The safe delivery of a healthy baby is a critical priority in perinatal care. However, maternal satisfaction also remains an important outcome, as it has the potential to influence psychological functioning and the mother-infant relationship (Goodman et al, 2004). Satisfaction with childbirth is a cognitive evaluation of the fit between a woman's personal preferences and her actual birth experience (Stevens et al, 2012). Such satisfaction is associated with a number of factors including participation in antenatal education (Phipps et al, 2009), positive expectations for the birth experience (Maggioni et al, 2006), a fit between expectations for the birthing experience and the actual birthing experience (Green et al, 1990), and maternal personality, including big five personality, coping and childbirth expectations (Catala et al, 2019).
Greater satisfaction during childbirth is also thought to be associated with healthcare factors, including being cared for by a midwife (Fair and Morrison, 2012; Sandall et al, 2016), lower levels of pain during labour (Green, 1993; Goodman et al, 2004), a physiological birth (Brown and Lumley, 1994), and a lack of complications during labour and birth (Fair and Morrison, 2012). Conversely, emergency caesarean section (c-section), postpartum haemorrhage, and having an Apgar score less than seven at five minutes have been identified as key risk factors for dissatisfaction with childbirth (Falk et al, 2019). Perceived levels of control during childbirth are consistently significantly predictive of satisfaction with childbirth (Hodnett and Simmons-Tropea, 1987; Green et al, 1990; Knapp, 1996; Green and Baston, 2003; Goodman et al, 2004; Fair and Morrison, 2012; Stevens et al, 2012; Inci et al, 2015; Preis et al, 2019) and have been proposed as the main variable related to childbirth satisfaction (Hodnett, 2002; Stevens et al, 2012).
A person's beliefs about personal or perceived control can influence their appraisal of events (Knapp, 1996). In the context of childbirth, perceived control is the ‘extent to which the mother believes her actions influenced or shaped the conditions of the birth environment’ (Stevens et al, 2012). Perceived control during childbirth is influenced by a variety of factors, including the provision of information (Cheung et al, 2007), the ability to actively participate in decision making during childbirth (Cheung et al, 2007), midwife care (Fair and Morrison, 2012), and birthing without complications and/or medical intervention (Cheung et al, 2007; Fair and Morrison, 2012).
Greater perceived control over the birth environment and process have been found to predict more positive emotions, less fear and guilt, and better perceived care in first-time mothers (Preis et al, 2019). While lower levels of perceived control during childbirth have been associated with postpartum depression (Green et al, 1990; Czarnocka and Slade, 2000; Sorenson and Tschetter, 2010; Gürber et al, 2012), anxiety in mothers who have recently given birth (Giakoumaki et al, 2009; Shlomi Polachek et al, 2014; Bell et al, 2015), lower levels of satisfaction with the childbirth experience (Weisman et al, 2010) and subsequent negative perceptions of her infant (Allen, 1998; Nicholls and Ayers, 2007; Weisman et al, 2010). Similarly, lower levels of satisfaction with the childbirth experience have been associated with postpartum depression (Neter et al, 1995; Saisto et al, 2001; Tuohy and McVey 2008; Sorenson and Tschetter, 2010; Gausia et al, 2012) and anxiety (Bell et al, 2015), and increases risks of the mother-infant forming a maladaptive attachment relationship (Egeland and Farber, 1984).
However, the concept of perceived control during childbirth is not backed by a uniform theoretical framework and has been measured using a variety of scales across studies (Stevens et al, 2012). Similarly, there is a lack of consensus on the operational definition of ‘satisfaction’ within the literature, resulting in various measures focusing on different aspects of satisfaction (Stevens et al, 2012). Further, much of the literature uses a single-item global measure of satisfaction with childbirth (Brewin and Bradley, 1982; Green et al, 1990; Brown and Lumley, 1994; Michels et al, 2013), which may not accurately reflect a woman's birthing experience (Stevens et al, 2012).
