Preterm birth is the leading cause of death in children younger than 5 years old worldwide1 and is associated with lifelong morbidity.2 Recent vital statistics signal an upward trend in preterm birth rates in the United States.3 Sleep disturbance may be a novel risk factor that is identifiable and modifiable,4–8 but there is a paucity of research examining sleep disorders and risk of preterm birth.
Sleep disorders are highly prevalent during pregnancy and represent heterogeneous conditions (eg, insomnia, sleep apnea, restless legs syndrome, narcolepsy, parasomnia, circadian rhythm disorder). Insomnia is evident in more than 50% of pregnant women,9 but previous studies have not examined associations with preterm birth. Prior research focused on sleep apnea and restless legs syndrome,10–12 but few investigations examined whether risk varied by gestational age at delivery and preterm birth type (spontaneous or medically indicated preterm birth). This refined examination is important because preterm birth is a multipathway endpoint, whereby causative mechanisms depend on preterm birth type. Additionally, morbidity and mortality vary significantly by gestational age at delivery.13 Finally, little is known about less common sleep disorders such as narcolepsy, parasomnia, and circadian rhythm disorder and risk of preterm birth.
We investigated a wide spectrum of sleep disorders and risk of preterm birth in a large cohort of nearly 3 million women. We hypothesized that women with a sleep disorder diagnosis during pregnancy would be at increased risk of preterm birth compared with women without a sleep disorder diagnosis and explored risk by preterm birth type, gestational age, and sleep disorder diagnosis.
MATERIALS AND METHODS
In this observational study, the sample was drawn from all California liveborn neonates between January 1, 2007, and December 31, 2012 (n=3,160,268). The sample was restricted to singleton births with gestations between 20 and 44 weeks of gestation linked to a hospital discharge database maintained by the California Office of Statewide Health Planning and Development (n=2,963,888). This database contains linked birth and death certificates as well as detailed information on maternal and infant characteristics, hospital discharge diagnoses, and procedures recorded as early as 1 year before delivery for the mother and as late as 1 year postdelivery for the mother and infant. Data files provided diagnoses and procedure codes based on the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM).14 Numerous studies have used this database to study risk of preterm birth,15,16 but we note that it is not possible to examine whether records accurately reflect true diagnoses. The study sample was further restricted to children without chromosomal abnormalities or major structural birth defects (n=2,892,756). Structural birth defects for the study were considered “major” if determined by clinical review as causing major morbidity and mortality that would likely be identified in the hospital at birth or lead to hospitalization during the first year of life.17 Finally, the sample was restricted to women without a mental illness or specific mental disorder complicating pregnancy ICD-9-CM code (n=2,770,207) to isolate the effects of sleep disorder diagnosis from the documented effects of mental illness on preterm birth.15
Sleep disorder during pregnancy was defined as an ICD-9-CM diagnostic code for a sleep disorder during a hospital admission during pregnancy or in the delivery hospital discharge record. Sleep disorders were classified into four categories: insomnia, sleep apnea, sleep-related movement disorder, and other (Table 1). Information about how sleep disorders were diagnosed was unavailable; nevertheless, we provide a brief description of each sleep disorder examined. Insomnia is defined as difficulty falling or staying asleep or waking much earlier than intended. Sleep apnea is when breathing is interrupted during sleep. There are two types of sleep apnea: obstructive sleep apnea and central sleep apnea. Obstructive sleep apnea is more common and is caused by airway blockage. Central sleep apnea occurs when the brain does not signal the muscles to breathe. Sleep-related movement disorders include restless legs syndrome, which is marked by urges to move the legs and accompanying unpleasant sensations that can be relieved by movement. The urges worsen during rest and at night, thus having a significant effect on sleep. Hypersomnia is defined as long sleep duration and excessive daytime sleepiness. In circadian rhythm disorder, the timing of sleep is affected; for example, patients may fall asleep very early in the evening and wake early in the morning. Parasomnias are abnormal experiences that occur during the transition to or from sleep or during sleep (eg, sleep walking, sleep terrors). Narcolepsy is marked by excessive daytime sleepiness and sometimes cataplexy (ie, sudden, uncontrollable, transient muscle weakness or paralysis).
