Predictors of Women’s Perceptions of the Childbirth Experience
Janet Bryanton, UPEI School of Nursing, 550 University Avenue, Charlottetown, Prince Edward Island, Canada, C1A 4P3.
Objective: To determine the factors that predict women’s perceptions of the childbirth experience and to examine whether these vary with the type of birth a woman experiences.
Design: Prospective cohort study.
Setting: The postpartum units of two eastern Canadian hospitals.
Participants: Six hundred fifty two women and their newborns.
Data Collection: Data were collected in hospital at 12 to 48 hours postpartum using self-report questionnaires and chart review.
Main Outcome Measure: Perception of the childbirth experience was measured for women having a vaginal and emergency cesarean birth using the Questionnaire Measuring Attitudes About Labor and Delivery and planned cesarean birth using the Modified Questionnaire Measuring Attitudes About Labor and Delivery.
Results: Of the 20 predictors of women’s childbirth perceptions, the strongest were type of birth; degree of awareness, relaxation, and control; helpfulness of partner support; and being together with the infant following birth.
Conclusions: Of the predictors of a quality birth experience, most were amenable to nursing interventions: enhancement of patient awareness, relaxation, and control; promotion of partner support; and provision of immediate opportunities for women to be with their babies.
Accordance with professional practice standards (Association of Women’s Health, Obstetric, and Neonatal Nurses, 2002), contemporary maternity care providers strive to create a childbirth experience that is safe for the mother and baby and positive and satisfying for the childbearing woman. Giving birth is an important life experience for women. The childbearing woman undergoes one of the most profound life changes she will ever experience, and there is always potential for psychological benefits or damage (Simkin, 1996). Women have reported gaining a sense of mastery, personal strength, and competency as they faced the challenges of labor and birth, and many have described a sense of elation and accomplishment (Callister, 2004). Enhanced maternal attachment and competence have also been associated with a positive experience (Mercer, 1986; Mercer & Ferketich, 1994).
In contrast, a negative birth experience can be very disempowering (Fenwick, Gamble, & Mawson, 2003) and can have negative effects on a woman’s self-esteem, self-efficacy, and mental health. Women have described themselves as failures (Callister, 2004) and have expressed feelings of anger, guilt, disappointment, loss of control, and inadequacy (Baker, Choi, Henshaw, & Tree, 2005). Negative or unsatisfying experiences can increase the risk for postpartum depression (Righetti-Veltema, Conne-Perreard, Bousquet, & Manzano, 1998) and posttraumatic stress disorder (Beck, 2004a, b). A negative or traumatic experience can also lead to fear of a subsequent birth (Waldenstrom, Hildingsson, & Ryding, 2006), can increase the likelihood of not having another baby (Gottvall & Waldenstrom, 2002), and can lead to problems with maternal/infant attachment (Reynolds, 1997).
Researchers have identified factors that promote a positive, satisfying experience, although there is less evidence about those that are most predictive of such an experience and whether the predictors vary with the type of birth a woman experiences. Nurses have the most contact with childbearing women during their birth experiences and are therefore in a unique position to shape its outcome. With clear evidence regarding the most significant predictors, nurses will be charged with modifying those that are amenable to change and reducing the impact of those that are not. This has the potential to have a positive influence on every woman’s birth experience.
Few studies have simultaneously investigated predictors of the childbirth experience to determine the relative importance of one predictor over another.
Perception of the childbirth experience is highly personalized, and women’s views vary regarding what constitutes a positive and satisfying experience. Satisfaction is a construct that is complex, multidimensional, and may change over time. It involves a positive affective response to an experience and a cognitive evaluation of the emotional response (Hodnett, 2002). It can be equated with a feeling that results after a positive evaluation of the labor experience (Bramadat & Driedger, 1993). Because satisfaction is multidimensional, women may be satisfied with some aspects of their experiences and dissatisfied with others, and positive and negative feelings can coexist (Waldenstrom, 1999). As well, being satisfied with care is not synonymous with feeling positive and satisfied with the experience but may be one factor contributing to it (Waldenstrom, Borg, Olsson, Skold, & Wall, 1996). A woman may be dissatisfied with the care she receives but feel pleased with her own behavior and positive about her experience (Green, Coupland, & Kitzinger, 1990).
There are many complex variables that influence women’s perceptions of their birth experiences (Hodnett, 2002). Numerous studies have investigated these factors, using a variety of research methods ranging from qualitative studies to randomized controlled trials and meta-analyses. The effects of these identified factors on birth perception were not consistent across studies nor were the outcomes measured.
