Sleep quality and associated factors among pregnant women attending antenatal care at Jimma Medical Center, Jimma, Southwest Ethiopia, 2020: cross-sectional study

Background Sleep is a natural physiological process vital for the physical and mental wellbeing of pregnant women and their fetuses. Even though poor sleep quality is a common problem among pregnant women, it is not studied in developing countries including Ethiopia. Therefore, this study was aimed to assess the poor sleep quality and associated factors among pregnant women attending antenatal care at Jimma medical center, Jimma, Southwest Ethiopia, 2020. Methods A cross-sectional study design was conducted among 415 pregnant women at Jimma Medical Center (JMC). The study subjects were selected using a systematic random sampling technique. Pittsburgh Sleep Quality Index (PSQI) was used to assess sleep quality using face-to-face interviews. SPSS version 25 was used to analyze data. Bivariate and multivariable logistic regressions were done to identify factors related to sleep quality. In multivariable logistic regression variables with a p-value less than 0.05 was considered significant and, adjusted OR (AOR) with 95% CI was used to present the strength of the association. Results The prevalence of poor sleep quality among pregnant women was found to be 30.8% (95% CI (26.5, 35.2). In multivariable analysis, age ≥ 30 years old (AOR = 1.94;95%CI:1.03,3.66), Multigravida (AOR = 1.90;95%CI:1.90,3.32),depression (AOR = 4.26;95%CI:2.54,7.14),stress (AOR = 1.85;95%CI:1.20,3.02) were variables significantly associated with poor sleep quality. Conclusion This study found a high prevalence of poor sleep quality among pregnant women. Older age, gravidity, depression, and stress were associated with poor sleep quality. It is better to have routine sleep pattern screening and teach sleep hygiene practice for pregnant women.


Introduction
Sleep is a physiological process and vital for the normal physical and mental well-being of an individual [1]. Sleep disorder is one of the most common type's mental illnesses, directly related to the thoughts, feelings, and behaviors of humans. Pregnancy is the term given to a woman while her fetus develops inside her uterus [2]. It causes many physical, social, economic, psychological, and hormonal changes contributing to develop different forms of mental illness for a short and long period [3] . In the first trimester period of pregnancy, night urination, shortness of breath, heartburn, forced body position in bed, breast tenderness, and itching are the physiological changes affecting sleep [4]. Quality of sleep is essential for the physical and mental health of both the mother and fetus. A pregnant mother should get at least eight hours of night sleep [5]. The most common sleep problems observed among pregnant women were resting leg syndrome, sleep apnea, insomnia, nocturnal gastroesophageal reflux, and Sleep-related breathing problems [6]. Nearly, 78% of pregnant women's had any forms of sleep disorder during their pregnancy time and complained of physical problems such as migraine headaches, gestational diabetes, obesity, cardiovascular conditions (hypertension and heart problems), and poor digestion [7]. A pregnant woman who slept less than six hours has a risk for premature birth, pre-eclampsia, prolonged labor, low progesterone level, abruption placenta, miscarriage, fetal death, and low birth weight [8,9]. Almost 70% of mental health conditions including depression, suicidal ideations, and postpartum psychosis happened during pregnancy are due to subjective and objective sleep pattern problems [10]. The prevalence of poor sleep quality among pregnant women in the USA (53-71%) [11], China (87%) [12], and Iran (96.4%) [13].
The determinants of poor sleep quality among pregnant women were socio-economic status, age, divorce, body mass index, first pregnancy, history of fetal death, history of prolonged labor, zinc and fluorine deficiency, gestational period, and having a history of chronic medical illness (DM, hypertension), lack of awareness about sleep hygiene practice, unwanted pregnancy, lack of social support, and history of mental illness [2,12,14].
The most effective intervention for sleep problems related to pregnancy was basic sleep hygiene practice and progressive muscle relaxation techniques [15]. The use of medication for sleep problems related to pregnancy is the last option and better not used in the first trimester of pregnancy [16]. Despite the above prevalence and impact of sleep on the physical and mental well-being of both a mother and a fetus, there is limited study evidence in African countries including Ethiopia. Therefore, this study aimed to assess the prevalence of poor sleep quality and its correlates among pregnant women attending perinatal service at Jimma Medical Center, Southwest Ethiopia,2020.

