Understanding farmers' safety behaviour towards

8 downloads 0 Views 1MB Size Report
Science of the Total Environment xxx (2017) xxx–xxx ..... Access to first-aid kit and fire extinguisher in the ..... Pesticide Action Network Asia and the Pacific.
STOTEN-24425; No of Pages 9 Science of the Total Environment xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran Rohollah Rezaei a,⁎, Christos A. Damalas b,⁎, Gholamhossein Abdollahzadeh c a b c

Department of Agricultural Extension, Communication and Rural Development, Faculty of Agriculture, University of Zanjan, Zanjan, Iran Department of Agricultural Development, Democritus University of Thrace, GR 682 00 Orestiada, Greece Department of Agricultural Extension and Education, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• A model for farmers' use of occupational safety and health (OSH) practices was developed. • The examined OSH practices included a range of behaviours grouped in four categories. • Almost half of the farmers (49.5%) showed totally unsafe behaviour in the use of PPE. • Farmers were prone to different occupational hazards in the agricultural sector. • Attitude, knowledge, and self-efficacy in safety affected the use of OSH practices.

a r t i c l e

i n f o

Article history: Received 2 September 2017 Received in revised form 19 October 2017 Accepted 19 October 2017 Available online xxxx Editor: D. Barcelo Keywords: Attitude Knowledge Occupational hazards Safety behaviour Self-efficacy

a b s t r a c t Preventive interventions for reducing occupational risks and health problems among farmers require the identification of factors contributing to unsafe behaviour, but research on this topic is rather limited. A theoretical model for studying factors affecting farmers' use of occupational safety and health (OSH) practices in Iran was developed. The model was empirically tested using data collected from a survey of 301 tomato farmers of Zanjan Province of Iran. The examined OSH practices encompassed a wide range of behaviours, grouped in four categories, i.e., use of pesticides, use of machinery, use of personal protective equipment (PPE), and applying ergonomic principles (i.e., fitting the task to the individual, designing the workplace based on human factors, taking into account the interaction between the workplace and the workers, exercising during work or rest). Almost half of the farmers (49.5%) showed unsafe behaviour in the use of PPE. Moreover, significant proportions of the farmers showed potentially unsafe behaviour in the use of pesticides (42.2%), in applying ergonomic principles (40.2%) and in the use of machinery (35.9%). Attitude towards OSH practices, knowledge on OSH practices, and selfefficacy in safety had a direct positive effect on farmers' use of OSH practices, explaining 73% of the variance in farmers' safety behaviour. Overall, findings contribute to a better understanding of the use of OSH practices among farmers, providing empirical evidence in the cognitive processing of farmers' with respect to safety behaviour in farming and offering practical information that can be incorporated into OSH intervention programs in Iran and other developing countries. © 2017 Elsevier B.V. All rights reserved.

⁎ Corresponding authors. E-mail addresses: [email protected] (R. Rezaei), [email protected] (C.A. Damalas).

https://doi.org/10.1016/j.scitotenv.2017.10.201 0048-9697/© 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

2

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

1. Introduction Agriculture is one the most hazardous economic sectors (Colémont and Van den Broucke, 2008; ILO, 2011), which along with mining and construction has the highest occupational hazard rate (Das et al., 2016). Data from ILO (International Labour Organization) and FAO (Food and Agriculture Organization) show that out of a total of 335,000 fatal workplace-related accidents worldwide, there are about 170,000 deaths among agricultural workers per year (Padilla, 2013). Thus, the rate of fatal accidents in farming is at least twice as high as in other occupations (Das et al., 2016). In fact, because of the difficulty and physical nature of agricultural activities and because agricultural work is typically performed in the open air (Kaewboonchoo et al., 2015), farmers can be exposed to various hazards and are susceptible to several occupational injuries. Toxicological hazards brought about by pesticides, also including fertilizers, fuels, and other organic substances, are common problems among most agricultural workers (Damalas and Eleftherohorinos, 2011; Padilla, 2013; Damalas and Koutroubas, 2016). Risk of slips, trips, and falls from heights as well as lifting heavy weights and other activities giving rise to musculoskeletal disorders are additional occupational hazards (ILO, 2011). Finally, other working conditions common to rural environments, such as exposure to extreme temperatures, inclement weather, and wild animals can be occupational hazards in the agricultural sector (ILO, 2011). A review of the literature and empirical studies in the field of OSH show that many factors can affect the use of OSH practices among farmers and that one of the most important of these variables is the attitude towards OSH (Brosseau and Li, 2005; Damalas et al., 2006b; Colémont and Van den Broucke, 2008; Raksanam et al., 2012; Adebola, 2014; Aluko et al., 2016). The attitude towards a particular behaviour reflects the overall positive or negative evaluation of an individual to perform that behaviour. In general, the more favorable the attitude towards the behaviour, the stronger should be the individual's intention to perform it (Armitage and Conner, 2001). A farmer will use safety measures only if he believes that the health benefits of using those measures are greater than the cost of using them (Kien, 2015). Therefore, the attitude can be considered as a crucial predictor of the behaviour of individuals (Ajzen, 2002; Peres et al., 2005). According to these issues, the first hypothesis of this study was developed as follows: Hypothesis 1. Attitude towards OSH has a positive and significant effect on farmers' use of OSH practices (Fig. 1). Another variable influencing the use of OSH practices that has been considered in various studies is knowledge on OSH practices (Damalas et al., 2006a; Atreya, 2007; Gupta et al., 2012; Raksanam et al., 2013; Kumari and Reddy, 2013; Mohanty et al., 2013; Kien, 2015). Generally, knowledge is defined as a fluid mix of framed experience, contextual

