This study examines how socio-demographic factors influence acute pesticide exposure and related health outcomes among pesticide applicators in 21 villages across Kadapa and Anantapur districts, Andhra Pradesh. It critically examines the intersecting roles of caste in determining land ownership status, landholding size, and the type of pest management (conventional, mixed and non-pesticide applicators) and how this influences exposure to hazardous chemicals used in agriculture. By situating pesticide use within broader agrarian structures, the study reveals how historical inequalities continue to determine who bears the health burden of chemical-intensive farming.

Using structured surveys, data were collected from 135 respondents across 21 villages in the Rayalaseema region (Kadapa, Anantapur, Chittoor and Nellore districts). The survey covered demographic variables (caste, age, gender, land ownership, and landholding size), pesticide application practices, use of bio-inputs, personal protective equipment (PPE), symptoms post-application, and health expenditures. Pesticide and bio-input usage were assessed in terms of frequency, dosage, and who recommends them. Regression analyses, like logistic, negative binomial, and ordinal regression, were used to examine associations between socio-demographic factors and reported symptoms, frequency of exposure, exposure pathways, and incurred health costs.

Results reveal that of 135 respondents, 55.5% (ie, 75 individuals) reported discomfort following pesticide application. Of the total respondents, 74% were conventional pesticide users, while only 7.4% used exclusively bio-inputs. Of the total pesticides recorded, 77% were insecticides, with organophosphates and pyrethroids being the most common. According to WHO classifications, 6% were classified as highly hazardous (Class Ib), 63% of the pesticides used were moderately hazardous (Class II), 7% were classified as slightly hazardous (Class III), and 23% were classified as unlikely to present acute exposure (Class U).

This study attempts to ask two questions: Does the incidence of acute pesticide exposure and its resulting health costs vary across caste groups, land ownership, and landholding size? Incidence of acute pesticide exposure is measured through the number of symptoms reported, the number of times acute exposure was reported in the previous year and the number of pathways of exposure. The results of the analysis show that non-pesticide applicators experienced significantly fewer symptoms and fewer exposure episodes than conventional pesticide users. APCNF, the state-funded project by Andhra Pradesh, was initiated in 2016 to transition conventional chemical agricultural practices to natural practices. The majority of the respondents who are natural farmers or mixed applicators are from the general category and are large landowners, consistent with existing literature that APCNF remains skewed toward socially privileged groups.

The results also showed that labourers experienced significantly greater exposure, they had higher predicted symptoms, more frequent exposure episodes and a higher probability of experiencing multiple pathways of exposure compared to owners and tenants. Additionally, landless respondents experienced significantly more symptoms compared to large landowners. SC/STs had higher odds of experiencing multiple pathways of exposure, and OBCs had higher predicted times of exposure in the previous year than GN. SC/STs have the highest proportion of labourers and landless individuals compared to other categories, and OBCs compared to GN category respondents. This disparity may be attributed to the lack of secure land rights, which compels them to work as agricultural labourers and bear a disproportionate health burden.

Tenants exhibited a diverse profile equally distributed across all landholding sizes, from landless and large landowners, and the latter often hire labour, which reduces personal exposure. This heterogeneity explains the broader range of exposure outcomes within the tenant group. These findings align with the literature on the increasing trend of reverse tenancy in Andhra Pradesh, when land is being leased out to large landholding farmers by smallholders.

The use of a higher-than-prescribed pesticide dose was associated with significantly worse outcomes. These respondents have higher symptom counts, more exposure episodes, and greater odds of experiencing multiple exposure pathways. This behaviour, common among growers of chilli, paddy, tomato, and cotton, may stem from pest resistance, risk-avoidance strategies, or a lack of measurement tools for accurate dosage.

Health cost data revealed further disparities. Although only 19 respondents reported monetary health costs, however, the distribution was highly skewed: one respondent accounted for the majority of the total reported costs. Many others opted for non-monetary coping strategies (e.g., oil application, rest days) and did not seek formal healthcare, suggesting that the true economic burden is underreported. Informal and unpaid labour substitution within families also masked indirect economic costs.

This study examines whether caste remains a key feature of agrarian inequality, determining land ownership, occupational roles, and ultimately deciding the extent of pesticide exposure. SC/ST respondents were overrepresented among landless labourers, underrepresented among bio-input users, and had higher predicted exposure risks. Similarly, landholding size and status linked with caste determine not just exposure frequency but the ability to mitigate it through hiring labour, PPE use, or adoption of natural farming. Without redistributive land reform, current agricultural interventions like APCNF may not adequately reach or benefit the most marginalised farmers. In conclusion, this study underscores that acute pesticide exposure in Rayalaseema is deeply embedded in caste, class, and land relations.