Childbirth experience in women participating in a continuity of midwifery care project

Background: Continuity models of care are rare in Sweden, despite the evidence of their bene ﬁ t to women and babies. Previous studies have shown certain factors are associated with a positive birth experience, including continuity of midwifery care. Aim: The aim was to investigate women's childbirth experiences in relation to background data, birth outcome and continuity with a known midwife, in a rural area of Sweden. Methods: An experimental cohort study. Participating women were offered continuity of midwifery care in pregnancy and birth, during selected time periods. Data were collected in mid-pregnancy and two months after birth. The Childbirth Experience Questionnaire was used to determine women's birth experiences. Result: A total of 226 women responded to the follow-up questionnaire. Not living with a partner, fear of giving birth, and a birth preference other than vaginal were associated with a less positive birth experience. Having had a vaginal birth with no epidural, no augmentation and no birth complication all yield a better birth experience. Women who had had a known midwife were more likely to have had a positive birth experience overall, predominantly in the domain Professional support. Conclusions: The results of this study showed that women who received care from a known midwife in labour were more likely to have a positive birth experience. The results also pointed out the bene ﬁ ts of a less medicalized birth as important for a good birth experience, and that some women may need extra support to avoid a less positive birth experience. ©


Background
A positive childbirth experience is important to women globally and has been acknowledged by the World Health Organization (WHO) [1] as an experience that fulfils women's sociocultural beliefs and expectations.Having a healthy baby is very important, but so is intrapartum care delivered with continuity and emotional support [1].Previous studies have identified certain aspects associated with a positive birth experience, such as women's prenatal expectations [2,3], having a normal vaginal birth [4,5], and using non-pharmacological pain relief methods [5,6].Other important factors are the possibility to communicate birth preferences beforehand [7] and support from the midwife [8].Taheri and co-workers reported, in a systematic review and meta-analysis of 20 trials comprising 22,800 participants from 12 countries, that support during labour, birth preparedness, and a minimal amount of intervention during labour were the most important factors leading to a positive birth experience [9].Support during labour has also been acknowledged by others to be important for a positive childbirth experience [5,10] and so have continuity models of midwifery care [10].
Continuity models of midwifery care have been proven beneficial to women and babies [12] and are highly recommended by WHO as one important factor when it comes to creating a positive birth experience [1].Continuity models usually provide woman-centred and individualized care from a known midwife or a small group of midwives, throughout pregnancy and birth.Such models generally create positive experiences of pregnancy and birth [12][13][14].In a recent systematic review and meta-synthesis of 13 qualitative studies, Perriman et al. [14] concluded what women value in continuity models.The midwife-woman relationship was the overarching theme underpinned by trust, personalized care, and empowerment.
Continuity models of midwifery care are growing internationally in countries like New Zealand [15,16], Australia [13,14,17,18], the United Kingdom [19], and Denmark [20,21], but are mainly accessible in metropolitan areas [22].Research about continuity models from rural areas is sparse, but some studies have found continuity models in rural areas that are considered safe and sustainable [23], and collaborative [24].In Sweden, where midwives are the primary caregivers for women with uncomplicated pregnancies and births, continuity during the antenatal period is good, with the majority of women meeting one or two midwives during pregnancy [25].Continuity throughout antenatal, intrapartum and postpartum periods is offered in only a few places.One such initiative is presented in this paper.
Barriers to the implementation of continuity models have been identified as medical dominance of maternity services [26], financial issues, and workforce shortages [27,28].It has been reported that staff who have not been exposed to the particulars of continuity models, such as caseload, can have difficulty understanding the value of such models [22,23].
The safety and quality of the service offered to pregnant women and their families is another fact that hampers women's birth options.In most industrialized countries there is an ongoing centralization of labour wards to tertiary facilities, with limited consideration about women's birth preferences [29,30].Research has shown that a safe birth environment for women includes the presence of a known caregiver and closeness to family and community support [27].The attempt to centralize births contradicts the fact that birth usually is safe for low-risk women provided with midwife-led models of care [31], including those in rural areas [23].Long travel distances to labour wards could cause poorer health outcomes for women and babies, when they are being forced to travel on dangerous roads under bad weather conditions.Roadside births without professional assistance [32] as well as planned unassisted births [33], might be consequences of the centralization.As noted in a study from Finland, living more than 35 km from the labour ward was one of the risk factors found in out-of-hospital births that could increase infant morbidity [34].

