Spatial variation in density of organisms can lead to challenges in estimation of population size. Associated vital rates responsible for this variation also may vary geographically and in response to local ecological conditions, with the result that subunits of a metapopulation may have different trajectories. Both temporal and spatial variation in population size occurs not only as a result of additions through birth and deletions through death, but also from gains and losses arising from immigration and emigration, respectively. Although virtually all organisms have evolved mechanisms for dispersal, the role of movement in population dynamics has received far less attention than have contributions from recruitment and losses to mortality. I used mark-recapture techniques to make inferences about the role of movement in local population dynamics of Rosss Goose (Chen rossii) colonies by estimating rates of movement between breeding subpopulations in the Queen Maud Gulf metapopulation. I also assessed decision-based philopatry (i.e., the role of previous nesting outcome; sensu Hoover 2003) and a potential cost of reproduction to female geese through experimental manipulation of nesting success. <p>Previous nest fate influenced intra-colony dispersal as failed nesters moved further between consecutive nest sites, but inter-colony movement was not affected by previous nest fate. Regardless of previous nest fate, Rosss Geese did not exhibit philopatry to nest sites, or to breeding territories, suggesting that philopatry occurs at a larger spatial scale. Breeding success accounted for a detectable, but only small amount of variation (<11%) in dispersal distance within colonies. I suggest that temporal variation in habitat availability favors flexibility in settling patterns by geese in a changing matrix of habitat availability, governed largely by receding snow cover. Such flexibility is necessary for nesting as early as possible, because recruitment is strongly linked to timing of breeding by arctic-nesting geese. Colonial philopatry may be important not only for favorable nesting but also for access to high-quality feeding areas adjacent to colonies. Such feeding areas represent a predictable food resource important not only to growing goslings, but also adult survival regardless of the outcome of their breeding attempt. <p>I concluded from experimental manipulation that successful reproduction was encumbered with a cost to survival of females. I argue that such a cost of breeding is more likely to be incurred when climatic conditions during incubation are harsh, and when the breeding population is larger. <p>I did not find evidence for geographic variation in survival, but rates of philopatry varied markedly among colonies. The substantial exchange of females among breeding colonies (1) underscores the potential for dispersal to alter breeding distribution, (2) demonstrates that the influence of immigration on colony-specific rates of population growth was nontrivial, and (3) provides behavioral evidence for extensive gene flow resulting from female dispersal. Estimates of emigration and survival from my studies were used in combination with those for fecundity parameters and colony-specific population growth rates (lambda) to interpolate the role of immigration from a simple balance equation. During years for which rates of movement were estimated, immigration constituted 9-20% of lambda at the Karrak Lake colony, suggesting that movement was an important contribution to population growth.

Spatial variation in density of organisms can lead to challenges in estimation of population size. Associated vital rates responsible for this variation also may vary geographically and in response to local ecological conditions, with the result that subunits of a metapopulation may have different trajectories. Both temporal and spatial variation in population size occurs not only as a result of additions through birth and deletions through death, but also from gains and losses arising from immigration and emigration, respectively. Although virtually all organisms have evolved mechanisms for dispersal, the role of movement in population dynamics has received far less attention than have contributions from recruitment and losses to mortality. I used mark-recapture techniques to make inferences about the role of movement in local population dynamics of Rosss Goose (Chen rossii) colonies by estimating rates of movement between breeding subpopulations in the Queen Maud Gulf metapopulation. I also assessed decision-based philopatry (i.e., the role of previous nesting outcome; sensu Hoover 2003) and a potential cost of reproduction to female geese through experimental manipulation of nesting success. <p>Previous nest fate influenced intra-colony dispersal as failed nesters moved further between consecutive nest sites, but inter-colony movement was not affected by previous nest fate. Regardless of previous nest fate, Rosss Geese did not exhibit philopatry to nest sites, or to breeding territories, suggesting that philopatry occurs at a larger spatial scale. Breeding success accounted for a detectable, but only small amount of variation (<11%) in dispersal distance within colonies. I suggest that temporal variation in habitat availability favors flexibility in settling patterns by geese in a changing matrix of habitat availability, governed largely by receding snow cover. Such flexibility is necessary for nesting as early as possible, because recruitment is strongly linked to timing of breeding by arctic-nesting geese. Colonial philopatry may be important not only for favorable nesting but also for access to high-quality feeding areas adjacent to colonies. Such feeding areas represent a predictable food resource important not only to growing goslings, but also adult survival regardless of the outcome of their breeding attempt. <p>I concluded from experimental manipulation that successful reproduction was encumbered with a cost to survival of females. I argue that such a cost of breeding is more likely to be incurred when climatic conditions during incubation are harsh, and when the breeding population is larger. <p>I did not find evidence for geographic variation in survival, but rates of philopatry varied markedly among colonies. The substantial exchange of females among breeding colonies (1) underscores the potential for dispersal to alter breeding distribution, (2) demonstrates that the influence of immigration on colony-specific rates of population growth was nontrivial, and (3) provides behavioral evidence for extensive gene flow resulting from female dispersal. Estimates of emigration and survival from my studies were used in combination with those for fecundity parameters and colony-specific population growth rates (lambda) to interpolate the role of immigration from a simple balance equation. During years for which rates of movement were estimated, immigration constituted 9-20% of lambda at the Karrak Lake colony, suggesting that movement was an important contribution to population growth.

