The relation between sleep quality during pregnancy and health-related quality of life—a systematic review

Abstract Background The majority of expectant mothers report sleep alterations during pregnancy and almost 40% report poor sleep quality. There is growing evidence that sleep quality (SQ) during pregnancy influences maternal health. This review focuses on how SQ during pregnancy relates to maternal health-related quality of life (HRQoL). The review also aims to identify whether this relation varies between pregnancy trimesters, and for different subdomains of HRQoL. Methods A systematic review was performed according to PRISMA guidelines and registered on Prospero in August 2021 with ID no: CRD42021264707. Pubmed, Psychinfo, Embase, Cochrane, and trial registries were searched up to June 2021. Studies with any design that investigated the relation between SQ and quality of life/HRQoL in pregnant women, published in English, and peer-reviewed, were included. Two independent reviewers screened titles, abstracts, and full texts, and extracted data from the included papers. The quality of the studies was evaluated using the Newcastle-Ottawa Scale. Results Three hundred and thirteen papers were identified in the initial search, of which 10 met the inclusion criteria. Data included 7330 participants from six different countries. The studies had longitudinal (n = 1) or cross-sectional designs (n = 9). In nine studies SQ was reported subjectively by self-report questionnaires. Actigraphic data was available from two studies. HRQoL was assessed by validated questionnaires in all studies. Due to high levels of clinical and methodological heterogeneity in included studies, a narrative synthesis was employed. Nine studies found that poor sleep quality was related to a lower overall HRQoL during pregnancy. Effect sizes were low to medium. This relation was reported most during the third trimester. Especially sleep disturbances and subjective low SQ seemed to be related consistently to lower HRQoL. Furthermore, an indication was found that SQ might have a relation with the mental and physical domain of HRQoL. The social and environmental domain may also be associated with overall SQ. Conclusion Despite the scarcity of studies available, this systematic review found evidence that low SQ is related to low HRQoL during pregnancy. An indication was found that the relationship between SQ and HRQoL during the second trimester might be less prominent.


Introduction
The majority of expectant mothers report sleep alterations during pregnancy [1][2][3]. and poor sleep quality occurs in almost 40% of all pregnant women, peaking in the third trimester, and persisting until about 50% two years postpartum [4,5]. During pregnancy, poor sleep quality is characterized by a decrease in mean sleep duration, an increase in sleep disturbances, and longer time to fall asleep. Gestational sleep disturbances seem to increase the risk of multiple adverse maternal/perinatal outcomes, including a higher chance of depression, cesarean birth, preeclampsia, gestational diabetes, preterm birth, lower birth weight, and offspring adiposity, and increased blood pressure [6,7].
The concepts Quality of life (QOL) and Healthrelated QOL (HRQoL) are used interchangeably in the literature. QOL is defined by the World Health Organization (WHO) as "individuals' perception of their position in life in the context of culture and value systems in which they live and in relation to their goals, expectations, standards, and concerns" [8]. HRQoL is a multi-dimensional concept and is commonly divided into different domains [9]. In recent years, the HRQoL has become increasingly important as a concept in the public health sector, such as prenatal care to detect unmet needs during pregnancy [10,11]. Due to its interchangeable use, this review will refer to QOL and HRQoL as HRQoL.
Poor sleep quality is found to be related to lower HRQoL in various populations, such as in the general population [12,13], postpartum women [14], and patients with Obstructive Sleep Apnea [15], cancer, diabetes, depression, Parkinson and chronic kidney diseases [16]. Previous articles also suggested a possible relationship between sleep quality (SQ) and HRQoL during pregnancy [17,18]. However, to date, no systematic review is available.
In 2010, Da Costa et al. [19] concluded that the HRQoL of pregnant women were comparable to patients with chronic disorders. However, frequently pregnant women are told a decrease in sleep is a collateral of pregnancy and therefore it is a natural and in principle healthy state which does not need attention. Consequently, it is interesting to see whether SQ has a similar impact on the HRQoL during pregnancy as in other populations and therefore may no longer be considered "just" collateral.
By clarifying the relation between SQ and the HRQoL further, we can better identify which aspects of sleep lead to poor perceived health and possibly expand knowledge on how to improve HRQoL during pregnancy. The aim of this review is to evaluate how sleep during pregnancy relates to HRQoL and whether this relation varies during different trimesters and/or on different subdomains of HRQoL.

