This is a sandwich thesis / In an interrupted time series (ITS) design, data are collected at multiple time points before and after the implementation of an intervention or program to investigate the effect of the intervention on an outcome of interest. ITS design is often implemented in healthcare settings and is considered the strongest quasi-experimental design in terms of internal and external validity as well as its ability to establish causal relationships. There are several statistical methods that can be used to analyze data from ITS studies. Nevertheless, limitations exist in practical applications, where researchers inappropriately apply the methods, and frequently ignore the assumptions and factors that may influence the optimality of the statistical analysis. Moreover, there is little to no guidance available regarding the application of the various methods, and a standardized framework for analysis of ITS studies does not exist. As such, there is a need to identify and compare existing ITS methods in terms of their strengths and limitations. Their methodological challenges also need to be investigated to inform and direct future research. In light of this, this PhD thesis addresses two main objectives: 1) to conduct a scoping review of the methods that have been employed in the analysis of ITS studies, and 2) to develop improved methods that address a major limitation of the statistical methods frequently used in ITS data analysis. These objectives are addressed in three projects. For the first project, a scoping review of the methods that have been used in analyzing ITS data was conducted, with the focus on ITS applications in health research. The review was based on the Arksey and O’Malley framework and the Joanna Briggs Handbook for scoping reviews. A total of 1389 studies were included in our scoping review. The articles were grouped into methods papers and applications papers based on the focus of the article. For the methods papers, we narratively described the identified methods and discussed their strengths and limitations. The application papers were summarized using frequencies and percentages. We identified some limitations of current methods and provided some recommendations useful in health research. In the second project, we developed and presented an improved method for ITS analysis when the data at each time point are aggregated across several participants, which is the most common case in ITS studies in healthcare settings. We considered the segmented linear regression approach, which our scoping review identified as the most frequently used method in ITS studies. When data are aggregated, heterogeneity is introduced due to variability in the patient population within sites (e.g. healthcare facilities) and this is ignored in the segmented linear regression method. Moreover, statistical uncertainty (imprecision) is introduced in the data because of the sample size (number of participants from whom data are aggregated). Ignoring this variability and uncertainty will likely lead to invalid estimates and loss of statistical power, which in turn leads to erroneous conclusions. Our proposed method incorporates patient variability and sample size as weights in a weighted segmented regression model. We performed extensive simulations and assessed the performance of our method using established performance criteria, such as bias, mean squared error, level and statistical power. We also compared our method with the segmented linear regression approach. The results indicated that the weighted segmented regression was uniformly more precise, less biased and more powerful than the segmented linear regression method. In the third project, we extended the weighted method to multisite ITS studies, where data are aggregated at two levels: across several participants within sites as well as across multiple sites. The extended method incorporates the two levels of heterogeneity using weights, where the weights are defined using patient variability, sample size, number of sites as well as site-to-site variability. This extended weighted regression model, which follows the weighted least squares approach is employed to estimate parameters and perform significance testing. We conducted extensive empirical evaluations using various scenarios generated from a multi-site ITS study and compared the performance of our method with that of the segmented linear regression model as well as a pooled analysis method previously developed for multisite studies. We observed that for most scenarios considered, our method produced estimates with narrower 95% confidence intervals and smaller p-values, indicating that our method is more precise and is associated with more statistical power. In some scenarios, where we considered low levels of heterogeneity, our method and the previously proposed method showed comparable results. In conclusion, this PhD thesis facilitates future ITS research by laying the groundwork for developing standard guidelines for the design and analysis of ITS studies. The proposed improved method for ITS analysis, which is the weighted segmented regression, contributes to the advancement of ITS research and will enable researchers to optimize their analysis, leading to more precise and powerful results. / Thesis / Doctor of Philosophy (PhD)