<p>Abstract</p> <p>Risky foraging is a male reproductive tactic in most polygynous mammals. It is speculated to result from intense intra-sexual reproductive competition. Consequently this behavior has been speculated to increase a male's reproductive competitiveness. However, individual males may differ in their propensity to take foraging risks. </p> <p>We therefore conducted a study on crop raiding behavior (a risky foraging strategy) in African elephants from the greater Amboseli ecosystem, in southern Kenya. We specifically examined the population sizes, gender and patterns of raiding elephants and investigated the effect of crop-raiding and genetic heterozygosity on male body size. We also examined the influence of age and genetic relatedness on observed patterns of association. Finally, we examined the role of life history milestones, association patterns and social structure on the acquisition of crop raiding behavior among wild free ranging male African elephants. With regard to the influence of association patterns on crop raiding behavior, we were specifically interested in understanding the mechanisms by which social learning might occur among male elephants.</p> <p>Our results showed that 241elephants from different populations in the ecosystem converged to raid farms. Approximately 35% of raiders were from Amboseli National Park, and the rest were other populations in the ecosystem. We observed only post-pubertal males but not females to raid. About one third of post-pubertal males from the Amboseli population were raiders. We found evidence of habitual raiding by some individuals. Crop raiding predicted post-pubertal male size, with raiders being larger than non-raiders. This result suggests that taking risks pays off for males. Our results also showed that other variables known to influence growth like genetic heterozygosity had no effect on size-for-age in male elephants, because low-heterozygosity males were rare. The probability that an individual male is a crop raider was greater for older individuals than young males. The probability that a male is a raider was greater when his two closest associates were raiders versus when they were not raiders and when a male's second closest associate was older, versus when his second closest associate was of similar age or younger. These results suggest that increasing energetic demands associated with life history milestones and social learning play a significant role in the initiation of crop raiding behavior. Raiders did not cluster into separate social units from non-raiders, probably due to the nature of social learning exhibited by this species and due to the diffuse nature of male elephant social units.</p> <p>These results have implications for understanding the evolution of risky foraging behavior in males, and for understanding the role of kin selection, dominance hierarchies and social learning in male elephant social systems. Results also have implications for understanding the spread of adaptive complex behavior in natural populations.</p> / Dissertation

A relatively recent increase in the popularity of evidence-based activism has created a higher demand for statisticians to work on human rights and economic development projects. The statistical challenges of revealing patterns of violence in armed conflict require efficient use of the data, and careful consideration of the implications of modeling decisions on estimates. Impact evaluation of a complex economic development project requires a careful consideration of causality and transparency to donors and beneficiaries. In this dissertation, I compare marginal and conditional models for capture recapture, and develop new hierarchical models that accommodate challenges in data from the armed conflict in Colombia, and more generally, in many other capture recapture settings. Additionally, I propose a study design for a non-randomized impact evaluation of the Millennium Villages Project (MVP), to be carried out during my postdoctoral fellowship. The design includes small area estimation of baseline variables, propensity score matching, and hierarchical models for causal inference.

