Assessment of pesticide use patterns among farmers in the Jamuna riverside regions of Tangail and Sirajganj districts, Bangladesh Authors: Md Rakibul Hasan, Shekh Md Hadiul Islam, Md Shah Alam, Era Mondal Jyoti, Sameya Rahman Sreya, Somaiya Islam Shuchy, Rokeya Begum, Md Azizul Haque Published: 2026 ABSTRACT AbstractPesticide use is increasing globally, particularly in developing agricultural systems like Bangladesh, raising concerns about occupational exposure and food safety. This study conducted a comparative cross-sectional survey of 600 farmers in Tangail and Sirajganj districts using structured questionnaires during the 2025 cropping season. It assessed pesticide use patterns, farmers’ knowledge, safety practices, and self-reported occupational health impacts. Tangail farmers primarily used insecticides, particularly lambda-cyhalothrin and cypermethrin, whereas Sirajganj farmers predominantly relied on fungicides, including propiconazole and azoxystrobin. Moderately hazardous pesticides dominated use in both districts. Three major concerns emerged from the study: widespread over-application of pesticides, low adherence to withholding periods, and the continued use of banned products. Over-application was reported by 54.7% of farmers in Tangail and 68.7% in Sirajganj, while WHP adherence remained low (46.3% and 21%, respectively), and banned pesticides were still used in 12 and 20% of cases respectively, indicating significant behavioral and regulatory gaps. Only 19% of farmers consistently used personal protective equipment, and occupational health complaints were more frequent in Sirajganj, with 50.3% reporting skin irritation. A key finding was a significant knowledge-practice gap, where WHP knowledge inconsistently increased the likelihood of unsafe practices and health complications. This study is novel in providing a comparative assessment of pesticide use and safety behaviors in riverine farming systems, highlighting behavioral inconsistencies despite awareness. These findings indicate that awareness alone is insufficient to ensure safe pesticide practices and highlight the need for behavior-focused interventions, strengthened regulatory enforcement, improved monitoring of pesticide distribution, and promotion of integrated pest management to reduce occupational exposure and food safety risks in similar agricultural settings.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-53466-8.Keywords: Pesticide, Withholding period, Personal protective equipment, Safety practices, Health risksSubject terms: Environmental sciences, Health care, Risk factors Introduction The increasing reliance on chemical pesticides in developing agricultural systems has raised significant concerns regarding unsafe application practices, occupational exposure, and food safety risks1,2. In many low- and middle-income countries, limited regulatory enforcement and inadequate farmer training contribute to the misuse of pesticides, including over-application, improper handling, and poor adherence to withholding periods (WHP)3–5. These unsafe practices not only compromise farmer health but also increase the risk of pesticide residues in food and environmental contamination. Pesticide exposure among farmers is widely recognized as a major global public health issue. Direct exposure occurs during mixing, loading, and spraying, while indirect exposure arises from contaminated soil, water, and agricultural produce6. Acute health effects include nausea, dizziness, and skin irritation, whereas chronic exposure has been associated with neurological disorders, endocrine disruption, reproductive complications, and cancer7–12. These risks are particularly pronounced in smallholder farming systems, where awareness and adoption of safety measures remain limited. Recent studies in developing countries have highlighted persistent gaps between farmers’ knowledge and actual practices, indicating that awareness alone does not necessarily translate into safe pesticide use13–16. Bangladesh has experienced a rapid increase in pesticide consumption over recent decades, rising from approximately 4,000 metric tons in the 1970s to over 40,000 metric tons in recent years, with usage growing at an estimated annual rate of around 10% and reaching about 1.8–1.9 kg per hectare of cropland, reflecting broader trends of agricultural intensification in developing regions17,18. A considerable proportion of pesticides used are moderately to highly hazardous, and their application is often inadequately regulated. Similar patterns have been reported in other South Asian countries, where unsafe pesticide handling, low use of personal protective equipment (PPE), and poor compliance with safety guidelines remain prevalent13,19–21. These issues are further exacerbated in ecologically sensitive environments such as riverine and floodplain agricultural systems, where frequent cropping and proximity to water bodies increase the risks of environmental contamination and human exposure. Despite growing evidence on pesticide use and farmer practices in Bangladesh, existing studies have primarily focused on general knowledge, attitudes, and practices (KAP)4,20,22–24, with limited emphasis on detailed usage patterns including types of pesticides used, types and concentration of active ingredients used, comparative assessments across regions, and associated health outcomes. In particular, there is a lack of comprehensive data on pesticide use behaviors in riverine farming systems, which are both agriculturally important and environmentally vulnerable. Moreover, the relationship between farmers’ knowledge, actual practices, and health outcomes remains insufficiently explored, especially in comparative contexts. This study is conceptually grounded in the Knowledge-Attitude-Practice (KAP) framework, which posits that knowledge influences