L’essor de la statistique en médecine est l’ultime étape de la quête d’une médecine scientifique poursuivie tout au long de l’histoire de la discipline. Cette méthodologie de production des preuves médicales est reconnue par tous comme un gage de qualité justifiant les prises de décisions médicales au niveau individuel et collectif.Initialement, le concept est une démarche pédagogique prônant l’autonomie de chaque praticien dans la recherche, l’analyse critique et l’application personnalisée des preuves disponibles. La démarche rejette le dogmatisme médical. Ce concept a été et demeure largement controversé. Pour autant, il a rapidement traversé les frontières pour devenir une exigence déontologique et juridique au risque d’une dérive normative.La priorité laissée à la qualité méthodologique des preuves médicales au détriment de leur applicabilité en vraie-vie pose question.L’hypothèse de ce travail est qu’il existe, aux différentes étapes du circuit de la preuve médicale, des biais cognitifs et/ou méthodologiques qui peuvent impacter l’exercice pertinent de la médecine malgré son alibi scientifique.Au travers de l’exemple des maladies respiratoires chroniques, ce travail épistémologique se propose de caractériser la preuve médicale. Il décompose le circuit de la preuve médicale pour analyser sa méthodologie de production, ses sources, vecteurs et cibles de diffusion, les conséquences de l’implication des agences réglementaires et de l’Etat dans son applicabilité mais aussi les biais cognitifs auxquels sont soumis médecins et patients. Il vise à déterminer avec transparence sur quelles preuves la médecine se fonde pour en permettre une utilisation pertinente. / The use of statistics in medicine is the final step for the development of scientific medicine pursued throughout the history of the discipline. This method of production of medical evidence is recognized by healthcare professionals, drug manufacturers and political institutions as a pledge of quality that justifies medical decision-making at the individual and collective levels. Initially, the concept is an educational approach advocating the autonomy of each practitioner in bibliographic research and critical appraisal of available evidence for their use in the context of personalized medicine. The approach rejects medical dogmatism. This concept has been and remains largely controversial. However, it has quickly become a deontological and legal requirement that could drift back to dogmatism. The priority given to the methodological quality of medical evidence while poor attention is paid to its real-life applicability raises concerns. In this work we hypothesize that, from the production to the use of medical evidence there are cognitive and/or methodological biases that may alter the relevance of medicine practice despite its scientific alibi. Through the example of chronic respiratory diseases, this epistemological work aims at characterizing the medical evidence. To do so, we analyzed its production methodology, the sources, vectors and targets for dissemination, the consequences of the involvement of regulatory agencies and governments in its applicability and the cognitive biases that may apply to physicians and patients. It aims at determining transparently on what type of evidence medicine is based to enable its relevant practice.

Femoral neck fracture (FNF) is a great challenge for today´s health care and is associated with high mortality and morbidity in the elderly.  In the short term several studies in the literature have demonstrated improved hip function, quality of life and fewer re-operations in elderly patients treated with total hip arthroplasty (THA) instead of internal fixation (IF). There are few reports on the long-term outcome comparing IF and THA. The vast majority of orthopaedic departments in Sweden use the direct lateral (DL) or posteriolateral (PL) approaches for hip arthroplasty. The PL approach has been linked to an increased risk of dislocation of the prosthesis and a higher rate of revision surgery in comparison to the DL approach. There are few reports focusing on radiological risk factors for prosthetic dislocation and patient reported hip function comparing the two surgical approaches for hip arthroplasty in FNF. The randomized controlled trial (RCT) is the gold standard for evaluating medical or surgical interventions. An RCT of high quality has to be internally and externally valid. Internal validity refers to a correct study design to avoid bias skewing the results. External validity (EV) refers to whether the results will be clinically relevant to a definable group of patients and can be extrapolated to the general health care situation. There are only a few reports in the orthopaedic literature focusing on the EV of published studies and none in the field of hip fractures. Study I: This is a RCT of 100 patients with a displaced FNF comparing THA and IF. Follow-up evaluations were performed at three months and 1, 2, 4, 11 and 17 years. It was found that the Harris hip score (HHS) was higher and the rate of reoperations lower for patients treated with THA. Study II: This is a prospective cohort study of 185 hips, comparing the DL and the PL approaches in patients treated with a hemiarthroplasty (HA) for a displaced FNF. Follow-up was after 1 year. There was no difference in patient reported outcome between the groups measured with the HHS and WOMAC index. The PL approach resulted in a higher re-operation rate while the DL approach in a higher incidence of limping. Study