The present study was part of the Neonatal and Obstetric Risk Assessment (NORA) study, a large prospective cohort study. Women with singleton pregnancies attending their first prenatal visit at KK Women’s and Children’s Hospital, Singapore, between September 1, 2010, and August 31, 2014, were invited to participate. All women with viable singleton pregnancies (confirmed by ultrasonography) and at less than 14 weeks of gestation were potentially eligible. Patients with chronic medical conditions such as renal disease and systemic lupus erythematosus, pregnancies complicated by aneuploidy or fetal anomalies, and pregnancies ending in termination, spontaneous abortion, or fetal death before 24 weeks of gestation were excluded. Pregnancy duration was determined from the menstrual history and was confirmed by fetal crown–rump length at the first prenatal ultrasonography examination. Enrolment details were described elsewhere22. Briefly, of the 3271 women identified, 1013 women consented to the study and 934 women completed all follow up visits (92%). 926 women had complete delivery record (96%). At recruitment, a detailed interview, a dating ultrasound scan and routine antenatal blood collection were done. At subsequent follow up visits, clinical and laboratory data were collected at 18 to 22 weeks, 28 to 32 weeks and 34 weeks and above. Following delivery, detailed information on pregnancy complications, labor, delivery and neonatal outcomes was collected. The present study mainly focused on patient demographics, sleep quality throughout pregnancy and BMI.
Information on socio-demographic and life style characteristic was collected during the 1st antenatal visit and updated during each subsequent visit when applicable. Included were age, BMI, race, parity, marital status, education level, occupation and monthly income. For assessment of effect of BMI on sleep quality, subjects were grouped into underweight (<18.5kg/m2), healthy weight (18.5 - <25 kg/m2), overweight (25 - < 30 kg/m2) and obese (30 kg/m2 and above) based on the BMI in the first visit.
Four study visits at the different gestational ages of pregnancy were conducted in the NORA Study (1st visit at 11-14 weeks, 2nd visit at 18-22 weeks, 3rd visit at 28-32 weeks and 4th visit after 34 weeks of gestation). During each visit, sleep quality was assessed via the Pittsburgh Sleep Quality Index (PSQI), which is a widely used and well-validated 19-item self-report questionnaire that measures sleep quality in adults23. It provides a global score ranging from no sleep difficulty to severe difficulties. Higher PSQI score indicates poorer sleep. Four main questions were asked in the questionnaire: (1) “During the past month, what time have you usually gone to bed at night”; (2) How long (in minutes) has it usually taken you to fall asleep each night”; (3) What time have you usually got up in the morning”; (4) How many hours of actual sleep did you get at night, which may be different from the number of hours you spent in bed”. Sleep duration is defined as question 4. Sleep latency is defined as question 2. Sleep efficiency is defined as sleep duration / hours in bed. Questions on whether there is difficulty in falling asleep, night awakening, bathroom use at night were also included in the questionnaire. PSQI score was calculated after adding the scores for each question.
Maternal demographics such as age and BMI were considered as continuous variables and were described as mean ± standard deviation. Other factors including race, education, marital status, occupation, income and parity were considered as categorical variables and were described as the sample number and frequency of each category. The mean PSQI, latency, duration and efficiency of sleep for each visit were described as mean ± standard deviation. Mixed model analysis was employed to study the change of sleep quality at different stages of pregnancy. Analysis was adjusted for age, race, education, coffee drinking, Smoking, exercise, SBP and parity. Repeated Covariance type of Mixed model analysis was Scaled Identity. Statistical analysis of the data was performed using SPSS 22.0 (IBM, Armonk, NY, USA).