The goal of this dissertation is to characterize the distribution and determinants of coccidioidomycosis in greater Tucson, Arizona, using landscape ecology and complex survey methods to control for environmental factors that affect <italic>Coccidioides</italic> exposure. Notifiable coccidioidomycosis cases reported to the health department in Arizona have dramatically increased since 1997 and indicate a potential epidemic of unknown causes. Epidemic determination is confounded by concurrent changes in notifiable disease reporting-compliance, misdiagnosis, and changing demographics of susceptible populations. A stratified, two-stage, address-based telephone survey of greater Tucson, Arizona, was conducted in 2002 and 2003. Subjects were recruited from direct marketing data by census block groups and landscape strata as determined using a geographic information system (GIS). Subjects were interviewed about potential risk factors. Address-level state health department notifiable-disease surveillance data were compared with self-reported survey data to estimate the true disease frequency.Comparing state surveillance data with the survey data, no coccidioidomycosis epidemic was detectable from 1992 to 2006 after adjusting surveillance data for reporting compliance. State health department surveillance reported only 20% of the probable reportable cases in 2001.Utilizing survey data and geographic coding, it was observed that spatial and temporal disease frequency was highly variable at the census block-group scale and indicates that localized soil disturbance events are a major group-level risk factor. Poststratification by 2000 census demographic data adjusted for selection bias into the survey and response rate. Being Hispanic showed similar odds ratio of self-reporting coccidioidomycosis diagnosis as of being non-Hispanic White race-ethnicity when controlled by other risk factors. Cigarette smoking in the home and having a home located in the low Hispanic foothills and low Hispanic riparian strata were associated with elevated risk of odds ratios for coccidioidomycosis. Sample stratification by landscape and demographics controlled for differential classification of susceptibility and exposures between strata.Clustered, address-based telephone surveys provide a feasible and valid method to recruit populations from address-based lists by using a GIS to design a survey and population survey statistical methods for the analysis. Notifiable coccidioidomycosis case surveillance can be improved by including reporting compliance in the analysis. Pathogen exposures and host susceptibility are important predictable group-level determinants of coccidioidomycosis that were controlled by stratified sampling using a landscape ecology approach.

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2018-05-08T17:29:10Z No. of bitstreams: 2 Dissertação - Juliana Brasiel da Silva - 2009.pdf: 3547451 bytes, checksum: 78ed50d3cc117bf024ed33818cba48a3 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-05-09T11:26:23Z (GMT) No. of bitstreams: 2 Dissertação - Juliana Brasiel da Silva - 2009.pdf: 3547451 bytes, checksum: 78ed50d3cc117bf024ed33818cba48a3 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-05-09T11:26:23Z (GMT). No. of bitstreams: 2 Dissertação - Juliana Brasiel da Silva - 2009.pdf: 3547451 bytes, checksum: 78ed50d3cc117bf024ed33818cba48a3 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2009-08-14 / The epidemiology of dengue fever in Brazil is recently characterized by an increase in hospitalizations, severe and unusual presentations of the disease and a shift towards children. This scenario demands improvements to the surveillance and to the health systems to timely detect and adequate management of these cases. The objectives of this study are to characterize the hospitalized cases of dengue fever and dengue hemorrhagic fever and to identify potential underreporting of these cases in the city of Goiânia, Central Brazil, between 2005 and 2008, based on two official information systems: SINAN and SIH/SUS. We conducted a capture-recapture study to estimate the number of hospitalized cases using a linkage between SINAN and SIH/SUS, after excluding duplicate records in each system. A 49.6% increase in hospitalized cases was estimated after the inclusion of SIH/SUS records that were not reported in SINAN. An underreporting of 70.4% of the information regarding hospitalization in SINAN was also detected, once these cases were reported but with no reference to hospitalization. Sensibility of the surveillance system to detect hospitalized cases was considered poor for the study period. The combined use