information, values and expert insight that provides a basis for evaluating and integrating new information and experiences (Davenport and Prusak, 1998; Kien, 2015). The level of knowledge can affect the health-related behaviour of individuals; the higher the knowledge level, the more likely the individual is to show safe behaviours (Glanz et al., 2008; Kien, 2015). According to the mentioned issues, the second hypothesis of this study was developed as follows: Hypothesis 2. Knowledge on OSH has a positive and significant effect on farmers' use of OSH practices (Fig. 1). Finally, in addition to the two variables mentioned, another main determinant of the use of OSH practices is self-efficacy in safety (DeJoy, 1996; Pettinger, 2000; Brown et al., 2000; Wagner et al., 2013). The concept of self-efficacy, derived from the cognitive social theory of Bandura (1997), refers to one's belief in his ability to mobilize cognitive resources, motivation, and a course of action, which are required to perform a particular task (Wood and Bandura, 1989); accordingly, selfefficacy in safety is defined as the beliefs about one's ability to follow indicated safety measures successfully (DeJoy, 1996). Since self-efficacy has an effect on the individual's subsequent effort and persistence in performing activities (Bandura, 1997), people who have a high sense of self-efficacy for a particular task, perform better than those who have a low sense of self-efficacy for that task (Gist and Mitchell, 1992). The level of confidence of individuals in their abilities to work safely, i.e., self-efficacy in safety, would determine to what extent those individuals actually engage in safe or unsafe practices in the workplace (Brown et al., 2000). Accordingly, the third hypothesis of the study was developed as follows: Hypothesis 3. Self-efficacy in safety has a positive and significant effect on farmers' use of OSH practices (Fig. 1). Despite the importance of OSH and the necessity of planning for reducing the occupational injuries of farmers, evidence shows that OSH in the Iranian agricultural sector is a neglected and almost unknown issue, as few farmers receive training in occupational health (Karami et al., 2016). Furthermore, relatively few provinces in Iran (i.e., about 35%) have agricultural health committees and centers for providing OSH services. In addition, there are several limitations for providing professional health services to agricultural workers through Primary Health Care (PHC) networks (Rafiei et al., 2015). On the other hand, studies about OSH are limited in Iran, particularly in the agricultural sector. Therefore, there is a big gap of knowledge regarding farmers' behaviour towards OSH, so that studies on farmers' behaviour towards OSH and factors affecting behaviour are necessary. The results of such studies can provide a basis for decision-making and development of preventive interventions to reduce occupational injuries among farmers. Considering the importance of this issue, the aim of this study was to examine the

Fig. 1. Theoretical research framework.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

current levels of farmers' use of OSH practices and determine factors affecting farmers' behaviour towards OSH. 2. Materials and methods

3

collected through face-to-face interviews with the farmers in a way of friendly discussion from April to July 2017.

2.3. Survey instrument

2.1. Study area The study area was confined to Zanjan Province, covering an area equal to 22,164 square km that is located in the northwest of central plateau of Iran (Fig. 2). About 37.5% of the total population of Zanjan Province resides in rural areas. Given that half of the province's area is agricultural land, this province has high potential in agriculture and is considered as one of the main agricultural poles in Iran. The main crops grown in the province are: wheat, barley, bean, garlic and various types of vegetables, including onions, potatoes, and tomatoes. Zanjan Province with a harvest level of 5004 ha and production rate of 231,881 t in the crop year of 2014–15 was one of the ten major provinces of tomato production in Iran. 2.2. Sampling The target population of this study was the tomato growers in Zanjan Province. According to the statistics of Zanjan Agri-Jihad Organization (2015), the total number of tomato growers in the study area was 3432. Based on the table of Bartlett et al. (2001), a sample of 275 farmers was required for this study. To ensure access to the required number of questionnaires, the sample was increased to 310 individuals. Of the 310 questionnaires that were collected, nine were dropped due to incomplete or inconsistence response and, therefore, 301 questionnaires were used for analysis. Given the heterogeneous characteristics of the population among the strata (i.e., counties) and the disproportionate distribution of farmers in the different counties of the province, the stratified random sampling method was used. Thus, the total number of tomato growers in Zanjan Province was divided into smaller groups (strata) in each county, based on farmers' distribution, and then a random sample from each stratum was taken in a number proportional to the stratum's size compared to the target population. These subsets of the strata were then pooled to form a random sample. Data were

Data were collected through a structured questionnaire, which included two parts. The first part included six questions capturing farmers' and farm's characteristics, including age, education, tomato farm size, farming experience, family size, and participation in training courses on OSH. The second part measured farmers' perceptions of each variable in the model with 38 items measuring the four latent variables 1-use of OSH practices (i.e., safety behaviour), 2-attitude towards OSH practices (6 items, Cronbach's alpha 0.85), 3-knowledge on OSH practices (6 items, Cronbach's alpha 0.86), 4-self-efficacy in safety (6 items, Cronbach's alpha 0.79). The variable of OSH practices as the dependent variable composed of four subsections, i.e., use of pesticides (6 items, Cronbach's alpha 0.81), use of machinery (4 items, Cronbach's alpha 0.83), use of PPE (5 items, Cronbach's alpha 0.85), and applying ergonomic principles (5 items, Cronbach's alpha 0.77). Cronbach's alpha is a measure used to assess the reliability (or internal consistency) of a set of scale or test items. It ranges from 0 to 1 and many methodologists recommend a minimum value between 0.65 and 0.8 (or higher in many cases). Application of ergonomic principles included fitting the task to the individual, designing the workplace based on human factors, taking into account the interaction between the workplace and the workers, exercising during work or rest. The tested items were inspired, developed, and combined from previous works. A list of measurement items and their sources of each part are presented in Tables 3 and 4. Before the initiation of the study, the questionnaire was sent out for review by six informant farmers and five local extension officers. This phase was used to refine the items and the constructs used in the study and to clarify the wording, the content, and the general layout of the survey instrument. Before officially issuing questionnaires, a pilot study of the survey instrument was conducted to ensure that farmers could understand the items and the measurement scales. To this end, the validated version of the questionnaire was sent to 30 farmers who had not been selected in the sample of the study. Completed questionnaires, feedback from the farmers, and observations by the

Fig. 2. Map of the study area.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

4

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

researchers resulted in minor changes in the survey instructions and refinement of the wording in some items.

