Objetivos: validar una escala para evaluar la contaminación de alimentos en el hogar y, medir la validez y confiabilidad. Métodos: se encuestaron madres o cuidadoras con niños menores a 5 años, en San Juan de Lurigancho en Lima Perú. La escala fue previamente sometida a una calificación de jueces expertos. Se realizó el análisis de factores con componentes principales, la consistencia interna fue medida con Kuder–Richardson Formula 20 (KR-20) y la validez de la prueba con el coeficiente Kaiser-Meyer-Olkin (KMO) y el test de esfericidad de Barlet. Resultados: la muestra tuvo 194 cuidadoras. Se definieron tres componentes (prácticas de limpieza, prácticas de almacenaje y contaminación del niño) con un total de 10 ítems. El ítem Prácticas de limpieza tuvo un KR-20=0.722 prácticas de almacenaje y contaminación de alimentos 0.518 y 0.5 respectivamente. La validez de la prueba tuvo un KMO = 0.549 y un test de Barlet 591.678 , gl =136, (p<0.05). Conclusiones: esta escala puede contribuir a la planificación de intervenciones en el aspecto de contaminación en el hogar y por lo tanto reducir episodios de infecciones en los niños. / Objectives: To validate a scale to assess contamination of food at home and measure the validity and reliability. Methods: mothers or caregivers with children under 5 years were surveyed in San Juan de Lurigancho in Lima Peru. The scale was previously subject to a rating of expert judges. Factor analysis with principal components was performed; the internal consistency was measured with Kuder-Richardson Formula 20 (KR-20) and the validity of the test with the coefficient Kaiser-Meyer-Olkin (KMO) and the sphericity test Barlet. Results: The sample was 194 caregivers. Three components (Practices cleaning, storage practices and pollution of the child) with a total of 10 items were identified. The item Cleaning Practices had a KR-20 = 0.722 storage practices and food contamination 0.518 and 0.5 respectively. The validity of the test was KMO = 0.549 and 591,678. Barlet test, df = 136, (p <0.05). Conclusions: This scale can contribute to the planning of interventions in the aspect of pollution in the home and thus reduce episodes of infections in children.

A preocupação constante com a inocuidade e a qualidade dos alimentos irradiados, aliada à tendência de se estudar os aspectos sensoriais desses alimentos, foram a motivação para o desenvolvimento deste trabalho que teve como objetivos verificar os efeitos da radiação ionizante produzida pelo 60Co na sobrevivência de Salmonella sp em hambúrgueres de carne de frango congelados e verificar a aceitação do produto irradiado durante o seu período de estocagem. Amostras de hambúrguer de carne de frango congelados, inoculados com quatro cepas diferentes de Salmonella sp foram expostos a níveis de radiação com doses variando entre O e 3,5 kGy para verificação da sobrevivência da população de Salmonella sp presente. Amostras não inoculadas e irradiadas com doses de 0,5 e 7 kGy foram submetidas à avaliação sensorial por painel não treinado, pelo período de 120 dias, com o intuito de se verificar o decaimento da qualidade do produto. Os valores D10 encontrados para Salmonella sp variaram entre 1,02 e 1,32 kGy, portanto, as doses de 5 e 7 kGy, utilizadas nas fases seguintes do experimento, seriam suficientes para reduzir a população de Salmonella sp no produto em 5-6 ciclos logarítmicos. A exposição às doses de 5 e 7 kGy não afetou as características sensoriais do produto. A vida de prateleira do hambúrguer de carne de frango congelado e irradiado com doses de 5 e 7 kGy foi de, no mínimo, 120 dias, ou seja, igual à do produto não irradiado. / Safety and quality of irradiated foods are still of great concern to researchers and consumers, in general. The aims of this research were to study the effects of ionizing radiation from 60Co on the population of Salmonella sp inoculated in frozen chicken meat patties as well as on their sensory characterisitcs during the storage period. Samples of frozen chicken meat patties, inoculated with a pool of four strains of Salmonella sp, were exposed to irradiation doses varying from 0 to 3.5 kGy in order to determine their sensitivity (D10). Non inoculated samples of frozen chicken meat patties exposed to 5 and 7 kGy and a non inoculated and non irradiated sample were submitted to sensory evaluation by a non-trained panel. 010 values varied from 1.02 to 1.32 kGy Samples exposed to 5 and 7 kGy did not showed appreciable changes during the storage period of 120 days. Therefore, irradiation can be applied to chicken meat patties in order to improve their safety during a storage period of, at least, 120 days.

