The Hygiene Hypothesis and the risk of Crohn’s disease : a case-control study utilizing prospectively-collected exposure data from an administrative database

La maladie de Crohn (MC) pédiatrique a des conséquences majeures sur la qualité de vie des patients atteints (troubles de croissance, absentéisme scolaire, etc). L’étiologie de la MC est inconnue. La théorie de l’hygiène (TH) stipule que les conditions de vie sanitaires des pays industrialisés préviennent l’exposition antigénique et empêchent le développement de la tolérance immunitaire chez les enfants. Ceci mènerait à une réaction excessive du système immunitaire lors d’expositions subséquentes et engendrerait le développement de maladies inflammatoires chroniques telles la MC. Objectif: Analyser l’association entre la fréquence, la temporalité et le type d’infections infantiles (indicateurs d’environnements pourvus d’antigènes) et le risque de MC pédiatrique. Une étude cas-témoin fût réalisée, les cas de MC provenant d’un centre hospitalier tertiaire montréalais. Les témoins, provenant des registres de la Régie d’assurance maladie du Québec (RAMQ), furent appariés aux cas selon leur âge, sexe et lieu de résidence. L’exposition aux infections fût déterminée grâce aux codes de diagnostic ICD-9 inscrits dans la base de données de la RAMQ. Un modèle de régression logistique conditionnelle fût construit afin d’analyser l’association entre infections et MC. Des ratios de cotes (RC) et intervalles de confiance à 95% (IC 95%) furent calculés. Résultats: 409 cas et 1621 témoins furent recrutés. Les résultats de l’analyse suggèrent un effet protecteur des infections infantiles sur le risque de MC (RC: 0,67 [IC: 0,48-0,93], p=0,018), plus particulièrement au cours des 5 premières années de vie (RC: 0.74 [IC: 0,57-0,96], p=0,025). Les infections rénales et urinaires, ainsi que les infections des voies orales et du système nerveux central (virale), semblent particulièrement associées à l’effet protecteur. Les résultats de l’étude appuient la théorie de l’hygiène: l’exposition aux infections infantiles pourrait réduire le risque de MC pédiatrique.

Desigualdades na diabetes mellitus e hipertensão arterial: um estudo ao nível de freguesia

RESUMO - Introdução: A diabetes mellitus e a hipertensão arterial são problemas de saúde de elevada prevalência em Portugal. A sua distribuição geográfica e social é pouco conhecida, comprometendo o desenho e implementação de políticas de saúde. Assim, este estudo teve como objetivo avaliar a existência das desigualdades socioeconómicas na prevalência de diabetes mellitus tipo 2 e de hipertensão arterial, na população residente na região Norte de Portugal, no ano de 2013. Métodos: Foi realizado um estudo ecológico que analisou as 2028 freguesias da região Norte. Os dados foram obtidos através do Sistema de Informação das Administrações Regionais de Saúde e do Censos 2011. A associação entre os indicadores socioeconómicos e a prevalência destas doenças foi medida através da diferença de prevalências, do risco atribuível populacional, do índice relativo de desigualdades e pelo coeficiente de regressão. Resultados: A prevalência de diabetes mellitus tipo 2 e hipertensão arterial foi de 6,16% e 19,35%, respetivamente, e apresentou uma distribuição heterogénea entre freguesias (variando entre 0%-23,7% para a diabetes e 2,8%-66,7% para a hipertensão). A prevalência de ambas as doenças estava significativamente associada com o baixo nível educacional, baixa atividade em sector terciário, desemprego e baixo rendimento (com diferença de prevalências entre decis opostos de até 1,3% na diabetes e até 5,3% na hipertensão). Os determinantes socioeconómicos foram responsáveis até 20% da prevalência destas doenças na população. Conclusão: Estes resultados demonstram a existência de uma distribuição socioeconómica e geográfica heterogéneas e a necessidade de criação de políticas de saúde que atuem nas freguesias menos favorecidas.