Table 1 Socio-demographic profile of the farmers. Variable

2.4. Data analysis To test the research hypotheses, a two-step procedure in structural equation modeling (SEM) was used. In the first step, the research measurement models (i.e., first-order confirmatory factor analysis/CFA) were evaluated to examine the goodness of fit for the research model and determine the validity and reliability of the model. In the second step, the relationships between the variables were estimated based on the structural model (Anderson and Gerbing, 1988). Various indices have been proposed to evaluate the fit of the model (Byrne, 2010; Hair et al., 2010). The indices employed in this study included relative chi-square (χ2/df), incremental fit index (IFI), goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), root mean square residual (RMR), comparative fit index (CFI), and root mean squared error approximation (RMSEA). The relative chisquare (also called the normed chi-square) equals the chi-square index divided by the degrees of freedom (χ2/df). This index might be less sensitive to sample size. The criterion for acceptance varies across researchers, ranging from less than 2 (Ullman, 2001) to less than 5 (Schumacker and Lomax, 2004). The convergent validity of the model was examined via three different criteria, including standardized factor loadings equal to or greater than 0.5, average variance extracted (AVE) equal to or greater than 0.5, and composite reliability (CR) equal to or greater than 0.7 (Hair et al., 2010). Moreover, to test the discriminant validity of the model, the value of the square root of the AVE of each latent variable needs to be greater than the correlation of that variable with the other latent variables, based on the approach suggested by Fornell and Larcker (1981). In addition to the validity of the model, CR was used to assess the reliability of the model, which requires that the value for each latent variable to be greater than 0.7 (Hair et al., 2010). In this section, the Graphic Software of AMOS20 was used. In the descriptive section, the collected data were analyzed using the statistical package for the social sciences (SPSS). Descriptive statistics, such as frequencies, percentages, and means were used. Regarding farmers' use of OSH practices, the average use score of the items in the four subsections of OSH practices (i.e., use of pesticides, use of machinery, use of PPE, and applying ergonomic principles) was used to discriminate safety behaviour of the farmers. To this end, farmers' performance was expressed as a value ranging between 0 and 1 by applying the following formula (Singh and Hiremath, 2010; Damalas and Abdollahzadeh, 2016):

Age (mean = 51.2 years) Up to 30 years From 31 to 45 years More than 45 years Education level No formal education (illiterate) 1–5 years of schooling (elementary education) 6–12 years of schooling (secondary education) Above 12 years of schooling (some college) Average tomato farm size (mean = 3.81 ha) Less than 3 ha From 3 to 6 ha From 6 to 9 ha From 9 to 12 ha Above 12 ha Average farming experience (mean = 28.2 years) Average family size (no. individuals) (mean = 5.2) Previous training on OSH Yes No

Frequency

%

– 78 100 123 – 125 100 57 19 – 190 64 15 21 11 – – – 47 254

– 25.9 33.2 40.9 – 41.6 33.2 18.9 6.3 – 63.1 21.3 5.0 7.0 3.6 – – – 15.6 84.4

farmers had low education level; a solid fraction (41.6%) had no formal education at all (i.e., never been in a school environment) and almost an equal proportion (33.2%) did not have more than formal education with rather poor reading and writing skills (Table 1). The average tomato farm size of farmers of the sample was 3.81 ha, but most farmers (63.1%) were in the category below 3 ha. The average farming experience was 28.2 years, revealing an experienced and rather conscientious average farmer in the survey sample (as assumed by age and experience). The average family size was 5.2 people. Most farmers (84.4%) did not receive any training on OSH (in the form of a seminar with classroom lectures or field demonstrations or both that required registration of participants and are organized by an official organization). Farmers' safety behaviour varied depending on type of OSH practices (Fig. 3). Almost half of the farmers (49.5%) had unsafe behaviour in the use of PPE. Moreover, a great proportion of the farmers (42.2%) behaved potentially unsafe in the use of pesticides (Fig. 3). Overall, the lowest percentage in the four groups of practices examined was related to farmers who showed potentially safe or safe behaviour, indicating that farmers were prone to different occupational injuries and hazards. Descriptive statistics of the main studied constructs are summarized in Table 2. The two variables, i.e., attitude towards OSH practices and knowledge on OSH practices had the highest and lowest average scores, respectively.

Performance ¼ ðactual value−minimum valueÞ=ðmaximum value−minimum valueÞ

where actual value = the average score of the items for each farmer, minimum value = the lowest score of the items for each farmer, and maximum value = the greatest score of the items for each farmer. Then, the safety behaviour of farmers was divided into five levels of 0.20 points each, totaling 1, following the five-point scale model of Damalas and Abdollahzadeh (2016), as below: -

Safe behaviour (excellent): 0.81–1.00; Potentially safe behaviour (good): 0.61–0.80; Intermediate behaviour (medium): 0.41–0.60; Potentially unsafe behaviour (poor): 0.21–0.40; Unsafe behaviour (bad): 0.00–0.20.

