乳品安全中的微生物风险评估

乳品安全日益成为社会和科学关注的焦点.本文介绍了乳品安全及微生物风险评估概念,并从影响乳品安全的危害因素出发,对乳品中的微生物危害来源及现状作了全面性综述.详细论述了国内外乳品中微生物风险评估情况和通用步骤,比较了微生物风险评估与化学危害风险评估的差异,最后对乳品微生物风险评估在中国的未来发展方向作了前瞻性展望.     

Paediatric HUS Cases Related to the Consumption of Raw Milk Sold by Vending Machines in Italy: Quantitative Risk Assessment Based on Escherichia coli O157 Official Controls over 7 years

A quantitative risk assessment (RA) was developed to estimate haemolytic‐uremic syndrome (HUS) cases in paediatric population associated with the consumption of raw milk sold in vending machines in Italy. The historical national evolution of raw milk consumption phenomenon since 2008, when consumer interest started to grow, and after 7 years of marketing adjustment, is outlined. Exposure assessment was based on the official Shiga toxin‐producing Escherichia coli O157:H7 (STEC) microbiological records of raw milk samples from vending machines monitored by the regional Veterinary Authorities from 2008 to 2014, microbial growth during storage, consumption frequency of raw milk, serving size, consumption preference and age of consumers. The differential risk considered milk handled under regulation conditions (4°C throughout all phases) and the worst time–temperature field handling conditions detected. In case of boiling milk before consumption, we assumed that the risk of HUS is fixed at zero. The model estimates clearly show that the public health significance of HUS cases due to raw milk STEC contamination depends on the current variability surrounding the risk profile of the food and the consumer behaviour has more impact than milk storage scenario. The estimated HUS cases predicted by our model are roughly in line with the effective STEC O157‐associated HUS cases notified in Italy only when the proportion of consumers not boiling milk before consumption is assumed to be 1%. Raw milk consumption remains a source of E. coli O157:H7 for humans, but its overall relevance is likely to have subsided and significant caution should be exerted for temporal, geographical and consumers behaviour analysis. Health education programmes and regulatory actions are required to educate people, primarily children, on other STEC sources.     

Food Safety Hazards and Microbiological Zoonoses in European Meat Imports Detected in Border Inspection in the Period 2008–2013

Microbiological contaminations and other food safety hazards are omnipresent within the European Union (EU) and a considerable risk for consumers, particularly in imported meat and meat products. The number of rejections at external EU borders has been increasing in recent years. Official authorities in each member state are therefore obliged to notify border rejections of food and animal feed due to a direct or indirect risk to human or animal health. This study explored the trends and temporal and spatial distribution of notifications on food safety hazards between January 2008 and December 2013 with a special emphasis on microbiological zoonoses in meat and meat products including poultry at border checks resulting from the rapid alert system for food and feed (RASFF). Results indicated that border rejection notifications are increasing exponentially, frequently due to Salmonella in poultry and shiga‐toxin‐producing E. coli in meat and meat products.     

国内外原料乳中金黄色葡萄球菌风险评估研究综述

近年来我国乳品消费量持续增加,乳品质量安全关乎着人们身体健康和生命安全.金黄色葡萄球菌及其肠毒素引起的乳品中毒事件时有发生,逐本求源地针对原料乳开展金黄色葡萄球菌风险评估研究是十分必要的.本文根据国际食品法典委员会拟定的微生物风险评估的原则和指导方针,从危害识别、危害特征描述、暴露评估、风险描述四方面综述了关于原料乳中金黄色葡萄球菌国内外风险评估的研究现状.通过国外成熟的定量风险评估案例,介绍关于原料乳中金黄色葡萄球菌定量风险评估的基本框架,提出未来应加强构建肠毒素预测模型,较为系统的对原料乳从牧场到餐桌进行全程监控和风险评估的建议.     

