This paper contains a literature review of the occupational injuries and ill-health in agriculture world-wide and a survey of the attempts that have been made to estimate the resulting economic and social costs.Agricultural workers suffer a wide variety of disorders as a result of their occupation. These range from minor (cuts, bruises) to more severe (deep wounds, fractures), permanent (amputation, spinal cord injury) and fatal injury. Ill-health as a result of contact with animals, micro-organisms, plant material dusts or chemicals are associated with certain types of agriculture. There is an underlying but unquantified incidence of pain, stress and injury as a result of ergonomic problems due to poor working procedures and conditions. Statistics from many countries or regions show that agriculture consistently has one of the highest accident and injury rates of the industrial sectors.There are many causes for the work related injury and ill-health in agricultural workers. In developed countries, tractors and other machinery cause a significant proportion of the accidents and are a major cause of occupational deaths. In less developed countries, accidents due to hand tools such as hoes, sickles and cutting instruments are most prevalent. Animals are a significant cause of injury and ill-health in many countries. Debilitating allergic reactions in the respiratory tract or the skin are caused by exposures to organic dusts, or by contact with allergenic plants in the field respectively. Where comparative data are available, occupational pesticide poisoning in agriculture is a small proportion (< 1–4%) of the total work related disorders.Because of the wide variety of occupational risks to agricultural workers, it is emphasised that if one type of agricultural practice is replaced by another then the risks from the alternative procedure need to be considered. If, for example, agrochemical pest control practices are replaced by methods involving the increased use of machinery, draught animals or manual operations, then an assessment of the resulting risks should be taken into account.Some of the economic costs of occupational injury and ill-health in agriculture can be quantified directly, such as medical costs, the cost of rehabilitation and loss of earnings. Other costs are more difficult to estimate such as loss of opportunity and income foregone for permanent and fatal injury and for the effect on a victim’s family. The estimation of the overall economic costs to farming communities and national agriculture requires further development. When one agricultural practice is replaced wholly or partly by another, for example agrochemical pest control by alternative control methods, then it is necessary to take into account the occupational health costs of the alternative procedure for realistic comparative assessment.There are a number of issues which require continued or increased attention by the relevant national and international authorities and by the agricultural industry. These include the improved collection and collation of occupational health statistics, a better understanding of the extent of ergonomic problems in agriculture, more realistic assessments of the cost of occupational injury and ill-health and the continued need to reduce occupational health disorders by appropriate training and education in agricultural practices and the use of agricultural equipment.

The interdependencies of parameters applied in the models of EUSES are visualised in a directed connectivity graph. The parameters (inputs, defaults, state variables, outputs) are represented by boxes (nodes) and their relations by lines (edges). The visualisation, on the one hand, clarifies the complexity of the models in EUSES and, on the other hand, creates an overview and transparency. The parameters’ relations to each other can be recognised faster, and the models can be better understood. The complexity was quantified by the number (variety), kind (substance parameter, physico-chemical parameter, concentration, other parameters), and depth (dimension) of the parameter and the number of relations (connectivity). The variety of EUSES (without the modelsSimple Treat andSimple Box whose interior structure is not documented and without the effect and risk characterisation) amounts to 466, the connectivity to 961, and the maximal dimension is 21.