Neurobehavioural effects of pesticide exposure among emerging farmers in the Western Cape

Doctoral Thesis


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

University of Cape Town

The number of emerging farmers is increasing in South Africa. This study investigated neurotoxic effects resulting from long-term low-level exposure to organophosphate pesticides amongst emerging farmers. A cohort study involving 319 (66% male and 34% female) emerging farmers in the Western Cape was conducted between 2009 and 2010. Testing included a questionnaire exploring demographic details, work history, medical history; the Q16 and the Brief Symptom Inventory (BSI). Additionally, neurobehavioural performance was assessed on the following: Digit span forward and backward, Digit Symbol, Santa Ana Pegboard, Pursuit Aiming, Benton Visual Retention and Vibration sense as measured by vibration sense threshold using a 256 Hz frequency tuning fork. Exposure to OP's was derived from three metrics: (i) OP applicator status, (ii) cumulative OP exposure weighted for task/activity, (iii) diagnosed past poisoning. Two sets of analyses were conducted: a cross-sectional analysis at baseline (N=319), and a cohort analysis (N=273), comparing change in neurobehavioural performance over the 12 month follow up period. The exposure variable for cumulative occupational OP exposure in the baseline cross-section was generated from the job history data in the farmer questionnaire based on the exposure obtained during the current job and preceding three jobs. This represented long-term past OP exposure. In both the baseline cross-section and the cohort analysis, the relationship investigated was between dichotomised outcomes and four models for OP exposure. Three models used each of the metrics for OP exposure and a fourth model used the metric of cumulative OP exposure adjusted for diagnosed past poisoning. The difference between the analysis in the cohort was that both cumulative exposure and acute poisoning were restricted to the 12 months of follow up, whereas the baseline, both cumulative and acute poisoning were based on any past exposure without a time period. Further, for the cohort study (n=273), the outcomes were the difference in scores (measured as the ratio of the repeat measure at one year to the baseline score) which were regressed in the four exposure models outlined above. All models were controlled for confounders including age, gender, schooling, language, current alcohol consumption, previous head injuries, psychological illness and socioeconomic status. Model building led to the inclusion of additional co-variates for some of the analyses.

Includes bibliographical references.