The aim of present study was to investigate the effect of lullaby intervention on anxiety and prenatal attachment in women with high-risk pregnancy. The study involved women with high-risk pregnancy who were assigned randomly to the intervention (n=30) or control (n=30) groups in a state hospital in Turkey. For two successive days, the intervention group listened to lullabies for 20 minutes once a day, meanwhile touched the abdomen and thought about their babies, but the control group did not. Data from the outcomes of anxiety and attachment were collected at baseline and at the end of the second day. Vital findings were measured on each study day, both before and after the lullaby intervention/usual care.

Women with high-risk pregnancy may experience anxiety and low prenatal attachment due to many problems with pregnancy and hospitalization. It is reported that anxiety can have negative effects on prenatal attachment. The anxiety during pregnancy may adversely affect the health of the pregnant woman and fetus.

During the study period, 777 pregnant women were admitted to the clinic and 125 of them with high-risk pregnancy were assessed for eligibility for the study on 2 days a week. The pregnant women (n=76) who met the inclusion criteria and agreed to participate in the study were recruited. Participants were randomized in equal numbers to the intervention (n=38) and control groups (n=38) by withdrawing computer randomized sampling numbers (using Microsoft Excel 2010) from concealed envelopes to determine their group status. Participants were numbered according to their order of admittance to the hospital. After obtaining informed content and collecting baseline data, main researcher randomly chose one code for each woman. Data was collected together until the sample size was reached.

Participants were recruited by simple random sampling from the perinatology clinic at a state hospital in Turkey. An estimation of sample size was calculated at an effect size of 2.43 with a 5% level of significance on the primary outcome of anxiety in a sample study. According to G*power 3.1 31, each group required 30 participants. During the study, 125 women were assessed for eligibility, 49 of them were excluded. Some participants withdrew due to preterm birth (n=5), referral to another hospital (n=1) or discharged from hospital (n=10). Finally 60 participants who completed the lullaby intervention were included in data analysis. As a result of the study, the power of the test (α=0.05, 95% confidence interval) was 99.9% for each group n=30.

The preliminary study included five women at risk of preterm delivery who met the inclusion criteria. They were subjected to a lullaby intervention, and their study variables were evaluated. The women in the preliminary study were not included in the main study.

The IG listened to lullabies, and accompanied by lullabies touched their abdomen and thought about their babies. The CG received only the usual care of the hospital.

The lullaby recital consisted of 10 different lullabies with a mean duration of two minutes, for a total of 20 minutes.The lullabies were among the best-known ones from the Turkish culture. At the place of study, an academician of the Music Education Department of the University, Fine Arts Faculty was consulted regarding the lullaby recital.The pregnant women in the intervention group listened to the lullabies for 20 minutes per day, for two successive days using an MP3 player. To provide a homogeneous intervention and to avoid bias, the personal lullaby preferences of the pregnant women were ignored.

This intervention was performed in single patient rooms during bed rest, when the participants were available and felt ready. The doors and windows of the room was closed, and the participants were left alone. A disposable ear-tip was used for each participant. Before the initiation of the recital, the MP3 player was checked and the participants were instructed in the operation of the MP3 player (turning on, turning off, adjusting volume). The tempo of the lullaby was concordant with the normal heart rate (about 60-70 beats per minute), the participant determined and adjusted the volume to suit.

The control group received only the usual care of the hospital, resting in reclined position. All of the participants in both groups were ensured to have empty bladders and closed their cellphones, and none received any medical interventions during the lullaby intervention/usual care.

The study data was collected between 11:30 and 13:30 after lunch when no medical care, diagnostic tests or routine follow-up were administered. Initially, sociodemographic and obstetric data were collected by face-to-face interviews with participants. Intervention effect outcomes were anxiety and attachment which were measured at baseline. Anxiety was measured by the State-Trait Anxiety Inventory (STAI) and attachment was measured by the Prenatal Attachment Inventory (PAI). Before pregnant women filled the STAI and PAI, the instructions were read aloud to them and then they responded the questionnaires in silence. The STAI and PAI were reassessed at the end of the study (day 2) after the lullaby intervention/usual care. Vital findings were also measured on each study day, both before and after the lullaby intervention/usual care. The systolic and diastolic blood pressures and pulse rates of all participants were measured using a digital sphygmomanometer. The respiratory rate was measured by observing the chest and count the number of respirations out loud for one full minute using a digital watch while at bed rest (the same device was used for each participant). The level of satisfaction with listening to lullabies was assessed after the lullaby intervention at the end of the study using a 10-point Likert-type scale.

After the purposes and procedures of this study was explained, signed informed consent was obtained from all participants. Participants were assured of privacy and that their identity was protected. The study was based on voluntariness. Usual care and treatment of pregnant women in each groups were not intervened during the study period.

Data were analyzed by IBM SPSS v.23. Categorical data were analysed using the chi-square test for homogeneity of characteristics between two groups. Parametric tests were used if the data was normally distributed, but if not, nonparametric tests were used. A Pearson Chi-square test, Fisher's exact test, and Yates' continuity correction were used to compare the categorical data. The continuous data was compared using a Student's t or Mann-Whitney U-tests. A Paired sample t-test or Wilcoxon Signed Rank test were used to compare the paired scores of the STAI and PAI and the vital findings within the groups. The relationship between these three parameters was analyzed using a Spearman-correlation test. Statistical significance was considered p<0.05.

There are also some limitations in this study. Firstly, only state anxiety of the pregnant women was assessed, and future studies would benefit from taking also trait anxiety into consideration. It was a single-center study; therefore, the generalizability of the results to other women with high-risk pregnancy is unclear. Since the perinatology clinic where the study was carried out was newly opened, there were some disruptions and most pregnant women were transferred from this clinic to other clinics of hospital and discharged from hospital before the required hospital stay; therefore, the study could only be performed with pregnant women hospitalized for 3 days.

Inclusion Criteria:

• Hospitalization due to high-risk pregnancy
• 19 years and older
• Pregnancy duration longer than 28 weeks.
• Singleton pregnancy.
• Duration of hospitalization 24 hours minimum.
• Ability to read and write and to comprehend.
• Speaking Turkish.

Exclusion Criteria:

• Hospital admission due to severe preeclampsia, eclampsia or ablatio placenta.
• Presence of mental or cognitive disorders.
• Being under psychiatric treatment.
• Vision or hearing impairments.
• Fetal deformity or anomalies such as congenital diseases