III: This is a retrospective cohort study of 373 patients with a cemented bipolar HA using a PL approach for a FNF with a follow-up ranging from 6 months to 7 years. Radiographs and all surgical records were reviewed regarding femoral offset (FO), leg length discrepancy (LLD) and Wiberg angle. Patients with recurrent dislocations had a decreased postoperative FO, LLD and shallower acetabulum on the operated side compared with their controls. Study IV: This is a prospective cohort study of 840 hips comparing patients included in a RCT with those that did not give their informed consent (NC) or did not fulfill the criteria for participating in the trial (MS). Patients in the NC and MS groups had an increased mortality rate in comparison to those included in the study. We did not find any differences in hip function between these groups. The main conclusions of this thesis are: Healthy and lucid elderly patients with good hip function preoperatively, should be treated with THA for a displaced FNF. The DL approach is favourable in treating displaced FNF with HA due to its decreased risk of reoperation but with an identical hip function outcome as the PL approach. Care should be taken to restore the LLD and FO otherwise this may increase the risk of recurrent dislocation of a HA. Our findings suggest that trial participants had a lower mortality rate than non-participants but the functional outcome of non-participants appeared to be satisfactory. This is important to take into consideration when extrapolating study results to a health care system. / Lårbenshalsfrakturer (FNF) är en utmaning för dagens sjukvård och förknippad med hög mortalitet i den äldre befolkningen. FNF orsakas främst av lågenergitrauma efter fall i samma plan hos patienter med benskörhet. Total höftledsprotes (THA) har visat sig ge en bättre höftfunktion, livskvalitet med färre omoperationer hos äldre patienter än sluten reposition samt fixering med skruvar (IF) under de första två-fyra åren efter operation. Få studier har publicerats med långtidsuppföljning av THA jämfört med IF. Vid behandling av felställd FNF med halv höftledsplastik (HA) eller THA finns olika kirurgiska metoder (snittföring) för implantation av protesen. Majoriteten av ortopedkliniker i Sverige använder direkt lateral (DL) eller posteriolateral (PL) snittföring. PL har kopplats till en ökad risk för luxation av protesen (urledsvridning), vilket i sin tur genererat en högre risk för omoperation jämfört med DL. Få studier har utvärderat höftfunktion med avseende på snittföring. Detsamma gäller vilka radiologiska faktorer som påverkar risken förprotesluxation. Det finns olika typer av studiedesigner för utvärdering av medicinska och kirurgiska behandlingsmetoder. Studiedesignen med högst bevisvärde anses vara den randomiserade kontrollerade studien (RCT). För att en RCT skall hålla hög kvalitet krävs intern och extern validitet. Intern validitet syftar på ett korrekt studieupplägg som mäter det som avses och undviker påverkan av icke önskvärda faktorer. Extern validitet (EV) innebär att erhållna resultat från studien är generaliserbara och kan extrapoleras till andra grupper än den studerade. Det inga rapporter i ortopediska litteraturen med fokus på EV med patienter som drabbats av FNF. Studie I: Prospektiv randomiserad studie av 100 patienter med en felställd FNF jämförs THA med IF med avseende på höftfunktion och reoperationer. Uppföljning gjordes efter tre månader samt 1, 2, 4, 11 och 17 år. HHS som mått på höftfunktion var högre och andelen reoperationer lägre för patienter som behandlades med THA. Studie II: Prospektiv kohortstudie med 183 patienter med felställd FNF behandlade med HA och antingen DL eller PL. Vi fann ingen skillnad i patientrapporterad höftfunktion utvärderat med HHS och WOMAC efter 1 år men en ökad förekomst av hälta vid DL. Studie III: Retrospektiv kohortstudie av 373 patienter opererade med en cementerad bipolär HA och PL. Postoperativa röntgenbilder granskades avseende femoral offset (FO), postoperativ benlägdsskillnad (LLD) samt acetabulär Wibergvinkel och relaterades till återkommande protesluxationer. Detta relaterades till återkommande luxationer. Uppföljningstiden varierade mellan 6 månader och 7 år. Patienter med återkommande luxationer hade en minskad postoperativ global FO, förkortad LLD och grundare acetabulum på den opererade sidan jämfört med kontroller. Studie IV: Prospektiv kohortstudie av 840 patienter med felställd FNF där patienter som ingick i en RCT jämförs med de som inte gav sitt samtycke för deltagande eller ej inkluderats i screeningprocessen. De patienterna som inte gav sitt samtycke hade ökad dödlighet, men ingen skillnad i höftfunktion jämfört med deltagare. Slutsatser • Friska och kognitivt vitala äldre patienter med god höftfunktion innan en felställd lårbenshalsfraktur bör behandlas med total höftledsprotes. • DLsnittföring ger ett lika bra patientrapporterad höftfunktion men en ökad förekomst av hälta jämfört med PL snitt vid behandling med halvprotes på grund av en felställd lårbenshalsfraktur. • Ansträngning bör göras för att återställa benlängdsskillnad samtfemoral offset och därigenom minska risken för återkommande luxation. • Våra resultat tyder på att patienter som deltar i en randomiserad studie har lägre dödlighet än patienter som avböjt att delta samt de som missats i screeningprocessen. Det är viktigt att ta hänsyn till när extrapolerar studieresultat till ett hälso-och sjukvårdssystem. Detta bör vägas in i värderingen av studieresultatens externa validitet.