This pilot study aims to generate knowledge that will support the birthing and postnatal experiences of women, particularly with regards to enhancing longer term maternal and infant outcomes. By measuring perceived control and overall satisfaction with childbirth, we sought to confirm and extend previous research regarding the association between factors present during delivery, childbirth, postpartum psychological functioning and the mother-infant relationship. The study utilises a newly developed measure of perceived control during childbirth to investigate the congruency between pre-birth expectations and post-birth perceptions of control (Stevens et al, 2012). Overall satisfaction with the childbirth experience is measured using a multi-item questionnaire designed to more accurately capture the experience of childbirth than previously used single-item measures (Stevens et al, 2012). It is predicted that parity, length and type of labour, and the presence of a midwife during antenatal care, labour and birth will be associated with perceived control during childbirth and satisfaction with the childbirth experience.
It is further predicted that there is a strong association between perception of control and overall satisfaction with the childbirth experience. An association between perceived control during childbirth, postpartum depression and anxiety is expected as is a relationship between these psychological outcomes and satisfaction with childbirth. Additionally, perceived control and satisfaction with childbirth is also expected to be associated with maternal attachment to the infant.
Methods
Participants
Following Health and Medical Human Research Ethics Committee approval from the University of Wollongong, members of the research team conducted information gathering and education sessions with midwives. Midwives then provided information about the study to pregnant women at their history-taking interview, which is a routine appointment at that particular hospital. A member of the research team was regularly present in the antenatal clinic waiting room to answer any questions that potential participants or midwives may have had.
Women were asked at that visit or subsequent visits whether they were interested in finding out more information regarding the study and, if they gave consent, their contact details were provided to the research team for a follow-up. The exclusion criteria were non-English speaking women or those planning on moving from the study area prior to birth. Full details of the study recruitment can be found in Townsend et al (2019). Of the participants, 48 women agreed to participate however, 10 women withdrew from the study for various reasons including pregnancy complications requiring hospitalisation (n=2), change of mind (n=7) and one woman experienced a miscarriage. All participating women provided written informed consent. The study was undertaken in 2014–2015.
Procedure
Women participated in data collection at approximately two months postpartum. Participants completed questionnaires regarding perceptions of control and satisfaction with childbirth, mental health and maternal attachment. Birth details were obtained from hospital record chart review as part of the approved protocol.
Measures
The measures were administered via online questionnaire to participants between 7–10 weeks postpartum. Participants were asked about the type of birth they had experienced, the care they had received antenatally and the care they received during birth. Women's primiparous or multiparous status was collected from the baseline survey and from the hospital medical records. There was no discrepancy between the two data sources.
Hospital record data was used to ascertain parity and length of labour. The total length of labour was obtained by summing the lengths of each of three stages of labour in minutes. Data on breastfeeding difficulties prior to discharge from hospital was also collected from hospital records.
Perceived control during childbirth
Perceived control during childbirth was measured using the perceived control in childbirth scale (Stevens et al, 2012). The scale consists of 12 items, rated on a six-point Likert scale from ‘strongly disagree’ to ‘strongly agree’ (eg ‘I was able to participate in making decisions about how to manage my labour and birth’). Total scores range from a minimum of six (reflecting low levels of perceived control) to a maximum of 72 (reflecting high levels of perceived control). Construct validity has been established for this measure (Stevens et al, 2012). The study reported high internal consistency in line with previous findings reported by Stevens et al (2012).
Satisfaction with childbirth
Satisfaction with childbirth was measured using the satisfaction with childbirth scale (Stevens et al, 2012). The scale consists of seven items, rated on a seven-point Likert scale from one at ‘strongly disagree’ to seven at ‘strongly agree’ (eg ‘The conditions of my childbirth experience were excellent’). Total scores range from a minimum of seven (indicating low levels of satisfaction) to a maximum of 49 (indicating a very high degree of satisfaction). Construct validity has been established for this measure (Stevens et al, 2012). The study reported good internal consistency in line with previous findings (Stevens et al, 2012).