Generally, the literature has suggested that a positive perception of childbirth, including satisfaction with the experience and care, is promoted by greater maternal age (Borjesson, Paperin, & Lindell, 2004); multiparity (Waldenstrom, 1999); prenatal education (Goodman, Mackey, & Tavakoli, 2004); a shorter labor (Nystedt, Hogberg, & Lundman, 2005); a home-like birth environment (Brown & Lumley, 1998); a vaginal birth (Hodnett, Downe, Edwards, & Walsh, 2005); fewer interventions such as inductions, forceps/vacuum extraction, and episiotomies (Creedy, Shochet, & Horsfall, 2000; Waldenstrom, Hildingsson, Rubertsson, & Radestad, 2004); increased maternal self-esteem/confidence (Mackey, 1995); decreased maternal stress and anxiety (Waldenstrom et al., 2006); nonseparation of the mother from her infant (Fenwick et al., 2003); maternal expectations being met (Soet, Brack, & Dilorio, 2003); increased maternal perception of control (Lundgren, 2005); decreased fear of pain (Soet et al.); participation in decision making (Waldenstrom et al., 1996, 2004); a positive perception of partner, nurse, midwife, and doula support (Hallgren, Kihlgren, & Olsson, 2005); and fewer birth complications (Righetti-Veltema et al., 1998).
Although numerous studies have investigated various factors influencing perception of the childbirth experience, few have studied several predictors simultaneously to determine the relative importance of one predictor over another. No large Canadian study has examined numerous predictors of perception of the childbirth experience simultaneously, although three early studies explored various factors associated with satisfaction with care (Canadian Medical Association [CMA], 1987; Crowe & von Baeyer, 1989; Seguin, Therrien, Champagne, & Larouche, 1989). Three Swedish studies (Waldenstrom, 1999; Waldenstrom et al., 1996, 2004) and three American studies (Goodman et al., 2004; Mercer, Hackley, & Bostrom, 1983; Soet et al., 2003) investigated numerous predictors of perception of the birth experience. Two British studies were more limited in their predictors (Green et al., 1990; Lavender, Walkinshaw, & Walton, 1999), and two large Australian studies (Brown & Lumley, 1994, 1998) and one Scottish study (van Teijlingen, Hundley, Rennie, Graham, & Fitzmaurice, 2003) explored satisfaction with care.
Results of these studies demonstrated that positive perception of caregiver and partner support (Lavender et al., 1999; Waldenstrom, 1999; Waldenstrom et al., 1996, 2004), involvement in decision making (Brown & Lumley, 1994, 1998; Green et al., 1990; Waldenstrom et al., 1996, 2004), having expectations met (CMA, 1987; Goodman et al., 2004; Green et al.; Soet et al., 2003), decreased perception of pain/increased pain relief (Lavender et al.; Soet et al.; Waldenstrom; Waldenstrom et al., 1996, 2004), vaginal birth (Mercer et al., 1983; Soet et al.; Waldenstrom; Waldenstrom et al., 1996), and access to information (Brown & Lumley, 1994; Green et al.; Seguin et al., 1989) were the most frequently identified factors that predicted a positive childbirth experience or satisfaction with care.
No studies were identified that examined predictors of birth perception by type of birth, although Seguin et al. (1989) developed models for vaginal and cesarean births to examine satisfaction with care. For vaginal births, pain, complications, satisfaction with medical services, participation in decision making, and length of labor predicted 44.8% of variance in satisfaction. Explanations, complications, and whether it was an emergency or planned cesarean predicted 52.1% of the variance for cesarean births.
In summary, positive outcomes of the childbirth experience include increased self-esteem, confidence, mastery, attachment, and maternal competence. Negative outcomes including disempowerment, postpartum depression, fear of childbirth, and posttraumatic stress disorder have also been reported. Many complex variables have been studied with respect to their influence on the childbirth experience, with inconsistent results. Last, only a limited number of studies have investigated several predictors simultaneously and the majority studied satisfaction with care rather than perception of the experience.
This study determined the factors that predict women’s perceptions of the childbirth experience and examined whether these vary with the type of birth a woman experiences.
- 1What are the predictors of women’s perceptions of their childbirth experience?
- 2Do these predictors vary by the type of birth (vaginal, emergency cesarean, and planned cesarean birth) a woman experiences?