Study protocol
This review was registered in August 2021 with the International Prospective Register of Systematic Reviews (PROSPERO; Registration no. CRD42021264707) and was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [20]. The checklist can be found in Appendix 1.

Search strategy
The search was conducted in July 2021 using PubMed, PsycInfo, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), and trial registries (https://clinicaltrials.gov and https://who.int/ictrp/en). The databases have been systematically searched from inception to June 2021. The database searches combined terms for pregnancy, sleep, and HRQoL concepts using the PICOS (participants, interest, comparison, outcome, and study design) strategy [21]. The search terms used are found in Appendix 2. Studies were included if (1) the population researched were pregnant women, (2) measured SQ either objective (e.g. polysomnography, actigraphy) or subjective (e.g. Pittsburgh Sleep Quality Index (PSQI) [22] or the Insomnia Severity Index Scale (ISI) [23] which is also often used as a measure of sleep quality [24]), (3) compared women with or without sleeping problems, (4) had an outcome of either QOL or HRQoL, and (5) were published as an original study (in journals, "in press" or in a trial registry). There were no restrictions on sample size, study design, or measurement tools. For example, sleep quality could be measured objectively as well as subjectively, to give a complete as possible answer to the research question.
Reference lists were screened to identify additional articles.

Study selection
Relevant records were identified through database searches. After removing duplicates, a two-step screening process was conducted. In the first step titles and abstracts were reviewed by two independent reviewers (MP, LV). In case of disagreement, papers were included. The second step included fulltext screening. Disagreement was resolved by discussion or a third reviewer (RM). Reference screening was performed afterward.

Quality appraisal, data extraction, and analyses
Data extraction was completed independently by both reviewers in Microsoft Excel and compared. Extracted data included author name, publication year, study design, country, sample size, trimester(s) observed, available data on sleep, HRQoL outcomes including subdomains, the relation of SQ and HRQoL, and methods of measurement.
The Newcastle-Ottawa Scale (NOS, see Appendix 3) was used to evaluate the quality of the included papers (with modifications to match the needs of this review). Two independent reviewers assessed the risk of bias (MP, LV). Disagreements were resolved by discussion with a third reviewer (RM). Cross-sectional studies with a score of !7 were considered high quality [25]. Longitudinal studies with a score of !2 in the selection category, !2 in comparability, and !2 in the outcome domain were considered fair quality [26].
Options to perform meta-analyses were explored. Figure 1 shows that the initial search identified 531 relevant records, of which 313 remained after the duplicate screening. After exclusion based on title and abstract, 19 full-text papers were screened, which led to 10 studies to be included in this review. Reference screening did not yield any extra eligible studies.

Study characteristics
Data across studies involved 7330 subjects ranging from 23 to 4352 participants per study (Table 1). Studies were published in English between 2010 and 2021. Seven studies included healthy pregnant women [19,28,29,31,32,34,35], one included women with prior sleeping disorders [27], one included women with ongoing mood disorders [33], and one study included women with high-risk pregnancies [30]. Eight studies  There is a significant relationship between sleep quality score and quality of life among high-risk pregnant women (r were cross-sectional [19,[27][28][29][30][31][32][33] and two were labeled as longitudinal study [34,35]. One study, however, labeled as a longitudinal, which included cross-sectional data pertaining to our research question has been treated as so [34]. Three studies focused on the entire pregnancy [29,32,35], one study on the second and third trimesters [28], one on only the first- [34], one on the second- [27], and four on the third trimester [19,30,31,33]. SQ was measured subjectively in nine studies, of which one also included actigraphic data [35]. One study used only actigraphic data [33]. The PSQI was used in six studies [27,28,30,31,34,35], two studies used the ISI [29,32] and one study [19] used the prime-MD patient health questionnaire (PHQ) [36].