Frequency of maternal characteristics for women with a recorded sleep disorder diagnosis was compared with women without a recorded disorder using χ2 testing. Potential confounders were selected from birth certificate records and included year of birth, race or ethnicity, maternal age at delivery, parity, prepregnancy body mass index (calculated as prepregnancy weight (kg)/[height (m)]2), maternal education, insurance coverage for delivery, participation in the Women, Infants and Children’s program, smoking during pregnancy, and previous preterm birth. Additional potential confounders selected from hospital discharge ICD-9-CM diagnoses included those that were noted by the coding hospital as “complicating pregnancy”: hypertensive disorder (ICD-9-CM 642—hypertension complicating pregnancy, childbirth, and the puerperium); infection (ICD-9-CM 646.6—infections of the genitourinary tract in pregnancy; 647—infectious and parasitic conditions in the mother classifiable elsewhere but complicating pregnancy, childbirth, or the puerperium), drug or alcohol dependence (ICD-9-CM 648.3—drug dependence; 304—drug dependence; 305.1–305.9—nondependent abuse of drugs; 303—alcohol dependence syndrome; 305.0—alcohol abuse). Logistic regression was used to create sleep disorder propensity scores in which maternal characteristics with frequencies differing with a P<.05 populated the model. A referent population of women without a recorded sleep disorder diagnosis was randomly selected at a one-to-one ratio using exact matching of propensity scores without replacement (Fig. 1). χ2 testing was used to compare the analyzed population of women with a recorded sleep disorder diagnosis and the unanalyzed population of women with a recorded sleep disorder diagnosis without a propensity score match.
Best obstetric estimate of gestation at birth was obtained from birth certificate records. Gestational age at delivery was categorized as before 34 weeks, between 34 and 36 weeks, and before 37 weeks of gestation. Preterm birth was defined as spontaneous or medically indicated. Spontaneous preterm birth was further subdivided into either preterm prelabor rupture of the membranes (PROM) or premature spontaneous labor with intact membranes. Medically indicated preterm births were those with induction or cesarean delivery before 37 weeks of gestation for whom there was no preterm PROM or premature spontaneous labor with intact membranes. Consistent with previous reports,15 all pregnancies with indication of preterm PROM in the neonate’s birth certificate or mother’s hospital discharge records were included in the preterm PROM group. Women with no indication of preterm PROM, but who had birth certificate or hospital discharge record indication of premature labor or tocolytic medication, were included in the spontaneous labor with intact membrane group. Pregnancies resulting in medically indicated birth were considered to be those without preterm PROM, premature labor, or tocolytic administration and for which there was a code for “medical induction” or “artificial rupture of membranes.” Additionally, medically indicated births included those for which there was a cesarean delivery without any of the aforementioned codes.
Logistic regression was used to calculate odds ratios (ORs) and 95% CIs for preterm birth by gestational age at delivery and by preterm PROM, spontaneous labor, and indicated preterm birth in women with and without sleep disorders compared with women without recorded sleep disorders.
All analyses were performed using SAS 9.4. Methods and protocols for the study were approved by the Committee for the Protection of Human Subjects within the Health and Human Services Agency of the State of California.
In our sample, 2,265 women had a diagnostic code for a sleep disorder during pregnancy. Of the women with a recorded sleep disorder diagnosis, 2,172 (95.9%) had an exact propensity score match among the population without a sleep disorder diagnosis. The analyzed population of women with a recorded sleep disorder diagnosis differed from the unanalyzed population of women with a recorded sleep disorder diagnosis but without an exact propensity score match. For example, the frequency of previous preterm birth, infection, hypertension, diabetes, smoking, drug abuse, and alcohol abuse was higher in the unanalyzed population of women (Table 2).
Women with and without a sleep disorder diagnosis differed statistically by maternal characteristics and obstetric complications. Women with sleep disorders were more likely to be of black race or ethnicity, older than 34 years of age, obese, and have more education than women without a sleep disorder diagnosis. Additionally, women with sleep disorders were more likely to have a history of preterm birth, have a hypertensive disorder, have diabetes or infection, and to smoke or abuse drugs or alcohol during pregnancy (Table 3).