Ações para garantir a produção de alimentos e a preservação do solo e da água são essenciais para a sustentabilidade do ambiente agrícola. O uso racional de fertilizantes aumenta a produtividade agrícola e permite que menos áreas sejam desmatadas, preservando os recursos naturais. Porém, fertilizantes fontes de fósforo e de zinco podem contaminar o solo com metais pesados, por exemplo, com cádmio (Cd). O objetivo do trabalho foi avaliar a transferência do Cd contidos em fertilizantes minerais e orgânicos para plantas de arroz, empregando-se o isótopo estável 111Cd como traçador no sistema solo-fertilizante-planta. O experimento foi realizado em casa de vegetação, utilizando vasos com 3 dm3 de amostra de Latossolo Vermelho-Amarelo distrófico. A planta teste foi o arroz, cultivar CONAI. A alteração isotópica do solo foi feita com CdCl2, com 95,2% de 111Cd. Avaliaram-se 8 fertilizantes, 5 minerais e 3 orgânicos. Como fertilizantes minerais, usaram-se uma rocha fosfática importada, 2 superfosfatos simples, um nacional e outro importado, um resíduo industrial óxido de zinco e um fertilizante sulfato de zinco. Como fertilizantes orgânicos, usaram-se 2 lodos de esgoto e um composto de lixo. Foram conduzidos dois tratamentos testemunhas, um com e outro sem adubação. A planta foi dividida em parte aérea (folhas + colmos), casca e grão de arroz. As amostras de planta foram digeridas com ácido nítrico, em sistema micro-ondas. A concentração do cádmio e abundância do 111Cd foram determinadas por espectrometria de massas com plasma (ICP-MS). A adição do 111CdCl2 não interferiu no desenvolvimento e nutrição das plantas de arroz. Com o uso da técnica isotópica, calculou-se que a transferência do Cd dos fertilizantes para as plantas de arroz pode passar de 10%. Os fertilizantes minerais acidulados apresentaram maior transferência de Cd para a planta, com máximo de 30%. Os fertilizantes orgânicos apresentaram transferência de mais de 5% do Cd para as plantas de arroz. O uso da técnica isotópica foi eficaz no cálculo de transferência de Cd do fertilizante para as plantas de arroz, enquanto que o método da diferença, dependendo do fertilizante, subestima ou superestima a transferência do Cd / Actions to improve food yield and conservation of soil and water are essential to maintain the sustainability of agricultural fields. The technical use of fertilizers increase crop yields and allows fewer areas are deforested, preserving natural resources. Nevertheless, phosphate and zinc fertilizers can cause soil contamination with heavy metals, for example, with cadmium (Cd). The objective of this study was to evaluate the transfer of Cd from mineral and organic fertilizers to the rice plants using the 111Cd stable isotope as a tracer in soil-fertilizer-plant system. The experiment was carried out in greenhouse using pots filled with 3 dm3 of a Typic Haploduox soil. The plant test was the rice, cultivar CONAI. The soil isotopic enrichment was made with CdCl2 spiked with 95.2% of 111Cd. Eight fertilizers were studied: 5 minerals and 3 organics. Among mineral fertilizers, one imported rock phosphate, 2 simple superphosphate, one national and one imported, one industrial waste zinc oxide and one zinc sulphate fertilizers were used. Among organic fertilizers, two sewage sludge and one compost were used. Two control treatments, one with and one without fertilizer, were used. The plant was separated into shoot (leaf + stem), straw and rice grain. The plant samples were digested with nitric acid using microwave system. The concentration of Cd and 111Cd abundance were determined by plasma mass spectrometry (ICP-MS). The addition of 111CdCl2 did not affect the growth and nutrition of rice plant. The isotope technique has permitted to estimate that the transfer of Cd from fertilizers to the rice plant can be over 10%. The acidulated mineral fertilizers had higher percentages of Cd transfer to plant, up to 30%. Organic fertilizers had over than 5% of Cd transferred to rice plant. The isotopic method was effective in calculating the transfer of Cd from fertilizer to rice plant, meanwhile the difference method had showed under or overestimation of the Cd transference to plant, depending upon the fertilizer

