Introdução: Estudos anteriores sugerem possível benefício do uso perioperatório das estatinas para redução de eventos cardiovasculares pósoperatórios. Metanálise previamente publicada demonstra que as estatinas reduzem mortalidade e infarto perioperatório em pacientes submetidos a cirurgia não-cardíaca. Entretanto, novas evidências em cirurgia cardíaca sugerem efeitos neutros ou mesmo deletérios das estatinas, sendo que metanálise recente demonstrou aumento da incidência de insuficiência renal aguda e tendência a maior mortalidade pós-operatória. Devido aos resultados conflitantes e escassez de evidências definitivas, foi realizada uma revisão sistemática e metanálise de estudos randomizados para avaliar os efeitos do uso perioperatório das estatinas em cirurgia cardíaca e nãocardíaca. Objetivo: Avaliar em pacientes adultos submetidos à cirurgia cardíaca e não-cardíaca a associação entre o uso perioperatório de estatinas e a incidência pós-operatória de infarto agudo do miocárdio, acidente vascular cerebral, insuficiência renal aguda e mortalidade. Metodologia: As bases de dados Pubmed, EMBASE e Cochrane foram avaliadas por 2 pesquisadores independentes até 1º de maio de 2018 para busca de artigos apropriados. Foram incluídos estudos randomizados que avaliaram o uso perioperatório de estatinas comparadas a placebo ou a nenhum tratamento em pacientes submetidos a cirurgia cardíaca e nãocardíaca. Foram calculados o risco relativo (RR) ou razão de chances (odds ratio, OR) e intervalo de confiança 95% (IC 95%) por meio de metanálise de efeitos fixos. Foi realizada a análise sequencial dos estudos (trial sequential analysis, TSA) para quantificar a confiabilidade estatística dos dados. A metodologia Cochrane foi utilizada. Os desfechos primários foram definidos como infarto agudo do miocárdio, acidente vascular cerebral, insuficiência renal aguda e mortalidade no período pós-operatório. Resultados: Foram incluídos 35 estudos randomizados totalizando 8200 pacientes. O uso de estatina comparado ao controle foi associado a menor incidência de infarto em pacientes submetidos a cirurgia não-cardíaca (OR=0,44 [IC 95%, 0,30 a 0,64], p < 0,0001), mas não em pacientes submetidos a cirurgia cardíaca (OR=0,93 [IC 95%, 0,70 a 1,24], p=0,61). O uso da estatina foi associado a maior incidência de insuficiência renal aguda em cirurgia cardíaca (RR=1,15 [IC 95%, 1,00 a 1,31], p=0,05), mas não em cirurgia não-cardíaca (RR=1,52 [IC 95%, 0,71 a 3,26], p=0,28). Não foi observado impacto do uso das estatinas na ocorrência de acidente vascular cerebral e mortalidade em ambos os grupos cirúrgicos. No entanto, a análise dos estudos com baixo risco de viés em cirurgia cardíaca demonstrou maior mortalidade com o uso de estatinas comparadas a placebo (OR=3,71 [IC 95%, 1,03 a 13,34], p=0,04). A análise sequencial dos estudos (TSA) não sugeriu conclusões definitivas sobre o assunto. Conclusão: As estatinas parecem ter efeito protetor contra infarto pós-operatório em cirurgia não-cardíaca, mas estão associadas a maior risco de insuficiência renal aguda em cirurgia cardíaca. Possíveis efeitos positivos ou negativos sobre a mortalidade não podem ser excluídos. Os dados dos estudos randomizados disponíveis até o momento ainda são insuficientes para conclusões definitivas sobre o uso perioperatório das estatinas. Estudos randomizados adicionais são necessários para avaliar o perfil de segurança e possíveis efeitos benéficos destas medicações nos desfechos pós-operatórios / Introduction: Previous studies supported potential beneficial effects of perioperative statin therapy to reduce postoperative complications. Accordingly, previous meta-analysis concluded that statin treatment decreases the perioperative incidence of mortality and myocardial infarction in non-cardiac surgery. In contrast, growing evidences on perioperative statins administration in cardiac surgery setting suggested neutral or even detrimental results. A recent systematic review and meta-analysis found that perioperative statin therapy in this population was associated with an increased incidence of postoperative acute kidney injury and a trend toward increased mortality. Due to the contrasting results and lack of definitive evidence, we performed a systematic review and meta-analysis of randomized controlled