Background: The burden of fatal and non-fatal injury for the New Zealand construction industry is larger than most other industrial sectors. Injury preventions efforts for construction have however been hampered because of insufficient, industry-specific, surveillance data that is essential for the effective targeting and evaluation of interventions. Aim: This thesis aimed to describe and test a feasible framework of Injury Surveillance for the New Zealand construction industry. Accordingly, the specific objectives to accomplish this aim were: To identify an optimal surveillance dataset for New Zealand construction injuries; To assess potential sources of data and collection methods; To describe an ideal study design for undertaking injury surveillance; To implement an operational design based on industry stakeholder input; To undertake and evaluate an injury surveillance trial; and To suggest how a viable surveillance system could be permanently established. Method: A trial injury surveillance system was developed by identifying known construction injury risk factors from the literature, reviewing the data collection practices of the New Zealand industry and other potential data sources and consulting with industry stakeholders about the most feasible collection methodology. This surveillance framework was then tested by combining national data from routine Government sources and data from 3 construction companies that employed approximately 720 workers between them. National construction injury data was obtained from the Accident Compensation Corporation, the Department of Labour and the Injury Information Manager. The trial Surveillance System was then evaluated in terms of its ability to collect the full range of an optimal dataset, the quality and completeness of information actually collected, the ability to identify and monitor injury priorities for the industry, and the future viability and acceptability of this surveillance design to the industry. Results: A total of 468 medically treated injuries were recorded by the participating companies, with 15 (3.2%) considered to be Serious Harm injuries as defined by the Health and Safety in Employment Act. The level of data completeness across companies was especially low, with on average 18 out of 34 data fields (53%) completely unrecorded. The data from one company was sufficiently complete (i.e. 63% across all fields) to allow individual risk factor analyses to be conducted, whereas the absence of complete denominator data prevented the completion of the same analyses for the other two companies. Viewed overall, Government agency data was sufficiently detailed to estimate national longitudinal trends, injury agency and mechanism priorities for specific occupations and industry subsectors, and allowed a rudimentary evaluation of a national intervention programme. However, questions about data accuracy, completeness and under-reporting were raised for each of the Government data sources used. Conclusions: Using data entirely from Government sources appears to be the most immediately viable framework of Injury Surveillance for the New Zealand construction industry. As such, the relevant range of analyses demonstrated by this study should be continued, expanded and improved. In contrast, obtaining injury surveillance data from companies in the manner that was tested does not appear to feasible, given the difficulty in recruiting companies and the poor data completeness of those companies that did participate. However, the increased range of prevention targets identified by the company that did largely contribute data as intended, demonstrated that company surveillance had merit relative to existing procedures. Suggested steps toward implementing viable construction injury surveillance within New Zealand are outlined, including a recommendation to the industry�s Health and Safety organisation, SiteSafe, to investigate the most feasible data collection protocol for its members.