Problem area
Previous research has shown that continuity models of midwifery care are associated with higher birth satisfaction, but research on continuity models in rural areas of Sweden is lacking.The aim of this study was to investigate women's childbirth experiences in relation to background data, birth outcome, and continuity with a known midwife, in a rural area of Sweden.

Design
This was an cohort study of women recruited in early pregnancy and followed up two months after birth.Participating women were offered continuity of midwifery care in pregnancy and birth, during selected time periods.

Context of midwifery in Sweden
In Sweden, midwives are the primary health providers for pregnant women during antenatal, intrapartum and postpartum care, regardless of risk and the care is free of charge.Antenatal care is usually offered in outpatient clinics and is organised within the primary health sector.During a normal pregnancy, women meet with their allocated midwives for 8-9 visits.The midwife working in antenatal care also meets women who are not pregnant when providing reproductive health care.For intrapartum care, midwives work in collaboration with obstetricians in hospital-based labour wards.Labour wards are organised within specialist hospital based care.There are few homebirths in Sweden and not financed by the government.A midwife working in a labour ward is independently responsible for normal births but is present during all births, including caesarean sections.In small and middle-sized hospitals, it is common that midwives to rotate between assisting women during labour and birth, postnatal care and taking care of gynaecological patients.Usually midwives work either in antenatal or in intrapartum/postpartum care.Continuity models of care do not exist.

Setting
The study was conducted in Sweden, at a hospital-based antenatal clinic in a rural area, and two referral hospitals with annual rates of 1700 and 750 births, respectively.Shortly before the start of the study, the labour ward situated at the rural hospital had been closed.The travel distances to the remaining hospitals were around 120 and 100 km, respectively, from the antenatal clinic.
Four midwives provided antenatal care from March 2017, and were on call for births every day between 7 a.m. and 11 p.m. during the period from 1 August 2017 to 30 June 2019.The reason for not being on call 24/7 was due to not having enough staff to cover on call throughout the night.After working 12 h the project midwife was replaced by a ward midwife, due to work-time regulations.The midwife on call had access to a mobile phone with excellent coverage (similar to mobile phones used by paramedics/health staff in the mountains) and a four-wheel-drive car with a birth kit, in case of emergency births.Women travelled with their partners in their own car, as midwives are not allowed to transfer 'patients' in their own car.In cases where the woman had a very quick birth, an ambulance was called.

Recruitment of participants
Information about the project was available in the antenatal clinic's waiting room and on web pages, and was frequently reported in the media.Women interested in the model of care contacted the project midwives by telephone and were provided with information.Those who consented to participate in the study were thereafter assigned a primary midwife and a time for the first booking visit.The project participants had the opportunity to meet the other project midwives at the visits during pregnancy or in parent education classes.All women who mastered the Swedish language well enough to be able to communicate by telephone were invited, but some women were highly prioritized, namely younger women, single women, women expecting their first baby, and women who feared childbirth.The basis for the priority was made due to previous studies indicating that these women would benefit more from the continuity [30].Another important intention was to invest in first-time mothers, as continuity models are recognized to improve the birth outcome in primiparous women [35].