This study aims to investigate the impact of central bank interest rate effects on the debt-to-equity ratio of major companies in the industrial sector. With an Interrupted Time Series analysis the research examines how changes in the Federal Reserve's policy rates between 2011 and 2019 have influenced corporate financial strategies. Data from 46 of the largest U.S-based industrial firms were analyzed to identify trends in debt ratio before and after interest rate changes. The method involved comparing debt-to-equity ratios over a period of 36 quarters capturing both pre-event and post-event data. The findings indicate an increase in interest led to an increase in the debt ratio among these firms. This suggests that companies may have acted proactively to secure lower debt cost ahead of anticipated rate increases contrary to traditional expectations that higher rates would reduce borrowing. The study concludes that such proactive financial behavior highlights the complexity of corporate decision making in response to macroeconomic policy changes. The results contribute to a deeper understanding of how industrial firms manage their capital structure under varying economic conditions. Future research should explore sector specific responses and the implications for smaller firms to provide a more comprehensive view of the impact of interest rate changes on corporate capital structure. / Denna studie undersöker hur centralbankernas räntehöjningar påverkar skuldsättningsgraden hos stora företag inom industrisektorn. Genom att använda en Interrupted Time Series (ITS)-analys undersöker forskningen hur förändringar i Federal Reserves styrräntor mellan 2011 och 2019 har påverkat företagens finansiella strategier. Data från 46 av de största industriföretagen baserade i USA analyserades för att identifiera trender i skuldsättningen före och efter ränteändringarna. Metodstycket innebar en jämförelse av skuldsättningsgrader över en period av 36 kvartal vilket omfattade både data före och efter ränteändringarna. Resultaten visar på en ökning av skuldsättningsgraden bland dessa företag efter räntehöjningarna. Detta tyder på att företagen kan ha agerat proaktivt för att säkra billigare lån innan de förväntade räntehöjningarna fullt ut realiserades i motsats till den traditionella förväntningen att högre räntor skulle minska belåningen. Studien drar slutsatsen att ett sådant proaktivt finansiellt beteende belyser komplexiteten i företagens beslutsfattande som svar på makroekonomiska politiska förändringar. Resultaten bidrar till en djupare förståelse för hur industriföretag hanterar sin kapitalstruktur under varierande ekonomiska förhållanden. Framtida forskning bör utforska sektorspecifika svar och implikationerna för mindre företag för att ge en mer heltäckande bild av hur ränteförändringar påverkar företagens kapitalstruktur.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第25156号 / 医博第5042号 / 京都大学大学院医学研究科医学専攻 / (主査)教授 川上 浩司, 教授 森田 智視, 教授 寺田 智祐 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Objectifs. L’objectif de ce mémoire est de parfaire nos connaissances quant à l’effet des actions policières sur les collisions routières au Québec. Ultimement, ce mémoire permettra d’identifier les conditions nécessaires pour que l’action policière influe sur les comportements des automobilistes. Pour se faire, deux études de cas sont employées. Dans un premier temps, nous évaluons l’effet d’un relâchement d’environ 60 % dans l’émission de constats d’infraction par les policiers de la ville de Québec sur les collisions avec blessures. Dans cet article, nous distinguons également les effets respectifs des constats d’infraction, des interceptions policières sans constat et des médias. Dans un second temps, nous évaluons l’impact d’une stratégie de sécurité routière mise en place conjointement par l’Association des directeurs de police du Québec (ADPQ) et la Société de l’assurance automobile du Québec (SAAQ). Dans les deux cas, un changement important est survenu dans l’émission de constats d’infraction émis en