Produção hospitalar : a importância da complexidade e da gravidade

RESUMO - A definição e medição da produção são questões centrais para a administração hospitalar. A produção hospitalar, quando se consideram os casos tratados, baseia-se em dois aspectos: a definição de sistemas de classificação de doentes como metodologia para identificar produtos e a criação de índices de casemix para se compararem esses mesmos produtos. Para a sua definição e implementação podem ser consideradas características relacionadas com a complexidade dos casos (atributo da oferta) ou com a sua gravidade (atributo da procura), ou ainda características mistas. Por sua vez, a análise do perfil e da política de admissões dos hospitais adquire um maior relevo no contexto de novas experiências previstas e em curso no SNS e da renovada necessidade de avaliação e regulação que daí decorrem. Neste estudo pretendeu-se discutir a metodologia para apuramento do índice de casemix dos hospitais, introduzindo- se a gravidade dos casos tratados como atributo relevante para a sua concretização. Assim, foi analisada uma amostra de 950 443 casos presentes na base de dados dos resumos de alta em 2002, tendo- -se dado particular atenção aos 31 hospitais posteriormente constituídos como SA. Foram considerados três índices de casemix: índice de complexidade (a partir do peso relativo dos DRGs), índice de gravidade (a partir da escala de mortalidade esperada do disease staging recalibrada para Portugal) e índice conjunto (média dos dois anteriores). Verificou-se que a análise do índice de complexidade, de gravidade e conjunto dá informações distintas sobre o perfil de admissões dos hospitais considerados. Os índices de complexidade e de gravidade mostram associações distintas às características dos hospitais e dos doentes tratados. Para além disso, existe uma diferença clara entre os casos com tratamento médico e cirúrgico. No entanto, para a globalidade dos hospitais analisados observou-se que os hospitais que tratam os casos mais graves tratam igualmente os mais complexos, tendo-se ainda identificado alguns hospitais em que tal não se verifica e, quando possível, apontado eventuais razões para esse comportamento.

O financiamento hospitalar e a definição de preços

RESUMO - Inseridos num contexto onde as unidades prestadoras de cuidados de saúde não possuem autonomia para fixar os seus próprios preços, estuda-se através do presente documento o processo de definição de preços do internamento hospitalar português entre 1993 e 2001, no sentido de aferir o grau de sinalização proporcionado ao mercado, uma vez que, na sua essência, um sistema de preços constitui um sistema de incentivos influenciador de comportamentos produtivos ao nível prestador. Utilizou-se a base de dados nacionais do sistema de classificação de doentes Grupos de Diagnóstico Homogéneos e foi caracterizada a evolução e tendência dos pesos relativos dos produtos do internamento hospitalar português. Foi também avaliada a relação da estrutura de preços com a quantidade produzida, a eficiência técnica e as necessidades em saúde da população portuguesa durante o período considerado. Apesar de existir alguma instabilidade durante o período analisado, o ponto de partida e de chegada não diferiu significativamente. Não foi encontrado nenhum padrão de variação que explicasse a instabilidade encontrada entre os produtos mais complexos, mais produzidos ou os que apresentaram maior/menor variação ao longo da década de 90. A hipótese relativa à correlação entre a evolução de preços e as variáveis quantidade, eficiência e necessidades também foi infirmada. Conclui-se que a utilização do sistema de preços constituiu na sua essência um sistema de incentivos neutro, não foi sinalizador para os prestadores e possuiu uma forte correlação com o nível de reposição de custos históricos dos hospitais portugueses.

Trends in the main diseases in an internal medicine ward, 2003-2011

Introduction : Population aging leads to a considerable increase in the prevalence of specific diseases. We aimed to assess if those changes were already reflected in an Internal Medicine ward. Methods : Anonymous data was obtained from the administrative database of the department of internal medicine of the Lausanne University Hospital (CHUV). All hospitalizations of adult (>=18 years) patients occurring between 2003 and 2011 were included. Infections, cancers and diseases according to body system (heart, lung...) were defined by the first letter of the ICD-10 code for the main cause of hospitalization. Specific diseases (myocardial infarction, heart failure...) were defined by the first three letters of the ICD-10 codes for the main cause of hospitalization. Results : Data from 32,741 hospitalizations occurring between 2003 and 2011 was analyzed. Cardiovascular (ICD-10 code I) and respiratory (ICD-10 code J) diseases ranked first and second, respectively, and their ranks did not change during the study period (figure). Digestive and endocrine diseases decreased while psychiatric diseases increased from rank 9 in 2003 to rank 6 in 2011 (figure). Among specific diseases, pneumonia (organism unspecified, code J18) ranked first in 2003 and second in 2011. Acute myocardial infarction (code I21) ranked second in 2003 and third in 2011. Chronic obstructive pulmonary disease with acute lower respiratory infection (code J44) ranked third in 2003 and decreased to rank 8 in 2011. Conversely, heart failure (code I50) increased from rank 8 in 2003 to rank 1 in 2011 and delirium (not induced by alcohol and other psychoactive substances, code F05) increased from below rank 20 in 2003 to rank 4 in 2011. For more details, see table. Conclusion : In less than 10 years, considerable changes occurred in the presentation of patients attending an Internal Medicine ward. The changes in diseases call for adaptations in hospital staff and logistics.