Data collection
Data were collected by means of two questionnaires, one in mid-pregnancy and the other two months after birth.The questionnaires were sent to each woman's home address with a prepaid response envelope.Two reminders were later sent by text messages.Women who did not respond to the second questionnaire after three months were invited to complete the follow-up questionnaire online, using the Netigate platform.The platform could be reached from a computer, mobile phone, or tablet.
Data collected from the first questionnaire comprised background questions (age, parity, country of birth, marital status, level of education), self-reported physical and mental health, birth preferences and questions about fear of childbirth.The Edinburgh Postnatal Depression Scale (EPDS) was used to identify women with depressive symptoms.EPDS is a 10-item self-report scale, where each item is assessed on a four-point scale (0-3) and the total ranges from 0 to 30.In the present study the cut-off of 13 or more was used, as recommended when used during pregnancy [36].Fear of birth was assessed using the Fear of Birth Scale (FOBS).The FOBS scale consists of two 100 mm Visual Analog Scales that are summed and averaged to get a score.When filling out the scale study participants are asked to respond to the question "How do you feel right now about the approaching birth?" and are instructed to place a mark on the two scales which have the anchor words calm/worried and no fear/strong fear.The cut-off point of 60 or more was used to classify women with fear of birth [37,38].
The second questionnaire was sent out two months after birth.Variables collected were gestational week of birth, onset of labour, mode of birth, birth complications, pain relief used, and augmentation, and whether the woman had a known midwife or not assisting.Women's birth experiences were investigated using the original version of the validated Childbirth Experience Questionnaire (CEQ) [39].The CEQ contains 22 items in four domains related to the birth experience: Own capacity,Professional support, Perceived safety, and Participation.Some items are rated on a 4-point Likert scale (1 = totally disagree, 4 = totally agree); three items are assessed with visual analogue scales, which were later transformed into categorical variables following the instructions of the CEQ's creators [39].The domain Own capacity contained eight items including women's feelings, such as 'I felt tired during labour and birth'.Professional support investigated the experience of midwifery care (five items) like 'My midwife understood my needs'.The domain Perceived safety focused on memories and the skill of the staff at the hospital and included six items.The domain Participation comprised three final items about decision-making and available choices.

Analysis
The scores were summed for each domain and also for the total score, with higher scores indicating higher satisfaction with birth.Differences in mean scores and standard deviations were calculated using Student's t-test or the analysis of variance (ANOVA) for the domains of the CEQ in relation to the women's background characteristics.Reliability was checked using Cronbach's alpha values for each domain, with a range of 0.81-0.91.
Effect sizes for explanatory variables that yielded statistically significant differences in the domains were calculated using Eta 2 [40], where 0.01 is a small effect, 0.06 a moderate effect, and 0.14 a large effect.SPSS version 25 was used in the analysis (SPSS Inc., Chicago, IL, USA).The study was approved by the Research Ethics Committee at the Regional Ethical Review Board, DNR 2017/ 120-31.

Results
During the study period, 314 women expressed an interest in the continuity project and consented to participate.Twenty-three women had a miscarriage, and 13 withdrew participation during pregnancy.From the 278 distributed follow-up questionnaires, 236 (85%) were returned, 192 on paper and 44 via the Netigate online database.They correspond to 75% of those who originally consented to participate.The 42 women not responding to the follow-up were more likely to be born in a country outside Sweden (p 0.000) and also more likely not to have completed the questionnaire in mid-pregnancy (p 0.000).No other background differences were found.Continuity of a known midwife during labour and birth was achieved by a total of 86 women, according to the midwives' notes, but only 77 women who had had a known midwife during labour and birth completed the questionnaire (34%).Seven of those nine women who had had continuity but did not return the questionnaire were born in a country outside Sweden.Of the 236 women who completed the follow-up questionnaire, 10 were excluded from further analysis.The reasons for exclusion were that four women gave birth prior to the on-call service started, four chose to give birth in a hospital outside the area, and two did not complete the Childbirth Experience Questionnaire (CEQ), leaving 226 questionnaires for analysis.

Background of participants
Table 1 shows the background characteristics from data collected in mid-pregnancy.The majority were 25-35 years old, living with a partner, and born in Sweden.The most common level of education was high school.Only a few women had had infertility problems, and the proportion of multiparous women was slightly higher (58.8%) than the proportion of primiparous women (41.2%).The majority rated their physical (94%) and emotional (88%) health as good.More than one in three women were classified with fear of birth in mid-pregnancy and 12% had depressive symptoms.The majority (88.6%) preferred to have a vaginal birth, 5.5% a caesarean section, and 5.5% did not express any birth preference.

Childbirth Experience Questionnaire
The mean value and standard deviations for the four domains and the total CEQ were Own capacity mean 2.72 (SD 0.70); Professional support mean 3.53 (SD 0.69); Perceived safety mean 3.10 (SD 0.71); and Participation mean 3.15 (SD 0.90).For the total CEQ, the mean was 3.04 (SD 0.65).