Background: Agricultural workers in New Zealand have high rates of occupational injury compared to most other occupational groups. They are also over-represented in work-related fatal injury statistics. While it is recognised that the personal and social costs of occupational injuries to agricultural workers are considerable, the ability to develop and evaluate evidence-based injury control strategies for this group has been limited by the lack of quality information on occupational exposures and injury events. Aim: The aim of this thesis was to develop and pilot a comprehensive occupational exposure and acute injury outcome data collection system for agricultural workers which will provide an evidence base for a public health approach to acute occupational injury control within the agricultural sector of New Zealand. The thesis objectives were therefore to: * Develop study methods to collect occupational exposure and injury outcome information. * Assess the likely validity of these study methods. * Determine the feasibility of implementing the study methods. * Suggest modifications to the study methods to enhance their validity and feasibility. Methods: Pastoral farms in the Waitaki region of New Zealand were identified using a database of New Zealand farm owners. The owners and workers on these farms were contacted and asked to participate. Participants were required to complete an Initial Questionnaire which included items on farm and personal characteristics, the farm environment, training, safety perceptions and attitudes and safety behaviour. Participants were then monitored for six months. During the monitoring period each participant completed a monthly log of their work activities during the preceding week. Any work-related injuries to workers on participating farms were also recorded and reported monthly. Participants who were injured were followed up for an interview to obtain detailed injury event information. At the end of the monitoring period a second Questionnaire was administered to assess change during the study. Participants were asked about any occupational injury events during the study as part of one of the monthly logs and the second Questionnaire to provide a comparison measure to the monthly reports. A random third of participating farms were visited at the end of the study to assess the validity of participants� reports on the farm environment. Results: Sixty-two farms were recruited into the study, a recruitment rate of 24%. This resulted in 82 study participants. Fifty-seven farms and 72 participants completed the study, resulting in retention rates of 92% and 88% respectively. Return of study items was high, with the lowest observed level of return being 92%. Levels of response error were low in most of the study items, with exceptions being the recording of the hours spent handling animals (37%) and total hours worked (22%). Most postal items (over 68%) were returned before a reminder call was made. Participants� reports about the farm environment closely matched the observations made during the visits, with little evidence of significant misreporting. The validity of reported injury events during the study could not be determined, as the two methods of capturing injury events identified different events. Conclusions: Within the limitations of the study, most of the study methods appeared to be feasible and have acceptable validity. The low recruitment rate and issues with validating the capture of injury events indicated that modifications to the study design were necessary to achieve acceptable validity and feasibility, however. Recommendations were made on how feasibility and validity might be improved.

La presente tesis va encaminada al estudio de la gravedad de los accidentes industriales con sustancias peligrosas mediante la utilización del análisis histórico de accidentes. Su objetivo final es utilizar el tratamiento cuantitativo de dicho análisis para desarrollar una metodología que permita predecir la gravedad de los accidentes.Los factores de riesgo que determinan la mayor o menor gravedad de los accidentes son tantos y tan complejos, que las aproximaciones determinísticas tradicionales no permiten desarrollar modelos adecuados que describan correctamente la realidad. Pretender establecer una correlación de determinados factores de riesgo para predecir la potencial gravedad de un accidente es, hoy por hoy, cuanto menos, difícil.Existen muchas publicaciones y trabajos de recopilación en los que se detallan innumerables situaciones accidentales. En los últimos años, el desarrollo de bases de datos informatizadas permite una utilización masiva de estas informaciones. Sin embargo, la utilización tradicional del análisis histórico de accidentes se ha encaminado prioritariamente a la detección cualitativa de factores de riesgo y de forma secundaria, al desarrollo cuantitativo de determinados modelos.El trabajo aquí presentado parte del estudio de diversas recopilaciones de accidentes de reconocido prestigio y de la aplicación de técnicas cuantitativas que permitan establecer o aproximar la potencial gravedad de los accidentes industriales en los que intervienen de manera directa sustancias peligrosas.Para ello, en el Capítulo 1, introductorio, se analiza la situación global de la seguridad industrial en España, se estudian las técnicas utilizadas en el análisis de riesgos y se presenta la orientación y el esquema general de este trabajo. En el Capítulo 2 se revisan los índices de riesgo más reconocidos, con el triple objetivo de conocer cuáles son los factores de riesgo determinantes en la gravedad de un accidente industrial, qué ponderación se hace de cada uno de ellos y qué relación matemática se