attitudes and, subsequently, practices; however, discrepancies between knowledge and behavior may arise due to contextual, economic, and institutional factors. By examining pesticide use patterns, safety behaviors, and health outcomes, this study also considers the behavioral risk pathway linking unsafe practices to occupational exposure and health effects. Given these gaps, the present study provides a comparative assessment of pesticide usage patterns, safety practices, and self-reported health impacts among farmers in two riverine districts of Bangladesh (Tangail and Sirajganj). The findings are expected to contribute to a broader understanding of pesticide use behaviors in riverine and smallholder farming systems, offering insights relevant to similar agricultural settings in other developing countries. By highlighting the gaps between knowledge, practices, and health outcomes, this study provides evidence that can inform policy, training programs, and risk reduction strategies at both national and international levels. Methodology Study design This study employed a cross-sectional survey design to assess pesticide usage patterns, safety practices, and exposure among farmers in the Jamuna riverside regions of Tangail and Sirajganj districts. A cross-sectional approach was chosen to capture a snapshot of prevailing pesticide handling behaviors, protective measures, and knowledge, attitudes, and practices (KAP) at a specific point in time. This design allowed for the simultaneous collection of information on socio-demographic characteristics, pesticide application practices, safety measures, and self-reported health effects. The study received ethical approval from the Ethical Review Committee of the Department of Food Technology and Nutritional Science under approval number FTNS/ERC/07/2025. All methods were performed in accordance with the relevant guidelines and regulations. Informed consent was obtained from all participants prior to data collection. Study area The study was conducted in Tangail and Sirajganj districts of central Bangladesh (Figs. 1 and 2), two agriculturally intensive regions characterized by year-round crop production and extensive reliance on chemical pesticides. These districts were purposively selected due to their high pesticide use, mixed and intensive cropping systems, and the predominance of small-scale farmers, many of whom have limited formal education and inadequate access to structured training or awareness programs on safe pesticide management. Both districts are located along the Jamuna river and share comparable agro-ecological and climatic characteristics, which allowed us to minimize environmental confounding and better isolate behavioral, knowledge-based, and management-related differences in pesticide use. At the same time, they differ in cropping intensity, dominant crop types, and pesticide use practices, making them suitable for a comparative assessment within a similar environmental context. Tangail (Dhaka Division; 24°01′-24°47′ N; 89°44′-90°14′ E), located on fertile floodplains of the Jamuna and Dhaleshwari rivers, supports diverse crops including vegetables, legumes, brinjal, paddy, wheat, and other cereals. Its tropical climate-hot summers, monsoons, and dry winters-along with frequent pest pressure, leads to repeated pesticide use25. Sirajganj (Rajshahi Division; 24°05′-24°50′ N; 89°15′-89°45′ E), near the Jamuna River, is a major rice-producing district, especially for Boro paddy, and also produces jute, mustard, lentils, and winter vegetables26. Compared to Tangail, Sirajganj has higher dominance of rice-based cropping systems alongside expanding vegetable cultivation, which may influence differences in pesticide types and application frequency. Map of Bangladesh with rivers and study districts (In red circle). Jamuna riverside study areas in (A) Tangail and (B) Sirajganj regions (Satellite imagery obtained from Google Maps). Tangail map source: https://www.google.com/maps/@24.2765251,89.753468,24186 m/data=!3m1!1e3; Sirajganj map source: https://www.google.com/maps/@24.5658371,89.7106891,40583 m/data=!3m1!1e3. Sampling strategy and sample size A mixed multistage sampling strategy, combining purposive district selection with random household selection was employed for this cross-sectional study. Initially, two agriculturally intensive districts of central Bangladesh-Tangail and Sirajganj-were purposively selected due to their extensive pesticide use and proximity to the Jamuna River, which strongly influences cropping intensity and pest pressure. From each district, two upazilas located adjacent to the Jamuna River were purposively chosen to capture farming systems characterized by frequent pesticide application. Subsequently, two unions were randomly selected from each upazila. Within each union, farming households were selected using systematic random sampling. The sample size was determined using a standard formula for cross-sectional studies, assuming a 50% prevalence of unsafe pesticide practices, a 95% confidence level, and a 5% margin of error, which yielded a minimum required sample size of approximately 384 respondents. To enhance the robustness of the analysis and enable subgroup comparisons, a total of 600 farmers were included in this study, comprising 300 respondents from Tangail and 300 from Sirajganj. Inclusion and exclusion criteria Farmers were eligible for inclusion if they were actively engaged in agricultural activities involving pesticide use within the Tangail or Sirajganj districts at the time of the survey, were present during field visits, and provided informed consent to participate voluntarily in the study. Only permanent or regular cultivators of the study areas were considered to ensure contextual relevance of farming practices. Farmers were excluded if they declined to participate, were