Sleep characteristics of Iranian people and their effects on daytime functioning: a population-based study

Sleep characteristics vary between populations. Detrimental sleep habits have cognitive consequences leading to daytime functioning debilitation. Until now no study has been done to investigate sleep characteristics in Iran thoroughly. In this study, we aimed to evaluate Iranians’ sleep characteristics and their association with daytime functioning. We conducted a population-based study from January 2017 to May 2019 on people more than 18 years old who lived in 11 urban destricts and 3 rural areas of Tehran, Iran. We randomly selected the participants using a multistage random stratified clustered sampling method. We obtained the participants’ demographic and anthropometric characteristics and details of bedtime, sleep duration, sleep onset latency, wake-up time and sleep impact on daytime functioning. Logistic regression model was used to assess the relationship between sleep characteristics) and daytime functioning. In total, 1830 people with a mean age of 40.83 years participated in the study. The gender distribution of the participants was even, and 70.98% of them were married. After adjusting for age and sex, the following three factors had a significant impact on daytime functioning: bedtime, sleep onset latency, and sleep duration. (OR = 1.12, P < 0.038, OR = 1.01, P < 0.011, and OR = 0.99, P = 0.01, respectively). We also found that longer sleep onset latency (P = 0.004) and shorter sleep durations (P = 0.029) significantly interfere with daytime functioning. Iranians’ sleep characteristics, especially their sleep duration and sleep onset latency, are associated with their daytime function. Interventions on people’s sleep hygiene are warranted to promote healthier sleep behaviors among Iranians, considering the high impact of current sleep characteristics on their daily lives.

www.nature.com/scientificreports/ deprivation is more prevalent in some occupations, such as healthcare workers 9 . Also, sleep characteristics vary in different populations, and findings in one country may not be applicable to other populations 5 . Sleep deprivation may cause fatigue, daytime sleepiness, and memory impairment. Anxiety, depression, and cardiovascular diseases are among disorders associated with sleep deprivation 2 . Studies have shown that insomnia increases the risk of car accidents by 2.5 to 4.5 times 10 . Also, an international study revealed that about 7% of people with insomnia had at least one episode of a car accident in the past year prior to the study. Furthermore, the car accidents were associated with several sleep problems, including difficulty initiating sleep, early morning awakening, non-restorative sleep, and poor sleep quality 11 . Sleep deprivation may lead to decreased productivity, and people with less sleep duration lose more productivity time 12,13 . Sleep characteristics other than sleep duration, such as rise time and bedtime, may also contribute to people's daily functioning. For instance, irregular bedtime may lead to decreased daytime functioning 14 .
Altogether, sleep characteristics differ between populations, and sleep problems have several negative cognitive, psychological, and functional impacts. Several studies investigated the sleep characteristics of the Iranian population. However, most of them focused on specific groups such as students, drivers, and people with health conditions [15][16][17][18] , and there is limited evidence on the sleep characteristics of the general Iranian population. Also, no study has evaluated the association between sleep characteristics and daytime functioning in Iran. To consider this matter, in this large-scaled study, we evaluated the sleep characteristics of Iranian people and evaluated its association with daytime functioning.