establece para determinar el nivel de riesgo asociado a cada instalación.Dado que para el desarrollo de este trabajo era preciso seleccionar una base de datos en la que se recopilaran numerosos accidentes con sustancias peligrosas, en el Capítulo 3 se incluye una descripción detallada de las bases de datos y de las referencias bibliográficas que han servido de base o apoyo al estudio realizado. En los Anexos 1 y 2 se detallan la estructura y características fundamentales de las bases de datos MARS y MHIDAS. Este análisis permite identificar también determinados factores de riesgo en el uso industrial de sustancias peligrosas. Por la mayor disponibilidad de acceso, facilidad de tratamiento de la información y cantidad y calidad de los datos recogidos, se optó en su momento por basar el estudio en los accidentes de la base de datos MHIDAS. No obstante, en el análisis de diversas situaciones accidentales, se ha recurrido en muchas ocasiones a otras fuentes, también descritas.El Capítulo 4 presenta el estudio de la gravedad de los accidentes industriales en función de los factores descriptores disponibles en la base de datos MHIDAS (muertos, heridos y pérdidas económicas). Supone una primera aproximación al problema mediante estadística descriptiva permitiendo ratificar y cuantificar, a la luz del análisis histórico de accidentes, lo que el juicio experto indica de forma más o menos intuitiva y establecer las probabilidades de ocurrencia de accidentes de una determinada gravedad.La limitación más importante observada en los métodos estadísticos descritos en el Capítulo 4, es que los análisis realizados son uni o bivariantes, por lo que en cada caso se observan las relaciones entre una única variable explicativa y una variable respuesta. Con el fin de generalizar este análisis y poder observar las relaciones entre unas y otras variables, se presentan en el Capítulo 5 los resultados más significativos obtenidos al aplicar a los datos recopilados en MHIDAS, diversos métodos estadísticos multivariante. Los métodos utilizados han sido los siguientes: análisis de correspondencias múltiples y análisis de conglomerados, análisis de regresión múltiple, modelos logit, árboles de clasificación o decisión y análisis discriminante.En la medida en que cada uno de ellos lo permite, se ha intentado conocer el potencial predictivo y/o explicativo que cada método aplicado aporta al ser utilizado con los datos de MHIDAS. Los resultados obtenidos pueden tener validez y utilidad en el análisis cuantitativo de riesgos al permitir estimar la gravedad de los accidentes a partir de unos pocos parámetros generales.El Capítulo 6 presenta tres estudios concretos derivados de los análisis anteriores en los que, para ciertas tipologías de accidentes, se predice la evolución de su coste económico con el tiempo, la gravedad de los mismos en función de la cantidad de sustancia interviniente en las explosiones accidentales y una primera aproximación a los árboles de sucesos para sustancias inflamables.Finalmente se resumen las conclusiones derivadas de la realización de esta tesis y se recopila la bibliografía utilizada como referencia a lo largo de todo este estudio. / The present work is a study of the severity of industrial accidents involving hazardous substances using historical analysis. The final objective is to use a quantitative methodology to develop a technique to estimate the accidents severity.The hazards that determine the severity of the industrial accidents are much and very complex. Then, a conventional approach using deterministic techniques to develop models to predict the importance of the consequences are inadequate.There are many compilations and publications where it is possible to find a complete accident description. In the last years, using databases and information systems it has been possible to study a large number of accidents from a qualitative point of view. The traditional use of historical analysis of accidents is focused mainly on the qualitative detection of hazards and to the development of quantitative models.This work presents a study of several prestigious compilations of accidents and the application of quantitative techniques to these recompilations to estimate the severity of industrial accidents involving dangerous substances.Chapter 1 (an introduction), analyse the global situation of the industrial safety in Spain and presents the most important techniques used in the risk analysis showing the general structure of this work.Chapter 2 presents a review of the most important risk index to know what factors of risk are very relevant, the influence of each factor in the total level of risk of the plant and the mathematical relation between all of them.To develop this work it has been necessary to select a database where a big number of accidents involving hazardous substances have been compilated. In Chapter 3, it is included a detailed description of the databases and other bibliographic references used to develop this work. Annexes 1 and 2 detail the most important properties of MARS and MHIDAS databases respectively. This description permits also identify several relevant factors of risk in industrial environments working with dangerous substances. For his disposability, easiness of use, quantity and quality of information compiled, this study is focused on the MHIDAS database accidents. Nevertheless, in the analysis of several scenarios, other sources of information have been used.Chapter 4 presents the severity of these accidents as function of several factors available in MHIDAS database system (number of deaths, number of injured, number of evacuated and damage in economic units). This Chapter is a first approximation to the