vertu du Code de la sécurité routière (CSR). Méthodologie. Afin d’évaluer l’effet de ces deux stratégies, nous avons agrégé les données sur les collisions et infractions au CSR sur une base mensuelle. Ces données proviennent principalement des rapports de collisions et des constats d’infraction remplis par les policiers et transmis à la SAAQ. Dans l’ensemble, nous avons utilisé un devis quasi-expérimental, soit celui des séries chronologiques interrompues. Résultats. Les résultats des deux articles démontrent que les policiers sont des acteurs clés en matière de sécurité routière. Les collisions avec blessures sont affectées par les fluctuations de leurs activités. La première série d’analyses établit qu’un relâchement d’environ 60 % dans le nombre de constats émis par les policiers se traduit par une hausse d’environ 10 % des collisions avec blessures, ce qui correspond à 15 collisions avec blessures supplémentaires par mois sur le territoire du Service de police de la ville de Québec. De plus, nos résultats montrent qu’une interception policière suivie d’un avertissement verbal n’est pas suffisante pour prévenir les collisions. De même, l’effet observé n’est pas attribuable aux médias. La deuxième série d’analyse montre que la stratégie conjointe de l’ADPQ et de la SAAQ, caractérisée par une hausse des constats émis et des campagnes médiatiques, fut suivie de baisses variant entre 14 et 36 % des collisions avec blessures graves. Interprétation. Les résultats démontrent que les actions policières ont une influence sur le bilan routier. Par contre, avant d’influer sur le comportement des automobilistes, certaines conditions doivent être respectées. Premièrement, l’intensité des contrôles policiers doit être suffisamment modifiée par rapport à son niveau initial. Deuxièmement, que ce soit une hausse ou une baisse, ce niveau doit être maintenu sur une période relativement longue (entre 12 et 24 mois environ) pour que les automobilistes soient exposés au message pénal et qu’ils considèrent ce changement dans le niveau de répression comme étant crédible. Troisièmement, l’émission de constats est un élément clé; la simple présence policière n’est pas suffisante pour prévenir les collisions. Enfin, les campagnes de sensibilisation semblent importantes, mais d’autres études sont nécessaires pour mieux apprécier leur rôle. / Objectives. The goal of this thesis is to further our understanding about the effect of police activities on traffic collisions in the Province of Quebec. The study also aims to pinpoint conditions that must be met to insure the effectiveness of such police interventions. To do so, we use two case studies. In the first place, we assess the impact of a 60% reduction in traffic citations issued by police officers on collisions with injuries. In this article, we are also able to estimate the respective effects of traffic citations, police interceptions not leading to the issuance of a citation and media coverage. In the second place, we evaluate a road safety program implemented by the Quebec Association of Chiefs of Police (ADPQ) and the Societé de l’assurance automobile du Québec (SAAQ). In both cases, there was a substantial change in the level of traffic citations handed down by police officers. Method. In order to assess the impact of both strategies, data on collisions and citations were aggregated on a monthly basis. These data come from police reports on traffic citations and collisions that are transmitted and managed by the SAAQ. In all cases, we used a quasi experimental design: interrupted time series. Results. Results from both articles show that police officers are key players in a road safety policy. Collisions with injuries vary according to the intensity of police enforcement activities. The first set of analyses establishes that a 60% reduction in the issuance of traffic citations is associated with a 10% increase in collisions with injuries (about 15 additional collisions per month on the Quebec City Police jurisdiction). Furthermore, simple interceptions (not leading to the issuance of a citation) as well as media coverage were not statistically linked to this increase. The second series of analyses demonstrate that the joint strategy of the ADPQ and SAAQ, characterized by an increase in police arrests and media campaigns, was associated with decreases varying between 14 and 36% in collisions with serious injuries in the Province of Quebec. Conclusion. Results from our analyses show that police activities have an important impact on the road toll. However, some conditions must be met in order to stimulate changes in driver behaviors. First, the intensity of police controls must be substantially leveled up relatively to the previous level. Second, this new level must be maintained for a long lasting period (between 12 and 24 months). By doing so, automobilists will have the opportunity to be exposed the legal threat and consider this threat as credible. Third, the issuance of traffic citations is a key component; the simple presence of police officers is not sufficient to produce a preventive effect on collisions. At last, media campaigns appear to be a central component of police enforcement programs but further studies are need to precisely estimate their role and contribution to collision prevention.