Over 30% of the population with a recorded sleep disorder diagnosis had insomnia and 57% had sleep apnea. Sleep-related movement disorders and other sleep disorders were much less common (7.5% and 5.4%, respectively). Women with a recorded sleep disorder diagnosis were at increased risk of preterm birth (Table 4). Nearly 15% of women with a recorded sleep disorder diagnosis delivered before 37 weeks of gestation (n=318) compared with 10.9% of women without a sleep disorder (n=237; OR 1.4, 95% CI 1.2–1.7). This association appeared to be driven by risk of preterm PROM and spontaneous preterm birth. Among women with a recorded sleep disorder diagnosis, 1.7%, 2.7%, and 0.9% had preterm PROM, spontaneous, and indicated preterm birth at less than 34 weeks of gestation, respectively; 2%, 4.7%, and 2.4% had preterm PROM, spontaneous, and indicated preterm birth 34–36 weeks of gestation, respectively. In contrast, among women without a recorded sleep disorder, 0.6%, 1.6%, and 0.6% had preterm PROM, spontaneous, and indicated preterm birth at less than 34 weeks of gestation, respectively; 1.4%, 4.4%, and 1.8% had preterm PROM, spontaneous, and indicated preterm birth 34–36 weeks of gestation, respectively (Table 4).
Risk varied by preterm birth type, gestational age, and by sleep disorder diagnosis (Table 4). Women with a recorded insomnia diagnosis were nearly twice as likely to deliver before 34 weeks of gestation (OR 1.7, 95% CI 1.1–2.6) compared with women without a sleep disorder diagnosis (Table 4). Risk was highest for preterm PROM at less than 34 weeks of gestation (OR 4.1, 95% CI 2.0–8.3). Sleep apnea was associated with a more than twofold higher risk of preterm birth at less than 34 weeks of gestation (OR 2.2, 95% CI 1.5–3.1) and significantly increased risk of preterm birth at less than 37 weeks of gestation (OR 1.5, 95% CI 1.2–1.8) but not at 34–36 weeks of gestation. More than 15% of women with sleep apnea delivered prematurely. Sleep-related movement disorders and other sleep disorders were rare, and both were only associated with increased risk of preterm PROM at less than 34 weeks of gestation (OR 3.9, 95% CI 1.3–11.9 and OR 4.0, 95% CI 1.1–14.2, respectively).
Women with a recorded sleep disorder diagnosis during pregnancy were at increased risk of preterm birth compared with women without a sleep disorder diagnosis, consistent with hypotheses, and risk varied by gestational age at delivery, preterm birth type, and sleep disorder diagnosis. The current study addresses a critical gap by examining insomnia diagnosis. Women with an insomnia diagnosis had nearly twofold higher risk for an early preterm birth (less than 34 weeks of gestation) compared with women without a recorded sleep disorder diagnosis. The importance of these findings may be considerable given studies showing that more than 50% of pregnant women experience clinically significant insomnia.9 The examination of other sleep disorders (eg, narcolepsy, hypersomnia, parasomnia, and circadian rhythm disorders) also represents a novel contribution to the literature. Considered together, these disorders are associated with increased risk of preterm PROM at less than 34 weeks of gestation.
Our findings reinforce recent meta-analytic and population-based work documenting associations between sleep apnea and preterm birth.10,11 We replicated previous work documenting associations with indicated preterm birth, but also associations with spontaneous preterm birth (less than 34 weeks and less than 37 weeks of gestation). Previous investigations were limited by small sample size18 and inability to control for major confounders.11 Converging with prior evidence, women with sleep-related movement disorders were at increased risk of preterm birth.12 The current study addressed a gap in the literature by examining risk by preterm birth type and found that women with a sleep-related movement disorder diagnosis were specifically at increased risk of preterm PROM before 34 weeks of gestation.
Current findings raise questions about pathways by which sleep disorders increase risk of preterm birth. Inflammation has been proposed as a potential mechanism.19 Sleep disturbance is linked to impaired immune functioning and elevated inflammation among nonpregnant and pregnant populations.20,21 High levels of inflammatory cytokines in amniotic fluid are associated with increased risk of preterm birth.22 Further research is needed to test this hypothesis and the extent to which treatment of sleep disorders during pregnancy may modulate inflammatory activity. Alternatively, sleep disorders may have a more indirect path to preterm birth through increased stress.