trials to examine the effects of perioperative statin therapy on postoperative outcomes in adult cardiac and non-cardiac surgery patients. Objective: To assess the association between perioperative statin therapy and postoperative myocardial infarction, stroke, acute kidney injury and mortality in patients submitted to cardiac and non-cardiac surgery. Methods: Pubmed, EMBASE, and Cochrane databases were searched through May 1st, 2018 for appropriate articles. Articles were independently assessed by 2 reviewers. Randomized controlled trials evaluating adult cardiac and non-cardiac surgery patients comparing perioperative statin therapy versus placebo or no treatment were included. Risk ratio (RR) or odds ratio (OR) and 95% confidence interval (CI) were obtained using fixedeffects meta-analyses. Trial sequential analysis (TSA) was performed to quantify the statistical reliability of data. The Cochrane methodology was used. Main outcomes were postoperative myocardial infarction, stroke, acute kidney injury, and mortality. Results: Data from 35 randomized controlled trials involving 8200 patients were included. Perioperative statin therapy was associated with lower incidence of postoperative myocardial infarction in noncardiac surgery compared to control (OR=0.44 [95% CI, 0.30 a 0.64], p < 0.0001), but not in cardiac surgery OR=0.93 [95% CI, 0.70 a 1.24], p=0.61). Higher incidence of acute kidney injury was evident in cardiac surgery patients receiving perioperative statins (RR=1.15 [95% CI, 1.00 a 1.31], p=0.05), but not in the non-cardiac surgery population (RR=1.52 [95% CI, 0.71 a 3.26], p=0.28). No difference in postoperative stroke and mortality was present in patients undergoing either cardiac or non-cardiac surgery. However, low risk of bias trials performed in cardiac surgery showed a higher mortality with statins versus placebo (OR=3.71 [95% CI, 1.03 a 13.34], p=0.04). Trial sequential analysis suggested no firm conclusions on the topic. Conclusions: Statins appear to be protective against postoperative myocardial infarction in non-cardiac surgery and associated with an increased risk of acute kidney injury in cardiac surgery. Possible positive or even negative effects on mortality could not be excluded. There is still insufficient randomized data for firm conclusions on perioperative statin therapy. Further randomized controlled trials should evaluate both the safety profile and possible effects on patients\' outcomes

Cette thèse porte sur le développement de méthodes semi-paramétriques robustes pour l'inférence de paramètres complexes émergeant à l'interface de l'inférence causale et la biostatistique. Ses motivations sont les applications à la recherche épidémiologique et médicale à l'ère des Big Data. Nous abordons plus particulièrement deux défis statistiques pour réconcilier, dans chaque contexte, estimation adaptative et inférence statistique. Le premier défi concerne la maximisation de l'information tirée d'essais contrôlés randomisés (ECRs) grâce à la conception d'essais adaptatifs. Nous présentons un cadre théorique pour la construction et l'analyse d'ECRs groupes-séquentiels, réponses-adaptatifs et ajustés aux covariable (traduction de l'expression anglaise « group-sequential, response-adaptive, covariate-adjusted », d'où l'acronyme CARA) qui permettent le recours à des procédures adaptatives d'estimation à la fois pour la construction dynamique des schémas de randomisation et pour l'estimation du modèle de réponse conditionnelle. Ce cadre enrichit la littérature existante sur les ECRs CARA notamment parce que l'estimation des effets est garantie robuste même lorsque les modèles sur lesquels s'appuient les procédures adaptatives d'estimation sont mal spécificiés. Le second défi concerne la mise au point et l'étude asymptotique d'une procédure inférentielle semi-paramétrique avec estimation adaptative des paramètres de nuisance. A titre d'exemple, nous choisissons comme paramètre d'intérêt la différence des risques marginaux pour un traitement binaire. Nous proposons une version cross-validée du principe d'inférence par minimisation ciblée de pertes (« Cross-validated Targeted Mimum Loss Estimation » en anglais, d'où l'acronyme CV-TMLE) qui, comme son nom le suggère, marie la procédure TMLE classique et le principe de la