Impact of Birthing Room Design on Maternal Childbirth Experience: Results From the Room4Birth Randomized Trial

Objective: To study the effect of the birthing room design on nulliparous women’s childbirth experience up to 1 year after birth. Background: Although it is known that the birth environment can support or hinder birth processes, the impact of the birthing room design on maternal childbirth experience over time is insufficiently studied. Methods: The Room4Birth randomized controlled trial was conducted at a labor ward in Sweden. Nulliparous women in active stage of spontaneous labor were randomized (n = 406) to either a regular birthing room (n = 202) or a new birthing room designed with more person-centered considerations (n = 204). Childbirth experiences were measured 2 hr, 3 months, and 12 months after birth by using a Visual Analogue Scale of Overall Childbirth Experience (VAS-OCE), the Fear of Birth Scale (FOBS), and the Childbirth Experience Questionnaire (CEQ2). Results: Women randomized to the new room had a more positive childbirth experience reported on the VAS-OCE 3 months (p =.002) and 12 months (p =.021) after birth compared to women randomized to a regular room. Women in the new room also scored higher in the total CEQ2 score (p =.039) and within the CEQ2 subdomain own capacity after 3 months (p =.028). The remaining CEQ2 domains and the FOBS scores did not differ between the groups. Conclusions: These findings show that a birthing room offering more possibilities to change features and functions in the room according to personal needs and requirements, positively affects the childbirth experience of nulliparous women 3 and 12 months after they have given birth. This study reports women’s childbirth experiences 2 hr, 3 months, and 12 months after participation in the Room4Birth randomized controlled superiority trial (RCT) in Sweden (Goldkuhl, Gyllensten et al., 2022). Women were randomized to give birth in either (i) a regular hospital birthing room (regular room, control group) or (ii) a refurbished room designed with more person-centered considerations (new room, intervention group). The trial was conducted in accordance with the Declaration of Helsinki (World Medical Association, 2013), registered at ClinicalTrial.gov ({“type”:”clinical-trial”,”attrs”:{“text”:”NCT03948815″,”term_id”:”NCT03948815″}}NCT03948815), and ethically approved by the Regional Ethics Board (Dnr:478-18). The study procedures for the RCT followed the CONSORT guidelines and are described in detail in a study protocol (Berg et al., 2019). Randomization was undertaken between January 2019 and October 2020 at a labor ward located in western Sweden and with an annual birth rate of around 4,000 births (Swedish Pregnancy Register, 2022). All randomized participants had access to the same level of professional support and labor analgesia regardless of allocated group. The birthing rooms in the study labor ward were also equally equipped with medico-technical devices. What differed between the two randomized groups was that the new room provided users with more opportunities to adjust features and functions in the room compared with in the seven, similarly designed regular birthing rooms (Figure 1). Women and companions in the new room could control different aspects of the room, such as the degree of the dimmable light and the position of the bed. The women also had access to a large bathtub and more options for upright birth position. Additionally, the new room was designed to induce the feelings of calmness and familiarity through a less clinical-like decor. For instance, the medico-technical devices in the room were covered by sliding wooden panels that could be raised if preferred or needed (Figure 2). Unlike in the regular rooms, users of the new room had access to programmed nature scenes displayed on two of the walls, combined with either classical music or nature sounds. There was also an entrance hall with a green-colored curtain that protected the new room from being seen from the hospital corridor outside. Photos of (A) one of the regular birthing rooms and (B) the new birthing room. New birthing room where the medico-technical devices were hidden behind wood-paneled walls (A) that could be rolled up if required (B). Eligible participants for randomization were women aged 18 or more, classified as Robson 1 (Robson et al., 2015); nulliparous with a single, live, cephalic fetus at term and with spontaneous onset of labor. Participants also needed to be in active stage of labor as defined in Sweden by the time of recruitment (two of the following three criteria fulfilled at labor ward admission: two or three labor contractions in 10 min, spontaneous rupture of membranes, and cervix dilated >3 cm or effaced and open >1 cm). Furthermore, only women who understood either Swedish, English, Arabic, or Somali or had access to an interpreter could be recruited. Women in induced or latent phase of labor or with a planned caesarean birth were excluded. Participant recruitment was carried out by the midwives and assistant nurses at the labor ward. Eligible women were asked to participate after labor ward arrival and confirmed active stage of labor, if both a regular room and the new room were vacant. All women gave signed, written consent to participate, and the time interval from labor ward arrival to randomization was set to be as short as possible to ensure an early allocation to the randomized room. The randomly computer-generated allocation list was managed by an independent agency. Information about the allocated room was printed in sealed, opaque envelopes sequentially placed in a study box at the labor ward. The recruiting care provider was not aware of the randomization sequence, and the independent agency ensured that the allocation list was followed. The women were provided with a four-digit ID code, which was printed on the sealed envelope. Participants’ demographic data were collected from obstetric records and through a self-reported, digital questionnaire 2 hr after birth (Follow-Up 1), while they were still in the birthing room. The questionnaire included a Visual Analogue Scale of Overall Childbirth Experience (VAS-OCE 1-10) and a modified Fear of Birth