Risk Assessment of Bacitracin Resistance as a Medicated Feed Additive

The aim of this study was to provide theoretical basis for regulatory departments to formulate risk management measures for bacterial resistance of animal origin.The point assessment method was applied to assess the risk of bacitracin resistance in animal origin foods of Chinese residents,according to the model of risk assessment analysis,based on the data from the monitoring report of the citizen's physiques in China in 2014,the fifth Chinese total diet study and the official website of European Medicine Agency.Dietary exposure to bacitracin was calculated by estimated daily intake,and hazard quotient (HQ) was used in risk characterisation to indicate risk level of health damage exposed to bacitracin.Firstly,through the way of hazard identification,the toxicity of bacitracin by oral ingestion was very low.But the pathogenic bacteria such as Enterococcus and Clostridium perfringens had developed resistance to bacitracin,with the resistance genes easily transferring to the human intestinal flora through bacteria commonness.Secondly,the hazard characterization indicated the toxicological ADI was 0.055 mg/(kg·BW),and the microbiological ADI was 3.9 μg/(kg·BW).Thirdly,exposure assessment showed that the maximum dietary exposure of bacitracin was in 2 to 7 age group among Chinese residents.And that of women through intake of dairy products was higher than men.Among 2 to 7 age group,dairy products contributed most to the dietary exposure of bacitracin,while meat was the largest contributor above 8 years old.Fourthly,risk characterization hazard quotient was 1.4901 for males and 1.4121 for females among Chinese residents aged 2 to 7 by calculation,and <1 among other age groups.Finally,the uncertainty analysis showed that the lack of monitoring data of bacitracin residues in animal foods,the lag of authoritative data collection and release,the absence of consumption data of aquatic products brought uncertainty to the results.In total,the risk of bacitracin resistance as a medicated feed additive through animal foods exposure was high in children aged 2 to 7 and moderate in other age groups.The maximum residue limit standard should be revised appropriately and risk management of bacitracin resistance need to be strengthened in imported animal foods.     

药物饲料添加剂杆菌肽耐药性风险评估

本研究旨在为监管部门制定动物源细菌耐药风险管理措施提供理论依据。利用2014年国民体质监测数据、第五次中国总膳食研究数据和欧洲药品管理局官方网站数据,采用抗菌药物风险评估模型,对中国居民经动物性食品摄入杆菌肽的耐药性风险进行点评估。用估算每日摄入量作为杆菌肽膳食暴露量,用危害指数（HQ）进行风险特征描述,反映杆菌肽耐药对人体健康损害的风险等级。结果表明,危害识别得杆菌肽经口毒性极低,但肠球菌、产气荚膜梭菌等致病菌已对杆菌肽产生耐药性,耐药基因可转移至人体肠道菌群。经危害特征描述可知,杆菌肽毒理学每日允许摄入量（ADI）为0.055 mg/（kg·BW）,微生物学ADI为3.9 μg/（kg·BW）。暴露评估发现,中国居民中2~7岁人群杆菌肽的膳食暴露量最高,女性通过摄入乳类的杆菌肽膳食暴露量高于男性。2~7岁人群中,杆菌肽的膳食暴露量以乳类贡献最大,8岁以上人群杆菌肽的膳食暴露量以肉类贡献最大。风险特征描述发现,中国居民中通过摄入肉、蛋、乳等动物性食品的杆菌肽耐药性风险,2~7岁人群HQ分别为男性1.4901、女性1.4121;其他年龄人群均<1。不确定性分析发现,结果不确定性源自缺乏动物性食品杆菌肽残留量监测数据、权威数据发布的滞后性、未考虑水产品等因素。综上动物性食品中药物饲料添加剂杆菌肽耐药性风险,2~7岁人群为高风险,其他年龄段人群为中等风险。针对杆菌肽的耐药性风险,必要时修订动物性食品中杆菌肽的最大残留限量,同时加强对进口动物性食品的杆菌肽耐药性风险管理。     

Quantitative Risk Assessment of Human Trichinellosis Caused by Consumption of Pork Meat Sausages in Argentina