3. Results 3.1. Descriptive results Mean age of the farmers was 51.2 years, with a tendency towards the category above 45 years (Table 1). Regarding education, most

3.2. Measurement models estimation for dependent and independents variables To test the construct validity, discriminant validity, CR, and fit of the model, seven individual measurement models were estimated through performing first-order CFA (Tables 3 and 4). The factor loading values for most observed variables were greater than 0.7 and so were significant, except for pesticide5, ergonomic5 (Table 3), knowledge6 and selfefficacy3 (Table 4) which had lower factor loading values of 0.44, 0.38, 0.41, and 0.28, respectively. These variables were therefore dropped from the measurement model, which was then retested. In addition, with regard to the AVE and CR values of each latent variable evaluated in this research, the values of all variables were higher than 0.5 and 0.7, respectively (Tables 3 and 4). Therefore, convergent validity and CR of the research model were evident. Discriminant validity with the square root of the AVE represented by the diagonal are presented in Table 5. In all cases, the Fornell and Larcker (1981) test is met for all pairs of latent variables. That is, there is discriminant validity; the latent variables are distinctly different from each other. Moreover, fit indices ranged from very good to excellent. For completeness, the ‘full

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

5

Fig. 3. Levels of farmers' safety behaviour.

measurement model’ is presented (Table 6), entering all items simultaneously in a CFA model with a predicted measurement model imposed (Cadogan et al., 2006). All fit heuristics were within acceptable cut-off ranges. 3.3. Structural model estimation The structural model of the study showed a direct relationship between the independent variables with the dependent variable of farmers' use of OSH practices. Although the estimated model based on the chi-square significant indicator lacked a good fitness, the model had an acceptable fitness based on other criteria (Fig. 4). The three independent variables explained about 73% of the variance of farmers' use of OSH practices (Fig. 4). Attitude towards OSH practices, knowledge on OSH practices, and self-efficacy in safety had statistically significant positive relationships with the variable of farmers' use of OSH practices; accordingly, research hypotheses H1, H2, and H3 were supported (Table 7). 4. Discussion This study was conducted to explain the factors affecting tomato farmers' use of OSH practices in Zanjan Province, Iran. The examined OSH practices encompassed a wide range of behaviours, mainly related to use of pesticides, use of machinery, use of PPE, and applying ergonomic principles. Farmers' safety behaviour varied depending on the type of OSH practices and was positively affected by attitude towards OSH practices, knowledge on OSH practices, and self-efficacy in safety. The study provided a framework for better understanding OSH of farmers in Iran, with emphasis on farmers' individual characteristics. Given that previous studies in this area had one-dimensional approach with emphasis on related practices to the use of pesticides, the present study investigated more comprehensively safety practices and Table 2 Descriptive statistics of the constructs. Constructs

Min.

Max.

Mean

SD

Attitude towards OSH Knowledge on OSH Safety self-efficacy

1 1 1

5 5 5

2.31 1.84 2.25

0.631 0.682 0.905

behaviours in the face of various occupational hazards, including OSH practices in four dimensions (as mentioned above). Overall, farmers used the practices related to all four dimensions at a low level and, accordingly, it can be preliminarily concluded that farmers have a great exposure potential to different occupational hazards. In addition, while the relationship between the three variables of attitude, knowledge, and self-efficacy with OSH practices has been examined separately in previous studies, this study provides empirical evidence for simultaneously investigating the effects of those three variables on OSH practices in the form of a comprehensive model. A great proportion of the farmers had unsafe behaviour with respect to PPE use. One possible reason for this result is that most farmers do not have enough access to PPE nor they have the necessary financial power to purchase the equipment. In addition, farmers often perceive PPE as a barrier that makes them feeling hot and inhibits their work (Damalas et al., 2006a; Kien, 2015). On the other hand, farmers had low perceptions of benefits by using PPE and, therefore, did not pay attention to PPE use. In this respect, the government should implement a project to provide basic PPE and help farmers notice the benefits by PPE use. Moreover, most farmers behaved potentially unsafe in the use of pesticides. One-fifth of the farmers had unsafe behaviour in the use of pesticides, probably due to their poor awareness about the harmful effects of unsafe use of pesticides, and most farmers were not familiar with principles of pesticide use. Thus, most farmers showed potentially unsafe behaviour in the use of pesticides. The importance of this issue is multiplied by the large amount of pesticides that are used in the area surveyed. Overall, a small number of the respondents showed safe behaviour as regards the use of pesticides, the use of machinery, and the use of PPE as well as applying ergonomic principles, showing that most farmers were prone to different occupational injuries and hazards. Attitude towards OSH practices had a positive and significant effect on farmers' use of OSH practices (supporting Hypothesis 1). This finding is consistent with the results of previous studies (Colémont and Van den Broucke, 2008; Raksanam et al., 2012; Adebola, 2014; Aluko et al., 2016). Attitudes are strong predictors of individuals' behaviour and the origin of the formation of any behaviour (Abdollahzadeh et al., 2015). Therefore, favorable attitudes of farmers towards OSH practices can provide the necessary conditions for improving occupational safety behaviour (Ajzen and Gilbert Cote, 2008). If individuals do not have the right attitude towards safety practices, all efforts to create an accident-

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

6

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

Table 3 Constructs, measurement items and reliability and validity tests for the dependent variables. Latent and observed variables Measurement model 1: Use of pesticides (Damalas and Eleftherohorinos, 2011; Adebola, 2014; Andrade-Rivas and Rother, 2015; Kien, 2015): CR = 0.851, AVE = 0.534 No eating, drinking or smoking while spraying pesticides (pesticide1) Using permitted pesticides with recommended dosage instructed on pesticide container labels (pesticide2) Spraying pesticides without strong winds and strong sunshine time (pesticide3) No reuse of empty pesticide containers or bottles for storing foodstuffs (pesticide4) Spraying only on the windless and less strong sunlight time (pesticide5) Avoiding the use of bare hands while mixing and spraying pesticides (pesticide6) Measurement model 2: Use of machinery (Colémont and Van den Broucke, 2008; Adebola, 2014; Andrade-Rivas and Rother, 2015): CR = 0.864, AVE = 0.616 Control and periodic service of machinery different parts, such as brakes, tires, etc. (machinery1) Regular and on time inspection of agricultural machinery (machinery2) Equipping tractors and other machinery with lights and proper lighting instruments (machinery3) Replacing the stiff fixed seat of tractor with seat which has soft and elastic material (machinery4) Measurement model 3: Use of PPE (Adebola, 2014; Andrade-Rivas and Rother, 2015; Damalas and Abdollahzadeh, 2016): CR = 0.889, AVE = 0.619 Dressing with gloves, hat, and warm clothing during cold (PPE1) Using PPE (e.g. gloves, boots, glasses, hat, and filter mask) while mixing and spraying pesticides (PPE2) Access to first-aid kit and fire extinguisher in the workplace (PPE3) Wearing cotton or linen clothes in summer (PPE4) Using protective equipment against sunlight, such as gloves, straw hat, sunglasses, sunscreen, etc. (PPE5) Measurement model 4: Applying ergonomic principles (Colémont and Van den Broucke, 2008; Adebola, 2014; Andrade-Rivas and Rother, 2015): CR = 0.816, AVE = 0.528 Avoiding monotonous repetitive motions during work (ergonomic1) No riding the tractor for a long and persistent time (ergonomic2) Avoid working with machinery that generate vibration (ergonomic3) Avoid lifting or carrying heavy loads individually (ergonomic4) Doing different exercises during work or rest (ergonomic5)

