards community-based care as a response to economic pressures and population aging, our work is the first to describe the relationship between health status and end-of-life care in frail community-living elderly.

Selected methods of mortality analysis focused on adults and the oldest age-groups Abstract Questions about human life span, longevity and mortality in general are natural to almost everyone. This Doctoral Thesis deals with one central question - whether some limit of human life span or of its improvements exists. It is rather a methodological work, therefore its aim is to introduce not only relevant theories but above all the methods usable in the mortality analysis focused on adults or the oldest-old. At the beginning the most important theories and opinions of scientist dealing with mortality are introduced. In the first half of the analytical part mainly the traditional and basic approaches are included. The theme of life tables is opened by an analysis of its construction in the Czech Republic, together with its possible modifications. As a result the independent macro code for the SAS software is attached in the electronic Appendix. This macro enables to calculate the unknown parameters of selected mortality laws by the method of weighted non-linear least squares and to produce the smoothed and extrapolated values of mortality rates. Using the individual life durations, life tables according to education attainment were constructed (also attached in the electronic Appendix). In the second half of the...

INTRODUÇÃO: A avaliação do desempenho físico pode ser realizada de maneira rápida e segura em idosos. A escala Short Physical Performance Battery (SPPB) é um instrumento prático que apresenta ótimos resultados na identificação precoce de idosos da comunidade em risco para desfechos adversos. Porém, ainda não foi estudada em idosos com condições agudas não internados. OBJETIVOS: Verificar o poder preditivo da escala SPPB em um ano para sobrevida, hospitalização, perda funcional e quedas, em idosos com condições agudas atendidos em Hospital Dia. E comparar a escala SPPB com a velocidade de marcha isolada. MÉTODOS: Estudo de coorte prospectivo com 512 idosos com doenças agudas ou crônicas descompensadas, capazes de deambular no plano (média de 79,5 ± 8,4 anos de idade; 63% mulheres), admitidos em um Hospital Dia Geriátrico (HDG) em São Paulo, Brasil. O desempenho físico foi realizado na admissão. Baixo risco foi definido como SPPB >= 9 pontos, médio risco entre 5 a 8 pontos e alto risco <= 4 pontos. Houve seguimento de um ano por contato telefônico mensal para aferição dos desfechos. Modelos de risco proporcionais de Cox foram calculados para cada desfecho após ajuste para fatores sociodemográficos (idade, sexo e raça), índice de comorbidades de Charlson, atividades de vida diária, cognição e sintomas depressivos. Foi analisado ainda, se a adição do teste de sentar e levantar e do teste do equilíbrio à velocidade de marcha melhoram a predição dos desfechos. RESULTADOS: Comparado aos de baixo risco, após ajuste para variáveis sociodemográficas e clínicas, participantes com médio e alto risco apresentaram maior incidência e risco de morte (18% vs. 6%, Hazard Ratio (HR)=2,5; 95%CI 1,2-5,5; 21% vs. 13%, HR=2,7; 95%CI 1,2-6,2, respectivamente), hospitalização (43% vs. 29%, HR=1,5; 95%CI 1,1-2.3; 44% vs. 29%, HR=1,7; 95%CI 1,1-2,7, respectivamente) e perda funcional (32% vs. 13%, HR=2,3; 95%CI 1,4-3,8; 58% vs. 13%, HR=3,9; 95%CI 2,2-6,6, respectivamente) em um ano de seguimento. Não houve associação da escala SPPB com o desfecho quedas. A adição do teste de sentar e levantar e do teste do equilíbrio à velocidade de marcha elhoraram a discriminação dos modelos de predição para morte (Harrell\'s C=0,67 vs 0,62; P=0,04), hospitalização (Harrell\'s C=0,60 vs 0,57; P=0,04) e perda funcional (Harrell\'s C=0,73 vs 0,70; P=0,01). CONCLUSÕES: a escala SPPB se mostrou um ótimo instrumento para o rastreamento de idosos com condições agudas sob risco de eventos adversos. A combinação dos três testes da SPPB teve uma melhor predição que a velocidade de marcha isolada / Background: Physical performance can be assessed quickly and safely in older adults. The Short Physical Performance Battery (SPPB) scale is a validated practical tool to identify community-dwelling older adults at high risk for adverse outcomes. However, has not been evaluated for outpatients in acute care conditions. Objective: To evaluate the value of the Short Physical Performance Battery (SPPB) in predicting 1-year survival, hospitalization, new dependence in basic activities of daily living (ADL) and falls among acutely ill older patients in a geriatric day hospital in São Paulo, Brazil. In addition, to estimate the contribution of the full SPPB over the gait speed alone in predicting the adverse outcomes. Methods: A prospective cohort study involving 512 with an acute medical problem or exacerbation of chronic disease, able to walk (mean age 79,5 ± 8,4 years; 63% female), consecutively admitted to a geriatric day hospital in São Paulo, Brazil. Physical performance was evaluated at admission. Low risk group was defined as SPPB >= 9 points, median-risk group as SPPB between 5 and 8 points, and high risk group as <= 4 SPPB points. We conducted 1-year followup by monthly phone contacts to assess the outcomes. Nested Cox proportional hazards models were calculated for each outcome after adjusting for sociodemographic factors (age, sex and race), Charlson comorbidity index, ADL, cognition and depressive symptoms. In addition, we estimated whether the chair-stands test and balance test improve the ability of gait speed to discriminate those at risk of outcomes. Results: Compared to patients with high risk, after adjusting for demographic and clinical variables, those with medium and low risk in SPPB presented a higher risk of 1-year death (18% vs. 6%, Hazard Ratio (HR)=2,5; 95%CI 1,2-5,5; 21% vs. 13%, HR=2,7; 95%CI 1,2-6,2, respectively), hospitalization (43% vs. 29%, HR=1,5; 95%CI 1,1-2.3; 44% vs. 29%, HR=1,7; 95%CI 1,1-2,7, respectively) and ADL dependence (32% vs. 13%, HR=2,3; 95%CI 1,4-3,8; 58% vs. 13%, HR=3,9; 95%CI 2,2-6,6, respectively). SPPB was not association with falls. The addition of chair-stands and balance tests to gait speed improved the discrimination of models to predict death (Harrell\'s C=0.67 vs 0.62; P=0.04), hospitalization (Harrell\'s C=0.60 vs 0.57; P=0.04) and new ADL dependence (Harrell\'s C=0.73 vs 0.70; P=0.01). Conclusions: The SPPB is as a powerful tool for identifying acutely ill older outpatients at high-risk of adverse outcomes. The combination of the three components of the SPPB resulted in better predictive performance than gait speed alone