validation croisée. L'estimateur CV-TMLE ainsi élaboré hérite de la propriété typique de double-robustesse et aussi des propriétés d'efficacité du TMLE classique. De façon remarquable, le CV-TMLE est linéairement asymptotique sous des conditions minimales, sans recourir aux conditions de type Donsker. / This dissertation focuses on developing robust semiparametric methods for complex parameters that emerge at the interface of causal inference and biostatistics, with applications to epidemiological and medical research in the era of Big Data. Specifically, we address two statistical challenges that arise in bridging the disconnect between data-adaptive estimation and statistical inference. The first challenge arises in maximizing information learned from Randomized Control Trials (RCT) through the use of adaptive trial designs. We present a framework to construct and analyze group sequential covariate-adjusted response-adaptive (CARA) RCTs that admits the use of data-adaptive approaches in constructing the randomization schemes and in estimating the conditional response model. This framework adds to the existing literature on CARA RCTs by allowing flexible options in both their design and analysis and by providing robust effect estimates even under model mis-specifications. The second challenge arises from obtaining a Central Limit Theorem when data-adaptive estimation is used to estimate the nuisance parameters. We consider as target parameter of interest the marginal risk difference of the outcome under a binary treatment, and propose a Cross-validated Targeted Minimum Loss Estimator (TMLE), which augments the classical TMLE with a sample-splitting procedure. The proposed Cross-Validated TMLE (CV-TMLE) inherits the double robustness properties and efficiency properties of the classical TMLE , and achieves asymptotic linearity at minimal conditions by avoiding the Donsker class condition.

Background and Purpose: To investigate the effect of an individualized exercise program versus a generic group-based exercise program on balance, gait, and functional performance of older adults categorized as having mild balance dysfunction and living in residential care facilities. Methods: Single blind randomized control design. One hundred-twenty residents fulfilled screening criteria for mild balance dysfunction based on the BioSwayTM balance and the Multi-Directional Reach Test (MDRT) primary outcome measures. Secondary assessment was completed using the Modified Physical Performance Test (PPT), hand-held dynamometer (lower-limb muscle strength testing), and gait speed analysis. Sixty subjects received individualized treatment from physical therapists (8 weeks). Another sixty subjects received generic group-based exercises (8 weeks). All outcome measures were collected at baseline and post-intervention (ninth week); and BioSwayTM and PPT measures at follow-up (thirteenth week) for the individualized group. Results: Individualized group (n=60) showed significant improvement compared to the group-based group (n=60) on the two BioSwayTM scores (limits of stability, p < .001; and postural stability, p = .016), the MDRT scores (forward reach, p < .001; backward reach, p = .007; right lateral reach, p < .001; and left lateral reach p < .001), the strength scores (hip flexors, p = .010; knee extensors, p = .002; hip abductors, p = .009; and ankle dorsiflexors, p = .025), the PPT outcomes (p < .001), and the gait scores (p = .012). Effect sizes ranged from small to large, with the largest sizes for limits of stability and MDRT. There were no significant differences between groups for the mCTSIB (p = .538). However, 96.7% of subjects in the individualized group scored within one SD of the reference mean, relative to 75% in the group-based group. At follow-up, the individualized group showed significant differences over time with medium to large effect sizes on the PPT (p < .001), limits of stability (p < .001), postural stability (p < .001), and mCTSIB (p = .005) measures. Post-hoc analysis revealed retention of gains for all measures at follow-up, except the mCTSIB. Conclusion: The individualized group showed significant improvements in the areas of balance, strength, mobility, and functional outcomes.

No description available.

No description available.