A prospective, cohort design was used for this study. A sample of 652 women and their newborns was recruited in the early postpartum period. As part of the larger study, a subsample of these mother-infant pairs was assigned to cohorts based on positive or negative childbirth perceptions and were followed at 1 month.
The in-hospital recruitment and data collection were conducted on the postpartum units of two general hospitals on Prince Edward Island (PEI), Canada. One 350-bed hospital has a postpartum unit of 10 private and semiprivate beds. Women labor, give birth, and recover in a birthing room and are then transferred to the postpartum unit. The second unit is in a 102-bed hospital that has six birthing suites where women labor, give birth, recover, and have their postpartum stay in one room. Both hospitals endeavor to provide one-on-one nursing support during active labor in a relatively low intervention environment.
All women who gave birth on PEI from October 2004 to December 2005 and their newborns were consecutively assessed for inclusion, except those who gave birth at home or were discharged early. Women were included if they were greater than or equal to 15 years of age; had a vaginal, planned, or emergency cesarean birth; and were able to read and speak English and provide consent. They were excluded if they had an unresolved illness, had received general anesthesia, were leaving the province before 1 month, or had given birth to a stillborn infant. Infants were excluded if they were less than 37 weeks gestation; weighed less than 2500 g; were multiples; had major congenital anomalies, traumatic birth injuries, or other serious illnesses; or were placed in foster care or adopted.
Participant flow Of the 1,443 women who gave birth on PEI during the time of the study, all except 1 gave birth in hospital. All women who gave birth in hospital and their newborns were assessed for eligibility, except for 31 who were discharged early before the research assistants were able to assess them. A total of 1,411 women and newborns were assessed for eligibility, and of those 195 were not eligible. Forty-six percent of the eligible women refused participation. The main reasons included not interested, disliked surveys, too busy, or woman/partner did not want a home visit. The remaining 652 women participated in the in-hospital data collection.
Description of the SampleTable 1 provides an overview of the maternal and newborn demographic and obstetric characteristics of the sample. The mothers ranged in age from 16 to 43 years, with a mean age of 28.4 years, and were primarily White. The majority (73.2%) were married or were in a common law relationship, had some/completed college or university education (76.6%), and had an adequate income (76.5%). (Based on Statistics Canada Low Income Cut-offs. Calculated using reported income category, family size, and size of population of residence, Canadian Council on Social Development, 2005). Over half (56.4%) of the mothers were multiparas. Almost three quarters (73.3%) experienced a vaginal birth. Of those women who had a cesarean birth (26.7%), 12.6% had an unplanned/emergency cesarean, 12.9% had a planned cesarean, and 1.2% had a planned cesarean with labor. Reasons for cesarean birth included repeat (34.5%), failure to progress (17.8%), cephalopelvic disproportion (CPD) (13.8%), fetal health/distress (12.1%), breech (11.5%), and other (10.3%).
Table 1. Characteristics of Sample
|Maternal age (years), M (SD)||28.4 (5.4)|
|Marital status (%)|| |
|Parity (%)|| |
|Education (%)|| |
|Income (SES) (%)|| |
|Type birth (%)|| |
|Exclusive breastfeeding in hospital (%)||71.5|
|Prenatal classes with present or past pregnancy (%)||67.9|
|Gender (%)|| |
|Gestational age (weeks) M (SD)||39.6 (1.1)|
|Birthweight (g) M (SD)||3, 598.9 (473.8)|
|Apgar 1 min M (SD)||8.4 (1.3)|
|Apgar 5 min M (SD)||9.3 (0.6)|
For those women who experienced a vaginal birth, the average length of labor was 7.5 hours (SD 4.3), and 71.3% had an analgesic or Entonox during labor. Almost one quarter (22.8%) of women who gave birth vaginally had an epidural and 7.7% had forceps or vacuum extraction. Of those women who had a vaginal birth or an emergency cesarean (n = 560), 28% were induced with oxytocin or a cervical gel and 23.9% were augmented during labor. For all women, 8.9% experienced complications. Of those who had complications (n = 58), 41.4% experienced a postpartum hemorrhage, 25.9% had pregnancy induced hypertension (PIH), 6.9% developed an infection, and 25.8% had other complications.