Quality assessment
Quality assessment is found in Figure 2. After discussion, agreement was met for the 10th article which was considered to be of unsatisfactory quality [33]. One study was found to be of satisfactory quality [27], four of good quality [29][30][31]35], and four of very good quality [19,28,32,34].
Five studies reported effect sizes on this relation. Two articles [29,32] reported a medium effect size [42,43] over the entire pregnancy. One study reported a low effect size for the second and third trimesters [28]. In the third trimester, two studies reported medium effect sizes [19,30].

The relation of sleep components and HRQoL
Four studies [27,28,30,35] explored the relation of HRQoL with various components of sleep. Sleep disturbances [27,28,30] and lower subjective SQ [27,30,35] were found to be associated with lower HRQoL in three of four studies. Two studies reported a higher daytime dysfunction [28,30] and lower sleep efficiency [30,35] to be related to a lower HRQoL. No relation was found for sleep latency.
The study using wrist actigraphy found daytime sleep duration to be associated with higher scores of HRQoL on the physical subdomain during the first trimester [35]. Shorter total nighttime sleep duration was also found to be related to lower mental HRQoL subdomain scores during the third trimester in this study. No relation was found between the total nighttime sleep duration and overall HRQoL [27,28,30].

The relation of overall SQ and domains of HRQoL
Seven studies reported on the relation of overall SQ with specific domains of HRQoL. Two studies measured HRQoL as only a physical and mental domain [31,35], and five included more domains (e.g. mental, physical, emotional, social, and environmental) [19,27,29,32,33].
Poor SQ was associated with both the mental and physical domains in the studies which investigated only two domains [31,35]. Tsai et al. [35] furthermore observed that the relation of SQ and the physical domain was strongest in the first trimester, weakening with progressing pregnancy; the relation of SQ and the mental component was found to be strongest in the first and third trimesters.
Rezaei et al. [27] and Kang et al. [33] reported on four HRQoL domains (physical, psychological, social, and environmental). Kang et al. [33] report no association between sleep and these domains. Rezaei et al. [27] found low SQ to be related to low scores in the psychological domain only.
Two articles reported on five domains of HRQoL (general, physical, psychological, social, and environmental health) [29,32]. Mourady et al. [29] found SQ to be associated with all. Davoud et al. [32] found only a relation of SQ with three domains, namely general health, physical health, and psychological health.
Da Costa et al. [19] report a significant relationship between lower SQ and lower scores in the domains of physical functioning, role physical, bodily pain, general health, vitality, social function, and mental health (psychological health). The domain of role emotion however did not have a relation with SQ.

Longitudinal effects of SQ on the HRQoL during pregnancy
Only one study described the longitudinal effects of SQ in the first trimester on the HRQoL of the second and third trimesters [35]. Lower SQ in the first trimester predicted lower physical HRQoL in the second trimester (b ¼ À0.87), and a lower mental HRQoL in both second and third trimesters (b ¼ À0.85). Higher total nighttime sleep duration in the first trimester predicted better mental HRQoL in the second and third trimesters (b > 1.38) but did not have a significant relation with physical HRQoL. Other sleep components, such as sleep latency, sleep efficiency, and total daytime sleep duration in the first trimester did not predict HRQoL in later trimesters.

Possibility of further analyses
To investigate the possibility of a meta-analysis a "Table of study characteristics illustrating similarity of PICO elements" [44] was constructed (see Appendix 4). Because of high clinical and methodological diversity [45] and as some scholars argue against doing metaanalysis on observational data anyway [46], we decided not to execute a meta-analysis.

Summary of findings
Nine studies show a relation between overall SQ and HRQoL during pregnancy (effect sizes: low to medium). One study reports no association between SQ and HRQoL.
Studies tended to show associations between sleep disturbances and HRQoL (3/4), and subjective SQ and HRQoL (3/4). Two studies did and two didn't find a relation between higher daytime dysfunction and sleep efficiency with lower HRQOL (2/4). No study found a relationship between sleep latency and HRQoL (0/4). One study, with actigraphic data found less daytime sleep and low total nighttime sleep duration to be associated with lower HRQoL. Three questionnaire studies, however, found no relation between total nighttime sleep duration and HRQoL.
The mental (psychological) and physical domains of HRQoL were found to be related to overall SQ scores in respectively 6/7 and 5/7 studies. The social and environmental domains of HRQoL were found to be associated with overall SQ scores in 2/5 studies.
One study found overall lower SQ and total nighttime sleep duration in the first trimester to be related to HRQoL in further trimesters.