Essa dissertação tem como objetivo analisar o comportamento manipulador relacionado à aplicação de instrumentos de mensuração de competências socioemocionais no ambiente escolar estimulado por duas abordagens que configuram contextos low-stake: o efeito da disseminação da informação sobre essas competências e suas relações com bem-estar e progresso social e da possibilidade dos professores receberem as respostas individuais dos alunos. Para isso, realizou-se um experimento com 2853 alunos de 10 escolas públicas da rede estadual de São Paulo. Os resultados indicam que os alunos são capazes de inflar suas respostas no sentido da desejabilidade social e que as duas abordagens fazem com que os alunos se sintam mais motivados em manipular suas respostas. Além disso, os resultados mostram que a inclusão de uma notificação sobre um mecanismo de detecção de respostas falsas no questionário cria uma barreira e desestimula o comportamento manipulador. / This dissertation aims to examine faking behaviour in personality assessments at schools. Two approaches that create low-stakes environments are investigated: information spreading about non-cognitive skills and their impact on well-being and social progress and the fact that teachers would receive individual students\' answers. Using a CRT technique with 2853 students from 10 São Paulo state public schools, we find that students are capable to fake their answers and these two approaches turn them more motivated to engage in faking. Moreover, we show that a warning about untruthful answers attached to the questionnaire imposes a cost, discouraging faking behaviour

INTRODUÇÃO: Pacientes com diagnóstico de parafilias e de transtornos relacionados às parafilias ocupam a atenção pública e clínica por se tratarem de indivíduos potencialmente inconvenientes, causadores de problemas à sociedade e a si próprios. Caracteristicamente, apresentam padrões invulgares de comportamento sexual e distinguem-se pela escolha de parcerias ou de contextos sexuais incomuns. JUSTIFICATIVA: O tratamento atual leva em consideração comportamentos, fantasias e necessidades sexuais parafílicas e procura modificá-los por meio de psicoterapia cognitivocomportamental, medicação antidepressiva e antiandrogênica. Apenas a supressão destes comportamentos tem se mostrado insatisfatória e ainda apresenta altas taxas de reincidência. Não há resultados publicados sobre a efetividade de um novo tratamento psicoterapêutico, cujo foco esteja voltado para a identificação e a reestruturação da identidade sexual e cujo principal objetivo seja o de promover maior contato destes pacientes com parcerias sexuais convencionais. MÉTODO: A intervenção foi testada por meio de ensaio clínico de tratamento, randomizado, placebo controlado e unicego, realizado com dois grupos paralelos: experimental (N=6) e controle (N=8). Todos os pacientes apresentavam diagnóstico de parafilias e foram medicados com inibidores seletivos da recaptação de serotonina. As intervenções psicoterapêuticas duraram dezesseis semanas com frequência de uma sessão por semana. As mudanças comportamentais foram monitoradas por meio de um diário e as comorbidades psiquiátricas, a qualidade de vida e a adequação social foram avaliadas por meio de escalas psicométricas, nos três períodos do estudo: início, término e seguimento pósintervenção. RESULTADOS: Não houve melhora estatisticamente significante dos escores de depressão, qualidade de vida e adequação social para ambos os grupos. O grupo experimental apresentou tendência de melhor resposta para depressão ao se tomar esta variável de forma categorial. Ambos os grupos apresentaram: maior incapacidade de adequação social (comparados a populações não clínicas); tendência de aumento do estado de ansiedade; melhora clínica global e redução dos índices de gravidade das parafilias. Houve diferenças estatisticamente significantes entre os grupos. Para o grupo experimental: aumento de fantasias e masturbações normofílicas; maior procura por contatos eróticos normofílicos e diminuição de fantasias parafílicas. Porém, houve diminuição de contatos íntimos normofílicos (beijos, abraços e carícias) para este grupo. Houve diferenças estatisticamente significantes intragrupos. Para o grupo experimental: redução de contatos íntimos parafílicos e totais (parafílicos e normofílicos). Para o grupo placebo: aumento de relações sexuais parafílicas e da totalidade das fantasias sexuais (parafílicas e normofílicas). CONCLUSÕES: Os defechos favoráveis observados no grupo de intervenção evidenciaram a importância em se identificar e trabalhar psicoterapeuticamente o impedimento que os parafílicos apresentam de se relacionar com parcerias sexuais normofílicas, uma vez que para o grupo experimental as mudanças positivas observadas no comportamento, fantasias e necessidades sexuais caminharam em direção a padrões