Le secteur de l'assurance automobile est confronté à plusieurs bouleversements règlementaires, financiers, comportementaux et technologiques. Afin de faire face aux défis résultant de ces changements et maintenir leur profitabilité, les assureurs doivent innover en matière de tarification. Dans ce contexte, nous développons dans cette thèse deux thématiques liées à la tarification en assurance automobile. La première thématique s'articule autour de l'optimisation des stratégies tarifaires, en souscription et au renouvellement. La deuxième thématique est orientée vers l'utilisation des expérimentations dans l'objectif de mieux appréhender les déterminants de la demande d'assurance.Tout d'abord, nous nous intéressons à l'optimisation tarifaire au renouvellement. Nous illustrons comment les modèles de demande empiriques reposant sur les données dont disposent l'assureur peuvent être utilisés afin d'optimiser sa rentabilité et la rétention de ses clients. Nous élargissons ensuite le cadre d'optimisation en tenant compte des dépendances inter-temporelles entre les décisions tarifaires actuelles et les profits générés au cours des périodes ultérieures. Ainsi, nous introduisons le cadre de Valeur Client qui permet à l'assureur d'adapter sa stratégie tarifaire en fonction des comportements des assurés au cours de leur vie client tout en tenant compte du cycle du marché. Les illustrations empiriques des deux premiers chapitres reposent sur des données naturelles observées par l'assureur.Dans la deuxième partie de la thèse, nous illustrons l'apport des expérimentions de terrain et de laboratoire à la compréhension de la demande d'assurance automobile. Une expérimentation de terrain nous permet d'affiner la mesure de l'élasticité prix des clients et de traiter le problème de tarification comme un problème de bandit contextuel. L'évaluation offline de plusieurs stratégies d'apprentissage par renforcement montre que celles appliquant une expérimentation tarifaire ciblée obtiennent de meilleures performances financières en comparaison à la stratégie myope, qui exclut toute possibilité d'expérimentation. Enfin, nous présentons les résultats d'une expérimentation de laboratoire dont l'objectif était de mesurer la valeur ajoutée des variables privées issues des modèles de décision dans le risque. En particulier, nous analysons le rôle de l'aversion au risque et la perception du risque dans l'explication des choix d'assurance automobile. La même expérimentation nous a permis d'analyser la validité externe en assurance expérimentale, c'est-à-dire la ressemblance des comportements des individus dans un contexte expérimental et dans le contexte économique réel du marché.En plus de la dualité expérimentation-optimisation dans le domaine de la tarification assurantielle, cette thèse illustre donc la dualité entre les données privées et les données publiques, ainsi que la dualité entre les modèles empiriques de demande d'assurance et les modèles théoriques. / The motor insurance sector currently confronts regulatory, financial, behavioral and technological challenges. Under these circumstances, insurers must uphold in improving their pricing strategies. Two topics related to pricing innovation are discussed in this thesis. We first take up the pricing strategy optimization for new businesses, as well as the renewals. Secondly, we highlight in the usage of experiments in leading us to a better understanding of insurance demand factors.On the first part of this thesis, we address pricing optimization at renewal, then illustrate how empirical demand models that rely on observable data could help the insurers to boost their profits and clients retention rate. We extend afterwards this framework by considering the impact of current pricing decisions on future cash-flows. Consequently, we introduce the Customer Value metric which allows insurers to reflect over the customers' behavior during their lifetime, when it comes to constructing their pricing strategy. The empirical illustrations of the first two chapters rely on natural data observed by the insurer.On the second part of this thesis, field and laboratory experiments will give us better comprehension of the motor insurance demand. Data from a field experiment refine the measure of clients' price elasticity. Offline assessment of several reinforcement learning algorithms shows how pricing experiments can achieve better performances compared with the myopic strategy which does not apply any kind of experiment. Laboratory experiments contribute to the understanding of demand models as well. In particular, we analyze the added value of risk aversion and risk perception in explaining the insurance choices. Furthermore, we examine the external validity of the experiment, i.e. the similarity between the behaviors of the customers in a lab environment versus their factual behaviors in the market.Aside from the duality between experiments and optimization, this thesis also illustrates the duality between private and public data, as well as the duality between empirical and theoretical insurance demand model.

Les études de séroprévalence portant sur les infections au SRAS-CoV-2 doivent souvent composer avec des échantillons non-aléatoires et non-représentatifs, limitant ainsi parfois la validité externe de leurs résultats lorsque ceux-ci sont appliqués à la population générale. Dans le cadre de ce mémoire, il s’agit d’investiguer la représentativité d’une cohorte pédiatrique d’une étude longitudinale de séroprévalence (Enfants et COVID-19 : Étude de séroprévalence) et d’évaluer dans quelle mesure ses estimations de séroprévalence peuvent s’appliquer à la population pédiatrique montréalaise en général. 1 632 enfants ont fourni au point de départ un échantillon sanguin afin de déterminer leur séropositivité aux anticorps contre le SRAS-CoV-2. À l’aide d’une modélisation par régression logistique et d’un procédé de « standardisation marginale », une pondération post-stratification calculée à partir des données du recensement canadien de 2016 a été appliquée à la population d’étude. Les variations dans les estimations de séroprévalence ont finalement été évaluées. D’importantes différences dans la distribution de certaines caractéristiques sociodémographiques peuvent être observées lorsqu’on compare la population d’étude et la population générale en se basant sur les données du recensement canadien de 2016. En comparaison des estimations non-pondérées, les estimations de séroprévalence générées à partir du procédé de « standardisation marginale » montrent une variation de plusieurs points de pourcentage, allant de -0,4% à +3,2%. La pondération n’a pas induit de changement dans l’estimation de mesures relatives comme les ratios de séroprévalence. Lorsque la population d’étude est non-représentative de la population-cible, il est nécessaire de pondérer les caractéristiques sociodémographiques associées à l’issue si l’on veut appliquer les résultats plus généralement. / Prevalence studies on SARS-CoV-2 infections have often based on study populations with non-random and non-representative samples, which limits the external validity of their results when applied to the general population. The aim of this thesis was to investigate the representativeness of a pediatric cohort of a longitudinal seroprevalence study (Children and COVID-19: Seroprevalence study) and to assess to what extent its baseline estimates of seroprevalence can be applied to the Montreal pediatric population. There were 1 632 children participants who provided a blood sample at baseline, which was used to determine their seropositivity to SARS-CoV-2 antibodies. Using logistic regression modeling and a "marginal standardization" method, post-stratification weights calculated from 2016 Canadian census data were applied to the study population. Variations in seroprevalence estimates were then assessed. Significant differences in the distribution of certain sociodemographic characteristics were observed when comparing the study population and the target population based on 2016 Canadian census data. Seroprevalence estimates were generated from the “marginal standardization” approach which differed to that of the non-standardized estimates, and the differences ranges from -0,4% to +3,2%. Weighting did not change relative measures estimates, such as seroprevalence ratios. When the study population is not representative of the target population, it is necessary to weight the sociodemographic characteristics associated with the prevalence estimates, if the results will be applied more broadly.