O presente estudo foi dividido em duas fases; a primeira visou avaliar o desempenho de diversos meios de cultura, para fungos, no ar de ambientes de produção de alimentos, através da resposta de contagem e identificação dos gêneros que podem conter espécies indesejáveis; também foi avaliada a condição ambiental juntamente com a contagem total de bactérias. O ambiente de duas áreas foi utilizado para a pesquisa, uma indústria de doces, produção de doce de leite e doce de amendoim, na Cidade de Ribeirão Preto (SP) e, outra, uma indústria de embutidos, nos setores de empacotamento e produção, na Região de Piracicaba (SP). Os meios de cultura utilizados foram: "ichloran Rose Bengal Chloramphenicol Ágar" RBC), "Sabouraud Dextrose a 4% Ágar", "Malt Ágar" (MA), "Malt Extract Agar - Yeast and Molds" (MEAYM), "Plate Count Ágar-Cloranfenicol" (PCA-Cloranfenicol), "Dichloran 18% Glycerol Ágar" (DG 18), Batata Dextrose Ágar (BDA) e "Oxytetracycline Glucose YeastAgar" (OGY); para a contagem de bactérias foi utilizado o meio "Plate Count Agar" (PCA). Cada um dos pontos foi amostrado em triplicata utilizando um amostrador de impactação linear. A segunda fase do projeto avaliou os meios de cultura "Dichloran Rose Bengal Chloramphenicol Ágar" (DRBC), "Sabouraud Dextrose a 4% Ágar" selecionados da primeira fase. Novas coletas foram realizadas amostrando cinco repetições em cada um dos pontos, que permaneceram os mesmos da primeira fase. Em paralelo foram utilizadas culturas puras de Aspergillus flavus, Aspergillus fumigatus, Aspergillus niger, Aspergillus parasiticus, Fusarium moniliforme, Fusarium verticulloides, Histoplasma capsulatum e Stachybotrys chartarum, inoculando os dois meios selecionados e avaliando o desenvolvimento desses fungos ao longo do período de incubação de sete dias, a 28ºC, para servir de parâmetro de comparação com as coletas de ar realizadas no mesmo período, a fim de se verificar se algum desses fungos indesejáveis estaria presente nas amostras de ar. De acordo com os resultados obtidos, os meios DRBC e Sabouraud Dextrose a 4% Ágar, foram, estatisticamente, melhores que os demais meios testados, apresentando um maior número de unidades formadoras de colônias/m3. O meio de Extrato de Malte diferenciou estatisticamente dos demais e teve o menor desempenho para a quantificação desses microrganismos no ar. A recomendação do melhor meio para identificação de fungos no ar de indústrias alimentícias não foi conclusiva, com exceção do PCA-Cloranfenicol que apresentou baixa diversidade de gêneros, sendo considerado reprovado. No geral foi observado um número elevado de unidades formadoras de colônias de fungos e bactérias/m 3 durante as amostragens. A avaliação quali-quantitativa dos ambientes estudados sugere que esses não se encontram em condições adequadas, sendo necessária a elaboração de um padrão referencial para o monitoramento da indústria alimentícia em nosso país. O amostrador utilizado nas coletas promoveu rapidez nas coletas e proporcionou a expressão dos resultados em medidas confiáveis, por ser um equipamento que permite calibração em seu sistema de aspiração de ar. / The present study consisted of two phases. The first one aimed at evaluating the performance of different culture media for fungi found in the air of food production environments, through the results of counting and identifying genders that may contain undesirable species. The environmental condition was also evaluated along with a total bacteria counting. The present work took place in two different places, an industry which produces sweets made from milk or peanuts, in Ribeirao Preto city (SP) and, in the packing and production sections of a meat encased products industry located in Piracicaba region (SP). The culture media used were: "Dichloran Rose Bengal Chloramphenicol Agar (DRBC)", "Sabouraud Dextrose 4% Agar (MA)", "Malt Extract Agar - Yeast and Molds (MEAYM)", "Plate Count Agar - Chloramphenicol (PCA-Chloramphenicol)", "Dichloran 18% Glycerol Agar (DG 18)", "Potato Dextrose Agar (BDA)" and "Oxytetracycline Glucose Yeast Agar (OGY)". The medium used for the bacteria counting was "Plate Count Agar (PCA)". Three replicates of each sampled spot were obtained by using a linear impacting sampler. The second phase of this project evaluated the following culture media, selected during the first phase of this study: "Dichloran Rose Bengal Chloramphenicol Agar (DRBC)" and "Sabouraud Dextrose 4% Agar". New sample collections (five replicates) were carried out for each of the spots selected. The sampled spots were the same for both phases. In a parallel manner, pure cultures of Aspergillus flavus, Aspergillus fumigatus, Aspergillus niger, Aspergillus