Differential effect of Charlson index and its components on in-hospital mortality and health costs

Introduction: The Charlson index (Charlson, 1987) is a commonly used comorbidity index in outcome studies. Still, the use of different weights makes its calculation cumbersome, while the sum of its components (comorbidities) is easier to compute. In this study, we assessed the effects of 1) the Charlson index adapted for the Swiss population and 2) the sum of its components (number of comorbidities, maximum 15) on a) in-hospital deaths and b) cost of hospitalization. Methods: Anonymous data was obtained from the administrative database of the department of internal medicine of the Lausanne University Hospital (CHUV). All hospitalizations of adult (>=18 years) patients occurring between 2003 and 2011 were included. For each hospitalization, the Charlson index and the number of comorbidities were calculated. Analyses were conducted using Stata. Results: Data from 32,741 hospitalizations occurring between 2003 and 2011 was analyzed. On bivariate analysis, both the Charlson index and the number of comorbidities were significantly and positively associated with in hospital death. Conversely, multivariate adjustment for age, gender and calendar year using Cox regression showed that the association was no longer significant for the number of comorbidities (table). On bivariate analysis, hospitalization costs increased both with Charlson index and with number of comorbidities, but the increase was much steeper for the number of comorbidities (figure). Robust regression after adjusting for age, gender, calendar year and duration of hospital stay showed that the increase in one comorbidity led to an average increase in hospital costs of 321 CHF (95% CI: 272 to 370), while the increase in one score point of the Charlson index led to a decrease in hospital costs of 49 CHF (95% CI: 31 to 67). Conclusion: Charlson index is better than the number of comorbidities in predicting in-hospital death. Conversely, the number of comorbidities significantly increases hospital costs.

Fallers in postacute rehabilitation have worse functional recovery and increased health services use.

OBJECTIVES: To determine characteristics associated with single and multiple fallers during postacute rehabilitation and to investigate the relationship among falls, rehabilitation outcomes, and health services use. DESIGN: Retrospective cohort study. SETTING: Geriatric postacute rehabilitation hospital. PARTICIPANTS: Patients (n = 4026) consecutively admitted over a 5-year period (2003-2007). MEASUREMENTS: All falls during hospitalization were prospectively recorded. Collected patients' characteristics included health, functional, cognitive, and affective status data. Length of stay and discharge destination were retrieved from the administrative database. RESULTS: During rehabilitation stay, 11.4% (458/4026) of patients fell once and an additional 6.3% (253/4026) fell several times. Compared with nonfallers, fallers were older and more frequently men. They were globally frailer, with lower Barthel score and more comorbidities, cognitive impairment, and depressive symptoms. In multivariate analyses, compared with 1-time fallers, multiple fallers were more likely to have lower Barthel score (adjOR: 2.45, 95% CI: 1.48-4.07; P = .001), cognitive impairment (adjOR: 1.43, 95% CI: 1.04-1.96; P = .026), and to have been admitted from a medicine ward (adjOR: 1.55, 95% CI: 1.03-2.32; P = .035). Odds of poor functional recovery and institutionalization at discharge, as well as length of stay, increased incrementally from nonfallers to 1-time and to multiple fallers. CONCLUSION: In these patients admitted to postacute rehabilitation, the proportion of fallers and multiple fallers was high. Multiple fallers were particularly at risk of poor functional recovery and increased health services use. Specific fall prevention programs targeting high-risk patients with cognitive impairment and low functional status should be developed in further studies.

Unplanned nursing home short stay : elderly patients characteristics and 3-month outcomes.

Background: Nursing home short stays (NHSS) in the canton of Vaud have been introduced for respite care purpose. However, a growing number of older patients are urgently admitted from home (within 24h) or directly after hospital discharge (58% of all admissions in 2010). NHSS appears therefore as an increasingly important component of the health care system, but the characteristics of admitted patients have not been previously described. A better knowledge would contribute to identify specific care needs and enhance their care. Objectives: 1) To describe the characteristics of patients admitted in unplanned NHSS ( after hospital stay or urgently from home); 2) To determine living disposition 3-month after NHSS discharge. Method: Over a 18-month period, elderly patients with unplanned NHSS admission to 2 facilities in Lausanne were identified. Demographic, social, health, and functional data, as well as main reason for admission were collected. Death and place of living at 3-months were collected using the administrative database. Results: Overall, 114 patients (mean age 83.1 ± 6.2 years, 77% women, 84% living alone) were assessed, 80% being admitted from hospital. Mean score in Lawton's instrumental ADL before NHSS admission was 4.6 ± 2.5 and 69% of the patients were home care recipients (median number of weekly visits: 5 ± 3). Patients reported going out 4.2 ± 1.3 times/week and 56% reported at least one fall over the past year. Among the 91 patients coming from the hospital, main reason for admission was injury/limb immobilization (58%), recuperation (13%) and functional impairment in basic ADL (10%). Mean score at Katz's Basic ADL at admission was 3.7 ± 1.9. Overall, 90% of patients were identified with gait and balance impairment, 78% with cognitive impairment and 70% with polypharmacy (>6 different drugs). At 3-month after NHSS discharge (N = 92), 72% patients were living at home, 16% had been admitted to long term care, and 6% died. Among patients living at home at follow-up, 11% had been readmitted to hospital during the follow-up period. Conclusion: Older patients with unplanned NHSS admission show a high prevalence of functional, mobility, and cognitive impairments, as well as other geriatric syndromes. Specific measures should be considered during these stays to prevent further functional decline and, possibly, hospital readmission. Patients admitted with basic ADLs impairment might be candidate for higher levels of care (rehabilitation).

