Pesticides are used globally in agriculture, and can have negative effects on human health and ecosystems, especially when not handled as intended. Still, an increasing number of smallholder farmers in low- and middle-income countries are using expensive pesticide products to increase their yield. Due to their low socio-economic status and educational level, smallholder farmers are particularly vulnerable to the negative impacts of yield losses, but also of those from pesticides. When confronted with pests, smallholder farmers develop a need for information and seek out appropriate sources. However, little is known about how smallholders go about this process, whether this process differs for organic and conventional pest management strategies and if farmers also focus on risks of pest management practices. Agro-input dealer are supposed to provide information on pesticide risks, but they often focus more on selling products than services. There is a knowledge gap in literature on how agro-input dealers give advice, what products they sell, whether they follow laws and recommendations on best practice, and whether their practices and shop organization prevent pesticide risk situations. This in turn leads to farmers not knowing about risks, or not considering them to be relevant. While farmers have been assessed regarding their pesticide knowledge, attitudes and practices, rarely these insights are compared across contexts, identifying differences and commonalities. This, together with a lack in pesticide training, results in farmers not always following good agricultural practices, thereby affecting their own health, their communities’ and ecosystems. These issues are not resolved in disciplinary silos, but only through cross-sectoral and participatory research and interventions. This idea of a transdisciplinary, border-crossing research project named ‘Comparative appraisal of pesticide use in tropical settings: exposure pathways, health effects and institutional determinants’ laid the foundation for this dissertation. The two study sites of the above research project were a market-oriented farming system in Zarcero County, Costa Rica and a subsistence-based farming system in Wakiso District, Uganda. In two cross-sectional surveys, this dissertation complemented a pesticide exposure and health assessment of farmers in both countries (Costa Rica in 2016, n=300 and Uganda in 2017, n =302), enrolling both farmers applying synthetic pesticides and such who follow other pest management practices. We found the majority of pesticides used in both case studies to be classified as highly hazardous by the World Health Organization. While a high awareness of negative health effects was identified, the use of personal protective equipment was rare, and hygiene and other safe use practices were not adopted by all farmers. Organic farmers were more likely to have been trained on safe pesticide use practices compared to users of synthetic pesticides. Pesticide use did not appear to drive household income. In a qualitative study in parallel to the cross-sectional survey in Uganda, we investigated pest-management information behavior from the perspective of smallholder farmers. Using an ethnographic approach, we conducted 46 semi-structured interviews and 15 on-farm observations in Wakiso District in 2017. The results indicated that farmers develop information needs when adopting new farming practices, or when presented with disruptive information (e.g. when new pests emerged), prompting farmers to seek information actively or be attentive to receiving information passively. Whether farmers used the new information depended on successful trial of the new pest management strategy, and on the credibility of the source. Furthermore, our results suggested that sources of information for conventional pesticides were well integrated into farmers’ daily lives, whereas information on organic strategies was provided through external sources (e.g. NGOs), but was not available at all times. To share the above results with the respective stakeholders in Uganda, we conducted a participatory workshop using the design thinking method. While assessing the knowledge gaps between academic and non-academic stakeholders, we found recommendations from non-academic stakeholders applying an inherently interdisciplinary and thus broader point of view, accounting for the roles of more different stakeholders in pesticide management, for example agro-input dealers and policy-makers. The non-academic knowledge was more fine-grained and detailed, exemplifying how a knowledge integration is essential to avoid a gap between what researchers investigate and what practitioners need. Following new insights from this workshop, we conducted an agro-input dealer study in Central and Western Uganda in 2019. We selected a mixed methods approach, using structured questionnaires and observations to study their knowledge, attitude and practices on pesticides (n=402), shop organization (n=392) and sales interaction (n=236). Actual behavior of agro-input dealers when selling pesticides was revealed through mystery shopping with local farmers buying pesticides (n=94). The findings revealed that most dealers saw advising customers as a responsibility, while only around a quarter of mystery shoppers received unsolicited advice when buying pesticides. Observations of sales interactions showed that the focus of discussion was on product choice and price, neglecting aspects of safe use. Most shops were lacking safety equipment and a quarter of shops sold repackaged products. Agro-input dealer showed limited understanding of pesticide safety labels and active ingredients. Around half the agro-input dealers held a certificate of competency, while only a minority was able to provide a government-approved up-to-date license. In conclusion, we found that the responsibility of why pesticides are not managed, handled and applied as intended is shared throughout actor levels. Crucial information does not reach the end-user, and where it does, the appropriate tools and equipment to follow the corresponding guidelines are missing. Meanwhile, a lack of awareness from farmers as well as conflicting interests prevent agro-input dealers from providing much needed advice. We recommend to make information on safe use, as well as alternatives to pesticides more continuously available in farmers’ daily lives, by leveraging the established information channels – the agro-input dealers and agricultural extension. Professionalization of both pesticide sellers and users allows to manage the negative effects of pesticides over the entire product life cycle, from purchase, via storage and application to residual and waste management. Bridging gaps and improving coordination and collaboration between stakeholders is crucial to align practice, research and policy in their quest for reaching a transition towards sustainable agriculture.