Using the Health Belief Model to Understand Why Making Oral HIV Self-Testing Available to Truck Drivers in Kenya Had Little Impact on Six-Month Testing

Research has found that offering HIV self-testing (HIVST) to truckers in Kenya increased testing rates at baseline but not over 6-month follow-up. We explored possible explanations based on the Health Belief Model by assessing HIV risk perception, self-efficacy, and fatalism as possible effect modifiers of the impact of offering HIVST (intervention n = 150) versus standard of care (SOC n = 155) on 6-month testing on the multiplicative and additive scales using log binomial and linear binomial regression and stratifying on significant modifiers. We found significant interaction between the intervention and fatalism on both the multiplicative (p = 0.020) and additive (p = 0.020) scales. In the stratified models, the HIVST intervention was associated with higher HIV testing among participants with low fatalism but lower testing among those with high fatalism (risk ratio [RR] = 1.30, p = 0.065 versus RR = 0.74, p = 0.072; risk difference [RD] per 100 = 14.00, p = 0.080 versus RD=-14.69, p = 0.086). Truckers in Kenya are described as being highly fatalistic, feeling lack of control over their lives and health. We found that fatalistic views negated the potential benefit of offering HIVST to truckers. For HIVST to have an impact among truckers, psychosocial interventions may be needed that address fatalistic views.