Schizophrénie et dépression : Recours aux services avant et après une hospitalisation pour tentative de suicide

Contexte : Une tentative de suicide antérieure et les troubles mentaux constituent d’importants facteurs de risque de suicide. Les services de santé ont un rôle important à jouer en matière de prévention. Objectif : Analyser les patrons de recours aux services à des fins de santé mentale avant et après une hospitalisation pour tentative de suicide chez les résidents montréalais qui ont reçu un diagnostic de schizophrénie ou de dépression. Méthode : Les données proviennent de la banque médico administrative jumelée de l’Agence de Montréal. Les caractéristiques des patients et les taux de contacts avec les services trois mois avant et après l’hospitalisation index furent comparées. L’échantillon représentatif compte 525 Montréalais hospitalisés à la suite d’une tentative de suicide (avr. 2003-déc. 2004) qui ont reçu un diagnostic de schizophrénie ou de dépression. Résultats : Le recours aux services a augmenté de manière significative suivant l’hospitalisation index. Les patients déjà en contact avec les services et les hommes avec comorbidité en termes d’abus de substances semblent en contact avec les services au cours des trois mois suivant leur hospitalisation contrairement aux femmes avec comorbidité. Le profil « urgence » de recours aux services semble prédire une absence de recours aux services. Conclusions : Les services de santé répondent aux tentatives de suicide, particulièrement chez les hommes avec troubles d’abus de substances. Cependant, des interventions sont requises en vue d’améliorer la coordination des services, principalement pour les personnes qui se présentent l’urgence, les femmes avec troubles d’abus de substances et les patients en marge du système avant leur hospitalisation.

A retrospective study of cholinesterase inhibitors for Alzheimer's disease : the effect of cerebrovascular disease on patient outcomes and the impact of biases on the results.

Introduction: La démence peut être causée par la maladie d’Alzheimer (MA), la maladie cérébrovasculaire (MCEREV), ou une combinaison des deux. Lorsque la maladie cérébrovasculaire est associée à la démence, les chances de survie sont considérées réduites. Il reste à démontrer si le traitement avec des inhibiteurs de la cholinestérase (ChEIs), qui améliore les symptômes cognitifs et la fonction globale chez les patients atteints de la MA, agit aussi sur les formes vasculaires de démence. Objectifs: La présente étude a été conçue pour déterminer si la coexistence d’une MCEREV était associée avec les chances de survie ou la durée de la période jusqu’au placement en hebergement chez les patients atteints de la MA et traités avec des ChEIs. Des études montrant de moins bons résultats chez les patients souffrant de MCEREV que chez ceux n’en souffrant pas pourrait militer contre l’utilisation des ChEIs chez les patients atteints à la fois de la MA et la MCEREV. L'objectif d'une seconde analyse était d'évaluer pour la première fois chez les patients atteints de la MA l'impact potentiel du biais de « temps-immortel » (et de suivi) sur ces résultats (mort ou placement en hebergement). Méthodes: Une étude de cohorte rétrospective a été conduite en utilisant les bases de données de la Régie de l’Assurance Maladie du Québec (RAMQ) pour examiner la durée de la période jusqu’au placement en hebergement ou jusqu’au v décès des patients atteints de la MA, âgés de 66 ans et plus, avec ou sans MCEREV, et traités avec des ChEIs entre le 1er Juillet 2000 et le 30 Juin 2003. Puisque les ChEIs sont uniquement indiquées pour la MA au Canada, chaque prescription de ChEIs a été considérée comme un diagnostic de la MA. La MCEREV concomitante a été identifié sur la base d'un diagnostic à vie d’un accident vasculaire cérébral (AVC) ou d’une endartériectomie, ou d’un diagnostic d'un accident ischémique transitoire au cours des six mois précédant la date d’entrée. Des analyses séparées ont été conduites pour les patients utilisant les ChEIs de façon persistante et pour ceux ayant interrompu la thérapie. Sept modèles de régression à risque proportionnel de Cox qui ont varié par rapport à la définition de la date d’entrée (début du suivi) et à la durée du suivi ont été utilisés pour évaluer l'impact du biais de temps-immortel. Résultats: 4,428 patients ont répondu aux critères d’inclusion pour la MA avec MCEREV; le groupe de patients souffrant seulement de la MA comptait 13,512 individus. Pour le critère d’évaluation composite considérant la durée de la période jusqu’au placement en hebergement ou jusqu’au décès, les taux de survie à 1,000 jours étaient plus faibles parmi les patients atteints de la MA avec MCEREV que parmi ceux atteints seulement de la MA (p<0.01), mais les différences absolues étaient très faibles (84% vs. 86% pour l’utilisation continue de ChEIs ; 77% vs. 78% pour la thérapie avec ChEIs interrompue). Pour les critères d’évaluation secondaires, la période jusqu’au décès était plus courte chez les patients avec la MCEREV que sans la MCEREV, mais la période jusqu’au vi placement en hebergement n’était pas différente entre les deux groupes. Dans l'analyse primaire (non-biaisée), aucune association a été trouvée entre le type de ChEI et la mort ou le placement en maison d'hébergement. Cependant, après l'introduction du biais de temps-immortel, on a observé un fort effet différentiel. Limitations: Les résultats peuvent avoir été affectés par le biais de sélection (classification impropre), par les différences entre les groupes en termes de consommation de tabac et d’indice de masse corporelle (ces informations n’étaient pas disponibles dans les bases de données de la RAMQ) et de durée de la thérapie avec les ChEIs. Conclusions: Les associations entre la coexistence d’une MCEREV et la durée de la période jusqu’au placement en hebergement ou au décès apparaissent peu pertinentes cliniquement parmi les patients atteints de la MA traités avec des ChEIs. L’absence de différence entre les patients atteints de la MA souffrant ou non de la MCEREV suggère que la coexistence d’une MCEREV ne devrait pas être une raison de refuser aux patients atteints de la MA l’accès au traitement avec des ChEIs. Le calcul des « personne-temps » non exposés dans l'analyse élimine les estimations biaisées de l'efficacité des médicaments.

