Investigating the prevalence of unsafe behaviors and the factors affecting it 1 using a cross-sectional study in a steel company in Iran

3 Background: Unsafe behaviors (UBs) are the most important cause of accidents (98 %). 4 Therefore, identifying the cause of UBs is also significant. The aim of this study was to investigate 5 the prevalence of UBs among the workers of a steel company in Iran and also to identify the factors 6 affecting these behaviors. 7 Method: This cross-sectional study was conducted on 270 people from the employees of a steel 8 company in Iran in 2016. The present study consists of two parts. 1) investigate the prevalence of 9 UBs of workers using safety behavior sampling method, 2) identify the factors affecting UBs using 10 a questionnaire (demographic sections, general health questionnaire, work-family conflict 11 questionnaire, and UK's HSE job stress questionnaire). The data from the questionnaires were 12 analyzed by SPSS 16. 13 Results: The frequency of observed UBs showed that out of 1310 observed behaviors, 531 cases 14 (40.5%) were unsafe. Also, UB among the sample group was significantly higher than the control 15 group (P < .05). The correlation between general health, work-family conflict, job stress, and UBs 16 showed that general health had a significant relationship with UBs (P < .05). 17 Conclusions: The relationship between support scale in job stress and accident showed that 18 organizational factors such as organizational climate, safety culture, management and cooperation 19 between people are among the factors affecting the accident and with optimal management of these 20 factors can play a significant role in future accidents. 21

The workforces have an important role in the service, survival, and success of organizations. Also, 2 the most people in the world (more than 58%) spend a third of their lives in the workplace (1). The 3 human resources in any organization are the most valuable resources of the organization, which 4 attracts a lot of attention (2). Today, occupational accidents are recognized and estimated as one 5 of the most important causes of disability and leading to death in developed and developing 6 countries and it is estimated that more than 300,000 people die each year, and many cases of 7 physical disability (3). Job accidents, no matter how severe they may be, cause great economic 8 and social problems in society. According to the World Health Organization (WHO), 350,000 9 workers lose their life each year, and job injuries accounts for 0.9% of the world's disability 10 adjusted life year (DALYS) (4). In the European Union, the number of days lost due to an annual 11 work accident is reported to be 150 million days (5).

12
Due to the great importance of accident control, various researches and models have been 13 presented in the field of cause detection of accidents, most of which are two factors of unsafe 14 behaviors (UBs) and unsafe conditions (UCs) as the most important causes of accidents. Studies 15 in industrialized countries indicate that about 90% of occupational accidents are caused by UBs 16 (6). In " Industrial accident prevention", Henrich states that more than 85% of occupational 17 accidents are due to UBs (7). These behaviors are derived from the inner and social characteristics 18 of individuals (8). Research on UBs has examined a variety of factors, including individual 19 characteristics, job stress, fatigue, and organizational factors such as safety culture. In addition to 20 these characteristics, mental health of individuals can also be considered (9). The combination of 21 UBs, UCs, and the inherent risk of the activity may lead to occupational accidents, and these risks  Another factor that is one of the social factors that has had a great impact on the mental health of In order to control occupational accidents, one must first look for the roots and causes of these 1 accidents. As mentioned above, the most important cause of accidents is UBs of workers.
2 Therefore, identifying the cause of UBs is also significant. Past studies have cited factors such as 3 stress, mental health, safety culture, and more. The aim of this study was to investigate the 4 prevalence of UBs among the workers of a steel company in Iran and also to identify the factors 5 affecting these behaviors.