There were important study limitations. A small percentage (less than 5%) of women with a recorded sleep disorder diagnosis did not have an exact propensity score match and thus were omitted from logistic regression analyses. Findings indicate that these women had more severe risk profiles and one third had a preterm birth (n=31). The sleep disorder prevalence in the current sample is lower than previous reports,9,23 possibly reflecting underdetection and underdiagnosis. Specifically, the sleep disorder prevalence in the current sample was 0.08% compared with previous work documenting prevalence rates of 12.6% for insomnia diagnosis, more than 50% for clinically significant insomnia symptoms, and 11.1% on a self-report measure of risk for sleep apnea.9,24–26 Those with a sleep disorder on their hospital discharge record may represent patients with the most severe presentation. Future work should examine the effect of disorder severity on preterm birth. By using ICD-9-CM diagnostic codes, we could only infer that the patient had the sleep disorder during pregnancy, but not whether the sleep disorder onset was before or during pregnancy. Future research should examine whether risk varies by disorder onset and chronicity. Information about diagnostic method was unavailable; thus, it is unknown whether clinicians utilized gold standard methods and established clinical criteria to diagnose sleep disorders. Treatment information was also unavailable; this major limitation may be less problematic in a cohort of pregnant women who may be undertreated, particularly for diagnoses perceived to be less urgent.
Although sleep disturbances are common during pregnancy, some women experience clinical disorders with serious implications. Current findings suggest that screening for more severe presentations, particularly insomnia and sleep apnea, would be prudent. Brief, patient-reported screening tools exist, such as the Berlin Questionnaire for sleep apnea (see sleepapnea.org/wp-content/uploads/2017/02/berlin-questionnaire.pdf)27 and the Insomnia Severity Index for insomnia (see https://www.ons.org/sites/default/files/InsomniaSeverityIndex_ISI.pdf).28 There are evidence-based, nonpharmacologic interventions such as continuous positive airway pressure for sleep apnea29 and cognitive behavior therapy for insomnia.30 Cognitive behavior therapy targets the underlying causes of insomnia instead of only treating its symptoms and recently was recommended as the first-line treatment for insomnia in the general population.31 It is available in a number of delivery formats including through the internet.32
In conclusion, findings indicate that sleep disorders during pregnancy are associated with increased risk of preterm birth. More work is urgently needed to test whether this is a causal relationship, identify biological mechanisms, and test the efficacy of interventions for sleep disorders during pregnancy and the effect on preterm birth.
1. Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE, et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet 2015;385:430–40.
2. D'Onofrio BM, Class QA, Rickert ME, Larsson H, Långström N, Lichtenstein P. Preterm birth and mortality and morbidity: a population-based quasi-experimental study. JAMA Psychiatry 2013;70:1231–40.
3. Hamilton BE, Martin JA, Osterman MJ. Births: preliminary data for 2015. Natl Vital Stat Rep 2016;65:1–15.
4. Micheli K, Komninos I, Bagkeris E, Roumeliotaki T, Koutis A, Kogevinas M, et al. Sleep patterns in late pregnancy and risk of preterm birth and fetal growth restriction. Epidemiology 2011;22:738–44.
5. Kajeepeta S, Sanchez SE, Gelaye B, Qiu C, Barrios YV, Enquobahrie DA, et al. Sleep duration, vital exhaustion, and odds of spontaneous preterm birth: a case-control study. BMC Pregnancy Childbirth 2014;14:337.
6. Li R, Zhang J, Zhou R, Liu J, Dai Z, Liu D, et al. Sleep disturbances during pregnancy are associated with cesarean delivery and preterm birth. J Matern Fetal Neonatal Med 2017;30:733–78.
7. Blair LM, Porter K, Leblebicioglu B, Christian LM. Poor sleep quality and associated inflammation predict preterm birth: heightened risk among African Americans. Sleep 2015;38:1259–67.
8. Okun ML, Schetter CD, Glynn LM. Poor sleep quality is associated with preterm birth. Sleep 2011;34:1493–8.
9. Dørheim SK, Bjorvatn B, Eberhard-Gran M. Insomnia and depressive symptoms in late pregnancy: a population-based study. Behav Sleep Med 2012;10:152–66.
10. Xu T, Feng Y, Peng H, Guo D, Li T. Obstructive sleep apnea and the risk of perinatal outcomes: a meta-analysis of cohort studies. Sci Rep 2014;4:6982.