Le biais de sélection par rapport au sexe (ou biais de sexe) représente une différence systématique des proportions d’hommes et de femmes entre un échantillon de participants et leur population, ce qui peut miner la validité d’une étude. La recherche sur le stress humain est susceptible au biais de sexe étant donné la présence de facteurs y étant généralement associés, principalement les protocoles invasifs – contenant des éléments douloureux, inconfortables ou menaçants pour les participants. La présente étude a vérifié si les proportions d’hommes et de femmes des études sur le stress varient selon 1) le niveau d’invasion d’une étude en stress (invasif ou non invasif) et selon 2) des facteurs exploratoires (p. ex. pays, méthode de recrutement). Deux domaines hors stress possédant des protocoles invasifs (douleur expérimentale) et non invasifs (mémoire) ont été utilisés comme domaines contrôles. Dans cette enquête transversale de la littérature, les proportions d’hommes et de femmes ont été recueillies dans 324 études contenant des protocoles invasifs ou non invasifs, représentant un total de 23 611 participants, dont 42,18 % d’hommes. La représentativité des sexes a varié selon le niveau d’invasion dans les domaines du stress et hors stress, les hommes davantage représentés dans les études invasives que dans les études non invasives. Les résultats indiquent que les facteurs exploratoires analysés peuvent tous être associés au sexe. Cette étude a identifié la présence de facteurs pouvant provoquer un biais de sexe en recherche sur le stress humain, ouvrant la voie aux recherches souhaitant approfondir la généralisation des résultats. / Sex selection bias (or sex bias) refers to a systematic difference in the proportions of men and women between a sample of participants and their population, which may undermine the validity of a study. Human stress research is vulnerable to sex bias, given the presence of factors typically associated with it, primarily invasive protocols — containing painful, uncomfortable, or threatening elements for participants. The present study has verified whether the proportions of men and women in stress studies differ by 1) the invasiveness of a study (invasive or non-invasive) and by 2) exploratory factors (e.g. country, recruitment method). Two non-stress domains with invasive (experimental pain) and non-invasive (memory) protocols were used as control domains. In a cross-sectional survey of the literature, the proportions of men and women were collected from 324 studies containing invasive or non-invasive protocols, representing a total of 23 611 participants, 42,18 % of whom were men. Sex representativeness differed across invasiveness levels in both the stress and non-stress domains with men being more represented in invasive than in non-invasive studies. Results indicate that the exploratory factors analyzed may all be associated with sex. This study identified the presence of factors that may cause sex bias in human stress research, opening the door to research wishing to further investigate the generalizability of results.

Introdução: Ruminação é a perseveração mal-adaptativa de pensamentos auto-centrados. Evidências sinalizam que ela está associada com início e manutenção de episódios depressivos, e ocorre em múltiplos transtornos mentais. A ruminação está associada com marcadores de desenvolvimento psicopatológico, como volumetria cerebral, memória, genes do BDNF e serotonina. É necessário aprofundar o conhecimento da ruminação enquanto traço dimensional, e conhecer melhor sua associação com variáveis sóciodemográficas, biológicas e clínicas para entender quando passa a ser um sintoma. Entretanto, aferi-la é um desafio, considerando que só existem escalas psicométricas. A mais utilizada, Ruminative Response Scale (RRS), foi validada em amostras não-clínicas. Objetivos: Avaliar ruminação transdiagnosticamente e determinar a validade de constructo da RRS em amostra clínica, buscando determinar fatores sócio-demográficos, clínicos e neurobiológicos associados a maiores escores de ruminação. Métodos: Estudo transversal, amostra não-probabilística. Foram convidados a participar 944 pacientes em atendimento psiquiátrico ambulatorial no HCPA entre março/2015 e junho/2016, maiores de 18 anos, que soubessem ler e escrever, e portadores de transtornos bipolares, depressão, esquizofrenia, esquizoafetivo, ansiedade generalizada, pânico, fobia específica e obsessivocompulsivo. Foram excluídos 373 com doenças que alteram resposta inflamatória, dependência química, gravidez, lactação, doenças neurológicas, vasculares e degenerativas. Recusaram-se a participar 254. Foram incluídos 317 pacientes, e 200 completaram a coleta de dados, que foi realizada em 4 etapas: 1) perfil sócio-demográfico e escalas auto-aplicáveis: ruminação, preocupação e funcionalidade; 2) amostras de sangue e entrevista clínica para aplicação das escalas de sintomas: depressão, mania, ansiedade e gravidade; 3) confirmação diagnóstica; e 4) processamento, armazenamento e análises bioquímicas das amostras de sangue. No primeiro artigo, revisamos sistematicamente a literatura sobre ruminação nos transtornos bipolares. No segundo, determinamos as validades de construto e externa da RRS. No terceiro, usamos machine learning para encontrar padrões de ruminação e determinar quais variáveis associadas preveem ruminação. Resultados: Ruminação está presente em todas as fases do transtorno bipolar, e é um sintoma estável independente do estado de humor, apesar de ter relação estreia com ele. Verificou-se também que mulheres ruminam mais que homens. Os escores de ruminação foram menores nos portadores de esquizofrenia que nos com depressão maior, bipolaridade e ansiedade. RRS apresentou boa confiabilidade, com 2-fatores correlacionados, brooding e ponderação, que apresentaram similaridade nas correlações com medidas clínicas, confirmando a validade externa transdiagnóstica. Por fim, encontrou-se que as variáveis associadas aos pacientes que mais ruminam são preocupação, sintomas de ansiedade generalizada e depressão, gravidade, nível socioeconômico e diagnóstico atual de pânico, sinalizando que ruminação pode ser um marcador de maior sensibilidade à ansiedade. Discussão: Ruminação parece ser um sintoma transdiagnóstico marcador de sofrimento. Os resultados desta tese contribuem para ampliar a discussão sobre diagnóstico psiquiátrico, agregando evidências para aprimorar as definições de limites e sobreposições diagnósticas