of SINAN and SIH/SUS in the surveillance routine greatly contributes to increase the sensibility of the system and to the timely detection of cases and outbreak response. / A epidemiologia de dengue no Brasil apresenta, recentemente, como principais características, o aumento no número de internações e de casos graves da doença e apresentações atípicas, especialmente em crianças. Esse cenário aponta para a necessidade de melhora da resposta do sistema de vigilância e dos serviços de saúde na detecção precoce e atendimento adequado desses casos. Este estudo teve como objetivos caracterizar o perfil epidemiológico dos pacientes hospitalizados por dengue e febre hemorrágica da dengue e identificar potenciais subnotificações desses casos no município de Goiânia entre os anos de 2005 e 2008, utilizando dados de dois sistemas de informação oficiais: SINAN e SIH/SUS. Foi utilizado o método de captura-recaptura no intuito de se obter uma estimativa do número de casos hospitalizados da doença a partir do pareamento dos dois sistemas, SINAN e SIH/SUS, após a exclusão das duplicidades. Após inclusão dos casos registrados no SIH/SUS que não estavam notificados no SINAN, houve um aumento de 49,6% na estimativa de internações por dengue no município. Também foi encontrado um subregistro de pelo menos 70,4% das informações sobre internações no SINAN, visto que, mesmo estando notificados, não havia nenhuma informação sobre a internação desses indivíduos. A sensibilidade do sistema de vigilância para a detecção de casos hospitalizados pode ser considerada baixa para o período avaliado. A utilização do SIH concomitantemente com o SINAN na rotina do serviço contribui bastante para aumentar a sensibilidade do sistema de vigilância epidemiológica da dengue na detecção dos casos graves e conseqüente tomada de decisões oportunas.

Introdução - O método de captura-recaptura vem sendo empregado em Epidemiologia desde meados do século XX, e se consolidou a partir dos anos 1990, quando se nota grande número de publicações sobre sua aplicação e desenvolvimento nesta área. O sub-registro de eventos vitais ainda se revela um entrave para o cálculo direto de indicadores como os de fecundidade e mortalidade infantil, forçando seu cálculo indireto através de métodos demográficos, cujos procedimentos não permitem estimação em níveis geográficos menores do que unidade da federação, em períodos intercensitários. Objetivo Estimar o sub-registro de nascidos vivos, aplicando o método de captura-recaptura para populações fechadas. Métodos - As bases de dados do Sistema Nacional de Informação sobre Nascidos Vivos (SINASC) e do Registro Civil do IBGE, nos segundo e terceiro trimestres de 2006 do estado de Sergipe, foram pareadas por relacionamento determinístico a partir do número da Declaração de Nascido Vivo. As desagregações geográficas adotadas foram as de microrregião e regional de saúde de residência da mãe. Os modelos de Huggins para populações fechadas foram aplicados para estimar as probabilidades de captura em cada uma das bases e o total de nascidos vivos ocorrido no período, dentro de cada desagregação geográfica O aplicativo utilizado para as estimações foi o Software MARK®. Resultados A aplicação do método de captura e recaptura para estimar sub-registro de nascidos vivos é factível, inclusive para desagregações geográficas menores do que unidade da federação. O relacionamento determinístico foi prejudicado em quatro microrregiões e em uma regional de saúde, devido à falta de preeenchimento do número da Declaração de Nascido Vivo na base do IBGE. O aplicativo MARK® apresenta interface amigável, o que facilitou a construção e seleção dos modelos estatísticos, permitindo identificar que a idade da mãe afeta a probabilidade de captura pelo Registro Civil, característica de heterogeneidade na população de nascidos vivos. Conclusões O relacionamento determinístico destas duas bases de dados oficiais viabiliza ações localizadas, porque acaba por identificar onde e quantas vidas mantêm-se no anonimato jurídico, devido ao sub-registro e registro tardio. O método de captura e recaptura mostrou-se uma alternativa acessível e barata para a estimação de sub-registro de nascidos vivos / Introduction Capture-recapture method has been used in Epidemiology since the middle of twentieth century and from the 1990s on, huge number of publications about applying and developing the method on this area have been noted. Underreporting of vital events is still a drawback for the direct calculation of health indicators like fertility and infant mortality, forcing indirect calculation with demographic methods, wich procedures do not allow estimation