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Orientador: Maria Cristina Pereira Lima / Resumo: Introdução: Os acidentes de trânsito (AT) fazem aumentar a população que possui alguma deficiência e os anos potenciais de vida perdidos; causam dor e sofrimento aos familiares das vítimas e geram gastos aos sistemas de saúde e previdenciário dos países. Objetivo: O objetivo deste trabalho foi avaliar os fatores associados aos óbitos causados por AT no Brasil à partir de uma série de quatro estudos produzidos com dados secundários. Método: O estudo 1 apresenta sínteses da prevalência do ato de beber e dirigir (%BD) no Brasil após a promulgação da Lei Seca de 2008 (LS08), obtidas por modelos de meta-análise de efeitos aleatórios, à partir de estudos selecionados nas bases Scielo, Medline, Embase e Web of Science. O estudo 2 é um “Interrupted Time Series” (ITS) que avalia o impacto da LS08 sobre o número mensal de óbitos por AT no Brasil entre 2002 a 2015 por meio de um modelo de séries temporais com resposta Binomial Negativa. O estudo 3 é um ITS que avalia o impacto da LS08 sobre a mortalidade por AT nos 27 estados brasileiros por meio de modelos de séries temporais da classe ARIMA, ajustados para cada um dos estados. O estudo 4 apresenta uma revisão sobre aspectos de planejamento e validade dos resultados de estudos conduzidos sob o delineamento ITS, à partir do referencial teórico de Campbell e Stanley (1963), com exemplos de estudos ITS aplicados no trânsito. Resultados: As sínteses obtidas no estudo 1 mostram que, em um período de 12 meses, 26,8% dos motoristas dirigem so... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: Traffic accidents (TA) increase the population with disabilities and the potential years of life lost, cause pain and suffering to the families of the victims, and generate expenses for the health and social security systems of the countries. Objective: The objective of this study was to evaluate the factors associated with deaths caused by TA in Brazil from a series of four studies produced with secondary data. Method: Study 1 presents a summary of the prevalence of drinking and driving (% BD) in Brazil after the enactment of the 2008 Dry Law (LS08), obtained by random effects meta-analysis models, based on studies selected at Scielo , Medline, Embase and Web the Science. Study 2 is an "Interrupted Time Series" (ITS) that assesses the impact of LS08 on the monthly number of TA deaths in Brazil between 2002 and 2015 through a time series model with Negative Binomial response. Study 3 is an ITS that assesses the impact of LS08 on TA mortality in the 27 Brazilian states through time series models of the ARIMA class, adjusted for each state. Study 4 presents a review on aspects of planning and validity of the results of studies conducted under the ITS design, based on the theoretical reference of Campbell and Stanley (1963), with examples of ITS studies applied in traffic. Results: The syntheses obtained in study 1 show that, in a 12-month period, 26.8% of the drivers drive under the influence of alcohol, 8.9% of the drivers are caught under the influence of alcoho... (Complete abstract click electronic access below) / Doutor

Objectifs. L’objectif de ce mémoire est de parfaire nos connaissances quant à l’effet des actions policières sur les collisions routières au Québec. Ultimement, ce mémoire permettra d’identifier les conditions nécessaires pour que l’action policière influe sur les comportements des automobilistes. Pour se faire, deux études de cas sont employées. Dans un premier temps, nous évaluons l’effet d’un relâchement d’environ 60 % dans l’émission de constats d’infraction par les policiers de la ville de Québec sur les collisions avec blessures. Dans cet article, nous distinguons également les effets respectifs des constats d’infraction, des interceptions policières sans constat et des médias. Dans un second temps, nous évaluons l’impact d’une stratégie de sécurité routière mise en place conjointement par l’Association des directeurs de police du Québec (ADPQ) et la Société de