not directly involved in pesticide handling or application, or were temporary residents or seasonal workers without sustained farming engagement in the selected regions. Data collection tools Data were collected through a structured, pre-tested questionnaire during face-to-face surveys with 600 farmers (300 each from Tangail and Sirajganj). Informed consent was obtained from all participants prior to data collection. The questionnaire comprised five sections: Section A captured socio-demographic and farming characteristics (e.g., age, education, household size, farming experience, and farm size); Section B focused on pesticide profiling, including commonly used pesticides, their regulatory status, and WHO hazard classification27; Section C assessed farmers’ knowledge, perceptions, and practices related to safe pesticide use, particularly withholding periods (WHP); Section D examined pesticide handling and safety behaviors, such as PPE use, mixing, and application methods; and Section E explored pesticide exposure and self-reported health effects, including acute symptoms and risk awareness. Statistical analysis Completed questionnaires were checked for consistency and completeness prior to data entry. Data were coded and analyzed using IBM SPSS Statistics, with Microsoft Excel used for preliminary data handling and graphical presentation. Descriptive statistics were computed, and categorical variables were summarized as frequencies and percentages. Associations between categorical variables were assessed using the chi-square (χ²) test where appropriate. Binary logistic regression analyses were performed to examine the association between independent variables and three binary outcome variables: unsafe pesticide practices, non-adherence to the withholding period (WHP), and self-reported health complications (coded as 1 = presence and 0 = absence). Separate multivariable logistic regression models were developed for each outcome to identify independent predictors while adjusting for potential confounders. Variables entered into the multivariable models were selected based on conceptual relevance, prior evidence, and their potential association with the outcomes. The independent variables included socio-demographic and farm-related characteristics (age, education, farm size, farming experience, type of farming, and location), as well as pesticide-related behavioral factors such as knowledge and perception of WHP, pesticide application practices, and use of personal protective equipment (PPE). A detailed table of all dependent and independent variables, including their definitions and coding, is provided as a supplementary document. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported, and statistical significance was set at p < 0.05. Results Sociodemographic characteristics Table 1 summarizes the socio-demographic characteristics of farmers in Tangail and Sirajganj. Age distribution differed significantly (p < 0.001), with most farmers aged 31–40 years (34.5%) and 41–50 years (23.7%). A higher proportion of farmers aged > 50 years was observed in Tangail. All respondents were male. Educational status varied significantly (p < 0.01), with 60% of farmers being illiterate, particularly in Sirajganj. Most farmers belonged to medium-sized households (67.3%), followed by large (22.8%) and small households (9.8%) (p < 0.001). Farming experience differed significantly (p < 0.001), with 25.3% having more than 20 years of experience, more common in Tangail, while Sirajganj had more farmers with 5–10 years of experience (31.3%). Table 1Socio-demographic information.LocationTotalp valueTangailSirajganjAge< 20 Y11 (26.8%)30 (73.2%)41 (6.8%)< 0.00121–30 Y40 (45.5%)48 (54.5%)88 (14.7%)31–40 Y103 (49.8%)104 (50.2%)207 (34.5%)41–50 Y62 (43.7%)80 (56.3%)142 (23.7%)> 50 Y84 (68.9%)38 (31.1%)122 (20.3%)GenderMale300 (50%)300 (50%)600(100%)EducationIlliterate161 (44.7%)199 (55.3%)360(60%)< 0.01Primary77 (61.6%)48(38.4%)125(20.8%)Secondary30 (45.5%)36 (54.5%)66(11%)Higher secondary17 (60.7%)11 (39.3%)28(4.7%)Graduate15 (71.4%)6 (28.6%)21(3.5%)Household sizeSmall (Up to 3)55 (93.2%)4 (6.8%)59(9.8%)< 0.001Medium (4–6)167 (41.3%)237 (58.7%)404(67.3%)Large (7 & more)78 (56.9%)59 (43.1%)137(22.8%)Farming experience< 5 years53 (41.7%)74 (58.3%)127(21.2%)< 0.0015–10 years63 (33.5%)125 (66.5%)188(31.3%)11–20 years74 (55.6%)59 (44.4%)133(22.2%)> 20 years110 (72.4%)42 (27.6%)152(25.3%)Farm size< 1 acre65 (33.0%)132 (67.0%)197(32.8%)< 0.0011–5 acres202 (58.7%)142 (41.3%)344(57.3%)6–10 acres18 (48.6%)19 (51.4%)37(6.2%)> 10 acres15 (68.2%)7 (31.8%)22(3.7%)Types of farmingVegetables61 (73.5%)22 (26.5%)83(13.8%)< 0.001Grains31 (36.5%)54 (63.5%)85(14.2%)Fruits1 (7.7%)12 (92.3%)13(2.2%)Mixed207 (49.4%)212 (50.6%)419(69.8%)Is agriculture your primary source of income?Yes64 (21.3%)216 (77.0%)280(46.7%)< 0.001No236 (78.7%)84 (28.0%)320(53.3%)Pesticide handling responsibilitySelf256 (56.8%)195 (43.2%)451(75.2%)< 0.001Family members25 (31.6%)54 (68.4%)79(13.2%)Hired labor18 (29.0%)44 (71.0%)62(10.3%)Other1 (12.5%)7 (87.5%)8(1.3%) Socio-demographic information. Farm size distribution also showed significant variation (p < 0.001), with most farmers operating 1–5 acre farms (57.3%). Small farms were more common in Sirajganj. Mixed farming was predominant (69.8%), followed by grains (14.2%) and vegetables (13.8%), with significant regional differences (p < 0.001). Agriculture was the primary income source for 46.7% of farmers, more commonly in Sirajganj (p < 0.001). Pesticide handling was mainly performed by farmers themselves (75.2%), with significant regional variation (p < 0.001). Distribution and regional differences in pesticide brand usage Figure 3 illustrates the distribution of pesticide brands used by farmers in Tangail and Sirajganj, derived from the data presented in Table 2. Pesticide use in