Materials and methods
Study design. This population-based study was conducted from January 2017 to May 2019 in Tehran, the capital of Iran. The study was approved and funded by the Tehran University of Medical Sciences' Ethics Committee (IR.TUMS.VCR.REC.1397.236). We confirm that all methods were performed in accordance with the relevant guidelines and regulations.
To describe the Iranian population's sleep characteristics, we conducted a population-based study on 2000 participants from the center of Tehran province, Tehran, chosen by the multistage random stratified clustered sampling method (1830 participants were included in the analysis, 91% Response rate). Tehran is Iran's capital city with about 12 million inhabitants and is divided into 22 districts administratively. The cluster sampling method was used to select a representative sample of Tehranian adults aged > 18 years old.
We randomly selected 11 districts in urban areas and 3 rural areas of Tehran. The clusters, each consisting of 10 males and ten females, were taken from each district proportional to Tehran's total population, and in total, we had about 100 clusters. After that, we randomly selected a healthcare center in each of the 11 urban districts and three rural areas. We randomly selected a household affiliated with the selected healthcare center for each cluster. We had two trained interviewers who went for the interviews in coordination with health care center staff. They went to the address of the selected household and gave explanations about the study goals and objectives. If they agreed to participate in the study, one male and one female who met the eligibility criteria, being Iranian and 18 years or older, were randomly selected from that household and were interviewed. After that, they went to the household on the right side of the selected household and repeated the same procedure. Interviewers continued sampling until they recruited 20 people for each cluster using the explained procedure.
One hundred seventy participants who had severe physical or psychological disabilities or who were taking sleep medications were not entered in the final analysis.
Measurements. Individuals after completing inform consent forms were interviewed by trained interviewers as main staffs in the study. Each interviewer staff passed a training workshop regarding the study aim, protocol and communications in households in each region for data collection.
The staff filled a questionnaire by interviewing each participant face-to-face. The questionnaire contained demographic (age, gender, marital status, and educational status) and anthropometric data (weight, height, and body mass index 19 , measured by interviewers using Matheo Digital Scale manufactured by Germany and NOVA NTT 3016 m manufactured by Iran), and questions regarding sleep characteristics (Bedtime, sleep onset latency, usual wake-up time, sleep duration on weekdays, sleep duration in weekends) 20 and sleep satisfaction (problems with waking up too early, satisfaction with current sleep pattern, and interference of sleep problems with daily functioning) in the last month 21 . We also asked them whether they think their sleep problems affect their daytime functioning or not.
We used Pittsburgh Sleep Quality Index (PSQI) to evaluate the participants' sleep characteristics. PSQI has 24 questions, 19 questions are self-rating questions, and a roommate or a bed partner scores the remaining five 20 . PSQI has been translated to Persian and proved to be a reliable, valid, sensitive, and specific tool evaluating sleep quality in Iranian People 22 . We included the first four questions of PSQI in our study evaluating bedtime, sleep onset latency, wake-up time, and sleep duration on weekdays and weekends. We used the crude numbers reported by the patients in our data analysis.
Insomnia Severity Index (ISI) is another questionnaire used for the evaluation of insomnia. This questionnaire consists of seven items, and each can be answered from 0 to 4 on a Likert scale where higher scores were indicative of more severe sleep problem 21 . ISI has been translated to Persian and was shown to be a reliable and valid tool evaluating insomnia severity in Iranian people 23 . In our study, we used three ISI questions evaluating problems waking up early, satisfaction with current sleep patterns, and interference of sleep problems with daytime functioning. We categorized the subjects based on the reported severity of the sleep problems and if they had daytime dysfunction due to sleep problem. www.nature.com/scientificreports/ Statistical analysis. Approximately, 4.17% of covariate data cells were missed and we used a single imputation method based on regression model for estimation of missing information by using mice package in R software. We used complex survey methods to enter cluster and sampling weight. All data analyses were weighted based on the 2011 Iran population census in Tehran based on gender and age categories in survey analysis.
To describe continuous and categorical variables weighted means and Standard Error (SE) and the weighted percentage were reported, respectively based on survey analysis. To compare means between two or more independent groups student t-test and one-way ANOVA were used, respectively and to qualitative variables between two or more groups x-square test was used.
In order to assess the relationship between sleep characteristics (as main exposures) and daytime functioning (as interesting outcome) logistic regression model was used and crude and age and sex-adjusted odds ratios were reported. To assess the trends of prevalence of interfere with daytime functioning, we generated quartile boundaries for the distribution of sleep characteristics and determined the prevalence of them in each quartile.
All data analyses were done by STATA software (R for imputation), and all p-values were 2-tailed and p-value ≤ 0.05 was considered significant.