problem using descriptive statistics and its results permits to quantify and verify, using historical analysis of accidents, that the expert judgement shows in a more or less intuitive manner and provides the probabilities of occurrence of accidents of a determined severity.The most important limitation observed in the use of the statistical methods in Chapter 4 is that the analysis developed are uni or bivariate (they work only with one independent variable and one dependent variable) while in reality they are multivariate. In order to generalise these analyses and observe the relations between both variables, Chapter 5 shows the most significant results obtained applying several multivariate statistical methods to the data compiled in MHIDAS. The methods used in these analyses are: Multiple Correspondence analysis, cluster analysis, multiple regression analysis, logit models, decision trees analysis and discriminant analysis.According to the interesting results of each method, an attempt to know the predictive and/or descriptive potential of each method when it is used with MHIDAS data has been made. The obtained results have interest and utility in the quantitative risk analysis because they permit to estimate the severity of these accidents using a few parameters.Chapter 6 presents three studies developed from the previous analysis. In these studies, for certain typologies of accidents, it is possible to predict the evolution of the economic cost in time, the severity of these accidents and a first approach to the event tree for flammable substances.Finally, the conclusions arising from this study have been summarised, as well as the bibliography taken as a referent in the making of this study.

Thesis (Ph. D.)--University of Alberta, 2010. / Title from pdf file main screen (viewed on Feb. 3, 2010). A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements for the degree of Doctor of Philosophy, Political Science, Public Health Sciences, University of Alberta. Includes bibliographical references.

Background: Agricultural workers in New Zealand have high rates of occupational injury compared to most other occupational groups. They are also over-represented in work-related fatal injury statistics. While it is recognised that the personal and social costs of occupational injuries to agricultural workers are considerable, the ability to develop and evaluate evidence-based injury control strategies for this group has been limited by the lack of quality information on occupational exposures and injury events. Aim: The aim of this thesis was to develop and pilot a comprehensive occupational exposure and acute injury outcome data collection system for agricultural workers which will provide an evidence base for a public health approach to acute occupational injury control within the agricultural sector of New Zealand. The thesis objectives were therefore to: * Develop study methods to collect occupational exposure and injury outcome information. * Assess the likely validity of these study methods. * Determine the feasibility of implementing the study methods. * Suggest modifications to the study methods to enhance their validity and feasibility. Methods: Pastoral farms in the Waitaki region of New Zealand were identified using a database of New Zealand farm owners. The owners and workers on these farms were contacted and asked to participate. Participants were required to complete an Initial Questionnaire which included items on farm and personal characteristics, the farm environment, training, safety perceptions and attitudes and safety behaviour. Participants were then monitored for six months. During the monitoring period each participant completed a monthly log of their work activities during the preceding week. Any work-related injuries to workers on participating farms were also recorded and reported monthly. Participants who were injured were followed up for an interview to obtain detailed injury event information. At the end of the monitoring period a second Questionnaire was administered to assess change during the study. Participants were asked about any occupational injury events during the study as part of one of the monthly logs and the second Questionnaire to provide a comparison measure to the monthly reports. A random third of participating farms were visited at the end of the study to assess the validity of participants� reports on the farm environment. Results: Sixty-two farms were recruited into the study, a recruitment rate of 24%. This resulted in 82 study participants. Fifty-seven farms and 72 participants completed the study, resulting in retention rates of 92% and 88% respectively. Return of study items was high, with the lowest observed level of return being 92%. Levels of response error were low in most of the study items, with exceptions being the recording of the hours spent handling animals (37%) and total hours worked (22%). Most postal items (over 68%) were returned before a reminder call was made. Participants� reports about the farm environment closely matched the observations made during the visits, with little evidence of significant misreporting. The validity of reported injury events during the study could not be determined, as the two methods of capturing injury events identified different events. Conclusions: Within the limitations of the study, most of the study methods appeared to be feasible and have acceptable validity. The low recruitment rate and issues with validating the capture of injury events indicated that modifications to the study design were necessary to achieve acceptable validity and feasibility, however. Recommendations were made on how feasibility and validity might be improved.