Standardized t-value loading

0.73

Fixed

0.63

9.81

0.78

12.61

0.73

11.91

Dropped



0.77

12.52

0.80

Fixed

0.87

16.37

0.80

14.86

0.65

11.52

0.66

Fixed

0.68

10.19

0.80

12.01

0.90 0.86

13.13 12.74

0.68

Fixed

0.76

10.49

0.67

9.54

0.79

11.08

Dropped



On a scale from 1 = very low to 5 = very high.

free workplace will be in vain (Sanaei Nasab et al., 2009). Individuals with a positive attitude are mentally more prepared to face occupational hazards and react more appropriately to them. Moreover, these individuals believe that safety and personal health are more important than work and high productivity (safety first, after work) and, therefore, they take more preventive measures in the workplace and more efforts to comply with safety principles, improving their own safety behaviour. Despite the importance of attitude as one of the main determinants of farmers' use of OSH practices, the descriptive results of this study indicated that the average score of the attitude towards OSH practices was relatively low and most farmers had a negative attitude towards OSH practices. One of the main reasons for this finding can be attributed to the low level of farmers' knowledge on OSH, because according to the rational model of health promotion, the attitude towards behaviour is influenced by knowledge and awareness about that behaviour (WHO, 2012).

Table 4 Constructs, measurement items and reliability and validity tests for the independent variables. Latent and observed variables

Standardized t-value loading

Measurement model 5: Attitude towards OSH (Aluko et al., 2016; Top et al., 2016): CR = 0.883, AVE = 0.561 Safety and health is more important than work and high production (safety first, after work) (attitude1). Prevention of injuries and occupational hazards is more important than their treatment (attitude2). Occupational hazards are considered as an inseparable part of agricultural work and avoiding them is impossible (attitude3). Use of personal protective equipment causes discomfort and reduces speed and accuracy while working (attitude4). Paying extra attention to occupational hazards is an unnecessary burden on farmers (attitude5). Training of farmers and provision of personal protective equipment is necessary for reducing exposure to occupational hazards (attitude6). Measurement model 6: Knowledge on OSH (Mostafa and Momen, 2014; Aluko et al., 2016): CR = 0.891, AVE = 0.623 Occupational hazards and their categories (e.g. physical, chemical, biological, ergonomic, and mechanical) (knowledge1). Occupational infections and their sources (knowledge2). Workplace health and wellness programs (knowledge3). Safety behaviours and protective measures (knowledge4). Personal protective equipment (knowledge5). General safety precautions in the workplace (knowledge6). Measurement model 7: Self-efficacy in safety (Brown et al., 2000; Pettinger, 2000; Lu et al., 2015): CR = 0.841, AVE = 0.519 Even if a safety procedure is long or complicated, I am able to follow it (self-efficacy1). I never follow safety procedures because they take too much time (self-efficacy2). I avoid trying to learn new safety procedures when they look difficult for me (self-efficacy3). I am confident in my ability to remove work place safety and health hazards (self-efficacy4). I am confident that I have the knowledge and skills to protect others and myself at work (self-efficacy5). In general, I can resolve most health and safety problems if I try hard enough (self-efficacy6).

0.67

Fixed

0.75

18.71

0.76

11.60

0.59

9.18

0.85

12.63

0.84

12.49

0.73

Fixed

0.88 0.85 0.66 0.80 Dropped

14.86 14.35 11.29 13.48 –

0.84

Fixed

0.76

14.02

Dropped



0.63

11.29

0.74

13.73

0.60

10.59

On a scale from 1 = very low to 5 = very high (or from 1 = strongly agree to 5 = strongly disagree).

The variable of knowledge on OSH practices had a positive and significant effect on farmers' use of OSH practices (supporting Hypothesis 2). This finding is consistent with the results of previous studies (Raksanam et al., 2013; Kumari and Reddy, 2013; Mohanty et al., 2013; Kien, 2015), but is not consistent with the results of Sanaei Nasab et al. (2009). Certainly, in any attempt to change behaviour, knowledge and information are considered as the main preconditions

Table 5 Discriminant validity matrix. Latent variable

1

2

3

4

5

6

7

1. Safe use of pesticides 2. Proper use of machinery 3. Use of PPE 4. Applying ergonomic principles 5. Attitude towards OSH 6. Knowledge on OSH 7. Self-efficacy in safety

0.73 0.64 0.41 0.55 0.66 0.41 0.58

0.78 0.46 0.50 0.55 0.32 0.46

0.79 0.26 0.54 0.39 0.39

0.73 0.38 0.15 0.45

0.81 0.39 0.51

0.79 0.19

0.72

Note: The figures corresponding to square root of AVE for each column latent variable is captured in bold along the diagonal. Other figures are the correlation between two latent variables.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx Table 6 Fit indices for the measurement models. Measurement model

Safe use of pesticides Proper use of machinery Use of PPE Applying ergonomic principles Attitude towards OSH Knowledge on OSH Safety self-efficacy Full measurement

