Factors associated with poor sleep quality among construction workers in Arba Minch town, Ethiopia: A cross‐sectional study

Abstract Background and Aims Sleep is critical for human physical and cognitive health. Even though poor sleep quality had a major negative impact on workers' health and safety, there is a scarcity of study that attempts to demonstrate its magnitude and causative factors in labor‐intensive environments. Methods A cross‐sectional study design was employed. A total of 423 individuals were selected using a simple random sampling technique, starting from April 1, 2020 to May 1, 2020. Interviews were administered using the validated and pretested Pittsburgh Sleep Quality Index (PSQI) tool. EpiData version 4.4.3.1 was used for data entry and SPSS version 25 for analysis. To associate factors with poor sleep quality binary logistic regression model was fitted at 95% confidence interval (CI). A p < 0.25 and p < 0.05 were used to declare association in bivariable and multivariable analysis, respectively. Results A total of 415 building construction workers take part in this study, with a response rate of 98.1%. A PSQI global score showed 66.3% (95% CI: [0.63, 0.71]) of construction workers suffering from poor sleep quality. Working hours, work experience, previous injury status, managerial support on occupational safety and health, cigarette smoking, and job satisfaction were the factors that were associated with poor sleep quality significantly. Conclusions Two‐thirds of construction workers suffer from poor sleep quality. Risk factors include shorter working years, longer working hours, prior injury, lack of managerial support, cigarette smoking, and low job satisfaction. Efficient health education and promotion through taking the identified associated factors into account could help reduce poor sleep quality.

Sleep is a basic behavior found throughout the animal kingdom, and it is essential for maintaining physical health and cognitive function in humans. 1 The ideal amount of sleep for adults to function well, prevent sleep debt, and avoid daytime drowsiness is around 7-8 h and at least 10 h for school-aged children and adolescents. 2One of the most essential things our body need is good sleep quality. 3[6] According to large-scale global research on sleep disorders conducted in eight Asian and African countries, 17% of the population in these developing countries suffers from sleep problems. 7In Tanzania, Ghana, and Kenya, overall rates of poor sleep quality varied from 8.3% to 12.7%, making it more common than several chronic disorders and infectious diseases. 7In Ethiopia, 53% of the population was estimated to have poor sleep quality. 8or sleep quality is a serious issue that construction workers face regularly.According to studies conducted in Korea and China, around 63% and 30%, respectively, of construction workers struggle with poor sleep quality. 9,10[13] Very sleepy workers are 70% more likely to be engaged in workplace accidents than their nonsleep-deprived coworkers. 14An increased risk of job injuries can also result from long workdays and inadequate sleep.Besides workers' health, poor sleep quality has far-reaching economic consequences on the nation's economy.Globally, lack of sleep and erratic sleeping patterns caused billions of dollar loss. 15gardless of the fact that poor sleep quality had a major negative impact on workers' health and safety, few research have attempted to demonstrate its magnitude and causative factors in labor-intensive environments.This lack of information makes it difficult to establish specific interventions and care programs to safeguard construction workers' overall health and safety.Therefore, this study aimed to close the gap of information regarding poor sleep quality prevalence among construction workers and the factors associated with it.

| Study design and setting
An institution-based cross-sectional study design was employed, and data were collected starting from April 1, 2020 to May 1, 2020.This study was conducted in Arba Minch town, the capital of Gamo zone, Southern Nation's Nationalities and Peoples Regional State of Ethiopia.It has a total population of 113,297 people.It lies 250 and 435 km from the regional city of Hawasa and the capital city of Ethiopia, Addis Ababa, respectively.According to information obtained from Gamo zone urban development and construction office, at the time of data collection, there were five buildings which have been under-construction by construction firms licensed with 1-8 grades.

| Population and eligibility criteria
All construction workers in Arba Minch town were the source population.
Selected construction workers who were involved in the construction process for the past 12 months were the study population.

| Sample size determination
The sample size was calculated using Epi Info version 7 and the single population proportion formula.A 95% confidence level, 5% marginal error, and 50% poor sleep quality proportion were assumed, as no previous studies had been conducted in this setting.A nonresponse rate of 10% was also considered, resulting in a sample size of 423.

| Sampling procedure
Of the 488 eligible construction workers at five construction sites, which were under construction by Grade 1-8 licensed construction firms in the study area, 423 were selected using a simple random sampling technique with a lottery method.The sample size was proportionally allocated to the five construction sites based on the number of workers at each site.