To assess for inclusion, the research assistants regularly reviewed the patient information kardexes on both postpartum units. Women who met the eligibility criteria were approached within 12 to 48 hours of childbirth. While in hospital, demographic and obstetric data were recorded from every consenting woman’s/newborn’s chart. Perceptions of the childbirth experience, the main outcome variable, and general self-efficacy, a predictor variable, were measured through self-report. Participants completed the questionnaires before discharge and returned them to the nurses’ station in a sealed envelope. Before initiating the study, a pilot was conducted to assess the adequacy of study procedures. Following the pilot, minor revisions were made to the consent to increase its clarity and women having a general anesthetic were excluded, as they were not able to complete the birth perception questionnaire. Pilot participants were not included in the final study sample.
Before the pilot was initiated, ethics approval of the full study protocol was obtained from two hospital and two university ethics review boards. During recruitment, the research assistants emphasized that participation was voluntary and that at any time women could choose not to answer questions or withdraw from the study without their or their infants’ care being affected. It was emphasized that their information would remain confidential and that there would be no way of identifying them. They were informed that there were no known risks in participating and that the results may contribute to better future care for mothers and infants.
The main outcome variable, birth perception, was defined as a woman’s perception of her childbirth experience with respect to the degree to which it was positive or negative. It was measured for women giving birth vaginally or by emergency cesarean using the Questionnaire Measuring Attitudes About Labor and Delivery (QMAALD). This 29-item questionnaire was adapted by Marut and Mercer (1979) from a 15-item tool developed by Samko and Schoenfeld (1975). The questionnaire measures attitudes about labor and birth on a 5-point, Likert-type scale. Eleven items, which are stated as questions, refer to labor only (e.g., confidence, breathing and relaxation, feeling of control, involvement in making decisions, partner support), 12 to birth only (e.g., confidence, relaxation, pleasant or satisfying feeling state, feeling of control, partner support), 2 to a combination of labor and birth (e.g., degree expectations met, involved in process as a team member), and 3 to initial contact with the infant following birth (e.g., holding and touching baby). The Cronbach alpha coefficient reliability has ranged from .76 to .87 for internal consistency (Cranley, Hedahl, & Pegg, 1983; Fawcett & Knauth, 1996; Marut & Mercer). The higher the total score, the more positively the childbirth experience is perceived, for a possible total score of 29 to 145.
Women giving birth by planned cesarean used the Modified QMAALD. This 29-item adaptation was developed by Cranley et al. (1983). Items related specifically to labor were replaced with ones measuring perception of the preoperative experience. This adaptation has alpha reliabilities ranging from .84 to .91 (Cranley et al.; Fawcett, Pollio, & Tully, 1992; Mercer & Stainton, 1984). Scoring for the MQMAALD is the same as for the original.
The potential predictor variables, derived from the literature, are listed in the maximum models in Table 2. The obstetric and demographic variables were collected through chart review and self-report. Sixteen of the variables were created by combining labor and birth items on the birth perception questionnaire that represented predictors addressed in the literature. General self-efficacy was measured using the General Self-efficacy Subscale of the Self-efficacy Scale (Sherer et al., 1982). The scale measures general self-efficacy that is not tied to specific situations or behaviors. In a sample of 376 students, a factor analysis yielded two subscales: General Self-efficacy (17 items) and Social Self-efficacy (6 items). The original two-factor structure was reconfirmed using a refined scale composed of 23 items retained from the original 36-item scale and 7 nonscored filler items, for a total of 30 items (N = 298 students). Sherer and Adams (1983) assessed the construct validity of the questionnaire, using three other personality scales, and refined the scale to a 5-point scale (N = 101 students). Criterion validity was established with a sample of 150 veterans. The coefficient alpha for the General Self-efficacy Subscale is .86 (Sherer et al.), and possible scores for the Subscale range from 17 to 85.
Table 2. Maximum Models
|Length of labor||X||X||X|| |
|Together with infant||X||X||X||X|
|Degree expectations met||X||X||X||X|
|Degree of awareness||X||X||X||X|
|Pleased with birth||X||b||X||X|
|Equipment not a bother||X||b||X||X|
|Degree of confidence||X||X||X||X|
|Degree of relaxation||X||X||X||X|
|Degree of control||X||X||X||X|
|Ability to make choices||X||X||X||X|
|Perception of pain||X||X||X||X|
|Perception of fear||X||X||X||X|
|Worry about infant||X||X||X||X|
|Enjoy holding infant||b||b||X||X|
|Type of birth||X|| |
|Induction|| ||X||X|| |
|Forceps/vacuum|| ||X|| |
|Analgesic/Entonox|| ||X||X|| |
|Epidural|| ||X||X|| |
|Perineal trauma|| ||X|| |
Sample size requirements
All calculations were based on an alpha of .05 and a power of .80. The total number of participants required was based on those needed for the subsample of 160 participants (80 per cohort), not reported on in this article. Because 14% of women rated their experience negatively, a total sample of 573 was required to obtain a negative cohort of 80. To account for losses at 1 month, an additional 79 women and newborns were recruited for a total of 652. For the regression analysis, 10 participants per predictor were required (Shannon & Davenport, 2001). The total sample of 652 could accommodate 64 predictors plus the intercept, whereas the data subsets could accommodate from 6 to over 50 predictors.