Fit indices χ2/df

P-value GFI

CFI

IFI

RMSEA RMR

2.820 2.597 2.872 2.902

0.018 0.014 0.022 0.031

0.979 0.981 0.985 0.995

0.985 0.986 0.992 0.997

0.985 0.986 0.992 0.997

0.680 0.620 0.079 0.041

0.041 0.380 0.017 0.028

2.223 2.145 2.082 1.982

0.008 0.006 0.005 0.000

0.991 0.990 0.992 0.843

0.996 0.996 0.997 0.919

0.996 0.996 0.997 0.920

0.048 0.056 0.061 0.057

0.032 0.440 0.049 0.043

for success (Khan and Damalas, 2015). In this context, knowledge and attitude of workers about safety are key elements for developing a safe behaviour, which allows the design of safety plans in the workplace (Feyer and Williamson, 1997). Accordingly, as long as farmers are not aware of the occupational hazards and their harmful consequences, the principles of OSH, and the necessity of using them, they cannot properly comply with OSH practices because of perceived lack of necessity and lack of sufficient motivation (Suklim et al., 2013; Kien, 2015). If farmers have good knowledge about the kind of pesticide they use and the health effects of the pesticide, they will probably prevent risky behaviour. In fact, when a farmer becomes aware of the negative effects of using pesticides on health, he formulates a special perception of the substance and then his cognitive process transforms that perception into mind-set so that, finally, he shows a particular behaviour (for example, the use of PPE during spraying) (Huang, 1993). On the other hand, improving the level of farmers' knowledge leads to a more favorable attitude towards OSH (WHO, 2012), which in turn increases their occupational safety behaviour.

7

The average score of the variable of knowledge on OSH practices was low and the importance of this issue is multiplied by the low level of farmers' literacy and the lack of participation in training courses on OSH. In general, such conditions do not favor appropriate occupational safety behaviour among farmers. However, the positive and significant effect of the variable of knowledge on farmers' use of OSH practices, coupled with farmers' low level of knowledge, emphasize the necessity of improving the use of OSH practices among farmers through largescale training and community awareness programs, particularly with the help of common mass media, such as radio and television. Training has been reported as a determinant of environmentally sound behaviour in pest control (Khan and Damalas, 2015). In addition, attention should be focused on extension mechanisms, such as holding meetings and educational-extension courses related to OSH, with a strong encouragement of farmers to participate in these courses. Verbal recommendations, extension brochures and bulletins, fact sheets, posters, and videos are other methods that can provide relevant information to farmers and increase their knowledge on OSH practices. Consistent with previous research (DeJoy, 1996; Pettinger, 2000; Brown et al., 2000; Wagner et al., 2013), the variable of self-efficacy in safety had a positive and significant effect on farmers' use of OSH practices (supporting Hypothesis 3). Evidently, safety performance requires both skills and belief in ability to perform those skills. Once an individual's self-efficacy belief forms a single behaviour or a set of specific behaviours, these beliefs then guide reactions, efforts, behaviours, and desires (Bandura, 1997). Moreover, one's beliefs and expectations have a direct effect on his behaviour, motivation, and performance in the workplace (Betz et al., 1996). In fact, when self-efficacy increases, farmers gain more motivation and self-confidence (DeJoy, 1996; Brown et al., 2000) and show better safety performance when working with hazardous resources, such as pesticides, machinery, and so on. In addition, self-efficacy in safety may increase the sense of farmers'

Fig. 4. Final structural model with standardized estimates.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

8

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx

Table 7 Estimates of the direct structural model. Hypothesized relationship Attitude → OSH practices (H1) Knowledge → OSH practices (H2) Self-efficacy → OSH practices (H3)

Unstandardized estimates

S.E.

Standardized estimates

Critical ratio

Sig.

Hypothesis test

0.562 0.186 0.231

0.073 0.044 0.032

0.486 0.196 0.391

7.686 4.203 7.293

0.000 0.000 0.000

Supported Supported Supported

responsibility for their own safety and health in the work. For example, as self-efficacy of an individual increases, the more likely is that this individual will use PPE (Wagner et al., 2013). On the other hand, interaction with the workplace is largely influenced by individuals' judgments about their abilities (i.e., self-efficacy), so that if people believe in their abilities, they perform better in the face of occupational hazards (Gist and Mitchell, 1992; Brown et al., 2000). In total, increasing the level of self-efficacy in safety, the more likely is that farmers will use OSH practices. Despite the importance of the variable of self-efficacy in safety, the descriptive results indicated that similar to the other two variables, i.e., attitude and knowledge, the average score of this variable was also relatively low, which is another reason for the low level of the farmers' occupational safety behaviour. Accordingly, provision of hands-on training or skill-building exercises in safety procedures to help farmers gain experience in performing related behaviours (at the individual level) as well as multifaceted safety interventions, such as safety campaigns, legislation or machines guarding (at the macro level) are possible interventions for increasing self-efficacy in safety and, consequently, farmers' occupational safety behaviour. This study provides useful information on farmers' behaviour regarding OSH, but has certain limitations that should be taken into account and addressed in future studies. First, a limitation of this study is the inclusion of data only from one crop in one province. Therefore, the study cannot be generalized to a national level. Thus, future studies should involve more participants of different crops in different provinces. Second, the study focused on a reduced set of psychological factors (i.e., attitude, knowledge, and self-efficacy) as the main factors affecting farmers' use of OSH practice according to the literature. Many more factors, such as social and personal/moral norms, beliefs and values, perceptions of risks, costs and benefits, behavioural control, and self-identity also worth investigation. On the other hand, variables related to the socio-economic status of farmers and farm characteristics were not included in the research model, hence, future work can examine the effects of contextual and social characteristics of farmers on safety behaviour. Furthermore, this study relied mainly on self-reports of farmers' safety behaviour and these behaviours were only partly validated against actual use. This might be considered as an inherent limitation of this study. Self-report studies may suffer limitations of this kind as people frequently want to report socially desirable behaviours (Damalas and Abdollahzadeh, 2016). However, the interviews in this study were conducted in a friendly way and there was good cooperation with the farmers without any refusals. Thus, the possibility of misreporting was assumed negligible.