| Sleep quality
Pittsburgh Sleep Quality Index (PSQI) was used to measure sleep quality.This tool has seven components, and each of the seven components is awarded a score ranging from 0 to 3 (0: being no problem).The global PSQI scores vary from 0 to 21.Finally, those with a global PSQI score of greater than 5 were deemed to have poor sleep quality. 16he tool for data collection was adopted from multiple works of literature, [16][17][18] and the questionnaire was administered by the interviewer.To collect the status of the outcome variable a validated tool (PSQI) was utilized. 18Three BSc nurses carried out the data collection procedure, and one BSc nurse professional supervised the procedure.The collected data were reported to the supervisor every 2 days.The completed questionnaires were sent to the principal investigator after the supervisor rechecked the filled questionnaires.

| Data quality control
Data collectors and supervisors were trained on data collection processes, methodologies, and ethical considerations.The questionnaire was pretested on 5% of the study's sample size among construction site employees in Mirab-Abaya, Southern Ethiopia.The pretest ascertained the consistency of the research tool.All the questionnaires were verified daily to see whether they were accurately filled or not.

| Data management and statistical analysis
Before analyzing the raw data with SPSS statistical software version 25.0, EpiData version 4.4.3.1 was used to code and enter the data into the computer.Then, it was cleaned and checked by running frequency.The data were described using frequency, proportion, mean, and tables.
Multicollinearity (variance inflation factor) and model fitness (Hosmer and Lemeshow) were checked before applying and interpreting the binary logistic regression model.In bivariable analysis, independent variables with a p < 0.25 were included in multivariable analysis, and variables were recognized as statistically significant if their p < 0.05.

| Ethical consideration
The Institutional Review Board of Arba Minch University, College of Medicine and Health Sciences, provided ethical clearance.Every research participant was given written informed consent after the study collector provide them a brief description of the importance of the study (Supporting Information S1: Appendix 1).For individuals under the age of 18, written informed consent was obtained from their parents.The participants' confidentiality was maintained throughout the whole investigation.

| Sociodemographic characteristics
From 423 construction workers, 415 responded with complete data making the response rate 98.1%.The majority (65.8%) of the respondents were males.The participants' mean (SD) age was 25.80 (±6.07) years, with the majority (47.7%) falling between the ages of 25 and 49.Most of the workers (40.2%) had only a primary school education (Table 1).

| Work-related characteristics
Of the total respondents, 59.3% of the workers were temporarily employed.Three-quarters of those interviewed had ≤2 years of work experience.Fifty-eight percent of the participants had poor managerial support on OSH.Sixty percent of the study participants had worked for more than 8 h/day (  3).

| Associated factors
In bivariable analysis, poor sleep quality was associated (p < 0. with their jobs (Table 5).