l’assurance automobile du Québec (SAAQ). Dans les deux cas, un changement important est survenu dans l’émission de constats d’infraction émis en vertu du Code de la sécurité routière (CSR). Méthodologie. Afin d’évaluer l’effet de ces deux stratégies, nous avons agrégé les données sur les collisions et infractions au CSR sur une base mensuelle. Ces données proviennent principalement des rapports de collisions et des constats d’infraction remplis par les policiers et transmis à la SAAQ. Dans l’ensemble, nous avons utilisé un devis quasi-expérimental, soit celui des séries chronologiques interrompues. Résultats. Les résultats des deux articles démontrent que les policiers sont des acteurs clés en matière de sécurité routière. Les collisions avec blessures sont affectées par les fluctuations de leurs activités. La première série d’analyses établit qu’un relâchement d’environ 60 % dans le nombre de constats émis par les policiers se traduit par une hausse d’environ 10 % des collisions avec blessures, ce qui correspond à 15 collisions avec blessures supplémentaires par mois sur le territoire du Service de police de la ville de Québec. De plus, nos résultats montrent qu’une interception policière suivie d’un avertissement verbal n’est pas suffisante pour prévenir les collisions. De même, l’effet observé n’est pas attribuable aux médias. La deuxième série d’analyse montre que la stratégie conjointe de l’ADPQ et de la SAAQ, caractérisée par une hausse des constats émis et des campagnes médiatiques, fut suivie de baisses variant entre 14 et 36 % des collisions avec blessures graves. Interprétation. Les résultats démontrent que les actions policières ont une influence sur le bilan routier. Par contre, avant d’influer sur le comportement des automobilistes, certaines conditions doivent être respectées. Premièrement, l’intensité des contrôles policiers doit être suffisamment modifiée par rapport à son niveau initial. Deuxièmement, que ce soit une hausse ou une baisse, ce niveau doit être maintenu sur une période relativement longue (entre 12 et 24 mois environ) pour que les automobilistes soient exposés au message pénal et qu’ils considèrent ce changement dans le niveau de répression comme étant crédible. Troisièmement, l’émission de constats est un élément clé; la simple présence policière n’est pas suffisante pour prévenir les collisions. Enfin, les campagnes de sensibilisation semblent importantes, mais d’autres études sont nécessaires pour mieux apprécier leur rôle. / Objectives. The goal of this thesis is to further our understanding about the effect of police activities on traffic collisions in the Province of Quebec. The study also aims to pinpoint conditions that must be met to insure the effectiveness of such police interventions. To do so, we use two case studies. In the first place, we assess the impact of a 60% reduction in traffic citations issued by police officers on collisions with injuries. In this article, we are also able to estimate the respective effects of traffic citations, police interceptions not leading to the issuance of a citation and media coverage. In the second place, we evaluate a road safety program implemented by the Quebec Association of Chiefs of Police (ADPQ) and the Societé de l’assurance automobile du Québec (SAAQ). In both cases, there was a substantial change in the level of traffic citations handed down by police officers. Method. In order to assess the impact of both strategies, data on collisions and citations were aggregated on a monthly basis. These data come from police reports on traffic citations and collisions that are transmitted and managed by the SAAQ. In all cases, we used a quasi experimental design: interrupted time series. Results. Results from both articles show that police officers are key players in a road safety policy. Collisions with injuries vary according to the intensity of police enforcement activities. The first set of analyses establishes that a 60% reduction in the issuance of traffic citations is associated with a 10% increase in collisions with injuries (about 15 additional collisions per month on the Quebec City Police jurisdiction). Furthermore, simple interceptions (not leading to the issuance of a citation) as well as media coverage were not statistically linked to this increase. The second series of analyses demonstrate that the joint strategy of the ADPQ and SAAQ, characterized by an increase in police arrests and media campaigns, was associated with decreases varying between 14 and 36% in collisions with serious injuries in the Province of Quebec. Conclusion. Results from our analyses show that police activities have an important impact on the road toll. However, some