In Argentina, there are three known species of genus Trichinella; however, Trichinella spiralis is most commonly associated with domestic pigs and it is recognized as the main cause of human trichinellosis by the consumption of products made with raw or insufficiently cooked pork meat. In some areas of Argentina, this disease is endemic and it is thus necessary to develop a more effective programme of prevention and control. Here, we developed a quantitative risk assessment of human trichinellosis following pork meat sausage consumption, which may be used to identify the stages with greater impact on the probability of acquiring the disease. The quantitative model was designed to describe the conditions in which the meat is produced, processed, transported, stored, sold and consumed in Argentina. The model predicted a risk of human trichinellosis of 4.88 × 10^?6 and an estimated annual number of trichinellosis cases of 109. The risk of human trichinellosis was sensitive to the number of Trichinella larvae that effectively survived the storage period (r = 0.89), the average probability of infection (P_Pinf) (r = 0.44) and the storage time (Storage) (r = 0.08). This model allowed assessing the impact of different factors influencing the risk of acquiring trichinellosis. The model may thus help to select possible strategies to reduce the risk in the chain of by‐products of pork production.     

Import risk analysis in animal disease control

At the Swiss Federal Veterinary Office risk analyses are conducted according to international standards. A risk analysis contains the elements risk management, risk assessment and risk communication. A risk assessment is based on risk profile, hazard identification and a pathway model. All available information is gathered, documented and assessed and the risk estimated. The question. "What is the probability that unprocessed wild boar meat imported to Switzerland from the federal state Mecklenburg Western Pommerania is contaminated with classical swine fever virus?" was answered by a release assessment. The hazard identification recognized classical swine fever virus and attenuated live virus vaccine used for oral immunization as hazards. The probability of contamination was estimated to be small. The question: "What is the likelihood to introduce Aujeszky's disease to Switzerland and infect the indigenous pig population with the disease, by means of importing pork and meat products?" was answered by assessing the release, exposure and resulting consequences. The risk of an infection of the indigenous pig population was estimated to be very small, as 80% of the imported products derive from countries or zones free from Aujeszky's disease. Furthermore the majority of the imported products are processed. The strict implementation of the regulations governing feeding of food wastes to pigs reduces the probability of exposure. In all assessments the risk management decides on a strategy to deal with the risk, taking into consideration the results and recommendations derived from the risk assessment as well as other relevant factors.     

Quantitative Risk Assessment for Zoonotic Transmission of Cryptosporidium parvum Infection Attributable to Recreational Use of Farmland

Cryptosporidiosis caused by Cryptosporidium parvum infection is a major cause of enteric illness in man and there is a significant reservoir in animals, particularly young ruminant species. To preliminary assess the magnitude of the risk posed by contact with faeces produced by infected livestock, two microbiological risk assessments have been developed: one for the risk of human infection with C. parvum while camping on contaminated land recently grazed by infected suckler cattle and a comparable risk assessment for camping on land recently spread with contaminated cattle slurry. Using a worst‐case scenario approach, the upper level of risk was estimated to be one infection in every 6211 person‐visits for a camping event on land recently grazed by infected cattle. Translated into camping events of 100 persons, this risk estimate would most likely lead to zero (98% likelihood) or one infection (1% likelihood). The results for cattle slurry model are similar despite different pathways. Sensitivity analysis was conducted for the grazing cattle model only. This suggested that the time between grazing and camping was the most important control strategy, but increasing hand‐washing frequency and the removal of cattle faeces before camping would also be beneficial. If the upper level of risk were to be judged unacceptable then further data would be required to more accurately estimate the risk of infection through these scenarios. Further research would also be required to assess the fraction of cases attributable to camping and/or environmental contact with Cryptosporidium oocysts.     