in geographic levels smaller than States, in intercensuses periods. Objective To estimate underreporting of live births, using the capture-recapture method for closed populations. Methods The National Live Birth Information System (SINASC) and the Civil Registry databases, in the second and third trimesters of 2006 of the State of Sergipe, Brazil, were linked through deterministic linkage using the Live Birth Declaration number as the connection key. Mothers micro-regions and health regions of residence were adopted as geographical subdivisions. Huggins closed capture models were used to estimate the capture probabilities for each database as well as the derived estimation of the total of live births in the selected time interval, at each of geographical subdivisions. MARK® Software was used to get all estimations. Results The capture-recapture method is feasible for estimation of live birth underreporting, even for geographical subdivisions smaller than States. Deterministic linkage was damaged due to lack of filling of the Live Declaration number in some of the Civil Registry databases registries, mainly at four mother residences micro-regions and at one health region. Program MARK® shows friendly interface, which favoured models building and selection, and allowed to point out mother ages influence in Civil Registry database capture probability, characterisc heterogeneity of live born population. Conclusions Deterministic linkage of these two official databases makes possible local actions, because it allows do identify where and how many lives are neglected legal registrations duty, due to underreporting and to late reporting. Capture-recapture method appears as an accessible and cheap alternative for live birth underreporting estimation

The red-backed salamander (RBS; Plethodon cinereus) has a large geographic range and, though it is common throughout, abundance varies spatially. It has been studied extensively for at least a century and as a result; its distribution, habitat use, behavioral ecology, reproductive ecology, and ecological role are well understood in comparison to other cryptic wildlife. Multiple sampling methods have been developed to monitor RBS, and I discuss in detail one common method – artificial cover objects (ACOs). Spatial capture-recapture (SCR) is uniquely suited to estimate population parameters, including true density, and is paired well with spatially referenced sampling methods, such as ACOs, and animals capable of retaining unique marks, such as RBS. In the introduction of this thesis, I review RBS, ACOs, and SCR independently in detail, and go on to discuss their compatibility for monitoring terrestrial salamander populations. Detailed, and often range-wide, monitoring of ecosystems is necessary to gather the information needed to achieve broad multiple-scale conservation objectives. Indicator species are recommended tools for filling in gaps of knowledge where these range-wide data do not exist. The use of indicator species is precluded by the lack of evaluation of candidate indicators and their relationship to indicated processes. In this thesis, I discuss in detail the practicality of indicator species as wildlife management tools, and challenges in their application – primarily their practicality when direct measurement of a variable of interest is possible. I advocate for integrative indicator species applications that make use of relationships to latent variables, review two conceptual models involving latent variables, and propose a modification to these models that makes relationships between variables more explicit. Inference of among-population variation to adaptive capacity, response to large-scale threats, and the condition of ecosystems is limited in part by unstandardized methods. Ecological relationships are made difficult to characterize by gaps in data - and this is especially true of links between indicator species their and related ecosystem processes. Using a candidate indicator species, I tested the congruence of population parameter estimates from study designs that varied. In Wendell State Forest, MA, I manipulated spatial arrangements of artificial cover objects (ACO) arrays and evaluated their use for monitoring terrestrial salamanders. ACOs mimic natural habitat - and attraction of RBS to traps may induce behavioral bias in parameter estimates if not accounted for. I sought to determine if variation in ACO design can be accommodated to make comparable estimates. I found that analyzing data from ACOs using spatial capture-recapture (SCR) modeling produces consistent within-population density estimates regardless of ACO configuration.