Scale (FOBS 0-100) (Haines et al., 2011; Hildingsson et al., 2017) and was available in Swedish, English, Arabic, and Somali (Supplementary Information 1). Data from Follow-Up 1 have been reported in an earlier publication (Goldkuhl, Gyllensten et al., 2022) but were also used in this study to measure the difference in childbirth experience over time. After 3 (Follow-Up 2) and 12 months (Follow-Up 3), all participants were asked to complete another online questionnaire (Supplementary Information 2) sent to their email address with two reminders to nonresponders. These questionnaires included VAS-OCE, FOBS, and the Childbirth Experience Questionnaire version 2 (CEQ2) (Dencker et al., 2020; Walker et al., 2020). A data collection overview is shown in Figure 3. The time intervals of follow-up were chosen based on research, showing that the experience of childbirth changes over time (Lundgren, 2005; Waldenström, 2003, 2004). It is, however, difficult to determine the most reliable time of measuring childbirth experience since it is a multidimensional and complex phenomenon. Reflections from a more long-term perspective may, nonetheless, be favorable since they tend to be more nuanced (Lundgren, 2005). Overview of data collection. Participants’ self-reported experiences on a Visual Analogue Scale of Overall Childbirth Experience (VAS-OCE), Fear of Birth Scale (FOBS), and Childbirth Experience Questionnaire version 2 (CEQ2). Assessing overall childbirth experience by using a VAS (continuous data) ranging from 1 to 10, where 10 is most positive, is routinely used in Sweden (Swedish Pregnancy Register, 2022). In the follow-up questionnaires 2 and 3, the participants self-reported their overall experience on a 100-mm linear analogue scale and not 1–10 as in Follow-Up 1, since it was in congruence with the linear scales in the other instruments (CEQ2 and FOBS). Since VAS most often is used to measure pain, we chose to name the variable VAS-OCE to clarify that we measure overall childbirth experience. To assess fear during birth and of a potential future birth, a modified FOBS was used. In its original form (Haines et al., 2011; Hildingsson et al., 2017), participants are asked to put a mark on two 100-mm linear analogue scales during pregnancy, which answers the question: How do you feel right now about the approaching birth? Answers are rated from calm to worried (0–100) and from no fear to strong fear (0–100), and the scales are summed and averaged to calculate a score. Since participants for this study answered the questionnaire after birth, we modified the FOBS to report the question retrospectively and prospectively: How do you rate worry and fear when you think about your completed labor and birth? and How do you rate worry and fear when you think about giving birth again?. Higher scores in the FOBS represent stronger childbirth fear. As suggested by the creators, a cutoff point of 60 and above was used to define the fear of childbirth (Hildingsson et al., 2017). The CEQ2 (Dencker et al., 2020; Walker et al., 2020) is a validated instrument containing 22 questions in four different domains (own capacity, perceived safety, professional support, and participation; Supplementary Information 2) concerning the childbirth experience. For 19 of the CEQ2 items, the response format is a 4-point Likert-type scale ranging from 1 = totally disagree, 2 = mostly disagree, 3 = mostly agree to 4 = totally agree. For the remaining three items (perceived pain, control, and sense of security), answers are assessed on a linear analogue scale (0–100), which are subsequently transformed to categorical variables: 1 = 0–40, 2 = 41–60, 3 = 61–80, and 4 = 81–100. The items produce mean scores in the four domains as well as a total CEQ2 score. Higher scores in the CEQ2 reflect more positive experiences. The few women who chose to answer the self-reported questionnaire at 2 hr in Arabic (two women in the new room and none in the regular room) or Somali (one woman in the new room and none in the regular room) received the questionnaires at 3 and 12 months in Swedish, since there were no validated translations for FOBS and CEQ2 in these languages. To compare differences between the two randomized groups, χ2 test was used for nonordered categorical variables and Fisher’s exact tests for dichotomous variables. For group comparisons of the continuous variables included in the instruments (VAS-OCE, FOBS, and CEQ2), Mann Whitney U test was used due to nonnormally distributed data. We also estimated effect sizes r, as proposed by Cohen (1988) to examine the degree of variance between the groups, where r > .1 is considered a small effect, r > .3 medium effect, and r > .5 large effect (Fritz et al., 2012). Results for categorical variables are presented with n (%), and for continuous outcome variables as mean (standard deviation), median, quartile 1 and 3, and 95% confidence intervals for mean. Cronbach’s α was used to evaluate the internal consistency reliability of the domains included in the CEQ2 questionnaire. A linear mixed effects model was used to describe the group differences in childbirth experience over the three time points. This regression model was chosen since it can provide information about the women’s individual change in childbirth experience over time but is also flexible since it has the capability to handle missing observations in the repeated measures data (Griffiths et al., 2017). The model examined the fixed effects of time and randomized group as well as the random effect of time on each of the dependent variables represented in VAS-OCE, FOBS, and CEQ2. A subsequent analysis controlling for the use of epidural analgesia and oxytocin augmentation during labor were conducted, due to the detected group differences in these variables and since these are factors known to affect the childbirth experience (Hildingsson et al., 2021). The analyses were performed using SPSS v.28 (SPSS IBM Statistics) for the comparisons between the two randomized groups and Stata (17.0, StataCorp LLC, College Station, TX) for the mixed effects model. All analyses were conducted according to the intention-to-treat methodology, and the significance tests were two-sided and had a significance level set at 5%.