Risk factors for herds to test positive for Mycobacterium avium ssp. paratuberculosis-antibodies with a commercial milk enzyme-linked immunosorbent assay (ELISA) in Ontario and western Canada

The objectives of this study were to identify risk factors associated with i) a Mycobacterium avium subsp. paratuberculosis (MAP)-antibody milk enzyme-linked immunosorbent assay (MAP milk ELISA)-positive herd status, and ii) the within-herd MAP milk ELISA-positive prevalence in Canadian dairy herds. This prospective cohort study was conducted between 2005 and 2009 on 226 herds in Ontario and western Canada, which participated in a voluntary risk assessment (RA)-based Johne’s disease control program. Two MAP milk ELISA and risk assessments and a previsit survey were available per herd. The overall farm RA scores alone could not be used to predict whether a herd would test positive for MAP antibodies. However, the results of this study indicated that increasing the likelihood of exposing calves to MAP through certain management practices, as assessed with the RA, increased the likelihood of a herd being test-positive for MAP antibodies.     

乳品中硫氰酸盐的来源及其风险分析

硫氰酸盐曾作为生鲜奶的保鲜剂被广泛使用,但因其毒性,已被我国列为乳及乳制品中的非法添加物质.鉴于硫氰酸盐的毒性作用,我们对市售乳品的硫氰酸盐含量进行风险监测,并对乳品中硫氰酸盐可能对人类健康造成的危害进行风险分析,结果发现852个监测样品中的硫氰酸含量均小于10mg/kg,其中25.9％样品的硫氰酸根含量低于1 mg/kg,仅有6.24％的样品含量在5～10mg/kg之间.由此,我们分析了乳品中硫氰酸盐的可能来源,认为852个市售乳制品中的硫氰酸盐主要来源于动物生鲜乳的自身本底,系非法添加的可能性极低.     

Prevalence of Mastitis and Brucellosis in Cattle in Awassa and the Peri‐Urban Areas of Two Smaller Towns

The prevalence of mastitis and brucellosis in urban and peri‐urban settings was studied in Awassa and two smaller nearby towns in southern Ethiopia, because milk‐born diseases are causing a risk for human health, besides direct impacts on animal production. Mastitis was investigated by examining 80 cows (320 udder quarters) using California mastitis test (CMT) and somatic cell count (SCC). The prevalence of brucellosis was assessed by sampling 177 cattle in Awassa and its peri‐urban areas using serological methods. Logistic regression was used to analyse risk factors associated with mastitis. Prevalence of clinical mastitis on quarter level was 0.9%, and 1.9% of quarters were non‐functional or blocked. Prevalence of sub‐clinical mastitis at quarter level in urban and peri‐urban areas was significantly different (P < 0.05). Cows in large herds and at advanced lactation number were associated with higher risk of infection. The percentage of quarters positive on CMT (42.5%) was close to the percentage‐positive detected by SCC (41.2%). Prevalence of brucellosis was 3.9% in the peri‐urban area, while no brucellosis cases were detected in Awassa. More frequent use of artificial insemination in the urban than in peri‐urban area might have contributed to the absence of brucellosis in the urban location. The extent of mastitis is, however, a threat to the dairy enterprise in and around Awassa. Pasteurization of milk and milk products is indicated in some parts of the area because of the danger of brucellosis.     

Human Pathogens in Marine Mammal Meat – A Northern Perspective

Only a few countries worldwide hunt seals and whales commercially. In Norway, hooded and harp seals and minke whales are commercially harvested, and coastal seals (harbour and grey seals) are hunted as game. Marine mammal meat is sold to the public and thus included in general microbiological meat control regulations. Slaughtering and dressing of marine mammals are performed in the open air on deck, and many factors on board sealing or whaling vessels may affect meat quality, such as the ice used for cooling whale meat and the seawater used for cleaning, storage of whale meat in the open air until ambient temperature is reached, and the hygienic conditions of equipment, decks, and other surfaces. Based on existing reports, it appears that meat of seal and whale does not usually represent a microbiological hazard to consumers in Norway, because human disease has not been associated with consumption of such foods. However, as hygienic control on marine mammal meat is ad hoc, mainly based on spot‐testing, and addresses very few human pathogens, this conclusion may be premature. Additionally, few data from surveys or systematic quality control screenings have been published. This review examines the occurrence of potential human pathogens in marine mammals, as well as critical points for contamination of meat during the slaughter, dressing, cooling, storage and processing of meat. Some zoonotic agents are of particular relevance as foodborne pathogens, such as Trichinella spp., Toxoplasma gondii, Salmonella and Leptospira spp. In addition, Mycoplasma spp. parapoxvirus and Mycobacterium spp. constitute occupational risks during handling of marine mammals and marine mammal products. Adequate training in hygienic procedures is necessary to minimize the risk of contamination on board, and acquiring further data is essential for obtaining a realistic assessment of the microbiological risk to humans from consuming marine mammal meat.     