conditions must be met in order to stimulate changes in driver behaviors. First, the intensity of police controls must be substantially leveled up relatively to the previous level. Second, this new level must be maintained for a long lasting period (between 12 and 24 months). By doing so, automobilists will have the opportunity to be exposed the legal threat and consider this threat as credible. Third, the issuance of traffic citations is a key component; the simple presence of police officers is not sufficient to produce a preventive effect on collisions. At last, media campaigns appear to be a central component of police enforcement programs but further studies are need to precisely estimate their role and contribution to collision prevention.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Submitted by Vinicius Peçanha (viniciusrpecanha@hotmail.com) on 2015-08-21T15:06:47Z No. of bitstreams: 1 dissertacao_ViniciusPecanha_biblioteca_final2 Copy.pdf: 2020205 bytes, checksum: cc440ff198ec3e166a25a3db500c5c89 (MD5) / Approved for entry into archive by BRUNA BARROS (bruna.barros@fgv.br) on 2015-08-24T13:07:46Z (GMT) No. of bitstreams: 1 dissertacao_ViniciusPecanha_biblioteca_final2 Copy.pdf: 2020205 bytes, checksum: cc440ff198ec3e166a25a3db500c5c89 (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-08-25T19:49:01Z (GMT) No. of bitstreams: 1 dissertacao_ViniciusPecanha_biblioteca_final2 Copy.pdf: 2020205 bytes, checksum: cc440ff198ec3e166a25a3db500c5c89 (MD5) / Made available in DSpace on 2015-08-25T19:49:14Z (GMT). No. of bitstreams: 1 dissertacao_ViniciusPecanha_biblioteca_final2 Copy.pdf: 2020205 bytes, checksum: cc440ff198ec3e166a25a3db500c5c89 (MD5) Previous issue date: 2015-04-29 / The paper quantifies the effects on violence and police activity of the Pacifying Police Unit program (UPP) in Rio de Janeiro and the possible geographical spillovers caused by this policy. This program consists of taking selected shantytowns controlled by criminals organizations back to the State. The strategy of the policy is to dislodge the criminals and then settle a permanent community-oriented police station in the slum. The installation of police units in these slums can generate geographical spillover effects to other regions of the State of Rio de Janeiro. We use the interrupted time series approach proposed by Gonzalez-Navarro (2013) to address effects of a police when there is contagion of the control group and we find that criminal outcomes decrease in areas of UPP and in areas near treated regions. Furthermore, we build a model which allows to perform counterfactuals of this policy and to estimate causal effects in other areas of the State of Rio de Janeiro outside the city.

Submitted by Marlene Santos (marlene.bc.ufg@gmail.com) on 2016-04-20T19:18:23Z No. of bitstreams: 2 Tese - Eliane Terezinha Afonso - 2015.pdf: 3480407 bytes, checksum: db146d8884a4aedac166a73ef268d89d (MD5) license_rdf: 19874 bytes, checksum: 38cb62ef53e6f513db2fb7e337df6485 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-04-25T15:00:31Z (GMT) No. of bitstreams: 2 Tese - Eliane Terezinha Afonso - 2015.pdf: 3480407 bytes, checksum: db146d8884a4aedac166a73ef268d89d (MD5) license_rdf: 19874 bytes, checksum: 38cb62ef53e6f513db2fb7e337df6485 (MD5) / Made available in DSpace on 2016-04-25T15:00:31Z (GMT). No. of bitstreams: 2 Tese - Eliane Terezinha Afonso - 2015.pdf: 3480407 bytes, checksum: db146d8884a4aedac166a73ef268d89d (MD5) license_rdf: 19874 bytes, checksum: 38cb62ef53e6f513db2fb7e337df6485 (MD5) Previous issue date: 2015-08-19 / BACKGROUND: Pneumonia causes substantial morbidity and mortality in all age groups around the world. The 10-valent pneumococcal conjugate vaccine (PCV10) was introduced into the routine infant immunization in Brazil, free of charge, in March 2010. The aim of this study was to evaluate the impact PCV10 vaccination on rates of all cause pneumonia hospitalizations one year and three years after its introduction in Brazil. METHODS: We conducted two interrupted time series analysis studies. The first evaluated only the direct effect of PCV10 vaccination, in five Brazilian cities (Belo Horizonte, Curitiba, Porto Alegre, São Paulo and Recife), and was conducted one year after starting the vaccination. The second study evaluated the direct and indirect impact (individuals not vaccinated) of PCV10 vaccination in Brazil, and was conducted three years after vaccination. We used data from the Brazilian Hospitalization System from 2005-2013. The main outcome was monthly rates of all-cause pneumonia hospitalizations identified by ICD-10 codes J12-J18. We used hospitalization rates for congenital malformations and non-respiratory