Study design and period
An institutional-based cross-sectional study was conducted in August 1-30/ 2020.

Study area
The study was conducted at Jimma medical center (JMC) antenatal care units. JMC is located in Jimma town, Oromia regional state, which is found in the southern part of Ethiopia 325 km far away from Addis Ababa. The center gives service to the catchment population of about 15 million people. There was about 9850 pregnant mother who had a follow-up in a year at antenatal care and on average of a typical month 848 pregnant mother visit antenatal care for follow up.

Source population
All pregnant women who had ANC follow up at Jimma Medical Center.

Study population
Pregnant women attending the antenatal care follow up during the study period.

Inclusion and exclusion criteria Inclusion criteria
Women with a gestational age of 4 weeks and above, and aged 18 and above were included in this study.

Exclusion criteria
Women who were critically ill and difficult to communicate were excluded from the study.

Sampling procedure and sampling techniques Sample size estimation
The sample size was estimated by using a single population proportion formula. Sample size with z-value of 1.96 and marginal error of 5% sample was calculated as: Assumption: n = initial sample size need for this study α = confidence interval (95%) p = proportion of =50% (0.5) d = marginal error of 5%, (z α/ 2 ) 2 = 1.96. To calculate the sample size, 50% of the proportion was used since there is no related study done to identify the prevalence of poor sleep quality in Ethiopia. We added a 10% (39) nonresponse rate; finally, the total sample size was 423.

Sampling procedure
Systematic random sampling was used to invite respondents within every two intervals while coming to ANC follow-up. The average monthly number of 848 pregnant women visited the hospital. The sampling interval was determined by dividing the total population who had follow-up during a month of data collection period in OPDs of the JMC ANC unit by the sample size. The selected skip interval was by taking total pregnant women of 848 and sample size 423. Therefore, the sampling fraction was 848/423 ≈ 2.

Dependent variable
Sleep quality: good /poor

Independent variables
Socio-demographic factors Age, religion, marital status, ethnicity, educational status, income status, employment, pre-pregnancy BMI, and residence.

Psychosocial factors and behavioral characteristics
Social support, stress, alcohol use, khat use, and smoking cigarettes.
Depression A total score of > 13 points using EPDS, was considered as maternal depression [18].
Anxiety Anxiety was assessed using the anxiety subscale adapted from the Depression, Anxiety, and Stress Scale (DASS-21), participants who scored ≥8 were considered as having anxiety [19].
Past and current medical illness Pregnant women who have known chronic medical illness and their diagnosis confirmed in any health institution either governmental or private that currently had follow-up for any chronic medical illness that was assessed by Yes or No questions.
Current substance use Use of at least anyone of substance in the past three months [20].
Perceived stress A cut-off value of 20 participants with a total PSS score more than or equal to 20 will be defined as stressed, meanwhile those with a score < 20 were grouped as non-stressed [21] .