parasiticus, Fusarium moniliforme, Fusarium verticulloides, Histoplasma capsulatum and Stachybotrys chartarum were employed to inoculate the two selected media. The development of these fungi was evaluated throughout a 7-day-incubation period, at 28ºC, to function as a comparative reference for the air samples obtained in the same period in order to verify if any of these undesirable fungi was present in the air samples. According to the results obtained, DRBC and Sabouraud Dextrose 4% Agar media were statistically superior than the others, presenting a higher number of colony forming units/m3. The Malt Extract medium was statistically different from the others and showed the worse performance as to the quantification of microorganisms present in the air. There was no conclusive recommendation as to the best medium for the identification of fungi found in the air of food industries, except for PCA-Chloramphenicol, which showed low diversity of genders and was discarded. In general, a high number of fungi and bacteria colony forming units/m3 was observed during the sampling procedures. The qualitative-quantitative evaluation of the environments studied suggests that they do not present adequate conditions, evidencing the need for the establishment of a referential standard in order to monitor food industries in our country. The sampler used in this work allowed a quick sample collection and a reliable expression of results, as this equipment enables the calibration of its air intake system.

Thesis submitted in fulfilment of the requirements for the degree Master of Technology: Chemistry in the Faculty of Applied Sciences at the Cape Peninsula University of Technology / Paperboard made from recycled fibres is being used more frequently in direct food packaging applications, in addition to its use as secondary and tertiary packaging. However, recent research has shown that there is a risk that harmful chemicals may migrate from the paperboard into the food. The simplest approach to reducing the migration of these contaminants is the use of barrier films. The barrier efficiencies of these various films can be examined by means of a migration test into a food simulant, followed by extraction in a suitable solvent. The extract can then be analysed by chromatographic techniques such as gas chromatography mass spectrometry (GC-MS) to determine the concentration of the specific contaminants. However on a production level, the availability of this type of highly specialised equipment is limited. A simple, cost effective method is needed to evaluate the barrier properties to specific chemical contaminants. The Heptane Vapour Transmission Rate (HVTR) test is a permeation test method for use at quality control level to determine barrier properties to the migration of organic vapours. The first part of the study focussed on establishing a universal correlation between HVTR and specific migration of diisobutyl phthalate (DiBP), dibutyl phthalate (DBP) and diethylhexyl phthalate (DEHP) that would be applicable to any type of functional barrier. However, experimental data demonstrated this was not possible as the correlation factor linking HVTR to specific migration was largely dependent on the type and morphology of the coating considered. The initial objective of the study was reconsidered in favour of building individual models specific to the nature of the coating and substrate considered. A correlation between HVTR and specific migration of DiBP, DBP and DEHP for a polyvinylidene chloride (PVDC) barrier polymer was constructed by varying the applied coating weight. The vapour transport mechanism for the HVTR test and the specific migration test were found to differ, showing that a direct correlation between HVTR and the specific migration was again not possible. However, an indirect correlation could be made. The HVTR method gives an indication of film integrity, whereas the coating weight could be used as an indicator of the specific migration. The correlation between the coating weight and the specific migration yielded an equation that can be used to calculate the specific migration through the PVDC barrier polymer, provided the quantity of the chemical contaminant originally present in the paperboard was known. This equation was specific to the type of barrier polymer, the specific chemical contaminant as well as the intended shelf-life of the food product to be packaged in the paperboard.