CEQ in relation to women's background characteristics
In Table 2, the domains of CEQ and the total score are presented in relation to women's background characteristics.In the domain Own capacity, women not living with a partner, those with a lower level of education, primiparous women, women with fear of birth, and women who either preferred a caesarean section or could not decide a preferred mode of birth were all less likely to score high in this domain.The differences were statistically significant, but they yielded small effect sizes (Eta 2 0.02-0.034).
In the domain Professional support, women not living with a partner and women who did not prefer a vaginal birth reported lower birth satisfaction (Eta 2 0.022-0.046).
The domain Perceived safety showed that women with a high fear of birth and those who preferred a caesarean section or could not decide on a birth preference were more likely to have a less positive experience.The effect sizes were small to moderate (Eta 2 0.05-0.069).
Higher age, single status, and country of birth other than Sweden were all negatively related to the domain Participation, with small effect sizes being found: Eta 2 0.028, 0.023, and 0.030, respectively.For the total CEQ, these background variables showed statistically significant differences in women's background: civil status, fear of birth and birth preference, with small-to-moderate effect sizes, namely Eta 2 0.048, 0.036, and 0.065, respectively.

CEQ in relation to birth outcome
In Table 3, birth outcome is presented for the domains of CEQ.The majority of women gave birth in gestational weeks 38-41, with spontaneous onset of labour (66%) and had a normal vaginal birth (79%).An epidural was used for pain relief in 35%, and 36% of the women needed augmentation of labour, with synthetic oxytocin.Nearly 40% reported some birth complication of various kinds from minor stitches to severe situations.The most commonly mentioned maternal complications were severe blood loss (13 comments), manual removal of the placenta (10 comments) and perineal ruptures (10 comments).For the baby, the complication was low heart frequency during labour (8 comments).Gestational week and onset of labour were not associated with any of the domains of CEQ.
The domain Own capacity was associated with mode of birth.The more the birth deviated from a normal vaginal birth, the lower the satisfaction.Use of an epidural, need of augmentation and selfreported complications were associated with a less positive birth experience.Women who had a known midwife during labour and birth, on the other hand, scored higher in this domain.There was a medium effect size for mode of birth (Eta 2 0.133) and for an epidural (Eta 2 0.073) and a small effect size for oxytocin augmentation, complications, and having a known midwife: Eta 2 0.045, 0.052, and 0.017, respectively.
In the domain Professional support, the only statistically significant variable was having had a known midwife, which increased satisfaction.However, the effect size was small (Eta 2 0.031).Having had a known midwife was also positively correlated to the domain Perceived safety, together with having had a normal vaginal birth, no epidural, no augmentation and no self-reported birth complications.All of these variables were associated with a more positive assessment in this domain.The effect sizes ranged from 0.016 for having had a known midwife to 0.102 for mode of birth.
Participation was associated with mode of birth, where those with a normal vaginal birth were most satisfied as were women without any birth complication, (small effect, Eta 2 0.037), whereas a large effect size occurred for mode of birth (Eta 2 0.174).For the total CEQ, three variables reached statistical significance: vaginal birth, no birth complications and having had a known midwife present.The effect size for mode of birth was large (Eta 2 0.168), and small for complications and having had a known midwife: Eta 2 0.35 and 0.028, respectively.