Data collection method and tools
A structured interviewer-administered questionnaire was used, which has different subunits, questionnaires to assess socio-demographic factors, Pittsburgh sleeping quality index (PSQI), factors that affect sleep quality including obstetrical factors/clinical factors, psychosocial factors /substance use factors.
The outcome variables were assessed using the Pittsburgh Sleep Quality Index (PSQI). The PSQI has 19 items, which are categorized into seven components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction over the last month. Each component scores ranging from 0 to 3 and then getting a global score with an interval from 0 to 21. This yields sensitivity and a specificity of 89.6 and 86.5% respectively [17]. The Cronbach alpha of PSQI in the current study was 0.78.
The Edinburgh Postnatal Depression Scale (EPDS) was used to assess the symptoms of maternal depression. It consists of 10 items and each question has four possible answers with an interval of 0-3, and a maximum score out of 30. Similar to the previous study, those who score 13 and above were considered as depressed mood [18]. EPDS is a common tool for screening depressive symptomatology; initially, for use during the postnatal periods, it is also additionally validated for use during the perinatal periods in different countries and settings The EPND is validated among the perinatal population in Ethiopia. The Cronbach alpha of EPDS in the current study was 0.91.
Adapted anxiety subscale from the Depression, Anxiety, and Stress Scale (DASS − 21) wasused to assess anxiety. Each item contributes 0 to 3 points to the sum score, resulting in a totalscore that intervals from 0 to 21, to considered anxiety, the score was 8 and above [19].
Substance use was assessed by ever use of alcohol, Khat, cigarette, and other particular substance for nonmedical purposes in life. The current use of the substance was assessed for the last three months [20]. The presence of a known chronic medical such as hypertension, diabetes mellitus, or others was assessed by yes/ no response.
The perceived stress scale (PSS) was used to assess stress during pregnancy. The PSS has 10 items of selfreport psychological instruments for measuring the perception of stress. Each item contains 0 to 4 points to the sum score resulting in a total score that intervals from 0 to 40, a higher score indicating greater perceived stress occurring one month before the interview [21]. Social support was assessed by the Maternity Social Support Scale (MSSS). MSSS has three categories; a score of less than 18 is low social support, 18-23 medium social support, and 24-30 high social support) [22].

Data collection procedure
Data were collected by five trained BSc psychiatry professionals and one supervisor from the MSc student in psychiatry was also trained on how to supervise data. Each section of the questionnaire was prepared in English and then translated into the Amharic and Afan Oromo, and to ensure its understandability and consistency, then back-translated to English by an independent person. Data collectors and a supervisor have received training for two days duration on the purpose of the study, tools, how to collect data, sampling techniques, and how to handle ethical issues including confidentiality. The pretest was done among 21 (5%) of the sample size pregnant women in Agaro General Hospital before the main study was done to identify impending problems in the proposed study such as data collection tools and to check the performance of the data collectors. The supervisor and principal investigator were made regular supervision to ensure that all necessary data were appropriately collected. Each day throughout the data collection period, the completed questionnaires were assured for completeness and consistency. The collected data were edited and entered into the computer, then checked twice and processed appropriately.

Data processing and analysis
The collected data were edited, coded, and entered into the Epi Data version 3.1, and then the data was exported to Statistical Package for Social Science (SPSS) 25.0 version for analysis. The bivariate logistic analysis was performed to select candidate variables. All variables with a p-value < 0.25 in the bivariate analysis were entered into the multivariable logistic regression model. Multivariable logistic regression analysis was employed to control for possible confounding effects and to determine the presence of a statistically significant association between independent variables and dependent variables. Hosmer and Lemeshow goodness was used to check the necessary assumptions. The strength of the association was presented by an adjusted odds ratio of 95% CI and Pvalue < 0.05 was considered as statistically significant. Descriptive statistics results containing frequency, percentages, and summary statistics (mean values and standard deviation) were presented to define the study population about relevant variables.

Socio-demographic characteristics of participants
In this study, a total of 415 participants were assessed, with an overall 98.1% of response rate. The mean (SD) age of the participants was 25.22 (±4.62) years. More than half (51.8%) of respondents were Muslim religious followers. Almost three fourth of the participants (75.4%) were married and 251(60.5%) of them were Oromo in their ethnicity. One-third (32.8%) of participants have attended college and above and the majority 163(39.3%) were housewives. Most of the respondents, 320(77.1%) lived in urban areas. More than three fourth, 326(78.6%) of the participants had a normal body mass index (BMI). More than one-third of (39.8%) women reported that their average monthly income is below 1000 Ethiopian birr (Table 1).

Obstetrics related characteristics of the participants
According to this study, nearly half of the study participants (47.0%) were in the third trimester in their gestational age. Almost two-thirds (64.8%) were multigravida and three hundred eight (74.2%) of the participants were multipara. Eighty-four participants (20.2%) had reported a previous history of abortion. More than two-thirds (68.0%) of the women had a planned pregnancy ( Table 2).