Dissertation submitted in fulfilment of the requirements for the degree Master of Technology: Environmental Health in the Faculty of Applied Sciences at the Cape Peninsula University of Technology, 2011 / The retail sector in South Africa is increasingly evolving into a dynamic industry, driven by changes in technology, saturating markets and globalisation. A major phenomenon in South Africa has been the evolution of hypermarkets, which sell large quantities of almost all consumer goods on a self-service basis. The South African consumers are becoming increasingly health conscious and, as such, the demand for wellness foods, health and convenience food has escalated. Convenience foods are expected to remain popular with consumers and supermarkets and will therefore increase the amount of ready-to-eat food items offered. As the retail industry has changed over the last two decades, so has the epidemiology of foodborne illnesses, with an increase in the incidence of bacterial infections caused by emerging organisms. In addition, there are certain food safety issues specifically associated with ready-to-eat foods. In recent years, incidences of enteric diseases associated with meat consumption have risen. The emergence of several new foodborne diseases has led to an increased focus attention on the issue of food safety by consumers and the industry. The most commonly implicated foods in these disease outbreaks have been meat and dairy products. The microbial load of eight convenience food manufacturing plants was determined by firstly sampling stainless steel food contact surfaces after they had been cleaned and sanitised at the end of a day‘s shift. The samples were analysed for Total Plate Count (TPC), Escherichia coli, Salmonella species and Staphylococcus aureus and Listeria. The results showed that 59 % of the total areas sampled for TPC failed to comply with the legal requirements for food surfaces specified in the South African Health Act (< 100 cfu.cm-2). Listeria was detected in 23 % of the samples taken and E.coli was found in 1.3 % of the samples, while S. aureus was not detected in any of the samples. Fifty percent of the plants applied conventional cleaning methods for cleaning and sanitation and the remaining 50 % used the low-pressure foam (LPF) method. The bacterial results of the two cleaning methods were statistically compared and a statistically significant difference (P ≤ 0.05) was found between the TPC means of the cleaning methods after cleaning. No statistically significant difference (P > 0.05) was found in terms of the Listeria species counts after both cleaning processes. The LPF method proved to be the superior cleaning option for reducing TPC counts. Secondly surface samples were collected from washed and sanitised dominant hands of food handlers and analysed for the presence of total plate counts, S. aureus and E. coli. The study aimed to evaluate the efficacy of hand washing practices and sanitation before commencing work. A total of 230 samples were collected, involving 100 % of the food handlers in selected convenience food outlets. The highest bacterial count taken from handswas 7.4 x 10-3 cfu.cm-2 and the lowest showed no detectable growth. Forty percent of the TPC analysed complied with the legal limit of < 100 cfu.cm-2 and only 18 % of the food handlers had no detectable bacteria present on their hands. One hand sample tested positive for E. coli, which is generally viewed as an indication of faecal contamination. S. aureus could not be detected on the hands of any of the food handlers. The results of this study indicated that hand hygiene is unsatisfactory and underlined the importance of further training to improve food handlers‘ knowledge of good hand washing practices. The study also aimed to present data on the food hygiene knowledge and practices of food handlers based on a representative sample from convenience food outlets in the Gauteng area. The management, as well as food handlers, were interviewed without prior announcement and managers were interviewed prior to starting their shifts, followed by food handlers, after they had passed through the change room and hand wash facilities. Although the majority of food handlers adhered to basic hygiene principles, the results highlighted a need for proper and continuous training in hygiene practices, not only for food handlers, but also for management. Furthermore, all food handlers should adhere to a formal cleaning schedule and specific courses should be planned for food handlers. Most training is done away from the workplace and the workers might find it difficult to translate theory into practice. Although food safety training programmes are essential, behavioural changes will not occur merely as a result of having received training but rather continuous development of food handlers. In conclusion, the popularity of convenience food is bound to increase with the growing appeal for modern foods. Consumers in South Africa nowadays demand good quality and safe products at a reasonable cost. Due to continuous time constraints, convenience food is the food of the future for the working mother. It is clear that managing foodborne disease is a challenge and an economic problem subject to various constraints. Food safety has too often become a hit-or-miss gamble, with parents obliged to roll the dice when it comes to the safety of their children‘s food and consumers in general. The food industry therefore needs to improve food safety processes to prevent the contamination of foods and use methods to ensure safe food for consumers. Better training, more testing and better methods of tracking food must be utilised to verify that the processes are working. This study endeavoured to add to the understanding and improvement of hygiene processes as well as food handlers‘ practices in the convenience food industry in the Gauteng Province.