Tangail was more diverse, with a larger number of brands reported compared to Sirajganj. The most frequently used pesticide in Tangail was Fighter 2.5 EC (20.3%), followed by Acamite 1.8 EC (11.7%), Rifit 500 EC (7.7%), Reload 18 SC (6.3%), Superheat 500 EC (6%), and Weednil 5 EC (6%). Several other pesticides were used at lower frequencies, generally below 5%. Distribution of pesticide brands used in (A) Tangail and (B) Sirajganj regions. Table 2Farmers’ knowledge and awareness of Withholding Period (WHP).LocationTotal(n = 600)p valueTangailSirajganjKnowledge about With Holding Period (WHP)Yes142 (57.7%)104 (42.3%)246 (41%)< 0.001No158 (44.6%)196 (55.4%)354 (59%)Sources of knowledge about WHPPesticide’s label28 (90.3%)3 (9.7%)31 (12.1%)Agricultural extension officer30 (66.7%)15 (33.3%)45 (17.6%)< 0.001News media0 (0%)21 (100%)21 (8.2%)Peers57 (73.1%)21 (26.9%)78 (30.5%)Others27 (33.3%)54 (66.7%)81 (31.6%)Perception about WHPIt is must to be maintained131 (83.4%)26 (16.6%)157 (26.2%)< 0.001It is optional74 (31.8%)159 (68.2%)233 (38.8%)It may possess health risk if harvested before53 (85.5%)9 (14.5%)62 (10.3%)nothing serious health hazard to be happened42 (28.4%)106 (71.6%)148 (24.7%)Knowledge about Consequences of not maintaining WHPYes155 (88.1%)21 (11.9%)176 (29.3%)< 0.001No145 (34.2%)279 (65.8%)424 (70.7%) Farmers’ knowledge and awareness of Withholding Period (WHP). In contrast, pesticide use in Sirajganj was more concentrated among fewer products. The most commonly used pesticide was Tilt (39.7%), followed by Amistar Top (13.7%), Ridomil Gold MZ (13.3%), Karate (11.7%), and Nitro (6.7%). Other pesticides were used at relatively low frequencies. Several pesticide brands were used exclusively in one region. For example, Fighter 2.5 EC (20.3%), Acamite 1.8 EC (11.7%), and Rifit 500 EC (7.7%) were reported only in Tangail, whereas Nitro (6.7%) and Caught (2.7%) were reported only in Sirajganj. Composition of pesticide active ingredients used by farmers Figure 4 presents the distribution of pesticide active ingredients used by farmers in Tangail and Sirajganj. In Tangail, insecticides were the most frequently reported pesticide group, particularly pyrethroids. The most commonly used active ingredients were lambda-cyhalothrin (24%) and cypermethrin (21.3%). Herbicides were also used, including pretilachlor (13.7%), quizalofop-p-ethyl (6%), and 2,4-D amine (1.7%). Distribution of active ingredients used in (A) Tangail and (B) Sirajganj region. In contrast, in Sirajganj, pesticide use was dominated by fungicides. The most frequently reported active ingredient was propiconazole (39.7%), followed by azoxystrobin (13.7%) and metalaxyl (13.3%). Neonicotinoids, including imidacloprid (3%) and thiamethoxam (6% in Tangail), were reported at lower frequencies in both regions. Chlorpyrifos was also reported in both regions, with higher use in Tangail (7.7%) compared to Sirajganj (1%). Use of registered and banned pesticides Figure 5 presents the regulatory status of pesticides used by farmers in Tangail and Sirajganj. The majority of pesticides used in both regions were registered products. However, banned pesticides were also reported in both districts, with a higher proportion in Sirajganj (20%) compared to Tangail (12%). Regulatory status of pesticides used by the farmers. Hazard classification of used pesticides Figure 6 presents the distribution of pesticides according to the World Health Organization (WHO) hazard classification in Tangail and Sirajganj. Moderately hazardous pesticides (WHO Class II) constituted the largest proportion of use in both regions, accounting for 68% in Tangail and 83% in Sirajganj. Highly hazardous pesticides (WHO Class Ib) were reported only in Sirajganj (2.7%) and were not observed in Tangail. Classification of pesticides used by farmers according to the WHO. Tangail showed a higher proportion of slightly hazardous pesticides (Class III) (16.3%) and pesticides unlikely to present acute hazard (15.7%) compared to Sirajganj, where Class III pesticides accounted for only 1%. Regional differences in types of pesticides used by farmers Figure 7 presents the distribution of pesticide types used by farmers in Tangail and Sirajganj. In Tangail, insecticides were the dominant pesticide type (55.3%), followed by herbicides (37%), while fungicide use was comparatively low (7.3%). In contrast, fungicides were the most frequently used pesticide type in Sirajganj (66.7%), followed by insecticides (25.7%) and herbicides (8%). Types of pesticides used by farmers. Correctness of purpose-specific pesticide use and farmers’ perceptions Figure 8 presents farmers’ self-reported correctness in using pesticides for their recommended purposes in Tangail and Sirajganj. A higher proportion of farmers in Sirajganj (72.7%) reported correct use of pesticides compared to Tangail (56%). Conversely, incorrect use was higher in Tangail (44%) than in Sirajganj (27.3%). Incorrect use included applying pesticides for purposes other than their intended function, such as using insecticides for fungal diseases or weed control. Adherence to recommended pesticide purpose. Farmers’ knowledge and perception of pesticide safety practices Table 2 presents farmers’ knowledge and perception regarding the withholding period (WHP) of pesticides. Overall, 41% of farmers reported being aware of WHP, with a significantly higher proportion in Tangail (57.7%) compared to Sirajganj (42.3%) (p < 0.001). Among those aware of WHP, sources of knowledge varied between regions. In Tangail, the most common sources were pesticide labels (90.3%) and peers (73.1%), whereas in Sirajganj, knowledge was primarily obtained from news media (100%) and other informal sources (66.7%). Agricultural extension officers were cited by 17.6% of respondents overall. Regarding perception of WHP, 26.2% of farmers reported that