Discussion
The main findings of this study were that some background variables were related to the domains of the CEQ.Not living with a partner, fear of birth, and a birth preference other than vaginal were associated with a less positive birth experience.Having had a vaginal birth without an epidural, augmentation, or birth complication all yield a better birth experience.Women who had had a known midwife were more likely to have had a positive birth experience overall, predominantly in the domain Professional support.
When comparing the results of the present study with other studies focusing on women's birth experience, we found some similarities.In studies using the CEQ, few studies have reported differences in women's background characteristics, with the exception of parity.Similar to the present study, a study from Spain [41] showed that multiparous women scored higher in Own capacity, which was also confirmed in a recent study from China [42].The original study [39], as well as a study from the United Kingdom that used the CEQ [43] included only first-time mothers.Some of the background characteristics associated with one or more of the domains of CEQ in the present study, such as age, country of birth, and level of education, have not been identified in other studies.The present study showed that women not living with a partner had a more negative birth experience, which was statistically significant in three out of four domains and the total CEQ.This was also found in a Swedish study including only women with fear of birth, but only in the domain Participation [44].
The result also showed that birth preference, when asked in mid-pregnancy, was associated with the majority of domains, and women who preferred a vaginal birth more often scored higher, with the exception of the domain Participation, where no statistically significant difference was found.This contradicts the result from Hildingsson and Rubertsson's study [44], with a sample of fearful women, where birth preference was only statistically significant in the domain Participation.Women who did not need any oxytocin augmentation and those who did not use an epidural were more likely to report higher birth satisfaction, similar to studies from Sweden [39] and China [42].Having a normal vaginal birth without any complications was also associated with a better birth experience, findings previously reported in many studies [39,[41][42][43].
The main focus of the present study was on the importance of continuity with a known midwife.The results showed that this was an important factor as the presence of a known midwife during labour and birth was associated with a better birth experience in the majority of the domains and the total CEQ score.The exception was the domain Participation.To our knowledge, there are no studies that have used the CEQ to address women's birth experience in continuity models.There is, however, strong evidence from a large Cochrane review [11] that continuity with a known midwife matters to women and creates a more positive birth experience.The result also showed that, compared to the large effect size of mode of birth in some of the domains, the effect size of continuity was small.This might be explained by the fairly limited access to a known midwife for the women in this study.The proportion of having a known midwife in this project was around 35%.The total response rate in this study was acceptable, as 85% actually returned the follow-up questionnaire.Of the 42 women not returning the questionnaire, the majority were foreign-born and might not have the language skills to understand and answer the questions.The inclusion criteria of mastering the Swedish language well enough to be able to communicate by telephone does not necessarily mean being able to understand written questions.
In continuity models with on-call services 24/7, usually 75%-80% of women have a known midwife [23].Midwives' opportunities to provide continuity in this study were also limited by other factors.First, the long distance to hospitals with labour wards, often in bad weather conditions, especially during winter time, was one situation that created difficulties.The problem with a long distance to hospital on dangerous roads has previously been reported by Dietsch and co-workers [32], in a study using in-depth interviews of women living in rural New South Wales, where the labour wards were closed.Women reported that the risk of travelling long distances was ignored in the obstetrics discourse.An Australian study interviewed clinicians and health policymakers in very remote areas [45] and highlighted the importance of women having the opportunity to give birth closer to their residence with midwife-led care, as there is evidence that births by women assessed as having a low risk of complications are safe with midwives as the primary caregiver [31].
Another issue was that the shortage of midwives made it impossible to offer continuity 24/7.A shortage of midwives has also been found in several studies as a problem for initiating continuity models [27,28].A third concern was the work-time regulations that stopped midwives from working more than 12 h.Work-time regulations were also reported in a study by Jepsen et al. [46] as a factor affecting continuity, where ward midwives had to take over the care when the caseload midwife had worked 12-16 h.
The present study is also compromised by its observational design and the self-recruitment of women to the project.A randomized controlled trial would have diminished someof these issues.However, it was not possible to conduct a randomized controlled trial, mainly due to the situation with staffing.Despite this, the results point in the same direction as the international literature [1,11], with higher satisfaction when having a known midwife.More research is needed in other areas of Sweden, that could form the basis of introducing continuity models as an evidence-based option.

Conclusions
The results of this study showed that, despite the fairly low percentage of women who actually had the opportunity to receive intrapartum care from a known midwife, those who received continuity were more likely to have a positive birth experience.This was most apparent regarding the professional support, which can be understood as saying the woman-midwife relationship really matters.The results pointed out the benefits of a less medicalized birth as important for a good birth experience, and it also showed certain characteristics of women, such as being single, having a fear of birth and preferring a caesarean section could need extra support in order to avoid a less positive birth experience.

Conflict of interest
None declared.

Table 1
Background of the participants.
a FOBS = the Fear Of Birth Scale.b EPDS = Edinburgh Postnatal Depression Scale.

Table 2
Domains of the Childbirth experience questionnaire in relation to women's background.

Table 3
Domains of the Childbirth experience questionnaire in relation to birth outcome.