Background: The burden of fatal and non-fatal injury for the New Zealand construction industry is larger than most other industrial sectors. Injury preventions efforts for construction have however been hampered because of insufficient, industry-specific, surveillance data that is essential for the effective targeting and evaluation of interventions. Aim: This thesis aimed to describe and test a feasible framework of Injury Surveillance for the New Zealand construction industry. Accordingly, the specific objectives to accomplish this aim were: To identify an optimal surveillance dataset for New Zealand construction injuries; To assess potential sources of data and collection methods; To describe an ideal study design for undertaking injury surveillance; To implement an operational design based on industry stakeholder input; To undertake and evaluate an injury surveillance trial; and To suggest how a viable surveillance system could be permanently established. Method: A trial injury surveillance system was developed by identifying known construction injury risk factors from the literature, reviewing the data collection practices of the New Zealand industry and other potential data sources and consulting with industry stakeholders about the most feasible collection methodology. This surveillance framework was then tested by combining national data from routine Government sources and data from 3 construction companies that employed approximately 720 workers between them. National construction injury data was obtained from the Accident Compensation Corporation, the Department of Labour and the Injury Information Manager. The trial Surveillance System was then evaluated in terms of its ability to collect the full range of an optimal dataset, the quality and completeness of information actually collected, the ability to identify and monitor injury priorities for the industry, and the future viability and acceptability of this surveillance design to the industry. Results: A total of 468 medically treated injuries were recorded by the participating companies, with 15 (3.2%) considered to be Serious Harm injuries as defined by the Health and Safety in Employment Act. The level of data completeness across companies was especially low, with on average 18 out of 34 data fields (53%) completely unrecorded. The data from one company was sufficiently complete (i.e. 63% across all fields) to allow individual risk factor analyses to be conducted, whereas the absence of complete denominator data prevented the completion of the same analyses for the other two companies. Viewed overall, Government agency data was sufficiently detailed to estimate national longitudinal trends, injury agency and mechanism priorities for specific occupations and industry subsectors, and allowed a rudimentary evaluation of a national intervention programme. However, questions about data accuracy, completeness and under-reporting were raised for each of the Government data sources used. Conclusions: Using data entirely from Government sources appears to be the most immediately viable framework of Injury Surveillance for the New Zealand construction industry. As such, the relevant range of analyses demonstrated by this study should be continued, expanded and improved. In contrast, obtaining injury surveillance data from companies in the manner that was tested does not appear to feasible, given the difficulty in recruiting companies and the poor data completeness of those companies that did participate. However, the increased range of prevention targets identified by the company that did largely contribute data as intended, demonstrated that company surveillance had merit relative to existing procedures. Suggested steps toward implementing viable construction injury surveillance within New Zealand are outlined, including a recommendation to the industry�s Health and Safety organisation, SiteSafe, to investigate the most feasible data collection protocol for its members.

Thesis (M.S.)--West Virginia University, 2000. / Title from document title page. Document formatted into pages; contains iii, 48 p. : ill. (some col.). Vita. Includes abstract. Includes bibliographical references (p. 43-45).

Thesis (Ph. D.)--University of Florida, 1983. / Description based on print version record. Typescript. Vita. Includes bibliographical references (leaves 104-111).