Neospora‐associated Abortion Episode over a 1‐Year Period in a Dairy Herd in North‐east Spain

This report describes a retrospective study based on serological screening, performed on a Neospora‐associated abortion episode over a 12‐month period in a dairy herd in north‐east Spain. During the 1‐year study period, the overall abortion rate for the herd was 23.2% (38 abortions of 164 diagnosed pregnancies). The data analysed were those derived from blood samples collected from the whole herd (n = 237) and from diagnosed pregnancies. Antibodies to Neospora caninum were found in 35.4% of the cattle with 44% of seropositive pregnant animals aborting over 1‐year period. Based on the odds ratio, the risk of abortion was 12.2 times higher (P < 0.0001) in the Neospora‐seropositive animals than in seronegative animals and significantly higher during the second term of gestation (P < 0.01) than during the first and third terms. Abortions were not found to be associated with parity status or season of pregnancy, and the common risk factors associated with pregnancy loss in the geographical area of the study. Age‐related differences in N. caninum seroprevalence were not statistically significant, indicating vertical transmission as the main route of infection. Indeed, a high percentage of congenitally infected offspring was observed (90.6%) and the farm had been free of dogs for the last 7 years. Our results suggest that, when a dairy herd shows an increased incidence of abortions due to Neosporosis, maternal serology can be a good indicator of the abortion risk in individual cows, and that the effects of factors normally related to abortion, such as parity and pregnancy season, may be masked.     

Prevalence and Determinants of Mastitis and Milk‐borne Zoonoses in Smallholder Dairy Farming Sector in Kibaha and Morogoro Districts in Eastern Tanzania

A study was carried out to establish the prevalence and determinants of mastitis and milk‐borne zoonoses in smallholder dairy farms in Kibaha and Morogoro districts (Tanzania). A total of 57 herds comprising 114 milking cows in Kibaha and 48 herds consisting of 96 milking animals in Morogoro were included in the study. A questionnaire survey was used to assess the socio‐economic determinants of mastitis, whereas California mastitis test (CMT) and microbiological assessment of milk was carried out to establish the status of mastitis and responsible aetiological agents. Seroconversion for brucellosis was evaluated using enzyme‐linked immunosorbent assay (ELISA). Animals were also tuberculin‐tested using a single comparative intradermal method and milk samples were cultured for isolation of Mycobacterium species. Based on CMT, the cow‐based prevalence of subclinical mastitis was 82.4% in Kibaha and 62.4% in Morogoro. Of the 919‐quarter milk samples cultured, 8.2% were positive for aerobic bacteria with predominant isolates being Staphylococcus epidermidis (2.8%), Staph. aureus (1.7%), Streptococcus agalactiae (1.2%) and Staph. intermedius (1.1%). There was a strong association between CMT positivity and bacteriological isolation [relative risk (RR) = 2.60; P = 0.02]. Fungal growth was observed in 21.8% (n = 881) of the samples and the isolates were yeast (19.2%), Mucor (2.5%) and Aspergillus (0.1%). Bucket feeding of calves was associated with increased risk of a quarter being CMT positive (RR = 1.24; P = 0.000), while residual calf suckling was associated with decreased risk of positivity (RR = 0.86; P = 0.015). Earth floor was associated with increased risk of CMT positivity at quarter level (RR = 1.13; P = 0.041) and Jersey breed was identified as a risk factor to mastitis. The prevalence of bovine tuberculosis was 0.4% (n = 259) and 1.7% (n = 181) in Kibaha and Morogoro, respectively. Similarly, the prevalence of brucellosis was 1% (n = 208) in Kibaha and 1.9% (n = 104) in Morogoro. Findings from this study have demonstrated a high prevalence of subclinical mastitis and existence of health risks to milk consumers despite the low prevalence of tuberculosis and brucellosis in the study herds.     