Discussion
The objective of this review was to evaluate the relationship between SQ and HRQoL during pregnancy. Our review suggests a relationship between poor SQ with lower HRQoL during pregnancy, backed up by all studies included, except for one. This difference in outcome may be due to a lack of power related to the small number of pregnant women included (n ¼ 23).
Effect sizes of SQ on HRQoL were considered low to medium. The studies which focused on all three trimesters or on only the third trimester found medium effects; the study that focused on the second and third trimesters combined found a low effect size, which might suggest a less prominent effect of SQ on the HRQoL during the second trimester, as is also suggested by the longitudinal study included in this review.
Subjective SQ and sleep disturbances seem associated with lower HRQoL in most studies. There also seems to be some evidence for daytime dysfunction and worsened sleep efficiency to be related to HRQoL in a negative way, but the findings were not consistent.
Our findings build further on the studies of Sut et al. [17] and Lagadec et al. [18]. Sut et al. [17] found that the SQ and HRQoL of pregnant women were worse than those of non-pregnant healthy controls. Lagadec et al. [18] did a systematic review of the determinants of HRQoL in pregnant women and concluded that sleep difficulties are one of these determinants.
To our knowledge, however, this is the only systematic review aimed to explore the relationship between SQ and HRQoL during pregnancy. The strength of this systematic review includes the use of multiple databases including trial registries and the use of two independent reviewers for screening and data extraction.
Findings from the present systematic review should however be considered in the light of several limitations. First, our search has only identified 10 studies, which limits the validity and generalizability of our conclusions. Second, the heterogeneity among studies and insufficient availability of data prevented executing a meta-analysis. Third, the included studies primarily consisted of cross-sectional designs, which limit the interpretation of causality. Only one study included examined the cross-sectional and longitudinal association between SQ and HRQoL in pregnant women. Fourth, some studies used HRQoL questionnaires containing direct questions about SQ, which directly affect the score of some subdomains. It was not clear whether these studies corrected this influence when discussing their results on the relation between SQ and HRQoL. However, these direct questions, such as "How satisfied are you with your sleep?" are not expected to explain all of the relations found between SQ and HRQoL. Lastly, the small number of included studies prevented us from testing for publication bias.
This review provides yet another argument to underline the importance of good sleep in pregnant women; adding to the previously published articles (see Introduction) about the relation of poor gestational sleep with the risk of adverse maternal/perinatal outcomes, such as depression, preeclampsia, preterm birth and offspring adiposity [6,7]. A low HRQoL is found to be related to low birth weight and preterm birth and is, therefore, an important factor, by itself, for optimizing maternal and neonatal health [47,48]. The WHO even considers optimizing HRQoL the number one priority of care during pregnancy [49]. So, despite the limited number of studies, this systematic review contributes to the general body of knowledge on HRQoL and sleep during pregnancy and may contribute to reaching the goal of the WHO's main priority of care during pregnancy. If HRQoL is indeed affected by SQ during pregnancy, sleep information should be considered as important as prenatal information given about prenatal supplements and food restrictions. Furthermore, screening to prevent the chronic aspect of sleep problems during and after pregnancy is to be considered [7]. Future studies should examine the effectiveness of evidence-based interventions to improve sleep during pregnancy and thus maybe prevent low quality of life and/or maternal and neonatal health issues. Furthermore, we recommend that future studies should try to combine both subjective and objective sleep measures. In case the used HRQoL scale also contains a sleep quality item, we recommend also analyzing the relation between SQ and HRQoL without this item.
By clarifying the relation between SQ and the HRQoL further, we can better identify which aspects of sleep quality lead to poor perceived health during pregnancy and expand knowledge on what aspects of sleep during pregnancy should be improved especially. This may also be helpful in guiding future interventions to improve maternal well-being and give children the best start in life.