Les lipides et les cancers urologiques - approches épidémiologiques. L’importance du temps immortel

Objectif: Définir l’effet des lipides et du traitement de la dyslipidémie sur les cancers de la prostate et de la vessie en utilisant différents devis d’étude et en tenant compte de la présence de plusieurs biais, particulièrement le biais du temps immortel. Devis: Le premier volet utilise un devis rétrospectif de type cas témoins. Un questionnaire semi-quantitatif de fréquence de consommation alimentaire validé a été utilisé. Le génotype COX2 de neuf polymorphisme nucléotidique unique (SNP) a été mesuré avec une plateforme Taqman. Des modèles de régression logistique non conditionnelle ont été utilisés pour comparer le risque de diagnostic d’un cancer de la prostate et l’interaction. Le deuxième volet utilise un devis rétrospectif de type cohorte basée sur les données administratives de la Régie de l’assurance-maladie du Québec (RAMQ). Des modèles de régression de Cox ont été employés pour mesurer l’association entre les statines et l’évolution du cancer de la vessie. Le troisième volet, porte un regard méthodologique sur le biais du temps immortel en examinant sa présence dans la littérature oncologique. Son importance est illustrée avec les données de la cohorte du deuxième volet, et les méthodes de correction possibles son appliquées. Résultats: L’étude du premier volet démontre qu’une diète riche en acides gras oméga-3 d’origine marine était fortement associée à un risque diminué de cancer de la prostate agressif (p<0.0001 pour la tendance). Le ratio de cote pour le cancer de la prostate du quartile supérieur d’oméga-3 était de 0.37 (IC 95% = 0.25 à 0.54). L’effet diététique était modifié par le génotype COX-2 SNP rs4648310 (p=0.002 pour l’interaction). En particulier, les hommes avec faible apport en oméga-3 et la variante rs4648310 avait un risque accru de cancer de la prostate (ratio de cote = 5.49, IC 95%=1.80 à 16.7), effet renversé par un apport en oméga-3 plus grand. L’étude du deuxième volet a observé que l’utilisation de statines est associée à une diminution du risque de progression du cancer de la vessie (risque relatif = 0.44, IC 95% = 0.20 à 0.96, p=0.039). Cette association était encore plus forte pour le décès de toute cause (HR = 0.57, 95% CI = 0.43 to 0.76, p=0.0001). L’effet des statines semble être dose-dépendant. L’étude du troisième volet démontre que le biais du temps immortel est fréquent et important dans les études épidémiologiques oncologiques. Il comporte plusieurs aspects dont certains sont mieux prévenus au stade du choix du devis d’étude et différentes méthodes statistiques permettent un contrôle de ce biais. Conclusion: 1) Une diète riche en oméga-3 aurait un effet protecteur pour le cancer de la prostate. 2) L’utilisation de statines aurait un effet protecteur sur la progression du cancer non invasif de la vessie. Les lipides semblent avoir un effet sur les cancers urologiques.