5. Conclusion This study provided a framework for understanding farmers' use of OSH practices in Iran, with an emphasis on farmers' characteristics. Overall, farmers used OSH practices related to all four dimensions at a low level and, accordingly, it can be preliminarily concluded that farmers have a great exposure potential to different occupational hazards. Attitude towards OSH practices, knowledge on OSH practices, and self-efficacy in safety had a direct positive effect on farmers' use of OSH practices, explaining 73% of the variance in farmers' safety behaviour. Therefore, improving these variables can promote OSH among farmers. Because few empirical studies have been conducted on OSH practices in Iran, this study helps strengthen the existing literature in

this field in Iran and fills a gap between science and policy making in OSH in agriculture.

Acknowledgements We would like to acknowledge the support of University of Zanjan for providing the funds for this research through project No. 9611. References Abdollahzadeh, G., Sharifzadeh, M.S., Damalas, C.A., 2015. Perceptions of the beneficial and harmful effects of pesticides among Iranian rice farmers influence the adoption of biological control. Crop. Prot. 75, 124–131. Adebola, J., 2014. Knowledge, attitude and compliance with occupational health and safety practices among pipeline products and marketing company (PPMC) staff in Lagos. Int. J. Med. Sci. 2, 158–173. Ajzen, I., 2002. Perceived behavioural control, self-efficacy, locus of control, and the theory of planned behaviour. J. Appl. Soc. Psychol. 32, 665–683. Ajzen, I., Gilbert Cote, N., 2008. Attitudes and the prediction of behaviour. In: Crano, W., Prislin, R. (Eds.), Attitudes and Attitude Change. Psychology Press, New York, USA, pp. 289–311. Aluko, A., Adebayo, A., Adebisi, T., Ewegbemi, M., Abidoye, A., Popoola, B., 2016. Knowledge, attitudes and perceptions of occupational hazards and safety practices in Nigerian healthcare workers. BMC Res. Notes 9, 71–85. Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a review and recommended two-step approach. Psychol. Bull. 103, 411–423. Andrade-Rivas, F., Rother, H., 2015. Chemical exposure reduction: factors impacting on South African herbicide sprayers' personal protective equipment compliance and high risk work practices. Environ. Res. 142, 34–45. Armitage, C., Conner, M., 2001. Efficacy of the theory of planned behaviour: a metaanalytic review. Brit. J. Soc. Psychol. 40, 471–499. Atreya, K., 2007. Pesticide use knowledge and practices: a gender differences in Nepal. Environ. Res. 104, 305–311. Bandura, A., 1997. Self-Efficacy: The Exercise of Control. Worth Publisher, New York, USA. Bartlett, J.E., Kotrlik, J.W., Higgins, C.C., 2001. Organizational research: determining appropriation sample size in survey research. Inf. Technol. Learn. Perform. J. 19, 43–50. Betz, N.E., Klein, K.L., Taylor, K.M., 1996. Evaluation of a short form of the career-decisionmaking self-efficacy scale. J. Career Assess. 4, 47–57. Brosseau, L., Li, S., 2005. Small business owners' health and safety intentions: a crosssectional survey. Environ. Health 4, 23–32. Brown, K., Willis, G., Prussia, G., 2000. Predicting safe employee behaviour in the steel industry: development and test of a sociotechnical model. J. Oper. Manag. 18, 445–465. Byrne, B., 2010. Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming. 2nd ed. Routledge Publisher, New York, USA. Cadogan, J.W., Cui, C.C., Morgan, R.E., Story, V.M., 2006. Factors facilitating and impeding the development of export market-oriented behaviour: a study of Hong Kong manufacturing exporters. Ind. Mark. Manag. 35, 634–647. Colémont, S., Van den Broucke, S., 2008. Measuring determinants of occupational health related behaviour in Flemish farmers: an application of the theory of planned behaviour. J. Saf. Res. 39, 55–64. Damalas, C.A., Abdollahzadeh, G., 2016. Farmers' use of personal protective equipment during handling of plant protection products: determinants of implementation. Sci. Total Environ. 571, 730–736. Damalas, C.A., Eleftherohorinos, I.G., 2011. Pesticide exposure, safety issues, and risk assessment indicators. Int. J. Environ. Res. Public Health 8, 1402–1419. Damalas, C.A., Koutroubas, S.D., 2016. Farmers' exposure to pesticides: toxicity types and ways of prevention. Toxics 4, 1–10. Damalas, C.A., Georgiou, E.B., Theodorou, M.G., 2006a. Pesticide use and safety practices among Greek tobacco farmers: a survey. Int. J. Environ. Health Res. 16, 339–348. Damalas, C.A., Theodorou, M.G., Georgiou, E.B., 2006b. Attitudes towards pesticide labelling among Greek tobacco farmers. Int. J. Pest Manag. 52, 269–274. Das, S., Bagchi, A., Gangopadhay, S., Ghosh, S., 2016. Work related health disorders of female agricultural workers correlated with sawing and ploughing in Bengal: an ergonomic profile. Int. J. Current Res. Acad. Rev. 4, 155–166. Davenport, T., Prusak, L., 1998. Working Knowledge: How Organizations Manage What they Know. Harvard Business School Press, Boston, USA. DeJoy, D.M., 1996. Theoretical models of health behaviour and workplace self-protection. J. Saf. Res. 27, 61–72. Feyer, A.M., Williamson, A., 1997. The involvement of human behaviour in occupational accidents: errors in context. Saf. Sci. 25, 55–65. Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res. 18, 39–50.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201