Psychosocial and substance-related factors of the participants
From the total of the respondents, about one-third (34.7) of the respondents had stress during pregnancy. Related to social support, more than half (53.3%), 141(34.0%), and 53 (12.8%) of the pregnant women had medium social support, high social support, and poor social support respectively. Regarding the current substance use, 19(4.6%),13(3.10%), and 3(0.7%) had used alcohol, chewing khat, and smoking a cigarette within the past three months before data collection time respectively ( Table 4).

Prevalence of sleep quality among pregnant women
In the current study, the prevalence of poor sleep quality among pregnant women was 30.8% (95% CI (26.5, 35.2). Among the total respondents, 98 (23.6%) rated their overall sleep quality as bad. Below one-half (42.7%) of      (Table 6).

Discussion
The finding of this study showed that the prevalence of poor sleep quality among pregnant women was 30.8% [(95% CI, 26.5, 35. 2)]. The study finding was higher as compared with a study done in Finland 15% [24], Peru 17% [25], {Venugopal, 2018 #34} and China 15.2% [26]. The variation might be due to the eligibility criteria and sample size. In Peru, pregnant women were between 24 and 28 weeks of pregnancy with 1298 participants while the current study was done on pregnant women including all trimesters of gestation, and the sample size was 415 [25]. Another discrepancy might be due to the difference in social support practice, sampling technique, the socio-cultural and demographic context of the women. However, the current study was lower than the meta-analysis study conducted by Sedov et al. 45.7% [27], Turkish 86% [28], Vietnam 41.2% [29], Iran 77% [30], Taiwan 60% [31], and meta-analysis study in China 54.2% [32]. This discrepancy might be due to variation in used inclusion criteria, sampling technique, study design, and sampling size. A study conducted in Iran included third-trimester pregnancy only, and a convenience sampling technique was used while the current study used a systematic random sampling technique [30]. Another possible reason might be the difference in study design, socioeconomic, sociocultural, and demographic characteristics in the populations.
Regarding factors associated with poor sleep quality, in the current study; older maternal age was nearly two times more likely to have poor sleep during the pregnancy period as compared to younger ones. Sleep quality among pregnant women decreases as the age of the mother increases, this finding was supported by other studies done in Taskiran (Turkish) (28, 29 and 45-yearold had worse sleep quality than the age group between 17 and 28), Taiwan [33]. China [12], Vietnam [29], and Meta-analysis [27]. The possible reason might be due to usually older mothers have a tendency to have care burdens, heavier domestic responsibilities, more likely to experience physical distresses, and slower recovery from delivery, this causes a probable reason for the decline of sleep quality [34].
This finding also revealed that pregnant women with multigravidas were nearly two times more likely to have poor sleep quality than those who are primigravida. We found that poor sleep quality during pregnancy was associated with multigravidas in cross-sectional analyses. Nevertheless, our findings were inconsistent with those finding from Francisco [35] which only found an association between sleep quality and primigravida. A possible explanation might be multigravidas pregnant mothers complained that their sleep pattern is influenced by their children's sleep habits. When children frequently wake up and/or cry at night, mothers are expected so they also wake up from their sleep to take care of theirs children.
The odds of having poor sleep quality among women who had depression were about 4.26 times higher when compared with the counterpart. This was consistent with the conducted in China [36]. This could be due to the presence of prenatal depression is one of the most possible psychological factors contributing to sleep disturbance during pregnancy [12].
The likelihood of developing poor sleep quality among pregnant women who have stress was 1.85 times higher as compared with no stress. A similar study was done in China [36]. The possible reason might be due to the direct effect of stress during pregnancy on sleep quality might be related to arginine vasopressin hormone, which is involved in the stress response and circadian regulation of the sleep-wake cycle [37].

Conclusion
Our study found that the prevalence of poor sleep quality among pregnant women was high as compared with studies done in the general population and many other countries. Older age, multigravidas, depression, and stress were statistically significantly associated with poor sleep quality in this study. It is a good alarm to be alert to give attention to routine screening of sleep patterns in pregnant women and to give special concern for pregnant women with the above-stated factors.