Évolution de la gestion pharmacologique de la polyarthrite rhumatoïde et impact sur le risque de fracture ostéoporotique non vertébrale

Au cours des dernières années, le développement des connaissances au niveau de l’étiologie de la maladie ainsi que l’arrivée de nouveaux médicaments et de lignes directrices guidant la pratique clinique sont susceptibles d’avoir entraîné une meilleure gestion de la polyarthrite rhumatoïde (PAR) et de l’ostéoporose, une comorbidité fréquente chez ces patients. Dans cette thèse, trois questions de recherche sont étudiées à l’aide des banques de données administratives québécoises (RAMQ, MED-ÉCHO). Une première étude documente l’utilisation des médicaments pour la PAR au Québec. À ce jour, il s’agit de la seule étude canadienne à rapporter les tendances d’utilisation des DMARD (disease-modifying antirheumatic drug) biologiques depuis leur introduction dans la pratique clinique. Au cours de la période à l’étude (2002-2008), l’utilisation de DMARD (synthétiques et biologiques) a augmenté légèrement dans la population atteinte de PAR (1,9%, 95% CI : 1,1 - 2,8). Cependant, malgré la présence de recommandations cliniques soulignant l’importance de commencer un traitement rapidement, et la couverture de ces traitements par le régime général d’assurance médicaments, les résultats démontrent une initiation sous-optimale des DMARD chez les patients nouvellement diagnostiqués (probabilité d’initiation à 12 mois : 38,5%). L’initiation de DMARD était beaucoup plus fréquente lorsqu’un rhumatologue était impliqué dans la provision des soins (OR : 4,31, 95% CI : 3,73 - 4,97). Concernant les DMARD biologiques, le facteur le plus fortement associé avec leur initiation était l’année calendrier. Chez les sujets diagnostiqués en 2002, 1,2 sur 1 000 ont initié un DMARD biologique moins d’un an après leur diagnostic. Pour ceux qui ont été diagnostiqués en 2007, le taux était de 13 sur 1 000. Les résultats démontrent que si la gestion pharmacologique de la PAR s’est améliorée au cours de la période à l’étude, elle demeure tout de même sous-optimale. Assurer un meilleur accès aux rhumatologues pourrait, semble-t-il, être une stratégie efficace pour améliorer la qualité des soins chez les patients atteints de PAR. Dans une deuxième étude, l’association entre l’utilisation des DMARD biologiques et le risque de fractures ostéoporotiques non vertébrales chez des patients PAR âgés de 50 ans et plus a été rapportée. Puisque l’inflammation chronique résultant de la PAR interfère avec le remodelage osseux et que les DMARD biologiques, en plus de leur effet anti-inflammatoire et immunosuppresseur, sont des modulateurs de l’activité cellulaire des ostéoclastes et des ostéoblastes pouvant possiblement mener à la prévention des pertes de densité minérale osseuse (DMO), il était attendu que leur utilisation réduirait le risque de fracture. Une étude de cas-témoin intra-cohorte a été conduite. Bien qu’aucune réduction du risque de fracture suivant l’utilisation de DMARD biologiques n’ait pu être démontrée (OR : 1,03, 95% CI : 0,42 - 2,53), l’étude établit le taux d’incidence de fractures ostéoporotiques non vertébrales dans une population canadienne atteinte de PAR (11/1 000 personnes - années) et souligne le rôle d’importants facteurs de risque. La prévalence élevée de l’ostéoporose dans la population atteinte de PAR justifie que l’on accorde plus d’attention à la prévention des fractures. Finalement, une troisième étude explore l’impact de la dissémination massive, en 2002, des lignes directrices du traitement de l’ostéoporose au Canada sur la gestion pharmacologique de l’ostéoporose et sur les taux d’incidence de fractures ostéoporotiques non vertébrales chez une population de patients PAR âgés de 50 ans et plus entre 1998 et 2008. Étant donné la disponibilité des traitements efficaces pour l’ostéoporose depuis le milieu des années 1990 et l’évolution des lignes directrices de traitement, une réduction du taux de fractures était attendue. Quelques études canadiennes ont démontré une réduction des fractures suivant une utilisation étendue des médicaments contre l’ostéoporose et de l’ostéodensitométrie dans une population générale, mais aucune ne s’est attardée plus particulièrement sur une population adulte atteinte de PAR. Dans cette étude observationnelle utilisant une approche de série chronologique, aucune réduction du taux de fracture après 2002 (période suivant la dissémination des lignes directrices) n’a pu être démontrée. Cependant, l’utilisation des médicaments pour l’ostéoporose, le passage d’ostéodensitométrie, ainsi que la provision de soins pour l’ostéoporose en post-fracture ont augmenté. Cette étude démontre que malgré des années de disponibilité de traitements efficaces et d’investissement dans le développement et la promotion de lignes directrices de traitement, l’effet bénéfique au niveau de la réduction des fractures ne s’est toujours pas concrétisé dans la population atteinte de PAR, au cours de la période à l’étude. Ces travaux sont les premiers à examiner, à l’aide d’une banque de données administratives, des sujets atteints de PAR sur une période s’étalant sur 11 ans, permettant non seulement l’étude des changements de pratique clinique suivant l’apparition de nouveaux traitements ou bien de nouvelles lignes directrices, mais également de leur impact sur la santé. De plus, via l’étude des déterminants de traitement, les résultats offrent des pistes de solution afin de combler l’écart entre la pratique observée et les recommandations cliniques. Enfin, les résultats de ces études bonifient la littérature concernant la qualité des soins pharmacologiques chez les patients PAR et de la prévention des fractures.