causes as a comparison groups. The time-series analysis was based on a generalized linear model. Pneumonia rates observed in the pre-vaccination period were used to estimate the hospitalization rates in the post-vaccination period of each study, adjusting for seasonality and secular trends. To estimate the direct (2-23 months of age) and indirect (≥5 years of age) impact of PCV10 vaccination, we calculated the percentage change in hospitalization rates, as the observed divided by the predicted rates of hospitalization in the post-intervention period minus one, with respective 95% CI and p values. The number of all-cause pneumonia hospitalizations averted by vaccination was calculated taking into account the difference between the predicted and observed number in the PCV10 post vaccination period. RESULTS: One year after introduction of PCV10 in Brazil, significant declines in hospitalizations for pneumonia in children aged 2-23 months were noted in Belo Horizonte (28.7%), Curitiba (23.3%), and Recife (27.4%). After three years of the introduction of PCV10, 461,519 pneumonia hospitalizations were averted in Brazil, and a significant decrease in rates of pneumonia hospitalization was observed in unvaccinated individuals aged 5-39 years, ranging from 14.1-17.4% (p<0.05). In contrast, an increased trend in pneumonia hospitalizations (p=0·004) was observed for elderly (≥ 65 years). CONCLUSION: Vaccination with PCV10 in Brazil was associated with reduction of pneumonia hospitalizations in vaccinated individuals. Herd effect was observed in individuals aged 5-39 years after three years of vaccination. Potential reasons for the increased trend in pneumonia hospitalization rates in the elderly should be investigated. / INTRODUÇÃO: As pneumonias contribuem com alta carga de morbimortalidades em todo mundo. No Brasil, a vacina pneumocócica conjugada 10 valente (PCV10) foi introduzida na rotina de imunização da infância em março de 2010. Este estudo teve como objetivo avaliar o impacto da vacinação nas taxas de hospitalizações por pneumonia no Brasil no curto e médio prazo do início da vacinação. METODOLOGIA: Dois estudos de séries temporais interrompidas foram conduzidos. O primeiro avaliou o efeito direto da vacinação em cinco capitais brasileiras (Belo Horizonte, Curitiba, Porto Alegre, São Paulo e Recife) e foi conduzido após um ano de introdução da PCV10 no país. O segundo estudo avaliou o impacto direto e indireto (população não vacinada) da vacinação em todo país e foi conduzido três anos após sua introdução. Os dados de hospitalizações foram obtidos do Sistema de Informações Hospitalares (SIH-SUS) de 2005 a 2013. O desfecho principal foi a taxa mensal de hospitalização por pneumonia definida pelos códigos J12-J18 da CID10. As taxas de hospitalizações por malformações congênitas e causas não respiratórias foram utilizadas como grupos de comparações. A análise de série temporal utilizou um modelo de regressão linear generalizado. As taxas de hospitalizações por pneumonia observadas no período pré-PCV10, ajustadas por tendência secular e sazonalidade, foram utilizadas para estimar as taxas no período pós-PCV10. O impacto da vacinação para cada faixa etária foi calculado como o percentual de mudança nas taxas de hospitalizações, dividindo-se as taxas observadas pelas taxas preditas do período pós PCV10, menos um. Os respectivos IC95% e os valores de p foram apresentados. O número de hospitalizações por pneumonia evitadas após três anos de vacinação foi estimado pela diferença entre os números de hospitalizações por pneumonia preditos e observados no período pós-vacinação. RESULTADOS: Após um ano de introdução da PCV10 no Brasil, observou-se significativo declínio nas taxas de hospitalizações por pneumonia em crianças de 2 a 23 meses em três das cinco capitais estudadas: Belo Horizonte (28,7%), Curitiba (23,3%), e Recife (27,4%). Após três anos da introdução da PCV10, 461.519 hospitalizações por pneumonia foram evitadas no Brasil e um significativo declínio nas taxas de pneumonia foi observado em indivíduos não vacinados de 5 a 39 anos variando de 14,1% a 17,4% (p<0,05). No entanto, observou-se um aumento significativo (9,9%, p=0,004) nas taxas de hospitalizações por pneumonia para idosos ≥65 anos. CONCLUSÕES: A vacinação com a PCV10 foi associada à significativa redução das hospitalizações por pneumonia na infância. Adicionalmente, o estudo evidenciou importante redução das hospitalizações por pneumonia em grupos etários não vacinados, sinalizando efeito indireto conferida pela vacina. A tendência de aumento das hospitalizações por pneumonias em idosos necessita de investigações para elucidação dos fatores envolvidos nesse fenômeno.