Data were analyzed using SAS© Version 9.1. Descriptive statistics were computed to determine the distribution of the variables, assess for outliers, and describe the sample. Before model building, bivariate statistics were calculated for all of the independent variables on the birth perception score (t-tests and correlations), and correlations were calculated on the independent continuous variables to assess for relationships and multicollinearity. Four multiple linear regression models were then developed using the same approach for each. Model 1 represented question 1 and included all women in the database (N = 652) minus the 8 who had a planned cesarean with labor (n = 644). These women’s birth scores were different than the other women who had cesarean births and were more comparable to those who had vaginal births. To address question 2, model 2 included the subset of women who had vaginal births (n = 478), model 3 represented those who had emergency cesareans (n = 82), and model 4 included women who had a planned cesarean (n = 84). For model 1, type of birth was a potential predictor, whereas for the other three it was not. Only predictors that made sense clinically were included. For each, a maximum model was developed with all independent variables and then a reduced model was built. The variables that had p values ≤ .05 were kept in the model and all other variables were taken out one at a time, starting with the one having the highest p value. Each time a variable was removed, the changes in all the remaining beta coefficients were assessed. If one of the coefficients changed by more than 1 point, the variable was placed back in the model.
Model 1: All births
The mean birth perception score for the 652 women and the subset of 644 women was 102.1 (SD 13.5), with a range from 57 to 137. The adjusted R2 for the model was .93. The maximum model consisted of 26 variables (see Table 2), whereas 17 variables plus the intercept remained in the final model presented in Table 3. Based on the ranking of beta estimates, the five variables most predictive of birth perception were degree of awareness, helpfulness of partner support, being together with the infant, degree of relaxation, and type of birth.
Table 3. Model 1: All Births (n = 577)a
|Degree of awareness||4.2||0.3||3.5-4.8||<.00|
|Together with infant||3.7||0.5||2.7-4.6||<.00|
|Degree of relaxation||3.5||0.3||3.0-4.1||<.00|
|Type of birth|
| Emergency cesarean||2.8||0.6||1.6-4.1||<.00|
|Degree of control||2.7||0.2||2.3-3.2||<.00|
|Degree expectations met||2.3||0.3||1.7-3.0||<.00|
|Perception of painb||2.3||0.2||1.9-2.6||<.00|
|Ability to make choices||2.2||0.2||1.8-2.6||<.00|
|Degree of confidence||2.1||0.3||1.6-2.6||<.00|
|Worry about infantb||2.0||0.2||1.7-2.3||<.00|
|Equipment not a bother||1.3||0.4||0.6-2.0||.00|
|Perception of fearb||1.0||0.2||0.7-1.3||<.00|
Model 2: Vaginal births
The subset for this model consisted of 478 women. The mean birth perception score for these women was 102.6 (SD 13.5), with a range from 57 to 134. The adjusted R2 for the final model was 0.94. The maximum model consisted of 28 variables (see Table 2), and 16 variables plus the intercept remained in the final model shown in Table 4. The five variables most predictive of birth perception were being together with the infant, degree of awareness, helpfulness of partner support, and degree of relaxation and control.