Made available in DSpace on 2016-01-13T13:27:40Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-06-18. Added 1 bitstream(s) on 2016-01-13T13:33:10Z : No. of bitstreams: 1 000851966.pdf: 2014959 bytes, checksum: 302671e03fe7fdb120d79533d609d563 (MD5) / A pesquisa teve como objetivo avaliar os benefícios da existência de refeitórios no local de trablaho e da utilização de Programas de Qualidade no ambiente de trabalho, com enfoque no em estar dos funcionários e índice de acidentes (no local de trabalho e de trajeto). Para consecução dos objetivos foi feita uma revisão bibliográfica referente ao tema e foram pesquisados 15 empresas no interior de São Paulo. As pesquisas nas empresas foram realizadas através de questionários para os responsáveis pelos recursos humanos e para os funcionários das empresas. A inteção foi analisar se existe a correlação entre os acidentes de trabalho típico e de trajeto, com a existência ou não de refeitório no local de trabalho e Programas de Qualidade. Também se buscou analisar a relação dos refeitórios com os aspectos ligados a Qualidade de vida no trabalho (QVT) / The objective of this research is evaluate the benefits of the existence of canteens inside the companies andimplementation of Quality Programs, focusing into the well-being of the employees and index of accidents (commuting and at work place). In order to achieve this objective a bibliographical revision about the subject was made and a study of casein 10 companies in the centre-west region of São Paulo. The study of case had been carried out through two questionnaires, one for the responsible for the human resources and other for the employees. The intention was to analyze if exists correlation among working accidents including commuting accidents and the existence or not canteens and Quality Programs. Also, analyze the relation of the refectories with on aspects the Quality of Life at Work (QLW)

Orientador: Maurício Cesar Delamaro / Banca: Messias Borges Silva / Banca: Jorge Muniz Junior / Banca: Fernando Augusto Silva Marins / Banca: Josimar Ribeiro de Almeida / Resumo: O objetivo desta tese foi identificar e analisar fatores relacionados com o ambiente profissional e pessoal do funcionário que possam contribuir para a ocorrência de quase-acidentes. Para isso foi realizado um estudo de caso em uma equipe operacional de logística de uma indústria química do Vale do Paraíba. O grupo pesquisado era composto de quarenta funcionários que participaram de um programa preventivo para redução de quase-acidentes em 2003. Foram conduzidas entrevistas semi-estruturadas que permitiram obter informações sobre o ambiente profissional e pessoal dos funcionários. Ao analisar o histórico de quase-acidentes ocorridos entre 2002 e 2009 foi possível definir dois grupos distintos: um grupo que não se envolveu com quase-acidentes e outro grupo que se envolveu com pelo menos um quase-acidente. Verificando as informações obtidas nas entrevistas e levando-se em conta os distintos grupos, foi possível identificar fatores com grande potencial para causar quase-acidentes. Analisando ainda os relatórios de segurança do departamento de logística verificou-se um crescimento dos quase-acidentes desde 2003, o que permitiu reaplicar um questionário no grupo pesquisado para verificar quais foram os fatores relacionados com o ambiente organizacional da logística que mais contribuíram para esse crescimento. Os resultados dessa tese confirmam o conceito de que acidentes não são fatalidades, pois as causas podem ser identificáveis eliminando, assim, a idéia de que o risco é um componente intrínseco do trabalho / Abstract: This thesis aims to identify and to analyze factors related with the employee's professional and personal environment that can contribute to the occurrence of near misses. For that a case study was done with an operational logistics team of a chemical industry of Vale do Paraíba. The researched group was composed of forty employees who participated of a preventive program for reduction of near misses in 2003. Semi-structured interviews were done, which allowed in obtaining information on the employees' professional and personal environment. After analyzing the history of near misses happened between 2002 and 2009 two different groups could be defined: a group that was not involved with near misses and other group that was involved with at least one near miss. Verifying the information obtained with interviews and considering the different groups, it was possible to identify factors with great potential to cause near misses. Analyzing the safety reports of the logistics department it was verified a growth of near misses since 2003, what allowed reapply a questionnaire with the group researched to verify which were the factors related with the logistics operational environment that have contributed more to that growth. The results of this thesis confirm the concept that accidents are not fatalities, as the causes can be identifiable, so eliminating the idea that the risk is an intrinsic component of the work / Doutor