Prevalence and Risk Factors for Bacterial Food‐Borne Zoonotic Hazards in Slaughter Pigs: A Review

The Hygiene Package and Regulation EC‐2160/2003 require information flow from farm to slaughterhouse to enhance European consumers protection in a ‘farm to fork’ approach. This obligation especially concerns food‐borne zoonotic hazards transmitted to humans through pork consumption, such as thermophilic Campylobacter spp., Listeria monocytogenes, Salmonella enterica and Yersinia enterocolitica. Prevalence estimates of these four hazards are affected by the sampling strategy and diagnostic procedure. Individual prevalence estimates for pig carriage (from digestive contents or lymph nodes collected at slaughterhouse) were higher than individual prevalence estimates for pig shedding (from faeces). Among risk factors described in the literature, poor pen cleaning and disinfection after pig departure to slaughterhouse and poor bio‐security measures are of major significance. Moreover, whereas wet feed increases the risk of pig infection by L. monocytogenes, dry feed is a risk factor for Salm. enterica. Mixing batches of pigs, notably in fattening herds, represents a risk for the transmission of Salm. enterica and Y. enterocolitica. Whereas small herds are more infected by thermophilic campylobacters and Y. enterocolitica, higher prevalence of Salmonella is observed in large herds due to a more frequent mixing of batches. Antibiotic treatment during the finishing period increases the risk of transmission of Salm. enterica. The forenamed elements should be taken into account to characterize farms in a risk assessment approach and to improve zoonotic hazard management in the pork food chain.     

The design of a Social Cost‐Benefit Analysis of preventive interventions for toxoplasmosis: An example of the One Health approach

Toxoplasma gondii infections cause a large disease burden in the Netherlands, with an estimated health loss of 1,900 Disability Adjusted Life Years and a cost‐of‐illness estimated at €44 million annually. Infections in humans occur via exposure to oocysts in the environment and after eating undercooked meat containing tissue cysts, leading to asymptomatic or mild symptoms, but potentially leading to the development of ocular toxoplasmosis. Infection in pregnant women can lead to stillbirth and disorders in newborns. At present, prevention is only targeted at pregnant women. Cat vaccination, freezing of meat destined for undercooked consumption and enhancing biosecurity in pig husbandries are possible interventions to prevent toxoplasmosis. As these interventions bear costs for sectors in society that differ from those profiting from the benefits, we perform a social cost‐benefit analysis (SCBA). In an SCBA, costs and benefits of societal domains affected by the interventions are identified, making explicit which stakeholder pays and who benefits. Using an epidemiological model, we consider transmission of T. gondii after vaccination of all owned cats or cats at livestock farms. To identify relevant high‐risk meat products that will be eaten undercooked, a quantitative microbial risk assessment model developed to attribute predicted T. gondii infections to specific meat products will be used. In addition, we evaluate serological monitoring of pigs at slaughter followed by an audit and tailor made advice for farmers in case positive results were found. The benefits will be modelled stochastically as reduction in DALYs and monetized in Euro's following reference prices for DALYs. If the balance of total costs and benefits is positive, this will lend support to implementation of these preventive interventions at the societal level. Ultimately, the SCBA will provide guidance to policy makers on the most optimal intervention measures to reduce the disease burden of T. gondii in the Netherlands.     