Table 4. Model 2: Vaginal Births (n = 436)a
|Together with infant||5.3||0.8||3.8-6.8||<.00|
|Degree of awareness||4.2||0.4||3.4-5.0||<.00|
|Degree of relaxation||3.1||0.3||2.6-3.8||<.00|
|Degree of control||2.8||0.3||2.3-3.4||<.00|
|Ability to make choices||2.7||0.2||2.2-3.1||<.00|
|Perception of painb||2.3||0.2||1.9-2.7||<.00|
|Degree of confidence||2.1||0.3||1.5-2.7||<.00|
|Worry about infantb||2.0||0.2||1.7-2.3||<.00|
|Degree expectations met||1.9||0.4||1.2-2.6||<.00|
|Perception of fearb||0.8||0.2||0.5-1.2||<.00|
Model 3: Emergency cesarean births
This subset consisted of 82 women. The mean birth perception score for these women was 97.4 (SD 10.3), with a range from 73 to 118. The adjusted R2 for the final model was 0.90. The maximum model consisted of 28 variables (see Table 2), whereas 15 predictors plus the intercept remained in the final model presented in Table 5. The five variables most predictive of birth perception were degree of awareness, less worry about the infant, degree of control, enjoyed holding infant, and pleased with birth.
Table 5. Model 3: Emergency Cesarean Births (n = 77)a
|Degree of awareness||3.5||0.9||1.6-5.4||.00|
|Worry about infantb||2.9||0.8||1.2-4.5||.00|
|Degree of control||2.8||0.7||1.4-4.1||.00|
|Enjoy holding infant||2.7||0.9||0.9-4.5||.00|
|Pleased with birth||2.6||0.9||0.9-4.4||.00|
|Degree of relaxation||2.3||0.8||0.8-3.9||.00|
|Degree of confidence||2.3||0.7||0.8-3.8||.00|
|Together with infant||2.3||0.8||0.6-3.9||.01|
|Ability to make choices||1.9||0.5||0.9-2.9||.00|
|Perception of painb||1.8||0.5||0.9-2.8||.00|
|Degree expectations met||1.1||0.4||0.3-1.9||.01|
|Perception of fearb||1.0||0.4||0.2-1.8||.01|
Model 4: Planned cesarean births
This subset consisted of 84 women. The mean birth perception score for these women was 103.6 (SD 15.3), with a range from 75 to 137. The adjusted R2 for the final model was 0.96. The maximum model consisted of 24 variables (see Table 2), and 14 predictors plus the intercept remained in the final model shown in Table 6. The five variables most predictive of birth perception were perception of fear, pleasantness experienced, being together with the infant, enjoyed holding infant, and helpfulness of nursing support.
Table 6. Model 4: Planned Cesarean Births (n = 69)a
|Perception of fearb||5.3||1.4||2.5-8.1||.00|
|Together with infant||3.6||0.9||1.9-5.4||.00|
|Enjoy holding infant||3.4||1.0||1.4-5.4||.00|
|Degree of relaxation||3.4||0.9||1.6-5.1||.00|
|Worry about infantb||3.2||1.1||1.1-5.3||.00|
|Degree expectations met||3.2||1.0||1.1-5.2||.00|
|Ability to make choices||3.1||0.9||1.3-4.9||.00|
|Degree of control||3.0||0.6||1.9-4.2||<.00|
|Perception of painb||2.8||0.9||1.1-4.5||.00|
|Equipment not a bother||2.6||1.0||0.6-4.6||.01|
|Degree of confidence||2.2||0.7||0.8-3.6||.00|
The findings are presented with respect to each of the strongest predictors across the models. When type of birth was entered into model 1, results demonstrated that women who had a planned cesarean birth scored significantly lower on birth perception than those who had an emergency cesarean or a vaginal birth. This finding is not supported by the literature. Although cesarean birth has been reported to be a strong predictor of a negative experience (CMA, 1987; Mercer et al., 1983; Soet et al., 2003), planned cesarean was not identified specifically. Emergency cesarean, on the other hand, has been shown to be a strong predictor (Seguin et al., 1989; Waldenstrom, 1999; Waldenstrom et al., 1996), with Waldenstrom et al. (2004) reporting it to be the most significant predictor of a negative birth experience.
The strongest predictor across the models was the degree of awareness of events during labor and birth.
The strongest predictor across the models was the degree of awareness of events during labor and birth, particularly in the all births and emergency cesarean birth models. It was the second strongest predictor in the vaginal birth model but did not predict perception in the planned cesarean model. This finding was not reported in the literature; however, awareness may have been represented in other studies by other variables such as degree of control or decision making. It is apparent that women having a vaginal or emergency cesarean birth wish to be aware of the events that are happening during labor and birth. This enables them to be more active participants and enhances their memory of the events that occurred. A clearer memory of the experience facilitates integration of the birth experience, which is important in the transition to mothering (Konrad, 1987).