convencionais / INTRODUCTION: Patients diagnosed with paraphilia and paraphilia-related disorders have public and medical attention drawn to them because they are potentially disturbing individuals who cause problems to society and to themselves. Typically, they show unusual patterns of sexual behavior and make unusual choices related to sexual partners or sexual contexts. RATIONALE: Current treatment considers paraphilic behaviors, fantasies, and urges, trying to change these symptoms using cognitive-behavioral psychotherapy, antidepressants, and anti-androgen medication. The elimination of these behaviors has proved unsatisfactory and still has high recurrence rates. We could not find in the literature any studies on the effectiveness of a new psychotherapeutic treatment focused on identifying and restructuring the patients\' sexual identity to allow for the promotion of contact between these patients and conventional sexual partners. METHOD: We conducted a randomized, placebo-controlled, single-blind clinical trial with two groups: experimental (N = 6) and control (N = 8). All patients were diagnosed with paraphilia and were treated with selective serotonin reuptake inhibitors. The psychotherapeutic interventions lasted sixteen weeks, and its frequency was one session a week. Behavioral changes were monitored using a diary, whereas psychiatric comorbidities, quality of life, and social adequacy were assessed using psychometric scales in the three study periods: baseline, end, and post-intervention follow-up. RESULTS: There was no significant improvement of depression, quality of life, and social adequacy scores for both groups. The experimental group showed a trend toward better response to depression when we considered this as categorical variable. Both groups showed higher inability of social adequacy (when compared with non-clinical populations); tendency of increased state of anxiety; improvement of overall clinical status, and reduced levels of paraphilia severity. There were statistically significant differences between the groups. The experimental group showed increased frequency of normophilic fantasies and masturbation; increased demand for normophilic erotic contact, and decreased paraphilic fantasies. However, there was a decrease in normophilic intimate contact (kisses, hugs, and caresses) in this group. There were statistically significant intragroup differences. The experimental group had reduced paraphilic and total (paraphilic and normophilic) intimate contact. The placebo group had increased paraphilic sexual intercourse and higher frequency of sexual fantasies (paraphilic and normophilic). CONCLUSIONS: The positive outcomes found in the intervention group showed the importance of psychotherapy in identifying and treating the paraphilic patients\' impediments that prevent them from having relationships with normophilic sexual partners. This was reinforced by the finding that the experimental group showed positive changes in terms of sexual behaviors, fantasies, and urges, which were closer to conventional standards

O envelhecimento populacional é uma preocupação mundial, acarretando prejuízos pulmonares. A cinesioterapia respiratória é muito utilizada, mas pouco relacionada com o meio aquático e idosos. Este estudo analisou os efeitos da cinesioterapia respiratória realizada bem dois meios - aquático e terrestre - na força muscular respiratória, flexibilidade e expansibilidade torácica de idosos. Completaram o estudo 59 sujeitos entre 60 e 65 anos, randomizados em três grupos. Dois deles realizaram exercícios respiratórios em meios aquático e terrestre. O terceiro atuou como controle. Comparando os grupos, observou-se uma melhora estatisticamente significativa da força muscular inspiratória do grupo que realizou atividades na água. As demais variáveis não sofreram alteração significativa / Aging of the population is a mundial concern. It causes damages in the pulmonary system. Respiratory exercises are too much used, but not so much related with the aquatic environment and aged. This study analised the effects of the respiratory exercise realized in two environments - aquatic and earth - in the respiratory muscle strength, thoracic mobility and flexibility in healthy aged. Completed the study 59 subjects between 60 and 65 years, randomized in three groups. Two of then realized respiratory exercises in earth and water and the third one acted as a control. Comparing the groups, there was a stastically significance improve in the inspiratory muscle strength in the water group. All of the other variables analised didn't change with the intervention