Maternal psychological outcomes
Maternal psychological health was measured using the depression anxiety stress scale (Lovibond and Lovibond, 1995) and the Edinburgh postnatal depression scale (Cox et al, 1987). The depression anxiety stress scale is a 42-item, self-report measure yielding three scales of 14 items, namely depression, anxiety and stress. Items are scored on a four-point Likert scale from 0–3, higher scores being indicative of higher levels of depression, anxiety or stress (Crawford et al, 2011). Total scores on each scale range from 0–42, with each scale scored into categories ranging from ‘normal’ to ‘severe’ (Lovibond and Lovibond, 1995). The depression anxiety stress scale exhibits high reliability and internal consistency on each scale (Crawford et al, 2011), consistent with findings from this study.
The Edinburgh postnatal depression scale is a widely used self-report measure of postpartum depression, consisting of 10 items regarding feelings over the last seven days. Items are scored on a four-point Likert scale from 0–3, higher scores being indicative of greater severity of depressive symptoms. Total scores range from 0–30 (Michels et al, 2013). The Edinburgh postnatal depression scale exhibits high reliability and internal consistency (Zubaran et al, 2010).
Maternal attachment
Maternal attachment was assessed using the maternal postnatal attachment scale (Condon and Corkindale, 1998). The maternal postnatal attachment scale is a self-report questionnaire containing 19 items loading onto three scales: absence of hostility (maximum score: 25), pleasure in interaction (maximum score: 25) and quality of attachment (maximum score: 45) (Condon, 2015).
Higher scores indicate a more positive maternal emotional response towards the infant. The measure demonstrates a high level of internal consistency and test-retest reliability (Condon and Corkindale, 1998). This scale has been administered at various time points over the first year of life and the authors suggest there has a reasonable degree of continuity in attachment scores over the first eight postnatal months (Condon and Corkindale, 1998). Quantitative data were analysed using the statistical package for the social sciences version 21.0 (Corp I, 2012).
Results
Perinatal analyses
Participants ranged in age from 22–41 years, with a mean age of 31 (SD=4.56). A minority of participants were born overseas (12%), and most were university educated (81%) and married (73%). Most participants reported their household income as greater than $80 000 USD per annum, indicating they were, as a group, more socio-economically advantaged (77%). Of the participants, 23 were primiparous (60.5%) and 15 (39.5%) were multiparous. Of the participants, 24 (63.2%) had a physiological birth, nine (23.7%) had an instrumental vaginal birth and five (7.9%) had a c-section birth. The mean length of labour was 323 minutes (SD=183.5), with a range between 3–759 minutes.
During the antenatal period, 19 (50%) women indicated they received midwife-led continuity of care, and 19 (50%) women received shared care (in this model of care, the woman's GP provides some of the antenatal care but does not attend labour and birth). During birth, 19 (50%) women received midwife-led continuity of care, and 19 (50%) women received hospital-based medical care. Of the women, 11 (28.9%) experienced breastfeeding difficulties prior to discharge from hospital, as recorded by hospital staff who were caring for the mother and child. Table 1 presents the means, standard deviations and ranges of all self-report measures used at two-months postpartum.