entre as doenças mentais em que a ruminação ocorre. Por fim, conhecer melhor os mecanismos bioquímicos e clínicos envolvidos na ruminação contribuem na compreensão sobre quando ela deixa de ser um traço normal e vira um sintoma que necessita de tratamento. / Introduction: Rumination has been described as maladaptive perseveration of self-centered thoughts. Evidence indicates that rumination is associated with onset and maintenance of depressive episodes, it’s present in several mental disorders. Rumination is associated with markers of development of psychopathology, such as cerebral volumetry, memory, BDNF and serotonin genes. Measuring rumination is a challenge, considering that are available only psychometric scales. The most used, the Ruminative Responses Scale (RRS), was validated on non-clinical samples. Objectives: To evaluate transdiagnostically the rumination and to determine construct validity of the RRS in outpatients, in order to determine which associated factors lead the patients to ruminate. Methods: Cross-sectional study, non-probabilistic sample. A total of 944 patients in psychiatric outpatient treatment at HCPA between March / 2015 and June / 2016, major than 18 years old, knowing read and write, presenting bipolar disorder, schizophrenia, schizoaffective disorder, generalized anxiety disorder, panic disorder, phobia specific and obsessive-compulsive disorder were invited to participate. We excluded 373 patients with diseases that alter inflammatory response, chemical dependence, pregnancy, lactation, neurological, vascular and degenerative diseases. Two hundred fifty four refused to participate, 317 were included, and 200 completed the data collection, which was performed in 4 stages: 1) socio-demographic profile and self-applicable scales: rumination, worry and functionality; 2) blood samples and clinical interview for the application of symptom scales: depression, mania, anxiety and severity; 3) diagnostic confirmation; and 4) processing, storage and biochemical analyzes of blood samples. In the first article, we systematically reviewed the literature on rumination in bipolar disorders. In the second, we evaluated construct and external validity of RRS. In the third, we used machine learning algorithms to find patterns of rumination and to determine which associated variables predict rumination. Results: Rumination is present in all phases of bipolar disorder, it is a stable symptom, independent of mood, despite it has close relationship with it. It has also been found that women ruminate more than men. Rumination scores were lower in patients with schizophrenia than in major depression, bipolarity and anxiety patients. RRS presented good reliability, with correlated 2-factors, brooding and pondering, which presented similar correlations with clinical measures, confirming the external transdiagnostic validity. Finally, it was found that the variables associated with the greater scores of rumination are worry, symptoms of generalized anxiety and depression, severity of symptoms, socioeconomic level and current diagnosis of panic, signaling that rumination may be a marker of greater sensitivity to anxiety. Discussion: Rumination seems to be a transdiagnostic symptom of suffering. The results of this thesis contribute to broadening the discussion about psychiatric diagnostic, adding evidence to improve the definitions of limits and diagnostic overlaps between mental illnesses in which rumination occurs. Finally, a better understanding of the biochemical and clinical mechanisms involved in rumination may contribute to understanding of when rumination ceases to be a normal trait and becomes a symptom that requires treatment.

Submitted by Alessandro Ribeiro de Carvalho Casalecchi (alercc@gmail.com) on 2017-07-03T21:17:55Z No. of bitstreams: 1 Tese_Alessandro_Casalecchi.pdf: 2174297 bytes, checksum: 27298549cf220c58b7eb52f7323446d7 (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2017-07-04T11:10:20Z (GMT) No. of bitstreams: 1 Tese_Alessandro_Casalecchi.pdf: 2174297 bytes, checksum: 27298549cf220c58b7eb52f7323446d7 (MD5) / Made available in DSpace on 2017-07-05T13:46:08Z (GMT). No. of bitstreams: 1 Tese_Alessandro_Casalecchi.pdf: 2174297 bytes, checksum: 27298549cf220c58b7eb52f7323446d7 (MD5) Previous issue date: 2017-05-25 / Os dois artigos desta tese, os capítulos 2 e 3, referem-se a testes de hipótese mas têm focos diferentes. O capítulo 2, intitulado "Improvements for external validity tests in fuzzy regression discontinuity designs," apresenta condições --- hipóteses de continuidade, monotonicidade estrita e convergência pontual --- sob as quais testes de qualidade de ajuste para duas amostras podem ser usados para testes de validade externa em modelos de tratamento-controle que sofrem de "compliance" imperfeito. Modelos com "compliance" imperfeito permitem a estimação de efeitos de tratamento apenas para a subpopulação de "compliers", sendo que tais estimativas não são necessariamente válidas para outras subpopulações ("always-takers" e "never-takers"). Sob as condições do capítulo 2, o uso do teste de qualidade de ajuste no lugar do teste de diferença de médias representa um avanço para testes de validade externa, uma vez que mais hipóteses alternativas são detectáveis pelo primeiro teste. Sugerimos combinar duas estatísticas de teste de qualidade de ajuste (uma para tratados e outra para não tratados) na forma de um teste múltiplo ao invés de um teste conjunto. O capítulo 3, intitulado "Higher-order UMP tests", sugere uma estratégia para se escolher, dentro de um conjunto de estatísticas de teste disponíveis, aquela que fornece o teste mais poderoso quando as funções de poder dos testes em questão não podem ser diferenciadas através de métodos assintóticos usuais, como análise de poder local ("local power analysis"). Propomos o uso de aproximações assintóticas de ordem mais alta, como expansões de Edgeworth, para se aproximar as densidades amostrais das estatísticas disponíveis e, com isso, verificar-se quais delas possuem a propriedade da razão monotônica de verossimilhança. Tal propriedade implica, pelo Teorema de Karlin-Rubin, que o teste é uniformemente mais poderoso (UMP) --- ao menos até certa ordem de aproximação --- se a estatística for suficiente para o parâmetro relevante. Para o caso em que as estatísticas sendo comparadas não são suficientes, argumentamos que frequentemente elas podem se tornar suficientes para uma família paramétrica de interesse após reparametrizações