INTRODUÇÃO: Pacientes com diagnóstico de parafilias e de transtornos relacionados às parafilias ocupam a atenção pública e clínica por se tratarem de indivíduos potencialmente inconvenientes, causadores de problemas à sociedade e a si próprios. Caracteristicamente, apresentam padrões invulgares de comportamento sexual e distinguem-se pela escolha de parcerias ou de contextos sexuais incomuns. JUSTIFICATIVA: O tratamento atual leva em consideração comportamentos, fantasias e necessidades sexuais parafílicas e procura modificá-los por meio de psicoterapia cognitivocomportamental, medicação antidepressiva e antiandrogênica. Apenas a supressão destes comportamentos tem se mostrado insatisfatória e ainda apresenta altas taxas de reincidência. Não há resultados publicados sobre a efetividade de um novo tratamento psicoterapêutico, cujo foco esteja voltado para a identificação e a reestruturação da identidade sexual e cujo principal objetivo seja o de promover maior contato destes pacientes com parcerias sexuais convencionais. MÉTODO: A intervenção foi testada por meio de ensaio clínico de tratamento, randomizado, placebo controlado e unicego, realizado com dois grupos paralelos: experimental (N=6) e controle (N=8). Todos os pacientes apresentavam diagnóstico de parafilias e foram medicados com inibidores seletivos da recaptação de serotonina. As intervenções psicoterapêuticas duraram dezesseis semanas com frequência de uma sessão por semana. As mudanças comportamentais foram monitoradas por meio de um diário e as comorbidades psiquiátricas, a qualidade de vida e a adequação social foram avaliadas por meio de escalas psicométricas, nos três períodos do estudo: início, término e seguimento pósintervenção. RESULTADOS: Não houve melhora estatisticamente significante dos escores de depressão, qualidade de vida e adequação social para ambos os grupos. O grupo experimental apresentou tendência de melhor resposta para depressão ao se tomar esta variável de forma categorial. Ambos os grupos apresentaram: maior incapacidade de adequação social (comparados a populações não clínicas); tendência de aumento do estado de ansiedade; melhora clínica global e redução dos índices de gravidade das parafilias. Houve diferenças estatisticamente significantes entre os grupos. Para o grupo experimental: aumento de fantasias e masturbações normofílicas; maior procura por contatos eróticos normofílicos e diminuição de fantasias parafílicas. Porém, houve diminuição de contatos íntimos normofílicos (beijos, abraços e carícias) para este grupo. Houve diferenças estatisticamente significantes intragrupos. Para o grupo experimental: redução de contatos íntimos parafílicos e totais (parafílicos e normofílicos). Para o grupo placebo: aumento de relações sexuais parafílicas e da totalidade das fantasias sexuais (parafílicas e normofílicas). CONCLUSÕES: Os defechos favoráveis observados no grupo de intervenção evidenciaram a importância em se identificar e trabalhar psicoterapeuticamente o impedimento que os parafílicos apresentam de se relacionar com parcerias sexuais normofílicas, uma vez que para o grupo experimental as mudanças positivas observadas no comportamento, fantasias e necessidades sexuais caminharam em direção a padrões convencionais / INTRODUCTION: Patients diagnosed with paraphilia and paraphilia-related disorders have public and medical attention drawn to them because they are potentially disturbing individuals who cause problems to society and to themselves. Typically, they show unusual patterns of sexual behavior and make unusual choices related to sexual partners or sexual contexts. RATIONALE: Current treatment considers paraphilic behaviors, fantasies, and urges, trying to change these symptoms using cognitive-behavioral psychotherapy, antidepressants, and anti-androgen medication. The elimination of these behaviors has proved unsatisfactory and still has high recurrence rates. We could not find in the literature any studies on the effectiveness of a new psychotherapeutic treatment focused on identifying and restructuring the patients\' sexual identity to allow for the promotion of contact between these patients and conventional sexual partners. METHOD: We conducted a randomized, placebo-controlled, single-blind clinical trial with two groups: experimental (N = 6) and control (N = 8). All patients were diagnosed with paraphilia and were treated with selective serotonin reuptake inhibitors. The psychotherapeutic interventions lasted sixteen weeks, and its frequency was one session a week. Behavioral changes were monitored using a diary, whereas psychiatric comorbidities, quality of life, and social adequacy were assessed using psychometric scales in the three study periods: baseline, end, and post-intervention follow-up. RESULTS: There was no significant improvement of depression, quality of life, and social adequacy scores for both groups. The experimental group showed a trend toward better response to depression when we considered this as categorical variable. Both groups showed higher inability of social adequacy (when compared with