An investigation into the neurological and neurobehavioural effects of long-term agrichemical exposure among deciduous fruit farm workers in the Western Cape, South Africa

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

It is increasingly being recognised that agrichemical exposure may have adverse chronic health effects in humans, particularly on central nervous system function. However, much of this evidence sterns from studies relating to the effects of acute intoxications (i.e. short-term high dose exposures) and little data exist on the chronic effects of long-term low-dose exposures to agrichemicals in the absence of acute poisoning. Such a finding would have substantial public health implications for prevention and control of chronic morbidity and mortality. This is particularly important in South Africa, where a sizeable portion of the rural population are employed in agricultural work, often under extremely unhealthy living and working conditions, and where occupational agrichemical exposures appear to be substantial. To address this question, this study investigated the prevalence of neurological and neurobehavioural abnormalities amongst 247 fruit farm workers in the Kouebokkeveld in the Western Cape, of whom 163 were current agrichemical applicators. Outcomes measured included neurological symptoms, peripheral vibration sense, motor tremor, as well as performance on the World Health Organisation Neurobehavioural Core Test Battery (WHO NCTB) and a set of neurobehavioural tests based on the Information Processing model of cognitive psychology. These latter tests have been developed in South Africa for subjects of low educational levels and aim to by-pass the powerful effects of culture that complicate traditional neuropsychological testing, which may mask the smaller effects due to occupational chemical exposures. Cumulative, and average lifetime intensity of exposure to organophosphates were estimated using a job- exposure matrix based on a combination of secondary industry data, interview reports and farmer records. Confounders measured included age, education, smoking and alcohol habits, non-occupational exposure to agrichemicals and other potential neurotoxins, past medical history and usage of personal protective equipment. The study results confirmed low levels of education and high alcohol consumption amongst the sample of farm workers. Multiple logistic and linear regression were used to identify exposure-effect relationships and to control for confounding. Neurological symptoms were significantly associated with a history of previous pesticide poisoning, although this may have arisen as a result of reporting bias. Vibration sense and the neurobehavioural tests exhibited associations with established covariates, and regression modelling of the WHO NCTB tests was remarkably similar to findings in another study of solvent-exposed factory workers in South Africa. None of the vibration sense, tremor or neurobehavioural outcomes were associated with past agrichemical poisoning in the sample, and only two tests showed significant relationships with long-term occupational exposure. These included the Pursuit Aiming subtest of the WHO NCTB and one of the tests of long-term semantic memory in the Information Processing tests. However, the strength of these the associations were small (partial r²s less than 2%) and these findings may have occurred due to chance arising from multiple comparisons. The neurobehavioural tests based on the Information Processing model appeared to offer little improvement on the WHO NCTB in terms of being less sensitive to cultural effects, although some evidence was present that tests of semantic access were able to detect occupational effects and were less sensitive to education. The absence of a demonstrable and consistent long-term agrichemical exposure-effect relationship appears to suggest that long-term agrichemical exposure is not associated with adverse chronic nervous system effects, although the lack of organophosphate specificity in construction of exposure indices in the job-exposure matrix may partly contribute to this finding. Recommendations to improve the characterisation of agrichemical exposures at farming work place are made, as well as suggestions concerning the role of biological monitoring for agrichemicals, improving working and living conditions on South African farms, and methods of neurological and neurobehavioural assessment in occupational health.