摘要 本文主要考慮群集樣本(clustered samples)的存活分析，而每一群集中又分為兩種組別(groups)。假定同群集同組別內的個體共享相同但不可觀測的隨機脆弱性(frailty)，因此面臨的是雙變量脆弱性變數的多變量存活資料。首先，驗證雙變量脆弱性對雙變量對數存活時間及雙變量存活時間之相關係數所造成的影響。接著，假定雙變量脆弱性服從雙變量對數常態分配，條件存活時間模式為韋伯迴歸模式，我們利用EM法則，推導出雙變量脆弱性之多變量存活模式中母數的估計方法。 關鍵詞：雙變量脆弱性，Weibull迴歸模式，對數常態分配，EM法則 / Abstract Consider survival analysis for clustered samples, where each cluster contains two groups. Assume that individuals within the same cluster and the same group share a common but unobservable random frailty. Hence, the focus of this work is on bivariate frailty model in analysis of multivariate survival data. First, we derive expressions for the correlation between the two survival times to show how the bivariate frailty affects these correlation coefficients. Then, the bivariate log-normal distribution is used to model the bivariate frailty. We modified EM algorithm to estimate the parameters for the Weibull regression model with bivariate log-normal frailty. Key words：bivariate frailty, Weibull regression model, log-normal distribution, EM algorithm.