Introdução: As doenças transmitidas por alimentos (DTA) constituem um grave problema de saúde pública a nível mundial. A Organização Mundial da Saúde estima que mais de um terço da população, incluindo a dos países desenvolvidos, é acometida por surtos de DTA anualmente, embora a maioria dos casos não seja notificada às autoridades sanitárias locais. Objetivo: Analisar a evolução das DTA nas diferentes regiões do Brasil entre 2000 e 2015. Metodologia: O presente estudo, de caráter descritivo, consistiu na análise de dados do Ministério da Saúde e de artigos científicos que tratam das DTA no contexto nacional, no período de janeiro de 2000 a dezembro de 2015. A primeira fase do estudo foi caracterizada pela coleta de dados sobre DTA disponíveis pelo Ministério da Saúde. Na segunda fase, foi realizado o levantamento e análise geral da produção científica brasileira nas seguintes bases de dados: LILACS, SciELO, Scopus, Web of Science, Pubmed e Embase. E na terceira fase do estudo foi traçado o panorama das DTA nas diferentes regiões do Brasil, relacionando os dados oficiais do Ministério da Saúde com os dados das pesquisas publicadas. Resultados: Entre 2000 e 2015 foram registrados no Ministério da Saúde 11.524 surtos de DTA com 219.909 doentes e 167 óbitos e publicados 50 artigos científicos sobre surtos de DTA no Brasil. Foi possível verificar que a região Sudeste apresentou maior prevalência no número de trabalhos publicados e de surtos de DTA. Na análise dos dados do MS observou-se que os alimentos mistos (12,4 por cento ) foram os mais envolvidos nos surtos, seguidos por aqueles preparados com ovos e produtos à base de ovos (8,7 por cento ); o agente etiológico não foi identificado em 57,8 por cento dos surtos e quando identificado, a Salmonella spp. foi o mais frequente (14,4 por cento ) e a maioria dos surtos ocorreram nas residências (36,6 por cento ). Considerando os dados dos artigos, a água foi o alimento que mais surtos de DTA causou (22 por cento ), o agente etiológico mais estudado foi a Salmonella spp. (22 por cento ) e as residências foram os locais mais envolvidos nos surtos (48 por cento ). Conclusões: Os dados obtidos indicam a necessidade de novas políticas públicas de incentivo a notificação dos surtos, bem como o desenvolvimento de programas de orientação e educação em boas práticas de manipulação de alimentos para a população e estabelecimentos produtores de alimentos / Introduction: Foodborne disease is a significant public health problem worldwide. The World Health Organization estimates that more than a third of the world\'s population, including those living in developed countries, is affected by foodborne disease, although many cases are not reported to the local health authority. Purpose: This study aimed to evaluate the occurrence of foodborne disease around different regions of Brazil between 2000 and 2015. Methodology: This is a descriptive study, based on analysis of data from the Brazilian Ministry of Health and from scientific manuscripts regarding foodborne disease at a national level, occurred between January 2000 and December 2015. Data from both sources were evaluated and compared to map the panorama of foodborne disease outbreaks in Brazil. Results: In the period of 2000 to 2015, a total of 11,524 foodborne outbreaks were reported to the Brazilian Ministry of Health, involving 219,909 cases and 167 deaths. In the same period, 50 manuscripts dealing with this topic were published. The largest prevalence of foodborne outbreaks was observed in the Southeast region of Brazil. Based on data from the Brazilian Ministry of Health, mixed foods were the most involved in the outbreaks (12.4 per cent ), followed by those foods prepared with eggs and egg products (8.7 per cent ). In most of the cases, the etiologic agent was not identified (57.8 per cent ). Among those identified, Salmonella spp. was the most frequent (14.4 per cent ), while homes were the main site of occurrence (36,6 per cent ). Based on data from the manuscripts, water was the most involved in the outbreaks (22 per cent ), while Salmonella spp. was the most reported pathogen (22 per cent ) and homes the main site of occurrence (48 per cent ). Conclusion: Notification of foodborne disease outbreaks is one of the fundamental duties to facilitate public health action and needs to be encouraged

No description available.

No description available.

Thesis (M.S.)--University of Missouri-Columbia, 2006. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file viewed on (Feb. 23, 2007). Includes bibliographical references.