Being together with the infant within 1 hour of birth was the next strongest predictor across the four models: strongest in the vaginal birth model, second in the all births model, and third in the planned cesarean model. This finding suggests that whether women have a vaginal or cesarean birth, perception of their experience is strongly predicted by whether they are able to be with their infant. This is supported in the literature (Fenwick et al., 2003; Mercer et al., 1983; Waldenstrom et al., 2004) and demonstrates the importance of nonseparation of mother and baby for all births, one of the principal tenants of Family-Centered Care (Health Canada, 2000).
Helpfulness of partner support, degree of relaxation, and degree of control were the next strongest predictors across the four models. Unlike the other models, partner support although predictive was one of the weakest predictors in the planned cesarean model. Partner support is well documented in the literature as being important to women during childbirth (Czarnocka & Slade, 2000; Hardin & Buckner, 2004) and was found to be the strongest predictor of birth perception by Mercer et al. (1983).
Degree of relaxation as a predictor was not reported in any of the studies that investigated several predictors of birth perception simultaneously. However, relaxation and breathing have been associated with satisfaction with the birth experience (Green, 1993; Green et al., 1990). Relaxation has been shown to decrease anxiety and perception of pain and to increase a woman’s ability to cope and is one of the skills taught to women during prenatal classes (Murray & McKinney, 2006). Stress and anxiety have been reported to have a negative influence on women’s perceptions of their childbirth experience (Waldenstrom, 1999). As well, relaxation may be associated with maintaining control (Mackey, 1995).
The findings demonstrated that maintaining control during childbirth for all women is important. Perceived personal control has been reported to be a predictor of childbirth perception/satisfaction (Goodman et al., 2004; Waldenstrom et al., 2004). The meaning of control has varied across studies and may be perceived as internal or external control. Some women pride themselves in their ability to maintain control over their behavior/body and are more likely to evaluate the experience as positive if they are satisfied with their own performance. Other women view control as being able to influence the environment in which they labor and give birth (Green & Baston, 2003; Hardin & Buckner, 2004).
Degree of fear during birth was the strongest predictor in the planned cesarean model, whereas it was the weakest predictor in the other models. Fear was not a predictor in any of the studies that examined several predictors simultaneously. It is possible that fear was represented in other studies by variables such as anxiety, loss of control, and worry about the baby. With the current trend toward elective planned cesareans, this is a revealing finding. It may reflect the psychological stress that some women who are having a planned cesarean are under or possibly some inherent differences in these women. Researchers have reported that women who fear childbirth are more likely to request a planned cesarean (Melender, 2002; Waldenstrom et al., 2006). One reason for this fear may be a negative or traumatic experience with a previous birth, which is usually a difficult vaginal birth or an emergency cesarean (Fenwick et al., 2003; Melender). It is also possible that other traumatic life events may precipitate fear.
The second strongest predictor of the planned cesarean model was pleasantness experienced (pleasant or satisfying feeling state during birth). This was also a predictor in the other models, but again not as strong. It is possible that women having a planned cesarean want to have a pleasant and satisfying experience, which may be different from a previous experience. Marut and Mercer (1979) reported that this variable was associated with women’s perceptions of their birth experience but planned cesareans were not included in their study.
Worry about the infant’s condition was the second strongest predictor in the emergency cesarean birth model. Although this factor was predictive in the other models, it was not as strong. This result makes sense clinically, as often the baby’s health is in jeopardy in an emergency cesarean. In the current study, 26% of emergency cesareans were due to fetal distress. Worry about the infant’s condition was not found to be a predictor of birth perception in any of the studies that examined several factors simultaneously, although anxiety was a predictor in Waldenstrom’s (1999) study. Marut and Mercer (1979) reported that women having an emergency cesarean worried significantly more about the infant’s condition than those having a vaginal birth, and this was one of the factors that had the greatest influence on their perception of their birth experience. Ryding, Wijma, and Wijma (1998) also found that the most common fear for women undergoing an emergency cesarean was concerns for the baby.
Although predictive in all models, the fact that nursing care/support was not rated more strongly except for in the planned cesarean model is also an interesting finding and does not concur with the literature. Nurse/midwife support was found to be one of the strongest predictors of birth perception/satisfaction and satisfaction with care in several studies (Brown & Lumley, 1994, 1998; Mercer et al., 1983; Waldenstrom, 1999; Waldenstrom et al., 1996, 2004). Other researchers have reported the significant role that nurses/midwives/doulas play in enhancing women’s perceptions of their birth experience (Hallgren et al., 2005; Hardin & Buckner, 2004; Hodnett, 2003; Lundgren, 2005). The reason that nursing care/support was not a stronger predictor is unclear, although the current study specifically investigated nursing support unlike many others that studied midwife and/or doula support. It is important to note, however, that many of the other predictors in the models are influenced by the care and support nurses provide.
