African pesticide intervention studies project

Approved 24/02/2021, Higher Degrees Research and Ethics Committee (HDREC) at Makerere University School of Public Health (New Mulago Hill Road, Mulago, Kampala, PO Box 7072, Uganda; +256 (0)393 291 397; hdrecadmin@musph.ac.ug), ref: HDREC 846

The co-primary outcomes of the study are:
1. Farmers’ knowledge, attitudes and practices (KAP) assessed based on three scores which contain 16 items each for knowledge, attitudes and practice at baseline and 12-month follow-up. The KAP items were developed based on other KAP assessments in the PESTROP survey (Staudacher et al., 2020) and another study (Gautam et al., 2017) and adapted in a way that they assessed KAP related to the training modules of the educational intervention curriculum that was tested in this study. For every correct answer, one point will be added to the knowledge, attitudes and practice score, resulting in final scores ranging between 0 and 16 (as proposed by Riccò et al., 2019 and Riccò et al., 2020).
2. Using PPE to cover all body parts, assessed via behavioral frequency measures, as common in behavioral science (e.g. Verplanken & Orbell, 2003): Farmers indicated how often they used 15 specific PPE items in the past year, such as gloves and gumboots, at baseline and 12-month follow-up. The response options ranged from "never using the respective PPE item" (1) to "always using the respective PPE item" (5). The PPE items will be categorized into eight groups, representing different body parts: eyes, mouth, hands, arms, legs, feet, trunk, and head. To calculate the total frequency of PPE use across all body parts, we will add up the average frequencies of using PPE for each category. For example, the total frequency of using PPE will be the average frequency of using items to cover hands, plus the average frequency of using items to cover mouth, and so on for each body part.

1. Psychosocial factors of PPE use (RANAS factors) assessed using the measurements proposed by the RANAS approach (Contzen & Mosler, 2015; Mosler, 2012). The RANAS model describes 17 psychosocial factors that are of potential relevance to change behavior that are grouped into five factors categories (Risk, attitude, norm, ability, and self-regulation). The items assessed these 17 RANAS factors asking participants to rate their degree of agreement or feeling related to certain RANAS questions, using a 5-point Likert scale: not at all (1) to very much (5) at baseline and 12-month follow-up. Risk factors contained a mean of two items on health vulnerability and a mean of two items on severity, e.g. “Imagine you are suffering from acute symptoms of pesticide usage; how severe do you rate the consequences for your own health?” – “not at all severe” to “very severe”. Attitudes factors contained five single items on beliefs about cost benefits. Norm factors contained a mean of two items on others’ behavior (descriptive norm), the mean of two items on others’ (dis)approval (injunctive norm), as well as the mean of two items on personal importance. Ability factors contained one item on how-to-do-knowledge and the mean of three items on confidence. Self-regulation factors contained two items on action planning, one item on barrier planning and the mean of three items on commitment.
2. Frequency of using gloves. As hands are the most exposed body parts during pesticide handling (Fuhrimann, Staudacher et al., 2020) the study assessed the mean frequency of using PPE covering hands using common behavioral frequency assessment (e.g. Verplanken & Orbell, 2003) at baseline and 12-month follow up. Participants were asked about the frequency of using PPE to cover their hands in the past 12 months, with answer options; never (1) to always (5).
3. Pesticide exposure measured using the exposure intensity scores developed by Fuhrimann, Staudacher et al. (2020) at baseline and 12-month follow-up. These scores were derived from calculating a semi-quantitative exposure algorithm developed and validated for low- and middle-income countries (Fuhrimann, Staudacher et al., 2020).
4. Exposure modified by PPE use assessed by using the modifying PPE variable from the pesticide exposure algorithm at baseline and 12-month follow-up (Fuhrimann, Staudacher et al., 2020). This variable accounts for the materials of PPE being used and the exposure risk of different body parts. For example, hands account for 40% of pesticide exposure, while eyes only account for 10% of exposure.
5. Farmer’s health assessed using a checklist for the frequency of 31 symptoms associated with neurological acute pesticide poisoning experienced e.g., dizziness, anxiousness, nausea, vomiting, sleeping difficulties, weakness of the limbs, and changes in taste and smell (as proposed by Farnham et al., 2021) at baseline and 12-month follow-up.