Though the American marten (Martes americana) is widely distributed across northern North America, habitat use and population abundance vary widely across the range. Due to its status as a furbearer, the species has been extensively researched, resulting in a large body of knowledge about the species’ ecology, distribution, and abundance, as well as drivers of population structure and dynamics. More recently, marten research has shifted focus to genetics, habitat associations, and estimation of population state variables. The rapid increase in estimation of states such as occupancy, abundance, and density has likely been driven by the increasing accessibility of noninvasive field technology, such as noninvasive genetic sampling and remote camera trapping, and by the statistical development of ecological hierarchical models. This convergence of advances in field and analytical methods is most apparent in the now widespread application of spatial capture-recapture, an approach that produces robust estimates of population densities and abundance that can be compared across time and space. These new models are especially valuable near the edges of marten distribution where populations are often recovering from historic overexploitation, and expanding into areas they have previously been absent from. In these areas, detailed, landscape-scale understanding of marten populations is necessary in order to establish current conditions, effectively monitor changes, and predict what effect management actions may have on marten populations. I utilized these models to study marten populations in New Hampshire where marten are a species of management interest, and recent recovery has led to their removal from the state endangered species list. Through a collaborative effort with New Hampshire Fish and Game Department in the winters of 2017 and 2018, marten were surveyed across northern New Hampshire using a novel camera trap design that allows for the identification of individuals. These data were analyzed using spatial capture-recapture models, allowing me to evaluate habitat associations that explain spatial variation in marten density and provide a population status assessment for the New Hampshire marten population. Marten densities are highest in the White Mountain National Forest, though other protected lands in northern New Hampshire also appear to support larger populations. The greatest population densities coincided with deeper snows, increased canopy closure, and intermediate boreal biomass. These results provide additional support for several hypotheses explaining marten habitat use across their range while also providing novel insight that will inform active management of both marten and the habitat they occur in. In addition to the population status assessment, I evaluated the relationship between estimates of occupancy and density in New Hampshire. Though utility of non-invasive methodology can decrease research costs, the need for individual identification in spatial capture-recapture models represents a cost increase over occupancy models. My results suggest that the two are positively correlated; however, occupancy is a poor predictor of the entire range of density, especially because the variables used to predict each of the state variables are different. Thus, occupancy is likely not a good proxy for density in New Hampshire, however it could be used to track general trends through time so long as density is re-evaluated periodically.

Introduction : Selon la publication conjointe par l’Organisation Mondiale de la Santé (OMS) et la Banque Mondiale du rapport mondial sur la prévention des traumatismes chaque année les accidents de la route sont responsables de plus de 1,2 million de morts dans le monde. 90% de ces décès surviennent dans les pays à revenu faible ou intermédiaire. Les accidents de la route concernent toutes les tranches d’âge, mais touchent plus particulièrement les jeunes actifs. La lutte contre ce fléau se heurte à la qualité des données sur les accidents de la route recueillies et rapportées, par le défaut de performance ou d’efficacité des outils de collecte des données existants et la non-représentativité des populations d’étude. Face à l’augmentation grandissante des accidents de la circulation dans les pays en développement et particulièrement dans notre pays, le Mali, et l’absence notoire d’un système de collecte de données fiable et extrapolable à la population générale, nous avons mis en en place un projet de recherche. Objectif : Etait de caractériser les accidents corporels de la route enregistrés dans les structures de santé et dans tous les commissariats de police du District de Bamako afin de proposer un système d’information efficace et durable permettant de doter le Mali d’un système de surveillance opérationnel et pertinent, à même d’aider à la prise de décision et aux actions opérationnelles de prévention et de prise en charge des accidents de la route favorisant ainsi la création d’un observatoire national de la sécurité routière basé sur la remontée et le codage des procès-verbaux. Méthode : A partir d’un appariement des données accidentelles corporelles de la route des structures de santé et de la police. Nous avons procédé à une estimation du nombre d’accidents par la méthode de capture-recapture sur une période de quatre mois (de Janvier à Avril) et assuré une surveillance épidémiologique sur 12 mois (de janvier à Décembre 2012) ne portant que sur les données collectées à la police. Résultats : L’estimation globale des accidentés a été nettement supérieur au dénombrement des 2 sources soit 8518 (8047-8990) victimes, contre 