| DISCUSSION
This is novel research to evaluate sleep quality and its related components among construction workers in Ethiopia using a validated PSQI tool.In this study, no participant reported using sleep medication in the previous month, and only one-third (33.8%) reported having very bad subjective sleep quality.These findings are similar to other studies conducted in Northern Ethiopia. 19This could be due to cultural or behavioral factors that people did not seek treatment for this kind of illness (sleep disorder) and did not want to take medications for sleep-related illnesses.
The PSQI global score revealed 66.3% of study participants struggle with poor sleep quality.This result lines up with research conducted in Jimma town among adults (65.4%) 20 and among pregnant women in Northern Ethiopia (68.4%). 19On the other hand, it is higher compared to studies conducted among chronic illness patients in South Wollo, Ethiopia (36%), 21 university students in Ethiopia (55.8%), 22 and Myanmar's migrant workers in Malaysia (62.5%). 23Moreover, this finding is higher than the study conducted among construction workers in Southern India (33.9%). 24This disparity might be explained by differences in research setting and socioeconomic level of the groups.For instant, in the study conducted in South Wollo, the study setting was a hospital and the study population was chronic patients.The other reason for the discrepancy could be due to cultural and lifestyle differences among the populations.Evidence suggests that lifestyle preferences have a significant association with sleep quality, and this is proven in the Southern India study. 24 the contrary, this study finding is lower than studies conducted among prisoners in Mettu town, Ethiopia (77.1%), 25 and among nurses in Northwest Ethiopia (75.5%). 26The found disparities between studies might be due to differences in living conditions and exposure to occupation hazards of the study population.Literature indicates that crowded living conditions could result in poor sleep quality than less crowded housing. 27The higher percentage of poor sleep quality among prisoners in Mettu town and nurses could be attributed to the above-mentioned condition.
This study found that increasing work experience was associated with reduced odds of poor sleep quality.A construction worker who had more than 2 years of work experience was 42% less likely to encounter poor sleep quality than a worker who had ≤2 years of work experience.A similar association was reported in a study done in China. 28Longer work experience would make workers adapt to their work environment, resulting in low work-related anxiety and fear, this would directly or indirectly improve their sleep quality. 29e current study also showed work hours as a significant factor in poor sleep quality.Workers with greater than 8 h of work hours/ day were three times more likely to experience poor sleep quality compared to workers who work for ≤8 h/day.This result aligns with the research carried out in America among Latino farmworkers, 30 and Myanmar migrant workers in Malaysia. 23Working long hours was linked to reduced sleep hours, trouble with falling asleep, and nonrestorative sleep.This phenomenon was corroborated and established by longitudinal research conducted on British civil servants. 31Furthermore, a higher duration of work might lead to accumulated fatigue that contributed to poor sleep quality. 32is study also indicated that prior history of work-related injury at the construction site was another risk factor for poor sleep quality.
Similar results were obtained from studies conducted in Shiraz, Iran 33 and Basel, Switzerland. 13Physical and psychological traumas caused by prior work-related injuries might affect sleep quality in negative ways.Additionally, workers who were injured might be on subsequent medications to alleviate pain, which might disturb their sleep patterns in consequence.
Construction workers who had poor managerial support on OSH were two times experiences poor sleep quality than workers who had support from their managers.The above result was similar to the study done in the United States. 34The reason might be that those who had less managerial support on OSH were linked to higher levels of anxiety and stress at work, and this leads them to poor sleep quality. 29 addition, this study showed that cigarette smokers had a high risk of poor sleep quality than nonsmokers.This was also strengthened by another study in Ethiopia. 35A similar result was also obtained from Jordan 36 and China. 37The presence of nicotine substances in cigarettes disrupts sleep by raising heart rate and increasing alertness. 38Furthermore, a study done in China indicated cigarette smoking might impair cognition and this directly impacts sleep quality. 37nally, this survey revealed that poor job satisfaction might increase the chance of experiencing poor sleep quality among workers.This finding is consistent with studies done in Iran 39 and Turkey. 40The possible reason for this might be due to low job satisfaction among employees had a detrimental influence on their daily activities, resulting in worse job performance and poor sleep quality.

| LIMITATIONS OF THE STUDY
This study might be prone to social desirability and recall bias.
Moreover, it could be vulnerable to healthy worker bias due to the fact that workers with injury might not be at the construction sites during the study period.Furthermore, a causal relation could not be sure for some variables due to the nature of our study design, crosssectional.

| CONCLUSION AND RECCOMENDATION
This study revealed that two-thirds of construction workers had poor sleep quality.Shorter working years (≤2 years), longer working hours/ day (>8 h), prior injury status, lack of managerial support on OSH, cigarette smoking, and lower job satisfaction were the risk factors for poor sleep quality among construction workers.Therefore, creating efficient health education and promotion by taking the identified associated factors into account might assist in reducing the burden of poor sleep quality among construction workers.In addition, further studies are recommended to identify the predicting factors of poor sleep quality among construction workers using advanced study methodologies.Moreover, factors such as living conditions, psychological and other health-related factors need to be assessed.

Table 2 )
.T A B L E 1 Sociodemographic characteristics of building construction workers in Arba Minch town, Southern Ethiopia, 2020.
Work-related characteristics of building construction workers of Arba Minch town, Southern Ethiopia, 2020.Behavioral characteristics of building construction workers of Arba Minch town, Southern Ethiopia, 2020.
work hours/day, previous injury status, managerial support on OSH, cigarette smoking, and job satisfaction were significantly associated with poor sleep quality.Construction workers who had work experience of more than 2 years were 42% (adjusted odds ratio [AOR] = 0.58, 95% confidence interval [CI]: [0.34, 0.98], p = 0.042) less likely to experience poor T A B L E 2 a Driver/operator, site engineer, painter, and welder/electrician.T A B L E 3 T A B L E 4 Sleep quality components scores among building construction workers of Arba Minch town, Southern Ethiopia, 2020.
T A B L E 5 Factors significantly associated with poor sleep quality among building construction workers in Southern Ethiopia, 2020.