apropriadas. Para fins de ilustração, nós aplicamos o método proposto para determinar o valor ótimo, em termos de poder, do parâmetro de suavização do estimador de densidade por kernel em bases de dados simuladas e concluímos que a ordem de aproximação usada nesta aplicação (segunda ordem) não é alta o suficiente para permitir a diferenciação das funções de poder associadas aos diferentes valores do parâmetro de suavização. / The two papers in this work, chapters 2 and 3, regard hypothesis testing but address different issues. Chapter 2, entitled "Improvements for external validity tests in fuzzy regression discontinuity designs", shows conditions --- assumptions of continuity, strict monotonicity and pointwise convergence --- under which two-sample goodness-of-fit (GOF) tests can be used to test for external validity in treatment-control models that suffer from imperfect compliance of units with respect to the assigned treatment. Imperfect compliance allows researchers to estimate only treatment effects for the subpopulation of compliers, and the validity of these estimates for other subpopulations (always-takers and never-takers) remains an open problem. Under the conditions in Chapter 2, the use of GOF tests in place of mean difference tests represents an improvement over other external validity tests in the literature, since more alternative hypotheses are detectable by the test statistic. We suggested to combine two GOF test statistics (one for the treated and one for the untreated) in a multiple test instead of a joint test. Chapter 3, entitled "Higher-order UMP tests", suggests a strategy to choose among candidate test statistics, according to a power criterion, when their power performances are not distinguishable by usual methods of asymptotic comparison like local power analysis. We propose the use of higher-order asymptotic expansions, like Edgeworth expansions, to approximate the sample densities of the candidate test statistics and verify which of them has the monotone likelihood ratio property. This property implies, by the Karlin-Rubin Theorem, that the test is uniformly most powerful (UMP) --- at least to an order of approximation --- if the statistic is sufficient for the relevant parameter. When the statistics under study are not sufficient, we argue that they can often be made sufficient for a desired parametric family after appropriate reparameterization. We applied the method to search for the power-optimal bandwidth for the kernel density estimator in simulated data sets, and concluded that the order of approximation that we used (second order) is still too low to allow us to distinguish among bandwidths.

Introdução: Ruminação é a perseveração mal-adaptativa de pensamentos auto-centrados. Evidências sinalizam que ela está associada com início e manutenção de episódios depressivos, e ocorre em múltiplos transtornos mentais. A ruminação está associada com marcadores de desenvolvimento psicopatológico, como volumetria cerebral, memória, genes do BDNF e serotonina. É necessário aprofundar o conhecimento da ruminação enquanto traço dimensional, e conhecer melhor sua associação com variáveis sóciodemográficas, biológicas e clínicas para entender quando passa a ser um sintoma. Entretanto, aferi-la é um desafio, considerando que só existem escalas psicométricas. A mais utilizada, Ruminative Response Scale (RRS), foi validada em amostras não-clínicas. Objetivos: Avaliar ruminação transdiagnosticamente e determinar a validade de constructo da RRS em amostra clínica, buscando determinar fatores sócio-demográficos, clínicos e neurobiológicos associados a maiores escores de ruminação. Métodos: Estudo transversal, amostra não-probabilística. Foram convidados a participar 944 pacientes em atendimento psiquiátrico ambulatorial no HCPA entre março/2015 e junho/2016, maiores de 18 anos, que soubessem ler e escrever, e portadores de transtornos bipolares, depressão, esquizofrenia, esquizoafetivo, ansiedade generalizada, pânico, fobia específica e obsessivocompulsivo. Foram excluídos 373 com doenças que alteram resposta inflamatória, dependência química, gravidez, lactação, doenças neurológicas, vasculares e degenerativas. Recusaram-se a participar 254. Foram incluídos 317 pacientes, e 200 completaram a coleta de dados, que foi realizada em 4 etapas: 1) perfil sócio-demográfico e escalas auto-aplicáveis: ruminação, preocupação e funcionalidade; 2) amostras de sangue e entrevista clínica para aplicação das escalas de sintomas: depressão, mania, ansiedade e gravidade; 3) confirmação diagnóstica; e 4) processamento, armazenamento e análises bioquímicas das amostras de sangue. No primeiro artigo, revisamos sistematicamente a literatura sobre ruminação nos transtornos bipolares. No segundo, determinamos as validades de construto e externa da RRS. No terceiro, usamos machine learning para encontrar padrões de ruminação e determinar quais variáveis associadas preveem ruminação. Resultados: Ruminação está presente em todas as fases do transtorno bipolar, e é um sintoma estável independente do estado de humor, apesar de ter relação estreia com ele. Verificou-se também que mulheres ruminam mais que homens. Os escores de ruminação foram menores nos portadores de esquizofrenia que nos com depressão maior, bipolaridade e ansiedade. RRS apresentou boa confiabilidade, com 2-fatores correlacionados, brooding e ponderação, que apresentaram similaridade nas correlações com medidas clínicas, confirmando a validade externa transdiagnóstica. Por fim, encontrou-se que as variáveis associadas aos pacientes que mais ruminam são preocupação, sintomas de ansiedade generalizada e depressão, gravidade, nível socioeconômico e diagnóstico atual de pânico, sinalizando que ruminação pode ser um marcador de maior sensibilidade à ansiedade. Discussão: Ruminação parece ser um sintoma transdiagnóstico marcador de sofrimento. Os resultados desta tese contribuem para ampliar a discussão sobre diagnóstico psiquiátrico, agregando evidências para aprimorar as definições de limites e sobreposições diagnósticas entre as doenças mentais em que a ruminação ocorre. Por fim, conhecer melhor os mecanismos bioquímicos e clínicos envolvidos na ruminação contribuem na compreensão sobre quando ela deixa de ser um traço normal e vira um sintoma que necessita de tratamento. / Introduction: Rumination has been described as maladaptive perseveration of self-centered thoughts. Evidence indicates that rumination is associated with onset and maintenance of