Most of the predictors identified are amenable to intervention by nurses, which may either prevent a negative experience or help alter the perception afterwards.
Implications for nursing practice
The evidence obtained through this study provides guidance for perinatal nursing practice. Its most important contribution to nursing practice is that, to the researchers’ knowledge, it is the first Canadian study to simultaneously investigate a large number of predictors of women’s perceptions of their birth experiences. It is also the first reported study to examine predictors of birth perception based on the type of birth. This elicits information regarding the needs of women with which nurses can endeavor to individualize care in order to promote positive experiences and prevent negative ones. Although not all predictors are modifiable, many are amenable to nursing intervention, which may either prevent a negative experience or help alter the perception afterwards.
Nurses can educate women prenatally about how awareness during childbirth may assist them in feeling positive about their experience. They can encourage women to ask to have their babies with them immediately following birth. Nurses can also educate partners about the significance of their role and how they can be most supportive during and after childbirth. They can provide women and their partners with a repertoire of relaxation techniques. In addition, nurses can educate them about the possibility of an emergency cesarean birth and what to expect should it happen. Prenatal care providers can also assess for fear in women having a planned cesarean and attempt to deal with the underlying cause. During labor and birth, it is suggested that nurses endeavor to support women’s need to be aware by informing them of events as they are unfolding. Nurses can also advocate for the newborn to be with the mother as soon as feasible following all types of birth. Based on the significant influence of the partner, it is suggested that nurses endeavour to enhance his/her support. If a partner is not supportive, then the nurse can make an extra effort to enhance the support she/he provides. Operating room personnel may require education about the significance of partner support and of partner and newborn presence in the recovery room.
Women will also benefit from nurses’ support in assisting them to relax no matter what method is chosen. As well, nurses need to consider the importance of control for women, regardless of whether it means internal control over their own bodies and behavior or external control over the environment. Careful assessment of each woman is important and assisting her to maintain control as she wishes is essential. Last, because of the importance women placed on being aware of events during birth, nurses present during labor and birth can encourage women to talk about their experience afterwards and help them fill in events that are not clearly remembered. This may assist them in modifying their perception if they have not been fully aware of the experience, which may promote integration of the experience and the transition to mothering. If a woman has had a negative experience, an intervention to deal with her feelings can be implemented.
When the sample of 652 women was compared with the 46% of eligible women who refused participation, it was underrepresentative of women having cesarean births by 7%. Clinically it makes sense that these women were less likely to participate due to a longer recovery time but this may have affected the results. The sample was also underrepresentative of women with complications by 10.7%. It is quite conceivable that these women felt less like completing the study; however, this difference may have had an effect on the predictors of birth perception. It is also possible that nonparticipants with complications may have been more negative about their experience, resulting in higher study mean birth perception scores. Last, the results are only generalizeable to women who have healthy, term, singleton infants weighing greater than or equal to 2500 g and who give birth in a relatively low intervention environment, where one-on-one support in active labor is the standard of care. They are not generalizeable to women having a general anesthetic or experiencing perinatal loss.
Recommendations for future research
Further qualitative research is required to explore the influence of fear on women who are having a planned cesarean. The impact of negative life experiences other than childbirth on this predictor also requires exploration. Research is needed to confirm the finding that women having planned cesarean births perceive their birth experience less positively than other women. As well, studies with more culturally diverse populations will help determine if the predictors of the childbirth experience are similar across cultures.
Of the 20 predictors of maternal perceptions of the childbirth experience, most were amenable to nursing interventions: enhancement of patient awareness, relaxation, and control; promotion of partner support; and provision of immediate opportunities for women to be with their babies. Nurses have a responsibility to use the knowledge obtained to guide their practice in an effort to enhance every woman’s birth experience.
Funded by the Groupe de recherche interuniversitaire en soins infirmiers de Montréal, Canadian Nurses Foundation Nursing Care Partnership Program, Isaac Walton Killam Health Centre, University of Prince Edward Island, Children’s Health & Applied Research Team, UPEI School of Nursing and Le Fonds de la recherche en santé du Québec.