WHP must be maintained, while 38.8% considered it optional, and 24.7% believed that harvesting before WHP does not pose serious health hazards (p < 0.001). Knowledge about the consequences of not maintaining WHP was reported by 29.3% of farmers overall, with a higher proportion in Tangail (88.1%) compared to Sirajganj (11.9%) (p < 0.001). Adherence to recommended WHP before harvest Figure 9 presents farmers’ practices regarding adherence to the recommended withholding period (WHP) before harvest in Tangail and Sirajganj. In Tangail, 46.3% of farmers reported adhering to the recommended WHP, while 53.7% did not follow the practice. In contrast, adherence in Sirajganj was considerably lower, with only 21% of farmers maintaining WHP and 79% not adhering to it. Practice of maintaining the Withholding Period (WHP) prior to harvest. Concentration of active ingredients in farmer-prepared pesticide solutions Figure 10 presents the concentration levels of active ingredients in pesticide solutions prepared by farmers in Tangail and Sirajganj. In Tangail, 54.7% of farmers applied pesticides at concentrations exceeding recommended levels, while 45.3% used diluted formulations. In Sirajganj, 68.7% of farmers applied over-concentrated solutions, whereas 31.3% used diluted formulations. Concentration levels of active ingredients. Practices and compliance in pesticide preparation among farmers Table 3 presents farmers’ practices regarding the preparation and application of pesticide solutions. Regarding instructions followed for pesticide preparation, the majority of farmers relied on retailers or distributors (65.2%), followed by experience-based practices (12.3%), while only 19.2% followed labeling instructions and 3.3% reported not following any instructions. Tangail farmers were more likely to follow labeling instructions (66.1%) compared to Sirajganj farmers (33.9%) (p < 0.001). Table 3Farmer practices in preparing and mixing pesticide solutions.LocationTotal(n = 600)p valueTangailSirajganjInstructions for preparation of pesticide solutionLabeling instructions76 (66.1%)39 (33.9%)115 (19.2%)< 0.001Retailers/Distributors182 (46.5%)209 (53.5%)391(65.2%)Based on experience33 (44.6%)41 (55.4%)74 (12.3%)Do not follow any instruction9 (45%)11 (55%)20 (3.3%)Compliance of Pesticide Solution Concentrations with Recommended Safety StandardsYes131 (91.0%)13 (9.0%)144 (24.0%)< 0.001No169(37.1%)287(62.9%)456(76%)Mixing multiple pesticides at a timeYes52 (19.5%)215 (80.5%)267 (44.5%)< 0.001No248 (74.5%)85 (25.5%)333 (55.5%)Using any bio-pesticidesYes69 (68.3%)32 (31.7%)101 (16.8%)< 0.001No231 (46.3%)268 (53.7%)499 (83.2%)Types of bio-pesticides usedAsh64 (67.37%)31 (32.63%)95 (15.8%)< 0.001Dhup5 (83.33%)1 (16.67%)6 (1%) Farmer practices in preparing and mixing pesticide solutions. Compliance with recommended pesticide solution concentrations was low, with only 24% of farmers preparing solutions correctly, while 76% did not comply with recommended standards. In Sirajganj, non-compliance was particularly high (62.9%) (p < 0.001). Mixing multiple pesticides was reported by 44.5% of farmers overall, with a higher proportion in Sirajganj (80.5%) compared to Tangail (19.5%) (p < 0.001). The use of bio-pesticides was limited, with only 16.8% of farmers reporting their use. Adoption was higher in Tangail (68.3%) compared to Sirajganj (31.7%) (p < 0.001). Among bio-pesticide users, ash (15.8%) and dhup (1%) were the most commonly reported types. Occupational safety practices during pesticide application Table 4 presents farmers’ practices related to safety protocols during pesticide application. Only 19% of farmers reported always using personal protective equipment (PPE), while 15% used it often, 38.2% sometimes, 16.9% seldom, and 10.9% never. Tangail farmers showed higher adherence compared to Sirajganj (p < 0.001). Regarding types of PPE, gas masks and disposable outer clothing were the most commonly used (53.5%), followed by rubber gloves (2.5%), while goggles and boots were rarely used (1.5%). A substantial proportion of farmers (42.5%) reported not using any protective equipment. Table 4Application of safety protocols during pesticide use among farmers.LocationTotalp valueTangailSirajganjHow often do farmers use Personal Protective Equipment?Always97 (85.1%)17 (14.9%)114 (19.0%)< 0.001Often76(84.4%)14(15.6%)90(15%)Sometimes82(35.8%)147(64.2%)229(38.2%)Seldom14(13.9%)87(86.1%)101(16.9%)Never30(46.2%)35(53.8%)65(10.9%)Types of protective equipment usedRubber Gloves9 (60%)6 (40%)15 (2.5%)< 0.001Gas Masks, Disposable outer clothing225 (70.1%)96 (29.9%)321 (53.5%)Goggles, Boots0 (0%)9 (100%)9 (1.5%)None66 (25.9%)189 (74.1%)255 (42.5%)Spraying pesticides in which direction of wind?With the wind0 (0%)126 (100%)126 (21%)< 0.001Against the wind24 (25.3%)71 (74.7%)95 (15.8%)No consideration of wind276 (72.8%)103 (27.2%)379 (63.2%)Methods for pesticide applicationDistribute tablets9 (56.3%)7 (43.8%)16 (2.7%)< 0.05In furrow/ banded2 (12.5%)14 (87.5%)16 (2.7%)Backpack287 (50.7%)279(49.35)566 (94.3%)Hand spray2 (100%)0 (0%)2 (0.3%)Repairing pesticide-related equipmentYes182 (47.3%)203 (52.7%)385 (64.2%)> 0.05No118(54.9%)97(45.1%)215(35.8%) Application of safety protocols during pesticide use among farmers. In terms of spraying practices, 63.2% of farmers reported no consideration of wind direction, 21% sprayed with the wind, and only 15.8% sprayed against the wind (p < 0.001). Backpack sprayers were the predominant method of pesticide application (94.3%), while other methods such as tablet distribution (2.7%), in-furrow application (2.7%), and hand spraying (0.3%) were rarely used (p < 0.05). Repair and maintenance of pesticide application equipment were reported by 64.2% of farmers, whereas 35.8% did not repair their equipment (> 0.05). Occupational health impacts of pesticide exposure among farmers Figure 11 