1432 pour la police et 3586 pour la santé. Nous pouvons conclure que la capture-recapture est une méthode sensible d’estimation de l’exhaustivité du nombre d’accidents de la route applicable dans les pays à revenu faible. Par contre avec la mortalité, cette sous-estimation diminue soit 99 versus 57 pour la police et 54 pour la santé. L’étude de la surveillance des AVP de Janvier à décembre 2012 a enregistré 2512 accidents avec 3905 victimes dont 407 blessés graves et 142 tués, impliquant 4434 véhicules. L’analyse des facteurs prédictifs de la gravité a révélé une vulnérabilité des moins de 15 ans OR=1,10(1,05-1,15) et des 60 ans et plus OR=1,15(1,06-1,25). Parmi les causes des accidents, l’imprudence du piéton a été le facteur le plus associé à la gravité des AVP OR=1,22 (1,06-1,40). Le défaut de lumière a été significativement associé à la gravité des accidents de la route à Bamako OR=1,05 (1,02-1,09). L’analyse des facteurs prédictifs de la létalité a élucidé un risque de mourir plus élevé chez les 60 ans et plus OR=1,06 (1,01-1,12). Les ménagères auraient un risque plus faible de décès que les autres couches socio professionnelles OR=0,97 (0,96-0,99). Parmi les causes des accidents de la route, l’imprudence du piéton a été le facteur le plus associé au risque de décès 1,11 (1,02-1,23) Les conditions climatiques surtout la saison des pluies et la saison froide ont été les périodes assez mortelles des accidentés de la route OR=1,08 (1,03-1,13). Conclusion : nous pouvons dire que cette étude a permis d'élucider les facteurs humains significativement associés à la gravité et la létalité des accidents de la route, mais aussi mis en évidence la nette sous-estimation du nombre des victimes de la route par la méthode de capture-recapture ayant comparé deux sources de données (police et hôpital). / Introduction: According to the joint publication by the World Health Organization (WHO) and the World Bankof the world report on the prevention of the traumas every year the road accidents are responsible for more than1.2 million deaths. 90 % of these deaths arise in low-income countries or intermediary. The fight against thisplague is impeded by the quality of the data which were collected and reported concerning road accidents, by thelack of performance and efficiency of the existing tools of data collection and the non-representativeness of thepopulations of study. In the face of the growing increase of the traffic accidents in developing countries andparticularly in our country, Mali, and the serious lack of a reliable and extrapolable data collection system in thegeneral population, we put in position a research project.Objective: It is to characterize the personal accidents of the road recorded in the structures of health and in allthe police stations of the District of Bamako in order to propose an effective and sustainable information systemallowing to endow Mali of an operational and relevant monitoring system which is able to help in the decisionmakingand in the operational actions of prevention and coverage of the road accidents thus, favoring thecreation of a national monitoring observatory of the road safety based on the rise and the coding of reports.Methods: we proceeded to an estimation of the number of accidents by the method of capture - recapture over aperiod of four months (from January to April, 2012) from a matching of the physical accidental data of the roadof the structures of health and the police. We also led an epidemiological surveillance over 12 months (fromJanuary to April 2012) concerning only the data collected in the police services.Results: The global estimation of the accident victims was absolutely upper to the enumeration of 2 sources thatis 8518 (8047-8990) victims, against 1432 for the police and 3586 for the health services. We can conclude thatthe capture - recapture is a sensitive method of estimation of the exhaustiveness of the applicable number of roadaccidents in low-income countries. On the other hand with the mortality, this underestimate decreases 99 versus57 for the police and 54 for the health services. The study of the surveillance of the RTA from January toDecember, 2012 recorded 2512 accidents with 3905 victims, among whom 407 badly wounded persons and 142persons killed, involving 4434 vehicles.The analysis of the predictive factors of the gravity revealed a vulnerability of individuals who were under 15OR=1.10 (1.05-1.15) and those who were 60 years old or above OR=1.15 (1.06-1.25).Among the causes of the accidents, the carelessness of the pedestrian was the factor the most associated with thegravity of the RTA OR=1.22 (1.06-1.40). The defect of light was associated significantly with the gravity of theroad accidents in Bamako OR=1.05 (1.02-1.09).The analysis of the predictive factors of the lethality showed arisk of dying higher with those were 60 years old and above OR=1.06 (1.01-1.12). The housewives had a lowerrisk of death more important than the other socio-professional classes OR=0.97 (0.96-0.99). Among the causesof the road accidents, the carelessness of the pedestrian was the factor the most associated with the risk of death1.11 (1.02-1.23). Weather conditions especially the rainy season and the cold season were the most lethal periodsof the road accident victims OR=1.08 (1.03-1.13).Conclusion: we can say that this study allowed to clarify human factors significantly associated to the gravityand the lethality of the road accidents, but it has also highlighted the important under estimation of the number ofthe victims of the road by the capture-recapture method which has compared two springs of data (police andhospital)