| Mean | SD | Range | Cronbach's (alpha) | |
|---|---|---|---|---|
| PCCh | 56.03 | 11.04 | 26−67 | 0.876 |
| SwCh | 35.29 | 12.48 | 7−49 | 0.777 |
| DASS | ||||
| Depression | 1.84 | 2.33 | 0−10 | 0.775 |
| Anxiety | 2.34 | 3.30 | 0−15 | 0.767 |
| Stress | 6.03 | 4.22 | 0−16 | 0.814 |
| EDPS | 4.82 | 2.99 | 0−14 | 0.685 |
| MPAS | ||||
| Quality of attachment | 41.37 | 2.94 | 33.2−45 | 0.572 |
| Absence of hostility | 19.78 | 3.13 | 14.6−25 | 0.596 |
| Pleasure in interaction | 21.00 | 3.30 | 10−25 | 0.629 |
PCCh = perceived control in childbirth scale; SwCh = satisfaction with childbirth scale; DASS = depression anxiety stress scale; EDPS = Edinburgh postnatal depression scale; MPAS = maternal postnatal attachment scale
A strong positive correlation was found between perceived control during childbirth and satisfaction with childbirth (r=0.779, p<0.001). An independent samples t-test indicated that there was a significant relationship between perceived control during childbirth and having a physiological birth, compared with an instrumental birth (being instrumental vaginal or c-section), t(36)=2.67, p=0.016. Further analyses indicated that a physiological birth was associated with higher levels of perceived control during birth. Independent samples t-tests further indicated that there was a significant relationship between perceived control during childbirth and receiving midwife-led continuity of care during the antenatal period (as compared with receiving shared care during this period), t(36)=4.348, p<0.05, and during birth (as compared with receiving hospital-based medical care, t(36)=4.348, p<0.05. Receiving care from a midwife during both the antenatal period and birth was associated with higher levels of perceived control during childbirth.
An independent samples t-test found a significant difference between levels of satisfaction with childbirth between those women who had a physiological birth compared with an instrumental birth, t(36)= 5.006, p<0.01. Women who had a physiological birth reported greater satisfaction with their childbirth experience. Further t-tests indicated a significant difference in satisfaction with childbirth for those women who received midwife-led continuity of care in the antenatal period compared with those who received shared care, t(36)=4.765, p<0.01. Those women who received midwife-led continuity of care reported more satisfactory experiences with childbirth than those receiving shared care during the antenatal period. Women who received midwife-led continuity of care during birth rated their birth experience as significantly more satisfactory than those who received hospital-based medical care during childbirth, t(36)=4.765, p<0.01. Those women who reported difficulty breastfeeding prior to hospital discharge reported significantly lower scores on the satisfaction with childbirth scale than those who did not report such difficulty, t(35)=-2.687, p=0.014. Parity and length of labour were not significantly associated with either satisfaction with childbirth or perceived control during childbirth (Table 2).
| PCCh | SwCh | |
|---|---|---|
| PCCh | - | 0.779** |
| SwCh | 0.779** | - |
| Variables present before and during birth | ||
| Physiological birth | 0.457** | 0.660** |
| Parity | 0.061 | 0.072 |
| Length of labour | −0.037 | −0.059 |
| Midwife care antenatal | 0.587** | 0.622** |
| Midwife care birth | 0.587** | 0.622** |
| Postnatal variables | ||
| Breastfeeding difficulties | 0.323 | −0.411* |
| EDPS | −0.362* | −0.151 |
| DASS | ||
| Depression | −0.200 | −0.120 |
| Anxiety | −0.267 | −0.379* |
| Stress | −0.074 | −0.119 |
| MPAS | ||
| Quality of attachment | 0.136 | 0.066 |
| Absence of hostility | 0.300 | 0.097 |
| Pleasure in interaction | −0.025 | −0.098 |
DASS = depression anxiety stress scale; EDPS = Edinburgh postnatal depression scale; MPAS = maternal postnatal attachment scale
Consideration of the postnatal implications at two-months postpartum of perceived control and satisfaction with childbirth yielded a moderate inverse correlation between perceived control during childbirth and depressive symptoms measured using the Edinburgh postnatal depression scale (r=-0.362, p=0.026). A moderate negative correlation was found between satisfaction with childbirth and anxiety symptoms as measured using the depression anxiety stress scale (r=0.408, p= 0.033). Maternal attachment at two-months postpartum was not significantly associated with either satisfaction with childbirth or perceived control during childbirth. Table 2 presents all correlational analyses. A linear regression analysis revealed perceived control during childbirth and type of birth to be significant predictors of satisfaction with the childbirth experience, together explaining 71.4% of the variance in those scores, F(2.33)=44.642, p< 0.001.