presents the self-reported health complications among farmers in Tangail and Sirajganj due to pesticide exposure. A higher proportion of farmers in Tangail (80.7%) reported no health complaints compared to Sirajganj (33.7%). In Sirajganj, the most commonly reported symptom was skin irritation (50.3%), followed by headache (9.7%) and nausea (4.7%). Other symptoms, including dizziness, respiratory issues, blurred vision, and gastrointestinal discomfort, were reported at lower frequencies in both regions. Self-reported health complications by the farmer. Logistic regression analyses Factors associated with unsafe pesticide practices Table 5 presents the multivariable logistic regression analysis of factors associated with unsafe pesticide practices among farmers. After adjustment, location and knowledge about withholding periods (WHP) were significantly associated with unsafe pesticide practices. Farmers from Tangail were less likely to engage in unsafe practices compared to those from Sirajganj (Adjusted OR = 0.47; 95% CI: 0.32–0.68; p < 0.001). Farmers with knowledge about WHP were more likely to report unsafe pesticide practices (Adjusted OR = 1.81; 95% CI: 1.26–2.61; p = 0.001). Other variables, including age, education, farm size, farming experience, and type of farming, were not significantly associated with unsafe pesticide practices (p > 0.05). Table 5Factors associated with unsafe pesticide practices among farmers.VariableAdjusted OR95% CIp-valueAge1.020.83–1.240.879Education0.910.76–1.080.286Farm size1.000.77–1.290.997Farming experience1.120.90–1.380.312Type of farming0.900.77–1.040.144Location (Tangail vs. Sirajganj)0.470.32–0.68< 0.001Knowledge about WHP1.811.26–2.610.001WHP: With holding periods. Factors associated with unsafe pesticide practices among farmers. WHP: With holding periods. Factors associated with non-adherence to withholding period (WHP) Table 6 presents the multivariable logistic regression analysis of factors associated with non-adherence to the withholding period (WHP) among farmers. Location and knowledge about WHP were significantly associated with non-adherence. Farmers from Tangail were more likely to report non-adherence to WHP compared to those from Sirajganj (Adjusted OR = 2.61; 95% CI: 1.75–3.90; p < 0.001). Farmers with knowledge about WHP were also more likely to report non-adherence (Adjusted OR = 2.75; 95% CI: 1.91–3.96; p < 0.001). Other variables, including age, education, farm size, farming experience, and perception about WHP, were not significantly associated with WHP non-adherence (p > 0.05). Table 6Predictors of non-adherence to Withholding Period (WHP).VariableAdjusted OR95% CIp-valueLocation (Tangail vs. Sirajganj)2.611.75–3.90< 0.001Age1.090.89–1.350.408Education0.970.81–1.160.753Farm size0.920.71–1.200.549Farming experience0.870.69–1.080.198Knowledge about WHP2.751.91–3.96< 0.001Perception about WHP1.140.96–1.350.129WHP: With holding periods. Predictors of non-adherence to Withholding Period (WHP). WHP: With holding periods. Determinants of self-reported health complications among farmers Table 7 presents the multivariable logistic regression analysis of factors associated with self-reported health complications among farmers. Location, knowledge about WHP, maintaining WHP, following safe pesticide concentrations, and use of bio-pesticides were significantly associated with health outcomes. Farmers from Tangail were less likely to report health complications compared with those from Sirajganj (Adjusted OR = 0.23; 95% CI: 0.13–0.41; p < 0.001). Farmers with knowledge about WHP were more likely to report health complications (Adjusted OR = 2.59; 95% CI: 1.64–4.08; p < 0.001). Table 7Determinants of self-reported health complications among farmers.VariableAdjusted OR95% CIp-valueLocation (Tangail vs. Sirajganj)0.230.13–0.41< 0.001Age0.850.68–1.080.180Education0.990.81–1.230.952Farming experience0.880.69–1.140.343Farm size1.270.94–1.740.124Knowledge about WHP2.591.64–4.08< 0.001WHP maintained0.510.31–0.820.005Safe concentration followed0.280.15–0.54< 0.001Use of bio-pesticides0.170.09–0.33< 0.001PPE use (type)0.830.67–1.030.094PPE use (frequency)0.840.70–1.020.077Mixing pesticides0.440.19–1.010.052Frequency of application1.070.91–1.240.419WHP: With holding periods, PPE: Personal protective equipment. Determinants of self-reported health complications among farmers. WHP: With holding periods, PPE: Personal protective equipment. Discussion This study reveals substantial regional variation in pesticide use patterns, safety practices, and associated health risks among farmers in the Jamuna riverside regions of Bangladesh. Differences between Tangail and Sirajganj were consistently observed across pesticide types, active ingredients, handling behaviors, and health outcomes, highlighting the critical role of agroecological conditions, cropping systems, and access to information and institutional support in shaping pesticide-use practices. The observed diversity of pesticide use in Tangail appears to be closely associated with its mixed and vegetable-based farming systems, where high pest pressure necessitates frequent and varied pesticide applications. Ali et al. and Khatun et al., have also reported similar high pest pressure in vegetable-based farming system in Bangladesh23,28. In contrast, the dominance of fungicides in Sirajganj reflects its cereal-based and floodplain agriculture, where environmental conditions such as high humidity, waterlogging, and seasonal flooding create favorable conditions for fungal diseases29. Similar dominance of fungicides in cereal-based agricultural system was also reported by Kongcharoen et al.30. These differences in study areas indicate that pesticide use is strongly driven by crop type and agroecological conditions31,32. However, the widespread use of pyrethroids in Tangail and fungicides