Conclusion
This review shows evidence of a relation between SQ and HRQoL during pregnancy in all trimesters, with most evidence found for the third trimester. The only longitudinal study in this review also indicates a longitudinal relation of sleep with the HRQoL during pregnancy, where SQ and sleep duration during the first trimester seem to be related to the HRQoL in the later trimesters. This review thus provides more evidence for the importance of good sleep during pregnancy. Future research should focus on developing evidencebased interventions to improve sleep during pregnancy and evaluate their effect on child and maternal health, including quality of life.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The author(s) reported there is no funding associated with the work featured in this article.

4-5
Selection process 8 Specify the methods used to decide whether a study met the inclusion criteria of the review, including how many reviewers screened each record and each report retrieved, whether they worked independently, and if applicable, details of automation tools used in the process.

5
Data collection process 9 Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and if applicable, details of automation tools used in the process.

5-6
Data items 10a List and define all outcomes for which data were sought. Specify whether all results that were compatible with each outcome domain in each study were sought (e.g. for all measures, time points, analyses), and if not, the methods used to decide which results to collect.

5-6
10b List and define all other variables for which data were sought (e.g. participant and intervention characteristics, funding sources). Describe any assumptions made about any missing or unclear information.
6 Study risk of bias assessment 11 Specify the methods used to assess risk of bias in the included studies, including details of the tool(s) used, how many reviewers assessed each study and whether they worked independently, and if applicable, details of automation tools used in the process.

5-6
Effect measures 12 Specify for each outcome the effect measure(s) (e.g. risk ratio, mean difference) used in the synthesis or presentation of results.

6
Synthesis methods 13a Describe the processes used to decide which studies were eligible for each synthesis [e.g. tabulating the study intervention characteristics and comparing against the planned groups for each synthesis (item #5)].

6
13b Describe any methods required to prepare the data for presentation or synthesis, such as handling of missing summary statistics, or data conversions.

N/A
13c Describe any methods used to tabulate or visually display results of individual studies and syntheses. Table 1 13d Describe any methods used to synthesize results and provide a rationale for the choice(s). If meta-analysis was performed, describe the model(s), method(s) to identify the presence and extent of statistical heterogeneity, and software package(s) used.

5
13e Describe any methods used to explore possible causes of heterogeneity among study results (e.g. subgroup analysis, meta-regression).

5
13f Describe any sensitivity analyses conducted to assess robustness of the synthesized results.

N/A
Reporting bias assessment 14 Describe any methods used to assess risk of bias due to missing results in a synthesis (arising from reporting biases).

4-5
Certainty assessment 15 Describe any methods used to assess certainty (or confidence) in the body of evidence for an outcome.

Results
Study selection 16a Describe the results of the search and selection process, from the number of records identified in the search to the number of studies included in the review, ideally using a flow diagram. 5-6, Figure 1 16b Cite studies that might appear to meet the inclusion criteria, but which were excluded, and explain why they were excluded. 5, Figure 1 (continued)  Figure 2 Discussion Discussion 23a Provide a general interpretation of the results in the context of other evidence.

10-13
23b Discuss any limitations of the evidence included in the review. 11-12 23c Discuss any limitations of the review processes used.
11-12 23d Discuss implications of the results for practice, policy, and future research.

12-13 Other information
Registration and protocol 24a Provide registration information for the review, including register name and registration number, or state that the review was not registered.
4 24b Indicate where the review protocol can be accessed, or state that a protocol was not prepared.

4
24c Describe and explain any amendments to information provided at registration or in the protocol.
N/A Support 25 Describe sources of financial or non-financial support for the review, and the role of the funders or sponsors in the review.

13
Competing interests 26 Declare any competing interests of review authors. 13 Availability of data, code and other materials 27 Report which of the following are publicly available and where they can be found: template data collection forms; data extracted from included studies; data used for all analyses; analytic code; any other materials used in the review.