7
The present cross-sectional study was conducted on 270 people from the employees of a steel 8 company in Iran in 2016. The present study consists of two parts. The first part was to investigate 9 the prevalence of UBs of workers using safety behavior sampling (SBS) method and the second 10 part was to identify the factors affecting UBs using a questionnaire. The data collection tools in 11 this study to identify the factors affecting UBs include a 5-part questionnaire consisting of 12 demographic sections (including age, work experience, level of education, marital status, family 13 head, shift working), general health questionnaire, work-family conflict questionnaire, and UK's 14 HSE job stress questionnaire. In this study, first the aims of the study were explained to the 15 participants and then they answered the questionnaires, which were anonymous. Written consent 16 was also received from the participants.  The second group was randomly selected from the same section that had no accidents during this 23 period.  Where P is the ratio of unsafe behavior to total behavior in the pilot study (equation 2), S accuracy 3 (5%) and Z reliability coefficient 95% according to the normal distribution table.  The data from the questionnaires were analyzed by SPSS 16. In analytical section, Mann-Whitney, 15 Chi-square test, Spearman correlation and linear regression were used.  Table   21 1).

22
The frequency of observed UBs in the two studied groups showed that out of 1310 observed 23 behaviors, 531 cases (40.5%) were unsafe. Also, UBs among the sample group was significantly 24 higher than the control group ( Figure 1). Also, checklist categories, the relative frequency of each 25 UBs showed that the most observed behavior (38%) is related to non-use or inappropriate use of 26 PPE, followed by awkward posture with 20.2% (Figure 2).
27 Table 2 shows the mean scores obtained from the questionnaires between the two sample and 28 control groups. As can be seen in the table, only in the stress and subscales section of the support 29 (management and colleagues), there was a significant difference between the two study groups (P 30 < .05).

31
The correlation between general health, work-family conflict, job stress, and UBs between the two 32 studied groups showed that the variables general health, work-family conflict, and job stress were 33 significantly corelated (P < .001). Also, general health had a significant relationship with unsafe 34 behavior (P < .05) ( Table 3).

1
The aim of this study was to investigate the prevalence of unsafe behaviors and the factors affecting 2 it in a steel company. The results of the study showed a significant relationship between general 3 health, job stress (support of managers) and unsafe behavior. Also, the results of SBS showed that 4 the most UBs was related to non-use or inappropriate use of PPE. In addition, the prevalence of 5 unsafe behavior observed for all observations was 39.81% and there was a significant difference 6 between the two studied groups.

7
According to various studies that have investigated UBs in various industries, the percentage of 8 unsafe behavior observed in the present study is the average of other studies, so that Azadeh et al.

9
up to 59% have also been shown (21). The study of Negahdari et al. in the electricity industry also 10 showed that 45% of all observed behaviors are unsafe and the most UBs is due to non-use of PPE.

11
They also found that there was a significant relationship between the two groups of accidents and absenteeism and loss of workforce (24).

28
The results of the present study showed that general health had a significant relationship with 29 occupational accidents in the industry in general. According to the results of general health 30 research, people who had an accident were in a more inappropriate situation than the other group.

31
In addition, the results showed that a higher mean score was observed among the sample group, Their results showed that general health with a mean score 3 of 29.4 in the accident group has a higher score than the accident-free group, and this difference is 4 significant. They also found that the depression score was the lowest and the social dysfunction 5 was the highest, but there was no significant difference between the two groups of had accident shows that reducing organizational support increases stress as well as increases the incidence of 24 accidents. In addition, they found that job stress was not significantly associated with accidents

34
The results of the present study showed that UBs and general health as well as job stress and and with optimal management of these factors can play a significant role in future accidents.
Correlation analysis between variables showed that social stressors such as work-family conflict 1 can also be a factor in reducing general health. Therefore, in order to reduce the incidence of 2 musculoskeletal disorders and accidents and also to increase the general health of employees, in 3 addition to considering job-related stress, industry management should also pay attention to the 4 social factor of work-family conflict and manage employees' stress with optimal management. 5 Implications 6 Unsafe behavior is one of the most important safety issues in industries. Identifying the factors 7 affecting unsafe behavior and controlling these behaviors can be effective in reducing accidents.

8
Organizations should record the prevalence of unsafe employee behavior. To do this, the method 9 presented in the present study can be applied. This study clearly has some limitations. The most important one lies in the fact that due to the 12 nature of the studied company; women were not included in the statistical population of the present 13 study. Therefore, work-family conflict, which is more common in women, was not examined 14 comprehensively, and in future studies, it is suggested that both genders be examined in order to 15 provide a more complete conclusion.