depressive episodes, it’s present in several mental disorders. Rumination is associated with markers of development of psychopathology, such as cerebral volumetry, memory, BDNF and serotonin genes. Measuring rumination is a challenge, considering that are available only psychometric scales. The most used, the Ruminative Responses Scale (RRS), was validated on non-clinical samples. Objectives: To evaluate transdiagnostically the rumination and to determine construct validity of the RRS in outpatients, in order to determine which associated factors lead the patients to ruminate. Methods: Cross-sectional study, non-probabilistic sample. A total of 944 patients in psychiatric outpatient treatment at HCPA between March / 2015 and June / 2016, major than 18 years old, knowing read and write, presenting bipolar disorder, schizophrenia, schizoaffective disorder, generalized anxiety disorder, panic disorder, phobia specific and obsessive-compulsive disorder were invited to participate. We excluded 373 patients with diseases that alter inflammatory response, chemical dependence, pregnancy, lactation, neurological, vascular and degenerative diseases. Two hundred fifty four refused to participate, 317 were included, and 200 completed the data collection, which was performed in 4 stages: 1) socio-demographic profile and self-applicable scales: rumination, worry and functionality; 2) blood samples and clinical interview for the application of symptom scales: depression, mania, anxiety and severity; 3) diagnostic confirmation; and 4) processing, storage and biochemical analyzes of blood samples. In the first article, we systematically reviewed the literature on rumination in bipolar disorders. In the second, we evaluated construct and external validity of RRS. In the third, we used machine learning algorithms to find patterns of rumination and to determine which associated variables predict rumination. Results: Rumination is present in all phases of bipolar disorder, it is a stable symptom, independent of mood, despite it has close relationship with it. It has also been found that women ruminate more than men. Rumination scores were lower in patients with schizophrenia than in major depression, bipolarity and anxiety patients. RRS presented good reliability, with correlated 2-factors, brooding and pondering, which presented similar correlations with clinical measures, confirming the external transdiagnostic validity. Finally, it was found that the variables associated with the greater scores of rumination are worry, symptoms of generalized anxiety and depression, severity of symptoms, socioeconomic level and current diagnosis of panic, signaling that rumination may be a marker of greater sensitivity to anxiety. Discussion: Rumination seems to be a transdiagnostic symptom of suffering. The results of this thesis contribute to broadening the discussion about psychiatric diagnostic, adding evidence to improve the definitions of limits and diagnostic overlaps between mental illnesses in which rumination occurs. Finally, a better understanding of the biochemical and clinical mechanisms involved in rumination may contribute to understanding of when rumination ceases to be a normal trait and becomes a symptom that requires treatment.

Introdução: Ruminação é a perseveração mal-adaptativa de pensamentos auto-centrados. Evidências sinalizam que ela está associada com início e manutenção de episódios depressivos, e ocorre em múltiplos transtornos mentais. A ruminação está associada com marcadores de desenvolvimento psicopatológico, como volumetria cerebral, memória, genes do BDNF e serotonina. É necessário aprofundar o conhecimento da ruminação enquanto traço dimensional, e conhecer melhor sua associação com variáveis sóciodemográficas, biológicas e clínicas para entender quando passa a ser um sintoma. Entretanto, aferi-la é um desafio, considerando que só existem escalas psicométricas. A mais utilizada, Ruminative Response Scale (RRS), foi validada em amostras não-clínicas. Objetivos: Avaliar ruminação transdiagnosticamente e determinar a validade de constructo da RRS em amostra clínica, buscando determinar fatores sócio-demográficos, clínicos e neurobiológicos associados a maiores escores de ruminação. Métodos: Estudo transversal, amostra não-probabilística. Foram convidados a participar 944 pacientes em atendimento psiquiátrico ambulatorial no HCPA entre março/2015 e junho/2016, maiores de 18 anos, que soubessem ler e escrever, e portadores de transtornos bipolares, depressão, esquizofrenia, esquizoafetivo, ansiedade generalizada, pânico, fobia específica e obsessivocompulsivo. Foram excluídos 373 com doenças que alteram resposta inflamatória, dependência química, gravidez, lactação, doenças neurológicas, vasculares e degenerativas. Recusaram-se a participar 254. Foram incluídos 317 pacientes, e 200 completaram a coleta de dados, que foi realizada em 4 etapas: 1) perfil sócio-demográfico e escalas auto-aplicáveis: ruminação, preocupação e funcionalidade; 2) amostras de sangue e entrevista clínica para aplicação das escalas de sintomas: depressão, mania, ansiedade e gravidade; 3) confirmação diagnóstica; e 4) processamento, armazenamento e análises bioquímicas das amostras de sangue. No primeiro artigo, revisamos sistematicamente a literatura sobre ruminação nos transtornos bipolares. No segundo, determinamos as validades de construto e externa da RRS. No terceiro, usamos machine learning para encontrar padrões de ruminação e determinar quais variáveis associadas preveem ruminação. Resultados: Ruminação está presente em todas as fases do transtorno bipolar, e é um sintoma estável independente do estado de humor, apesar de ter relação estreia com ele. Verificou-se também que mulheres ruminam mais que homens. Os escores de ruminação foram menores nos portadores de esquizofrenia que nos com depressão maior, bipolaridade e ansiedade. RRS apresentou boa confiabilidade, com 2-fatores correlacionados, brooding e ponderação, que apresentaram similaridade nas correlações com medidas clínicas, confirmando a validade externa transdiagnóstica. Por fim, encontrou-se que as variáveis associadas aos pacientes que mais ruminam são preocupação, sintomas de ansiedade generalizada e depressão, gravidade, nível socioeconômico e diagnóstico atual de pânico, sinalizando que ruminação pode ser um marcador de maior sensibilidade à ansiedade. Discussão: Ruminação parece ser um sintoma transdiagnóstico marcador de sofrimento. Os resultados desta tese