R. Rezaei et al. / Science of the Total Environment xxx (2017) xxx–xxx Gist, M.E., Mitchell, T.R., 1992. Self-efficacy: a theoretical analysis of its determinants and malleability. Acad. Manag. Rev. 17, 183–211. Glanz, K., Rimer, B., Viswana, K., 2008. Health Behaviour and Health Education: Theory, Research, and Practice. John Wiley & Sons, San Francisco, USA. Gupta, C.D., Gupta, V.K., Pallavi, N., Patel, J.R., 2012. Gender differences in knowledge, attitude and practices regarding the pesticide use among farm workers: a questionnaire based study. Res. J. Pharm. Biol. Chem. 3, 632–639. Hair, J., Black, C., Babin, J., Anderson, E., 2010. Multivariate Data Analysis. Prentice Hall Publisher, New Jersey, USA. Huang, C.L., 1993. Simultaneous-equation model for estimating consumer risk perceptions, attitudes and willingness-to-pay for residue-free produce. J. Cons. Affairs 27, 377–396. ILO, 2011. Safety and health in agriculture. Research Report. Labour Protection Department, International Labour Organization, Switzerland. Kaewboonchoo, O., Kongtip, P., Woskie, S., 2015. Occupational health and safety for agricultural workers in Thailand: gaps and recommendations, with a focus on pesticide use. Environ. Occup. Health Policy 25, 102–120. Karami, G.H., Bijani, M., Salamat, E., 2016. Investigating attitude toward safety issues among agricultural Jihad professionals with an emphasis on safety training. J. Health Saf. Work 6, 43–58. Khan, M., Damalas, C.A., 2015. Factors preventing the adoption of alternatives to chemical pest control among Pakistani cotton farmers. Int. J. Pest Manag. 61, 9–16. Kien, A., 2015. Factors Influencing Safety Pesticide Use Behaviour Among Farmers in Thainguyen Province, Vietnam. MSc Thesis. Faculty of Nursing, Burapha University, Thailand. Kumari, P.L., Reddy, K.G., 2013. Knowledge and practices of safety use of pesticides among farm workers. J. Agric. Vet. Sci. 6, 1–8. Lu, L., Shi, L., Han, L., Ling, L., 2015. Individual and organizational factors associated with the use of personal protective equipment by Chinese migrant workers exposed to organic solvents. Saf. Sci. 76, 168–174. Mohanty, M., Behera, B., Jena, S., Srikanth, S., Mogane, C., Samal, S., Behera, A., 2013. Knowledge attitude and practice of pesticide use among agricultural workers in Puducherry, South India. J. For. Legal Med. 20, 1028–1031. Mostafa, N., Momen, M., 2014. Occupational health and safety training: knowledge, attitude and practice technical education students. Egyptian J. Occup. Med. 38, 153–165. Padilla, A., 2013. Occupational Health and Safety of Agricultural Workers: ILO Conventions and Gaps. Research Report. Pesticide Action Network Asia and the Pacific (PAN AP), Malaysia.

9

Peres, F., Rozemberg, B., de Lucca, S., 2005. Risk perception related to work in a rural community of Rio de Janeiro State: pesticides, health and environment. Cad. Saude Publica 21, 1836–1844. Pettinger, C., 2000. Improving Occupational Safety and Health Interventions - A Comparison of Safety Self-Efficacy & Safety Stages of Change. PhD Dissertation. Department of Psychology, Virginia Polytechnic Institute and State University, Virginia, USA. Rafiei, M., Ezzatian, R., Farshad, A., Sokooti, M., Tabibi, R., 2015. Occupational health services integrated in primary health care in Iran. Ann. Glob. Health 81, 561–567. Raksanam, B., Surasak, T., Siriwong, W., Robson, M.G., 2012. Factors associated with pesticide risk behaviours among rice farmers in rural community. Thailand J. Environ. Earth Sci. 2, 32–39. Raksanam, B., Suklim, N., Songthap, A., 2013. Model development to reduce pesticide risk behaviours among rubber farmers in Khogyang Community, Trang, Thailand. Eur. J. Res. Educ. 2, 101–108. Sanaei Nasab, H., Ghofranipour, F., Kazemnejad, A., Khavanin, A., Tavakoli, R., 2009. Evaluation of knowledge, attitude and behaviour of workers towards occupational health and safety. Iran. J. Public Health 38, 125–129. Schumacker, R.E., Lomax, R.G., 2004. A Beginner's Guide to Structural Equation Modeling. 2nd ed. Lawrence Erlbaum Associates Mahwah, New Jersey, USA. Singh, P.K., Hiremath, B.N., 2010. Sustainable livelihood security index in a developing country: a tool for development planning. Ecol. Indic. 10, 442–451. Suklim, N., Raksanam, B., Songthap, A., 2013. Risk behaviours related agrochemical use among rubber farmers in Southern of Thailand. Eur. J. Res. Educ. 2, 109–115. Top, Y., Adanur, H., Öz, M., 2016. Comparison of practices related to occupational health and safety in microscale wood-product enterprises. Saf. Sci. 8, 374–381. Ullman, J.B., 2001. Structural equation modeling. In: Tabachnick, B.G., Fidell, L.S. (Eds.), Using Multivariate Statistics, 4th ed. Allyn & Bacon, Boston, USA. Wagner, H., Kim, A., Gordon, L., 2013. Relationship between personal protective equipment, self-efficacy, and job satisfaction of women in the building trades. J. Constr. Eng. Manag. 139, 1–8. WHO, 2012. Health Education: Theoretical Concepts, Effective Strategies and Core Competencies: A Foundation Document to Guide Capacity Development of Health Educators. Research Report. World Health Organization, Cairo, Egypt. Wood, R.E., Bandura, A., 1989. Social and cognitive theory of organizational management. Acad. Manag. Rev. 14, 361–384. Zanjan Agri-Jihad Organization, 2015. Statistical yearbook of cropping products. Deputy of Planning and Economic Affairs, Office of Statistics and Information Technology. Zanjan Agri-Jihad Organization, Iran.

Please cite this article as: Rezaei, R., et al., Understanding farmers' safety behaviour towards pesticide exposure and other occupational risks: The case of Zanjan, Iran, Sci Total Environ (2017), https://doi.org/10.1016/j.scitotenv.2017.10.201