We conducted a cross-sectional survey. A self-composed questionnaire was distributed via internet between February and March 2018. It was dedicated to pregnant Polish-speaking women, regardless of inhabitancy and was distributed by web pages and Facebook groups, which were designed for pregnant women. Web pages written in Polish language provided general medical and social information on pregnancy and were available without any limits. The questionnaire was administered by Google Forms (Google LLC, Mountain View, California, US). Link to the survey was presented on web pages providing general information on pregnancy and pregnancy ailments Facebook groups were accessible for women declaring being pregnant without any special requirements. Neither Facebook groups nor web pages were advertised at the time of the survey. On websites and Facebook group, a short information on sleep problems in pregnancy and its importance was presented with an invitation to complete the survey.
Women aged over 18 years old and pregnant at the time of the survey were included in the study. Only questionnaires which were filled in completely were considered in the analysis. Women declaring having sleep problems before the current pregnancy were also included in the analysis. Women in multiple pregnancies were excluded.
The first part of the questionnaire included questions regarding sociodemographic data, information on the current pregnancy, including pregnancy ailments, and ones that characterized sleep patterns like sleep duration, occurrence of daytime naps or factors deteriorating sleep quality. Gestational age was calculated on the basis of the first day of the last menstrual period or the transfer day in assisted reproduction technique procedures reported by the respondents. In case of differences in gestational age exceeding 5 days after the verification of the crown–rump length (CRL) measured on the first-trimester scan, gestational age calculated on the basis of CRL was used in further analyses.
The second part of the questionnaire contained PSQI, which is commonly used to measure sleep quality and patterns of sleep in adults [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]. The PSQI differentiates “poor” from “good” sleep by measuring seven components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction over the past four weeks. The PSQI includes questions concerning usual time in bed, number of minutes needed to fall asleep, usual time of getting up, hours of sleep per night, and factors which deteriorated sleep quality over the past four weeks. Medication use, daytime alertness, snoring and pauses between breaths are also assessed. The answers are given using a 4-grade scale, which describes the frequency of incidents: 0—the were no incidents over the past 4 weeks, 1—less than once a week, 2—once or twice a week, 3—three times a week or more. The results are evaluated regarding the above seven components, assessed from 0 to 21 points in total. Results higher than 5 points indicate low sleep quality and differentiates patients into with “poor” and “good” sleep [11, 14, 16,17,18]. The PSQI questionnaire is characterized by high in-ternal consistency, with Cronbach’s alpha index of 0.83 . Its high reliability and validity were observed in numerous populations of pregnant women around the world: Hungarian, Italian, Brazilian, Peruvian, Serbian, Persian, Pacific-Northwest or Georgian (USA) [18,19,20,21,22,23,24]. The Polish version of the PSQI was validated and previously used by other researchers .
Life stress level and satisfaction with the current relationship were assessed with a visual numerical scale from 0 to 10. Women over the age of 18, currently pregnant and speaking Polish were included in the study. No specific criteria other than singleton pregnancy at the time of the survey were used for recruitment for the study. The survey was voluntary and anonymous—no questions regarding personal data that would enable the identification of participants were included and only the authors of the study had access to the collected information. Access to the questionnaire was granted after login only once. The respondents were asked to describe their sleep experience over the past four weeks. Only fully completed questionnaires were comprised in the analysis. The answers were double-checked by the researchers and there were no identical records.
Data were reported as absolute numbers and percentages or means and standard deviations. Statistical analyses were performed using Statistica version 13.3. The Mann–Whitney test or Fisher exact tests were used to compare the analyzed variables. All tests were two tailed and p < 0.05 was considered significant. Logistic regression analysis was performed to determine the independent risk factors of sleep problems during pregnancy.
The study protocol obtained the approval of the Ethics Committee of the Medical University of Warsaw (no. 124/2018). The committee retreated from the obligation of having a written or verbal consent to participate in the study. Completing the questionnaire was deemed tantamount to giving consent.
A power analysis was performed to assess the minimal number of respondents. For the 90% power, with a value of the zero hypothesis 90% (as basing on the available published Polish data, we assumed that 90% of pregnant women declare having any sleep problems), 1000 women should be included in the study (G*Power 18.104.22.168, Franz Faul, Christian-Albrecht University, Kiel, Germany).
The results of the study concerning the subjective assessment of sleep quality conducted in the same study population were presented in previously published paper .