Discussion
This study sought to investigate the relationships between factors present prior to birth and perceived control, and satisfaction with childbirth, as well as psychological correlates within two months of birth. The study found that length of labour did not significantly impact on perceived control during childbirth. This is concordant with some previous findings (Green and Baston, 2003) but is inconsistent with other literature in this area (Bélanger-Lévesque et al, 2014). Further research may be needed on the effects of parity and length of labour in order to ascertain more conclusively whether these have an effect on perceived control during childbirth.
Type of labour and the presence of midwife-led continuity of care during the antenatal and birthing periods had a significant effect on a woman's level of perceived control during childbirth. These findings are consistent with the literature, which suggests that instrumental intervention during childbirth (as compared with a physiological birth) can significantly decrease levels of perceived control during childbirth (Cheung et al, 2007; Fair and Morrison, 2012). This may be partially explained by instrumental intervention resulting in women feeling a loss of control during labour as control is placed in the hands of medical professionals (Fair and Morrison, 2012).
Further, the presence of a midwife during birth has been found to have positive effects on levels of perceived control (Fair and Morrison, 2012), and the related construct of maternal satisfaction with childbirth (Sandall et al, 2016). The current study has extended this finding to the antenatal period, suggesting that support from a midwife prior to birth can result in women feeling more prepared for, informed and in control during their labour, resulting in higher satisfaction with their childbirth experience. Further research is recommended to replicate this finding, particularly with women who experience pregnancy or health complications (Sandall et al, 2016).
Due to a lack of consistency in measuring women's satisfaction and assessing the cost of various maternity models, these outcomes were reported narratively. The majority of included studies reported a higher rate of maternal satisfaction in midwife-led continuity models of care. Similarly, there was a trend towards a cost-saving effect for midwife-led continuity care compared to other care models. Perceived control during childbirth was significantly correlated with satisfaction with the childbirth experience. This result was expected and is consistent with the literature, which has found that perceived control during childbirth is one of the most significantly predictive factors influencing levels of satisfaction with childbirth (Hodnett and Simmons-Tropea, 1987; Green et al, 1990; Knapp, 1996; Green and Baston, 2003; Goodman et al, 2004; Fair and Morrison, 2012; Stevens et al, 2012; Inci et al, 2015).
This may be because a woman's perception of her ability to control an event, such as childbirth, influences her appraisal of that event, and subsequently her level of satisfaction (Knapp 1996). Further analyses indicated that perceived control during childbirth, along with the type of birth, was one of the most significantly predictive factors influencing satisfaction with the childbirth experience, and that these factors together explained the majority (71.4%) of the variance in satisfaction scores.
These findings add to our current understanding of good clinical care and have implications for the provision of healthcare services both during the antenatal period and during birth itself. In order to improve overall satisfaction with the experience of childbirth, healthcare providers need to ensure that women feel in control of their birth environment. In order to increase perceptions of control, it is encouraged that midwives are involved during early stages of pregnancy. Further, educating women about the possibility and consequences of instrumental intervention before birth may result in higher levels of perceived control during instrumental births. Such education may also result in a better fit between a woman's expectations and the reality of her birth, and a subsequently higher level of satisfaction with the childbirth experience. Offering a debriefing after instrumental births may also support women's understanding of their experience (Ayers et al, 2006) and studies have shown that women appreciate this opportunity (Selkirk et al, 2006), although the evidence to support effectiveness is limited (Rowan et al, 2007).
The study found that prior to discharge from hospital after giving birth, 28.9% of women experienced difficulties breastfeeding their infant. There was no association between length of hospital stay and incidence of breastfeeding difficulties. However, there was a significant relationship between feeling dissatisfied with the childbirth experience and difficulty in breastfeeding. There have been similar findings in this area (Reynolds, 1997). Additional supportive breastfeeding guidance (Saeed et al, 2011) and compassionate encouragement may be necessary in the first 24 hours to ensure that early breastfeeding attempts are successful (Hobbs et al, 2016) for women who felt unsatisfied with their childbirth experience.