non-clinical populations); tendency of increased state of anxiety; improvement of overall clinical status, and reduced levels of paraphilia severity. There were statistically significant differences between the groups. The experimental group showed increased frequency of normophilic fantasies and masturbation; increased demand for normophilic erotic contact, and decreased paraphilic fantasies. However, there was a decrease in normophilic intimate contact (kisses, hugs, and caresses) in this group. There were statistically significant intragroup differences. The experimental group had reduced paraphilic and total (paraphilic and normophilic) intimate contact. The placebo group had increased paraphilic sexual intercourse and higher frequency of sexual fantasies (paraphilic and normophilic). CONCLUSIONS: The positive outcomes found in the intervention group showed the importance of psychotherapy in identifying and treating the paraphilic patients\' impediments that prevent them from having relationships with normophilic sexual partners. This was reinforced by the finding that the experimental group showed positive changes in terms of sexual behaviors, fantasies, and urges, which were closer to conventional standards

La méta-analyse (MA) en réseau, en généralisant la MA conventionnelle, permet d'évaluer toutes les comparaisons deux à deux possibles entre interventions. Les biais de publication ont reçu peu d’attention dans ce contexte. Nous avons évalué l’impact des biais de publication en utilisant un réseau de 74 essais randomisés évaluant 12 antidépresseurs contre placebo enregistrés à la FDA et un réseau de 51 essais parmi les 74 dont les résultats étaient publiés. Nous avons montré comment les biais de publication biaisaient les quantités d'effet estimées et le classement des traitements. L'effet du biais de publication peut différer entre MA en réseau et MA conventionnelle en ce que les biais affectant un traitement peuvent affecter le classement de tous les traitements. Nous avons ensuite généralisé un test de détection des biais à la MA en réseau. Il est basé sur la comparaison entre les nombres attendu et observé d’essais avec résultats statistiquement significatifs sur l’ensemble du réseau. Nous avons montré par des études de simulation que le test proposé avait une puissance correcte après ajustement sur l’erreur de type I, excepté lorsque la variance inter-essais était élevée. Par ailleurs, le test indiquait un signal significatif de biais sur le réseau d’essais d’antidépresseurs publiés. Enfin, nous avons introduit deux modèles d’analyse de sensibilité des résultats d'une MA en réseau aux biais de publication: un modèle de méta-régression qui relie la quantité d’effet estimée à son erreur standard, et un modèle de sélection dans lequel on estime la propension d’un essai à être publié puis l’on redresse le poids des essais en fonction de cette propension. Nous les avons appliqués aux réseaux d’essais d’antidépresseurs. Ce test et ces modèles d'ajustement tirent leur force de tous les essais du réseau, sous l’hypothèse qu'un biais moyen commun opère sur toutes les branches du réseau. / Network meta-analysis (NMA), a generalization of conventional MA, allows for assessing all possible pairwise comparisons between multiple treatments. Reporting bias, a major threat to the validity of MA, has received little attention in the context of NMA. We assessed the impact of reporting bias empirically using data from 74 FDA-registered placebo-controlled trials of 12 antidepressants and their 51 matching publications. We showed how reporting bias biased NMA-based estimates of treatments efficacy and modified ranking. The effect of reporting bias in NMAs may differ from that in classical meta-analyses in that reporting bias affecting only one drug may affect the ranking of all drugs. Then, we extended a test to detect reporting bias in network of trials. It compares the number of expected trials with statistically significant results to the observed number of trials with significant p-values across the network. We showed through simulation studies that the test was fairly powerful after adjustment for size, except when between-trial variance was substantial. Besides, it showed evidence of bias in the network of published antidepressant trials. Finally, we introduced two methods of sensitivity analysis for reporting bias in NMA: a meta-regression model that allows the effect size to depend on its standard error and a selection model that estimates the propensity of trial results being published and in which trials with lower propensity are weighted up in the NMA model. We illustrated their use on the antidepressant datasets. The proposed test and adjustment models borrow strength from all trials across the network, under the assumption that conventional MAs in the network share a common mean bias mechanism.