A cross-sectional analytical study was conducted to evaluate the association between health literacy and behaviors aimed at preventing exposure to plant pest control chemicals among Thai farmers. The variables measured included general information characterizing the sample group, health literacy, and behaviors aimed at preventing exposure to plant pest control chemicals.
Population and Sample Group
The population consists of male and female farmers aged 18 years and older engaged in agriculture, with agricultural households registered with the agriculture office totaling 521,593 people [17]. The required sample size was calculated using G*Power software[18], which was created from Cohen's (1988) [19] formula. The calculation was performed according to the statistics used in the data analysis, with a 95% confidence level, a margin of error of .05, a test power of .80, and an effect size of .05[18]c. This resulted in a required sample size of 464 people. Multistage stratified cluster random sampling was used, and the data were divided according to the characteristics of the agricultural areas via a stratified random sampling method; the results were divided into 3 groups: 1) paddy farming only, 2) paddy and upland rice farming, and 3) paddy farming and cash crop growing. Cluster random sampling was used to select a representative sample from each of the 3 groups, resulting in 1 subdistrict per group, which was sufficient to represent the population because farmers in the same group showed high homogeneity. The data were then collected from the sample group using simple random sampling until the required number of participants was reached.
Research Ethics
We obtained ethical approval from the Ethical Review Committee of the Public Health Office of Nakhon Ratchasima Province, Thai Ministry of Public Health (Reference No: KHE 2020-053).
Research Tools
The study included a questionnaire and risk screening, which were conducted using a fingerprick blood test. The details are as follows:
The questionnaire consisted of three parts:
Part 1
General information. The questions included sex, age, education, average annual household income, history of chemical pest control substance use, sources of chemical pest control substance use, use of chemical pest control substances in rice cultivation, history of chemical pest control substance exposure, and history of training on the use of chemical pest control substances.
Part 2
A questionnaire was used to measure health knowledge related to the use of chemical pest control substances across six aspects. The questionnaire was divided into two subsections
Part 2.1 covered one aspect: knowledge and understanding. This portion consists of 13 questions with closed-ended response options of yes, no, and unsure. A correct answer was scored as 1 point, while wrong or unsure answers were scored as 0 points. A participant’s total score in the area of health knowledge could be classified as high (a score above 80%), moderate (a score between 60% and 80%), or low (a score below 60%) using Bloom’s cutoff point[20, 21].
Part 2.2 covered five aspects: access to information and services (7 items), communication (8 items), self-management (6 items), media and information literacy (6 items), and decision-making (10 items). The responses were closed-ended, with five answer choices describing the frequency with which the participant performed that behavior: regularly (all seven days in a week), frequently (5–6 days in a week), sometimes (3–4 days in a week), infrequently (1–2 days in a week), or never (not at all). For the 30 positive questions, responses were scored from 5 − 1, and for the seven negative questions, responses were scored from 1–5.
Part 3 covered preventive behavior aimed at reducing exposure to chemical pesticides and consisted of 30 items divided as follows: behaviors performed in preparation for (10 questions), during (9 questions), and after (11 questions) the spraying of chemical pesticides. The responses were closed-ended with five answer choices, 13 questions were scored positively, and 17 questions were scored negatively. Response options and scoring were performed as described in section 2.2. Scores from parts 2.2 and 3 were used to interpret the participants’ level of health knowledge or behavior. Total scores could be classified as high (75–100%), medium (26–74%), or low (0–25%) using quartile calculations.
The risk of exposure to chemical pesticides was determined by finger-prick blood tests using cholinesterase reactive paper on farmers following the procedure set in place by the Bureau of Occupational and Environmental Diseases within Thailand’s Ministry of Public Health. The blood test results of the farmers were divided into four groups according to the level of AChE: safe (≥ 100.0 U/ml), normal (87.5–99.9 U/ml), at-risk (75.0–87.4 U/ml), and unsafe (< 75.0 U/ml) [23].
Quality Assessment of the Research Tool
The research tool used in this study was tested for accuracy and content coverage by three qualified experts, namely, a health knowledge expert, an occupational health and safety professional, and a researcher specializing in the health of farmers. Only questions with an index of item objective congruence (IOC) greater than 0.60 were selected. Suggestions were then implemented, and the tool was piloted with a group of 30 farmers similar to those in the research sample group. The reliability of each questionnaire item and part was evaluated separately. The knowledge and understanding component was analyzed using the Kuder-Richardson 20 (KR20) method, which yielded a reliability of r = 0.79. The parts covering access to information and services, communication, self-management, and media literacy yielded a Cronbach's alpha coefficient (α) of 0.86. For decision-making, the reliability was α = 0.85; for the questionnaire on behaviors aimed at avoiding exposure to chemical pesticides, the reliability was α = 0.88.
Data collection process
1. Field research assistants were trained to collect data in the communities where the sample group resided. The lead investigator distributed a data collection manual for the study, which included the details of the research project, the purpose of the research, the guidelines for introducing the data collectors, an explanatory letter and research consent form, the characteristics of the sample group for data collection, the content of each part of the tool, and steps on how to use the tool to collect the data. To ensure that the data collectors understood and could correctly use the tool according to the research procedures, the lead investigator organized a practical training session and demonstrated how to collect the data in the field. This training included providing information on how to introduce themselves, how to explain the purpose of seeking cooperation when providing information, how to correctly record data and how to check for complete responses. The data collectors also practiced actual data collection with farmers who were similar to the individuals in the research sample group, and their performance was evaluated until they could fully follow the study procedures.
2. A letter was sent to the Subdistrict Administrative Organization leader to request permission for data collection within the chosen area. Once approval was received, the researchers met with the village leader and village health volunteers (HVVs) to explain the details of the research, emphasizing the ethical principles of the research to protect the rights of the sample group throughout the study period.
3. The date, time, and location of each visit to the area for data collection were coordinated and agreed upon with the sample group of farmers.
4. Each data collector proceeded to the field, and data were collected from the sample. If a member of the sample group was unable to read the questionnaire by themselves, the data collector read the questionnaire and let the participants respond by themselves.
5. The lead investigator checked the accuracy of the collected data again and digitally inputted the obtained information into the database for analysis.
Data analysis
The analysis was performed using SPSS to present the results. Descriptive statistics are presented as values, frequency distributions, percentages, averages, standard deviations, and minimum and maximum values. Inferential statistics were used to assess the association between health knowledge and exposure prevention behaviors using the Spearman rank correlation coefficient (or Spearman's rho [rs]). As the data obtained from the questionnaire were not normally distributed, binary logistic regression was performed using the stepwise method, which consisted of selecting the independent variable with the highest correlation value for the first analysis, followed by variables with successively lower correlation values.