Does Immediate Access to Birth Control Help Prevent Pregnancy? A Comparison of Onsite Provision Versus Off Campus Referral for Contraception at Two School-Based Clinics

PURPOSE: The purpose of this study was to assess the impact of different policies on access to hormonal contraception and pregnancy rates at two high school-based clinics. METHODS: Two clinics in high schools (Schools A and B), located in a large urban district in the southwest US, provide primary medical care to enrolled students with parental consent; the majority of whom have no health insurance coverage. The hormonal contraceptive dispensing policy of at School clinic A involves providing barrier, hormonal and emergency contraceptive services on site. School clinic B uses a referral policy that directs students to obtain contraception at an off-campus affiliated family planning clinic. Baseline data (age, race and history of prior pregnancy) on female students seeking hormonal contraception at the two clinics between 9/2008-12/2009 were extracted from an electronic administrative database (AHLERS Integrated System). Data on birth control use and pregnancy tests for each student was then tracked electronically through 3/31/2010. The outcomes measures were accessing hormonal contraception and positive pregnancy tests at any point during or after birth control use were started through 12/2009. The appointment keeping rate for contraceptive services and the overall pregnancy rates were compared between the two schools. In addition the pregnancy rates were compared between the two schools for students with and without a prior history of pregnancy. RESULTS: School clinic A: 79 students sought hormonal contraception; mean age 17.5 years; 68% were > 18 years; 77% were Hispanic; and 20% reported prior pregnancy. The mean duration of the observation period was 13 months (4-19 months). All 79 students received hormonal contraception (65% pill and 35% long acting progestin injection) onsite. During the observation period, the overall pregnancy rate was 6% (5/79); 4.7% (3/63) among students with no prior pregnancy. School clinic B: 40 students sought hormonal contraception; mean age 17.5 years; 52% > 18 years; 88 % were Hispanic; and 7.5% reported prior pregnancy. All 40 students were referred to the affiliated clinic. The mean duration of the observation period was 11.9 months (4-19 months). 50% (20) kept their appointment. Pills were dispensed to 85% (17/20) and 15% (3/20) received long acting progestin injection. The overall pregnancy rate was 20% (8/40); 21.6% (8/37) among students with no prior pregnancy. A significantly higher frequency of students seeking hormonal contraception kept their initial appointment for birth control at the school dispensing onsite contraception compared to the school with a referral policy for contraception (p<0.05). The pregnancy rate was significantly higher for the school with a referral policy for contraception compared to the school with onsite contraceptive services (p< 0.05). The pregnancy rate was also significantly higher for students without a prior history of pregnancy in the school with a referral policy for contraception (21.6%) versus the school with onsite contraceptive services (4.7%) (p< 0.05). CONCLUSION: This preliminary study showed that School clinic B with a referral policy had a lower appointment keeping rate for contraceptive services and a higher pregnancy rate than School clinic A with on-site contraceptive services. An on-site dispensing policy for hormonal contraceptives at high school-based health clinics may be a convenient and effective approach to prevent unintended first and repeat pregnancies among adolescents who seek hormonal contraception. This study has strong implications for reproductive health policy, especially as directed toward high-risk teenage populations.

Leaving and Coming Back to Canada : Evidence from Longitudinal Data (Working Paper 32)

The paper exploits the unique strengths of Statistics Canada's Longitudinal Administrative Database ("LAD"), constructed from individuals' tax records, to shed new light on the extent and nature of the emigration of Canadians to other countries and their patterns of return over the period 1982-1999. The empirical evidence begins with some simple graphs of the overall rates of leaving over time, and follows with the presentation of the estimation results of a model that essentially addresses the question: "who moves?" The paper then analyses the rates of return for those observed to leave the country - something for which there is virtually no existing evidence. Simple return rates are reported first, followed by the results of a hazard model of the probability of returning which takes into account individuals' characteristics and the number of years they have already been out of the country. Taken together, these results provide a new empirical basis for discussions of emigration in general, and the brain drain in particular. Of particular interest are the ebb and flow of emigration rates observed over the last two decades, including a perhaps surprising turndown in the most recent years after climbing through the earlier part of the 1990s; the data on the number who return after leaving, the associated patterns by income level, and the increases observed over the last decade.