Results
In total, 1830 people (907 men and 923 women) with a mean age of 40.83 ± 0.08 years participated in the study. Response rate was 91%. Most of the sample population lived in the urban areas (94.37%) and were married (70.89%). Table 1 demonstrates the demographic and anthropometric characteristics of the studied population.
Disturbution of bedtime, sleep onset latency, sleep duration, and wake-up time are depited in Figs. 1-4 among men, wemon, and whole population, respectively. We found that older and married participants with higher BMI, illiterates, and those who lived in urban areas had more problems waking up too early (P < 0.05). Also, participants with higher BMI and females we more satisfied with their current sleep pattern (P < 0.05). We found that sleep problems significantly interfered with the daytime functioning of people with higher BMI and people living in urban areas (P < 0.05). Sleep problems negatively impacted the daytime functioning of the urban population and divorced participants (P < 0.05). Table 2 shows the association between demographic and anthropometric characteristics and sleep problems.
Mean bedtime and wake-up time were 24.06 ± 1.45 and 7.53 ± 0.07, respectively, with a mean sleep duration of 431.31 ± 5.59 min. We found that older and illiterate adults and divorced people go to bed sooner (P < 0.05). Younger, thinner males, higher educated, and single adults, and those with shorter neck circumferences woke up later (P < 0.05). We found that males have a shorter sleep onset latency than females (22.6 vs. 35.54, P < 0.0001), but there was no significant association between other demographic and anthropometric characteristics and sleep onset latency (P > 0.05). Younger adults, females, and urban residents had longer sleep duration on workdays (P < 0.05). On the weekends, younger adults, those with lower BMI and shorter neck circumferences, higher educated, and single adults had a longer sleep duration (P < 0.05). Table 3 shows the association between sleep and demographic and anthropometric characteristics.
After adjusting for age and sex, late bedtime (OR = 1.12, 95% CI = 1.01-1.21) and longer sleep onset latency (OR = 1.01, 95% CI = 1.02) had significant negative impacts on daytime functioning. In contrast, longer sleep duration had a reverse impact (OR = 0.99, 95% CI = 0.98-0.99). Sleep problems significantly interfered with the daytime functioning of those with longer sleep onset latency (P = 0.004) and shorter sleep durations (P = 0.029). Association between sleep characteristics and negative impacts of sleep problems on daytime functioning and interference of sleep problems with daytime functioning are shown in Tables 4 and 5, respectively.  We found that late bedtime impacts daytime functioning negatively (OR = 1.12, P = 0.038). In a study on adolescents, Wolfson et al. found that students with late bedtime habits had a bad academic performance 29 . Fernandez-Mendoza et al. evaluated the impact of chronotype on sleep behaviors and its association with daytime functioning, cognition, and mood. E chronotype people (evening type) tend to have late bedtime and impaired daytime functioning due to fatigue 30 . Our findings regarding the worse daily functioning in people who have later bedtimes may be due to several reasons; first, E-types have more irregular sleep patterns than M-types (morning type), which lead to sleepiness during the day 30   www.nature.com/scientificreports/ on daytime functioning. We also found that higher educated and younger Iranians have late bedtimes. Such sleep patterns may be troublesome as younger adults are more functional than older adults, and people with higher educational levels have more critical roles in the community. Late bedtime in these groups, as the functional part of the community, and the negative impacts of such sleep habits need to be addressed in future health policies to enhance Iranian people's work efficacy. We found that longer sleep onset latency is significantly associated with daytime functioning impairment (OR = 1.01, P = 0.011). Ohayon et al. also found that longer sleep latency is associated with sleep dissatisfaction 32 . People's dissatisfaction and perception of their sleep may be the reason for attributing their functional impairment  Table 3. Demographic and anthropometric characteristics of participants with different sleep characteristics. BMI Body Mass Index.

Waking-up time Sleep onset latency (min) Sleep duration (min)
Weekend sleep duration (min)

Mean (SD)/ r P-value Mean (SD)/r P-value Mean (SD)/r P-value Mean (SD)/r P-value Mean (SD)/r P-value
Age ( www.nature.com/scientificreports/ to sleep 33 . On the other hand, people with higher sleep onset latency are more affected by adverse stimuli, further affecting their daytime function perception 34 . Prolonged sleep onset latency happens in people with mood disturbances 35 , thus impairing their daytime functioning. Sleep duration also affects daytime functioning. Males, older adults, and rural region residents had shorter sleep duration (P < 0.05). Early morning awakening was more common among older adults, reflecting in their wake-up time, as those older adults went to bed earlier and woke up in earlier hours. Our findings are in line with previous studies regarding a higher prevalence of early morning awakening in older adults 36 . In a similar study, Ferrie et al. found that people with higher job positions sleep more 37 . The association between sleep duration and daytime sleepiness is well studied in previous works, which, in turn, may have consequences such as the increased risk of road accidents 38,39 . Sleeping for less than six hours a day is associated with poor cognitive function 37 . Sleepiness and poor cognitive function may be reasons for impaired daytime functioning. At-risk groups such as males, older adults, and people living in rural areas may benefit from interventions to increase sleep duration. Treating the elderly for early morning awakening may be beneficial in this regard 40 .
In our study, most participants (65.08%) were dissatisfied with their current sleep patterns. In other countries, the prevalence of sleep dissatisfaction ranged between 8 and 18.5% 33 , significantly less than our findings. Several factors are independently associated with sleep satisfaction, including feeling refreshed, sleep calmness, ease of falling asleep, and awakening frequency 41 . As people's perceptions of their sleep quality and satisfaction with their sleep are important factors regarding interpretation of sleep impacts on their daytime functioning 33 , future studies on the Iranian population are needed to determine and address the factors contributing to sleep dissatisfaction. In this study, higher educated people were not satisfied with their sleep. As they have more critical roles in society, their daytime functioning impairment may be more prominent.

Limitations.
We found an association between sleep characteristics and daytime functioning in the Iranian population, but due to the study's cross-sectional design, a causal relationship cannot be interpreted from our findings. Future longitudinal studies are required to evaluate this relation further. People's psychological characteristics are other factors affecting daytime functioning. Future studies evaluating more variables will help determine the factors affecting people's daytime functioning and the factors mediating these effects.

Conclusion
Iranians' sleep characteristics, especially their sleep duration and sleep onset latency, are associated with daytime functioning. As we found high levels of sleep dissatisfaction among our participants, interventions to promote sleep hygiene are needed.