11. Bin YS, Cistulli PA, Ford JB. Population-based study of sleep apnea in pregnancy and maternal and infant outcomes. J Clin Sleep Med 2016;12:871–7.
12. Oyieng'o DO, Kirwa K, Tong I, Martin S, Antonio Rojas-Suarez J, Bourjeily G. Restless legs symptoms and pregnancy and neonatal outcomes. Clin Ther 2016;38:256–64.
13. Slattery MM, Morrison JJ. Preterm delivery. Lancet 2002;360:1489–97.
14. American Medical Association. International classification of diseases, 9th revision, clinical modification. Chicago (IL): American Medical Association; 2008.
15. Baer RJ, Chambers CD, Bandoli G, Jelliffe-Pawlowski LL. Risk of preterm birth by subtype among Medi-Cal participants with mental illness. Am J Obstet Gynecol 2016;215:519.e1–9.
16. Yang J, Baer RJ, Berghella V, Chambers C, Chung P, Coker T, et al. Recurrence of preterm birth and early term birth. Obstet Gynecol 2016;128:364–72.
17. Baer RJ, Norton ME, Shaw GM, Flessel MC, Goldman S, Currier RJ, et al. Risk of selected structural abnormalities in infants after increased nuchal translucency measurement. Am J Obstet Gynecol 2014;211:675.e1–19.
18. Louis JM, Auckley D, Sokol RJ, Mercer BM. Maternal and neonatal morbidities associated with obstructive sleep apnea complicating pregnancy. Am J Obstet Gynecol 2010;202:261.e1–5.
19. Okun ML, Roberts JM, Marsland AL, Hall M. How disturbed sleep may be a risk factor for adverse pregnancy outcomes. Obstet Gynecol Surv 2009;64:273–80.
20. Irwin M. Effects of sleep and sleep loss on immunity and cytokines. Brain Behav Immun 2002;16:503–12.
21. Okun ML, Luther JF, Wisniewski SR, Wisner KL. Disturbed sleep and inflammatory cytokines in depressed and nondepressed pregnant women: an exploratory analysis of pregnancy outcomes. Psychosom Med 2013;75:670–81.
22. Wei SQ, Fraser W, Luo ZC. Inflammatory cytokines and spontaneous preterm birth in asymptomatic women: a systematic review. Obstet Gynecol 2010;116:393–401.
23. Mindell JA, Cook RA, Nikolovski J. Sleep patterns and sleep disturbances across pregnancy. Sleep Med 2015;16:483–8.
24. Kizilirmak A, Timur S, Kartal B. Insomnia in pregnancy and factors related to insomnia. ScientificWorldJournal 2012;2012:197093.
25. Okun ML, Buysse DJ, Hall MH. Identifying insomnia in early pregnancy: validation of the Insomnia Symptoms Questionnaire (ISQ) in pregnant women. J Clin Sleep Med 2015;11:645–54.
26. Facco FL, Kramer J, Ho KH, Zee PC, Grobman WA. Sleep disturbances in pregnancy. Obstet Gynecol 2010;115:77–83.
27. Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med 1999;131:485–91.
28. Morin CM. Insomnia: psychological assessment and management. New York (NY): Guilford Press; 1993.
29. Sharples LD, Clutterbuck-James AL, Glover MJ, Bennett MS, Chadwick R, Pittman MA, et al. Meta-analysis of randomised controlled trials of oral mandibular advancement devices and continuous positive airway pressure for obstructive sleep apnoea-hypopnoea. Sleep Med Rev 2016;27:108–24.
30. Trauer JM, Qian MY, Doyle JS, Rajaratnam SM, Cunnington D. Cognitive behavioral therapy for chronic insomnia a systematic review and meta-analysis. Ann Intern Med 2015;163:191–204.
31. Qaseem A, Kansagara D, Forciea MA, Cooke M, Denberg TD; Clinical Guidelines Committee of the American College of Physicians. Management of chronic insomnia disorder in adults: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2016;165:125–33.
32. Ritterband LM, Thorndike FP, Ingersoll KS, Lord HR, Gonder-Frederick L, Frederick C, et al. Effect of a web-based cognitive behavior therapy for insomnia intervention with 1-year follow-up: a randomized clinical trial. JAMA Psychiatry 2017;74:68–75.