“There are risks to be taken and some just push it too far”: A mixed methods exploration of human risk factors in agricultural quad-bike incidents in New Zealand

Quad-bike incidents are a major cause of occupational injury and fatality on New Zealand farms, warranting health and safety attention. Factors pertaining to the rider, quad-bike and environment have been identified as contributors in agricultural quad-bike incident risk. However, how farmers make behavioural decisions regarding quad-bike risk is unknown. Relying on farmers’ knowledge of the risks as a mechanism for adopting more precautionary behaviours appears ineffective. The overall aim of this research was to explore how human factors impact on farmers’ personal risk perception of experiencing a quad-bike incident whilst working. This research was approached in two phases, moving from quantitative to qualitative inquiry. Phase 1 involved a face-to-face structured survey of 216 farmers. The relationship between the number of self-reported quad-bike loss of control events (LCEs) experienced in a working lifetime and standardised measures of unrealistic optimism (UO), fatalism and propensity to risk-taking was investigated using Poisson regression analysis. Open-text data relating to participants’ descriptions of LCEs and how they perceived quad-bike training courses were analysed using qualitative content analysis (QCA). Phase 2 used grounded theory methods to explore how farmers perceived risk using semi-structured interviews with eight farmers purposively sampled from those who took part in Phase 1. In Phase 1, the regression model of best fit to predict LCE included UO (Incident rate ratio (IRR) 0.84, 95% CI 0.75 to 0.94), impulsive sensation-seeking (IRR 1.08, 95% CI 1.01 to 1.16), younger age (IRR 0.98, 95% CI 0.97 to 0.99) and being male (IRR 4.00, 95% CI 2.15 to 7.44). Unexpectedly, farmers with higher levels of UO and, thus, stronger beliefs that “it won’t happen to me”, were less likely to report LCEs. QCA of participants’ LCE descriptions further illustrated the constructs of UO, fatalism and risk-taking, with the latter identified more frequently and suggestive of farmers finding themselves in risky situations, which they then had to deal with, rather than risk-seeking behaviour per se. Farmers wanted quad-bike training courses that address risky scenarios in realistic environments with credible trainers. In Phase 2, farmers’ perception of personal quad-bike incident risk depended on how previous quad-bike incidents impacted on the individual, their personal attributes, having responsibilities (to others and/or the job at hand) and being familiar with the quad-bike, terrain and task being performed. A grounded theory model was developed to explain a temporal aspect to risk perception: the sudden realisation that things are going wrong (‘in the moment’) versus contemplating risk beforehand (‘reflective’) which has implications for safety intervention. The combined findings suggest there is a complex interaction of factors impacting on farmers’ perception of quad-bike incident risk. Although perceiving personal susceptibility contributed to farmers adopting precautionary behaviours, this did not necessarily relate to fewer incidents. Farmers needed a degree of confidence in their ability to manage risk but too much confidence, or too little, could be problematic. ‘In the moment’, however, pressing on with the job despite (or in denial of) risk often took priority over personal safety