Heart failure in young adults

Heart failure (HF) is a major health concern affecting 15 million people in Europe and around 900 000 people in the U.K. HF predominantly affects the elderly, with the mean age of patients with a diagnosis of HF between 70 and 80 years. Most previous HF studies have accordingly focused on older patients. Although HF is less common in younger adults (<65 years), 15% to 20% of patients hospitalised with HF are younger than 60 years of age. Very few studies have described the characteristics of younger adults with HF and its outcome. The aims of this thesis are to describe the clinical characteristics of younger adults with HF, explore the epidemiology of HF in younger adults and determine their short- and long-term outcomes. This was made possible by access multiple databases consisting of large patient cohorts with HF. The first chapter is a systematic literature review of younger adults with HF. Gaps in the current literature were identified and the thesis focused on some of these. The CHARM study allows detail characterisations of younger adults with HF. It recorded characteristics of patients with HF, including symptoms and signs of HF, electrocardiographic changes, chest radiographic findings, and also left ventricular ejection fraction. HF hospitalisations and its precipitating factors were also recorded systematically. Younger adults were more likely to have a third heart sound and hepatomegaly, but less likely to have pulmonary crackles and peripheral oedema. Similarly, radiological findings in younger adults were less likely to show interstitial pulmonary oedema or pleural effusion. Interestingly, younger adults aged <40 years not only have similar HF hospitalisation rate to older patients, however during their presentation with decompensated HF, they were less likely to have clinical pulmonary oedema and radiological signs of HF. Physicians managing younger adults with HF need to be aware of this. Younger adults were also less compliant with medications and lifestyle restriction resulting in hospitalisation with decompensated HF. Fortunately, despite these challenges, mortality rates in younger adults with HF were lower compared to older patients. To further substantiate the findings from the CHARM study, the MAGGIC study, a meta-analysis consists of over 40 000 patients with HF from large observational studies and randomised controlled trials, was examined. In both databases, the commonest aetiology of HF in younger adults was dilated cardiomyopathy. The ejection fraction was the lowest in younger adults. Similar to the CHARM study, mortality rates in younger adults were lower compared to older patients. However, in the MAGGIC study, by stratifying mortality into patients with preserved ejection fraction and with reduced ejection fraction, younger patients with preserved ejection fraction have a much lower mortality rate compared to patients with reduced ejection fraction. Findings from clinical trials are not always reflective of the real life clinical practice. The U.K. Clinical Practice Research Datalink (CPRD), a large and well-validated primary care database with 654 practices contributing information into the database representing approximated 8% of the U.K. population, is a rich dataset offering a unique opportunity to examine the characteristics, treatments, and outcomes of younger adults with HF in the community. In contrast to the CHARM and MAGGIC studies, younger adults aged <40 years were stratified into 20-29 and 30-39 years in the CPRD analysis. This is possible due to the larger number of younger adults with HF. Further stratifying the younger age groups demonstrated heterogeneity among younger adults with HF. In contrast to previous data showing younger adults have lower co-morbidities, the proportions of depression, chronic kidney disease, asthma, and any connective tissue disease were high among patients aged 20-29 years in the analysis from the CPRD. Surprisingly, the treatment rates for angiotensin converting enzyme (ACE) inhibitor, and aldosterone antagonist were the lowest in patients aged 20-29 years. With the exception of patients aged ≥80 years, treatment rate with beta-blocker was also the lowest in patients aged 20-29 years. With over two decades of follow up, long-term mortality rates in younger adults with HF can be determined. The mortality rates continued to decline from 1988 to 2011. Physicians managing younger adults with HF can now use this contemporary data to provide prognostic information to patients and their family. A hospital administrative database is the logical next platform to explore younger adults with HF. The Alberta Ministry of Health database links an outpatient database to a hospitalisation database providing ample data to examine the relationship between outpatient clinic visits and hospital admissions in younger adults with HF. Following a diagnosis of HF in the outpatient setting, younger adults were admitted to the hospital with decompensated HF much sooner than older patients. Younger adults also presented to emergency department more frequently following their first hospitalisation for HF. In conclusion, this thesis presented the characteristics and outcomes of younger adults with HF, and helped to extend our current understanding on this important topic. I hope the data presented here will benefit not only physicians looking after younger adults with HF, but also patients and their family.