such as propiconazole in Sirajganj raises important concerns, as these compounds have been associated with neurotoxicity, endocrine disruption, hepatotoxicity, and reproductive effects33–36. Furthermore, the concurrent use of multiple pesticide types suggests the potential for cumulative and synergistic exposure, which may amplify both acute and chronic health risks37. The continued detection of banned pesticides, particularly in Sirajganj, points to significant gaps in regulatory enforcement and pesticide governance. According to the research report of Bangladesh Resource Center for Indigenous Knowledge (BARCIK), about 68% of affected individuals had used pesticides (e.g., paraquat, carbaryl, aluminum phosphide etc.) officially banned in Bangladesh38. According to the Pesticide Action Network, nearly 75% of farmers in the Asia-Pacific region-including Bangladesh, India, Laos, and Vietnam-reported using highly hazardous pesticides (HHPs) or chemicals that are banned in one or more countries39. Although most pesticides reported were registered in current study, the persistence of banned products are still available due to the weak supply chain monitoring, informal distribution channels, and limited enforcement40,41. Given that many banned pesticides fall under the category of highly hazardous pesticides (HHPs), their use presents serious risks, including acute poisoning and long-term health consequences such as carcinogenicity and reproductive health disorders42,43. This situation reflects a disconnect between policy frameworks and field-level implementation, particularly in rural and riverine areas where regulatory oversight is limited. The predominance of moderately hazardous (WHO Class II) pesticides in both regions further highlights the potential for occupational exposure. These pesticides are widely used due to their effectiveness and affordability but are also associated with a high burden of pesticide-related poisoning in low- and middle-income countries44,45. The detection of highly hazardous (Class Ib) pesticides in Sirajganj is particularly alarming, as these compounds can cause severe toxicity even at low exposure levels27. The relatively higher use of lower-hazard pesticides in Tangail may suggest better access to safer alternatives or greater exposure to agricultural extension services, whereas the reliance on more hazardous compounds in Sirajganj indicates structural constraints, including limited awareness, availability, and regulatory enforcement. This disparity highlights the need for equitable access to safer pesticide options across regions. Farmers’ pesticide uses practices revealed multiple critical gaps in knowledge and behavior. Misuse of pesticides for incorrect purposes, particularly in more intensive farming systems, reflects limited understanding of pesticide specificity and mode of action. The current results align with previous studies from Bangladesh46, Pakistan47, and Ethiopia48, indicating that misuse of pesticides, including application on non-recommended crops and use against unintended pest targets, is widespread. Knowledge of withholding periods (WHP) was generally low, and misconceptions regarding its importance were widespread in both study areas, increasing the risk of pesticide residue contamination in food products49,50. A significant knowledge-practice gap was evident, as awareness did not consistently lead to safe practices, especially regarding WHP adherence. Consistent findings have been reported by Istriningsih et al.51 and Uikey et al.52, highlighting that high levels of knowledge do not necessarily ensure the proper implementation of Good Agricultural Practices (GAP) among farmers. This suggests that pesticide use decisions are influenced not only by knowledge but also by economic pressures, perceived risk, and experiential practices53,54. Unsafe pesticide handling practices were prevalent across both regions. Over-application, improper dilution, and mixing of multiple pesticides were commonly reported, reflecting a reliance on informal knowledge and perceived effectiveness rather than scientific recommendations. These practices are consistent with findings from South Asia, where farmers often adjust pesticide dosages based on personal judgment, cost considerations, or peer advice55–57. Over-concentration increases the risk of acute toxicity, environmental contamination, and crop damage, while under-dilution reduces efficacy and contributes to pesticide resistance58,59. The mixing of multiple pesticides without knowledge of compatibility further increases the risk of chemical interactions, synergistic toxicity, and cumulative exposure. Occupational safety practices were also found to be inadequate. The low use of personal protective equipment (PPE), unsafe spraying practices, and poor maintenance of application equipment significantly increase exposure risk. Damalas et al.60 and Gebretsadkan et al.2, also reported lower use of PPE and unsafe pesticides application among the farmers of northern Greece and Ethiopia. These are probably due to the high cost of PPE, discomfort in hot and humid climates, lack of awareness, and underestimation of risks61,62. In addition, the reliance on manual spraying equipment in smallholder farming systems increases direct contact with pesticides, further elevating exposure risks. Environmental factors, such as high temperatures and humidity, may also discourage consistent PPE use, exacerbating occupational hazards. These unsafe practices are reflected in the observed health outcomes. Farmers in Sirajganj reported a significantly higher prevalence of symptoms such as skin irritation, headache, and nausea, indicating greater pesticide exposure. Similar symptoms have been associated with exposure to commonly used pesticide classes, including pyrethroids, organophosphates, and carbamates63,64. The