contribuem para ampliar a discussão sobre diagnóstico psiquiátrico, agregando evidências para aprimorar as definições de limites e sobreposições diagnósticas entre as doenças mentais em que a ruminação ocorre. Por fim, conhecer melhor os mecanismos bioquímicos e clínicos envolvidos na ruminação contribuem na compreensão sobre quando ela deixa de ser um traço normal e vira um sintoma que necessita de tratamento. / Introduction: Rumination has been described as maladaptive perseveration of self-centered thoughts. Evidence indicates that rumination is associated with onset and maintenance of depressive episodes, it’s present in several mental disorders. Rumination is associated with markers of development of psychopathology, such as cerebral volumetry, memory, BDNF and serotonin genes. Measuring rumination is a challenge, considering that are available only psychometric scales. The most used, the Ruminative Responses Scale (RRS), was validated on non-clinical samples. Objectives: To evaluate transdiagnostically the rumination and to determine construct validity of the RRS in outpatients, in order to determine which associated factors lead the patients to ruminate. Methods: Cross-sectional study, non-probabilistic sample. A total of 944 patients in psychiatric outpatient treatment at HCPA between March / 2015 and June / 2016, major than 18 years old, knowing read and write, presenting bipolar disorder, schizophrenia, schizoaffective disorder, generalized anxiety disorder, panic disorder, phobia specific and obsessive-compulsive disorder were invited to participate. We excluded 373 patients with diseases that alter inflammatory response, chemical dependence, pregnancy, lactation, neurological, vascular and degenerative diseases. Two hundred fifty four refused to participate, 317 were included, and 200 completed the data collection, which was performed in 4 stages: 1) socio-demographic profile and self-applicable scales: rumination, worry and functionality; 2) blood samples and clinical interview for the application of symptom scales: depression, mania, anxiety and severity; 3) diagnostic confirmation; and 4) processing, storage and biochemical analyzes of blood samples. In the first article, we systematically reviewed the literature on rumination in bipolar disorders. In the second, we evaluated construct and external validity of RRS. In the third, we used machine learning algorithms to find patterns of rumination and to determine which associated variables predict rumination. Results: Rumination is present in all phases of bipolar disorder, it is a stable symptom, independent of mood, despite it has close relationship with it. It has also been found that women ruminate more than men. Rumination scores were lower in patients with schizophrenia than in major depression, bipolarity and anxiety patients. RRS presented good reliability, with correlated 2-factors, brooding and pondering, which presented similar correlations with clinical measures, confirming the external transdiagnostic validity. Finally, it was found that the variables associated with the greater scores of rumination are worry, symptoms of generalized anxiety and depression, severity of symptoms, socioeconomic level and current diagnosis of panic, signaling that rumination may be a marker of greater sensitivity to anxiety. Discussion: Rumination seems to be a transdiagnostic symptom of suffering. The results of this thesis contribute to broadening the discussion about psychiatric diagnostic, adding evidence to improve the definitions of limits and diagnostic overlaps between mental illnesses in which rumination occurs. Finally, a better understanding of the biochemical and clinical mechanisms involved in rumination may contribute to understanding of when rumination ceases to be a normal trait and becomes a symptom that requires treatment.

Meta-analysis was used to investigate the relationship between visuo-spatial ability and performance in remote environments. In order to be included, each study needed to examine the relationship between the use of an ego-centric perspective and various dimensions of performance (i.e., identification, localization, navigation, and mission completion time). The moderator analysis investigated relationships involving: (a) visuo-spatial construct with an emphasis on Carroll’s (1993) visualization (VZ) factor; (b) performance outcome (i.e., identification, localization, navigation, and mission completion time); (c) autonomy to support mission performance; (d) task type (i.e., navigation vs. reconnaissance); and (e) experimental testbed (i.e., physical vs. virtual environments). The process of searching and screening for published and unpublished analyses identified 81 works of interest that were found to represent 50 unique datasets. 518 effects were extracted from these datasets for analyses. Analyses of aggregated effects (Hunter & Schmidt, 2004) found that visuo-spatial abilities were significantly associated with each construct, such that effect sizes ranged from weak (r = .235) to moderately strong (r = .371). For meta-regression (Borenstein, Hedges, Figgins, & Rothstein, 2009; Kalaian & Raudenbush, 1996; Tabachnick & Fidell, 2007), moderation by visuo-spatial construct (i.e., focusing on visualization) was consistently supported for all outcomes. For at least one of the outcomes, support was found for moderation by test, the reliability coefficient of a test, autonomy (i.e. to support identification, localization, and navigation), testbed (i.e., physical vs. virtual environment), intended domain of application, and gender. These findings illustrate that majority of what researchers refer to as “spatial ability” actually uses measures that load onto Carroll’s (1993) visualization (VZ) factor. The associations between this predictor and all performance outcomes were significant, but the significant iv variation across moderators highlight important issues for the design of unmanned systems and the external validity of findings across domains. For example, higher levels of autonomy for supporting navigation decreased the association between visualization (VZ) and performance. In contrast, higher levels of autonomy for supporting identification and localization increased the association between visualization (VZ) and performance. Furthermore, moderation by testbed, intended domain of application, and gender challenged the degree to which findings can be expected to generalize across domains and sets of participants.