Regarding longer-term outcomes, the results indicate that perceived control during childbirth is inversely correlated with depressive symptoms at two-months postpartum, concordant with previous studies (Green et al, 1990; Czarnocka and Slade, 2000; Sorenson and Tschetter, 2010; Gürber et al, 2012). However, the study found no association between perceived control and postpartum anxiety, unlike prior studies (Giakoumaki et al, 2009; Shlomi Polachek et al, 2014; Bell et al, 2015). Interestingly, the study found that there was no association between satisfaction with the childbirth experience and depressive symptoms two-months postpartum, but did find that satisfaction with childbirth was related to anxiety symptoms at that period, consistent with previous findings (Bell et al, 2015). These results may be explained by the use of multiple instruments used to measure anxiety and depressive symptoms. Despite these limitations, the results from the current study are broadly consistent with literature, suggesting that perceived control and satisfaction with the childbirth experience are associated with long-term maternal psychological wellbeing.
The current study did not find a relationship between perceived control or satisfaction with childbirth and maternal attachment, as measured by the maternal postnatal attachment scale. This is somewhat inconsistent with the literature, which suggests that low levels of perceived control may result in a mother harbouring negative perceptions of her child (Allen, 1998; Nicholls and Ayers, 2007). However, the maternal postnatal attachment scale is a more holistic measure of maternal attachment than those used in the aforementioned literature (Allen, 1998; Nicholls and Ayers, 2007). There is a small but growing evidence base for the impact of childbirth, particularly childbirth-related post-traumatic stress, on maternal attachment; however, the influence of maternal attachment on childbirth-related post-traumatic stress has not been investigated long-term (MacKinnon et al, 2018). Further research is therefore needed to better understand the impact of childbirth on longer-term maternal attachment.
The findings from this study may have important implications for healthcare providers. The presence of depressive and anxiety symptoms in women at two-months postpartum, and the significant relationship between these symptoms and perceived control, and satisfaction with childbirth, indicates the need for provision of services which facilitate higher levels of perceived control and satisfaction with childbirth. The results suggest that in order to minimise the effects of low perceptions of control or satisfaction with childbirth further support is needed in the postnatal period. Potentially, debriefing sessions after childbirth and continued support from midwives in the postpartum period may reduce the prevalence of these symptoms. There are inconsistencies in the literature as to the impacts of certain factors (for example, parity and length of labour) on perceptions of control and satisfaction with childbirth, and these inconsistencies may be exacerbated by the use of different measures of perceived control and satisfaction between studies.
The results of this study suggest that the two newly developed measures of satisfaction with childbirth and perceived control during childbirth have utility (Stevens et al, 2012). The perceived control in childbirth scale was developed in light of inconsistencies and problems encountered by previous measures (Stevens et al, 2012), and is thus considered a good measure of perceived control. While the satisfaction with childbirth scale is a multi-item measure, enabling a more detailed picture of satisfaction with childbirth than more commonly used one-item measures (Stevens et al, 2012). The results of this study broadly conform to the literature, indicating that these measures may be adequate for use in future studies. Further confirmation of the results using these measures may be required as well as validation of the measures on different populations.
Limitations
The study utilised a sample of 38 women from the Illawarra region in New South Wales, Australia. This is a small sample size and is restricted to a small geographic area. The results may not be generalisable to a larger population. The study did not specifically investigate the possible impact of demographic factors, such as age, income, marital status, educational attainment or pre-existing mental health issues on the outcomes measured as these would require a larger sample than this pilot. This may mean that some of these factors had an unknown effect on the results obtained.
Conclusion
The results of this study accord with the previous literature, suggesting that the two newly developed measures of perceived control during childbirth and satisfaction with childbirth are adequate measures. The results suggest that type of birth and perceived levels of control during childbirth are highly predictive of satisfaction with childbirth. Midwife-led continuity of care significantly affected perceived levels of control. Thus, healthcare providers should consider the broader provision of this model, as well as adequate debriefing opportunities for instrumental births and additional supportive breastfeeding guidance, in order to ensure higher levels of satisfaction with childbirth. Such services and care should extend into the postpartum period in order to ensure that maternal psychological wellbeing is supported.