cumulative effect of unsafe practices-including overuse, poor PPE compliance, incorrect application, and non-adherence to WHP-likely contributes to these health disparities. Although this study focuses on self-reported acute symptoms, chronic exposure to pesticides may lead to long-term health consequences, including neurodegenerative disorders, reproductive toxicity, and endocrine disruption7. These findings highlight the broader public health implications of unsafe pesticide use in agricultural communities. The multivariable analyses further reinforce the importance of behavioral and contextual determinants. Geographic location consistently influenced unsafe practices, WHP non-adherence, and health outcomes, emphasizing the role of regional context. Interestingly, knowledge of WHP was positively associated with unsafe practices and adverse health outcomes, further confirming the presence of a knowledge-practice gap. The absence of significant associations with socio-demographic variables suggests that unsafe practices are widespread and not confined to specific demographic groups, indicating the need for broad-based interventions. Conversely, adherence to recommended practices-such as maintaining WHP, using appropriate pesticide concentrations, and adopting bio-pesticides-was associated with a lower likelihood of health complications. These findings highlight the protective role of safe pesticide handling and sustainable pest management practices. The promotion of integrated pest management (IPM) and bio-pesticides can reduce reliance on hazardous chemicals while maintaining agricultural productivity and environmental sustainability. Practical and Policy Implications The findings of this study have important implications for farmers and agricultural stakeholders. Improving pesticide safety requires a shift from knowledge-based approaches to behavior-focused interventions, including hands-on training, farmer field schools, and demonstration-based learning. Enhancing access to affordable and climate-appropriate PPE is essential to reduce occupational exposure. Simplifying pesticide label instructions and improving communication through locally understandable formats may further enhance compliance. Additionally, promoting bio-pesticides and integrated pest management (IPM) practices can provide safer and more sustainable alternatives to conventional pesticide use. At the policy level, the continued use of banned and highly hazardous pesticides highlights the need for stronger regulatory enforcement and improved pesticide governance. Strengthening supply chain monitoring, regulating agrochemical retailers, and restricting unauthorized pesticide sales are critical. Expanding agricultural extension services, particularly in underserved regions such as Sirajganj, is essential to improve farmer knowledge and practices. Policies should also support the adoption of safer alternatives, including subsidies and incentives for bio-pesticides and IPM. Integrating pesticide safety into national agricultural and public health strategies can further reduce occupational exposure and enhance food safety. Conclusion This study identifies a critical knowledge–practice gap as the central finding, where farmers’ awareness of pesticide safety-particularly regarding withholding periods (WHP)-does not consistently translate into safe practices, leading to unsafe pesticide use and increased health risks. Regional differences further indicate that contextual factors such as farming systems, access to extension services, and local practices influence pesticide behavior and exposure. These findings highlight that improving knowledge alone is insufficient and emphasize the need for behavior-focused interventions, including hands-on training, farmer field schools, and improved risk communication strategies. Enhancing access to affordable personal protective equipment (PPE), promoting adherence to WHP, and encouraging the adoption of bio-pesticides and integrated pest management (IPM) are essential to reduce occupational and environmental risks. From a policy perspective, strengthening regulatory enforcement-particularly controlling banned pesticides and monitoring pesticide distribution-is crucial. In addition, the presence of moderately hazardous and banned pesticides underscores the need for improved market surveillance and stricter regulation at distribution and retail levels. Improving labeling systems with clear, locally understandable instructions and strengthening monitoring of pre-harvest practices are also necessary to enhance WHP compliance and reduce food safety risks. Furthermore, increasing the affordability and accessibility of PPE, along with incentive-based programs, and promoting IPM and bio-pesticide adoption can support safer and more sustainable agricultural practices. As a cross-sectional survey, it captures data at a single point in time, which limits the ability to establish causal relationships between pesticide use practices and reported health outcomes. Future research should employ longitudinal designs, incorporate biomonitoring or residue analysis, and expand to diverse agro-ecological settings to better assess exposure pathways and intervention effectiveness. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Author contributions M.R.H. has supervised the research and reviewed the manuscript.S.M.H.I., M.S.A, E.M.J, S.R.S have collected and analyzed the data and also wrote the draft. S.I.S. has prepared the figures ana reviewed the manuscript.R.B. has reviewed the manuscript.M.A.H. has reviewed the manuscript. Funding No funding was received for conducting this research. Data availability The datasets analyzed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets analyzed during the current study are available from the corresponding author on reasonable request.