Metagenomic profiles of the rumen microbiota during the transition period in low‐yield and high‐yield dairy cows

We investigated potential relationships between rumen microbiota and milk production in dairy cows during the transition period. Twelve dairy cows were divided into a low‐yield (LY) or high‐yield (HY) group based on their milk yield. Rumen samples were taken from dairy cows at 3 weeks before parturition, and at 4, 8, and 12 weeks after parturition. 16S rDNA‐based metagenomic analysis showed that diversities of rumen microbiota in both groups were similar and the number of operational taxonomic units (OTUs) was lower in the postpartum than prepartum period in both groups. The abundance of Bacteroidetes and ratio of Bacteroidetes:Firmicutes was higher in the HY than the LY group. OTUs assigned to Prevotella bryantii, Fibrobacter succinogenes, Ruminococcus albus, Butyrivibrio fibrisolvens, and Succinivibrio sp. were abundant in the HY group. These OTUs were significantly related to the propionate molar proportion of rumen fluids in the HY group. OTUs assigned to Lachnospiraceae, Bifidobacterium sp. and Saccharofermentans were dominant in the LY group. Predictive functional profiling revealed that abundance of gene families involved in amino acid and vitamin metabolism was higher in the HY than the LY group. These results suggest that the community structure and fermentation products of rumen microbiota could be associated with milk production of dairy cows.     

Risk Factors Associated with Detection of Salmonella in Broiler Litter at the Time of New Flock Placement

In this study, we investigated risk factors associated with the probability to detect Salmonella in samples of litter collected within 2 h prior to new flock placement in 76 grow‐out houses on 38 conventional broiler farms located in the US states of Mississippi, Alabama and Texas. We evaluated characteristics of location and layout of the farm; area adjacent to and surrounding the house; house construction; condition and type of equipment in the house; litter management and other production, sanitation, visitation and biosecurity practices; non‐broiler animal species on the farm; and weather conditions on the 3 days leading up to flock placement. Logistic regression was used to model the relationships between probability to detect Salmonella in litter and potential risk factors. In the screening process, each risk factor was evaluated as a single fixed effects factor in a multilevel model that accounted for variability among the sampled farms and their production complexes and companies. Of almost 370 risk factors screened, 24 were associated with the probability to detect Salmonella in litter. These were characteristics of the surroundings of the house, house construction and conditions, litter management, length of downtimes between flocks in the house, biosecurity and farm location. After investigation of collinearity between these variables and building of models for important risk factor categories, the list of candidate variables for the final model was refined to eight factors. The final model demonstrated that a higher probability of detecting Salmonella in litter was strongly associated with the use of wood to construct the base of the walls or to cover the inside of the broiler house foundation, and with the use of fresh wood shavings to top‐dress or completely replace the litter between flocks.     

Participatory probabilistic assessment of the risk to human health associated with cryptosporidiosis from urban dairying in Dagoretti,Nairobi, Kenya

We carried out a participatory risk assessment to estimate the risk (negative consequences and their likelihood) from zoonotic Cryptosporidium originating in dairy farms in urban Dagoretti, Nairobi to dairy farm households and their neighbours. We selected 20 households at high risk for Cryptosporidium from a larger sample of 300 dairy households in Dagoretti based on risk factors present. We then conducted a participatory mapping of the flow of the hazard from its origin (cattle) to human potential victims. This showed three main exposure pathways (food and water borne, occupational and recreational). This was used to develop a fault tree model which we parameterised using information from the study and literature. A stochastic simulation was used to estimate the probability of exposure to zoonotic cryptosporidiosis originating from urban dairying. Around 6 % of environmental samples were positive for Cryptosporidium. Probability of exposure to Cryptosporidium from dairy cattle ranged from 0.0055 for people with clinical acquired immunodeficiency syndrome in non-dairy households to 0.0102 for children under 5 years from dairy households. Most of the estimated health burden was born by children. Although dairy cattle are the source of Cryptosporidium, the model suggests consumption of vegetables is a greater source of risk than consumption of milk. In conclusion, by combining participatory methods with quantitative microbial risk assessment, we were able to rapidly, and with appropriate ‘imprecision’, investigate health risk to communities from Cryptosporidium and identify the most vulnerable groups and the most risky practices.