An educational intervention to improve women’s ability to cope with childbirth


  • Wan-Yim Ip,

    1. Authors: Wan-Yim Ip, PhD, RN, RM, RTN, Associate Professor, Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, Hong Kong; Catherine SK Tang, PhD, Professor, Deputy Head, Department of Psychology, National University of Singapore, Singapore; William B Goggins, ScD, Assistant Professor, Nethersole School of Nursing and School of Public Health, Chinese University of Hong Kong, Shatin, Hong Kong
    Search for more papers by this author
  • Catherine SK Tang,

    1. Authors: Wan-Yim Ip, PhD, RN, RM, RTN, Associate Professor, Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, Hong Kong; Catherine SK Tang, PhD, Professor, Deputy Head, Department of Psychology, National University of Singapore, Singapore; William B Goggins, ScD, Assistant Professor, Nethersole School of Nursing and School of Public Health, Chinese University of Hong Kong, Shatin, Hong Kong
    Search for more papers by this author
  • William B Goggins

    1. Authors: Wan-Yim Ip, PhD, RN, RM, RTN, Associate Professor, Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, Hong Kong; Catherine SK Tang, PhD, Professor, Deputy Head, Department of Psychology, National University of Singapore, Singapore; William B Goggins, ScD, Assistant Professor, Nethersole School of Nursing and School of Public Health, Chinese University of Hong Kong, Shatin, Hong Kong
    Search for more papers by this author

Wan-Yim Ip, Associate Professor, Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, Hong Kong. Telephone: +852 26096229. E-mail:


Aim and objective.  To test the effectiveness of an efficacy-enhancing educational intervention to promote women’s self-efficacy for childbirth and coping ability in reducing anxiety and pain during labour.

Background.  The evidence of the effective application of the self-efficacy theory in health-promoting interventions has been well established. Little effort has been made by health professionals to integrate self-efficacy theory into childbirth care.

Design.  Randomised controlled trial.

Methods.  An efficacy-enhancing educational intervention based on Bandura’s self-efficacy theory was evaluated. The eligible Chinese first-time pregnant women were randomly assigned to either an experimental group (n = 60) or a control group (n = 73). The experimental group received two 90-minute sessions of the educational programme in between the 33rd–35th weeks of pregnancy. Follow-up assessments on outcome measures were conducted within 48 hours after delivery. The short form of the Chinese Childbirth Self-Efficacy Inventory was used to measure maternal self-efficacy prior to labour. Evaluation of pain and anxiety during the three stages of labour and performance of coping behaviour during labour were measured by the Visual Analogue Scale and Childbirth Coping Behaviour Scale respectively.

Results.  The experimental group was significantly more likely than the control group to demonstrate higher levels of self-efficacy for childbirth (p < 0·0001), lower perceived anxiety (p < 0·001, early stage and p = 0·02, middle stage) and pain (p < 0·01, early stage and p = 0·01, middle stage) and greater performance of coping behaviour during labour (p < 0·01).

Conclusions.  The educational intervention based on Bandura’s self-efficacy theory is effective in promoting pregnant women’s self-efficacy for childbirth and reducing their perceived pain and anxiety in the first two stages of labour.

Relevance to clinical practice.  Relief of pain and anxiety is an important issue for both women and childbirth health professionals. The efficacy-enhancing educational intervention should be further developed and integrated into childbirth educational interventions for promoting women’s coping ability during childbirth.


Although researchers have reported that prenatal childbirth education provided by midwives does have an impact on women’s ability to cope with labour, some findings must be interpreted with caution due to their methodological weaknesses. Additionally, many studies were not guided by a theoretical framework to test and build upon existing educational theories. Self-efficacy has been used as a theoretical framework to explore, explain and predict health behaviour in a variety of health promoting research (Lenz & Shortridge-Baggett 2002). It is an important component of many therapeutic programmes aimed to reduce negative emotions and empowering coping behaviours. Previous experimental studies have demonstrated methods of manipulating self-efficacy and their effect on initiating and maintaining health behaviour (Dijkstra & Wolde 2005, Elfhag & Rossner 2005). Given the gaps in studies applying self-efficacy theory to childbirth education, an efficacy-enhancing childbirth programme needs to be developed and its evaluated effectiveness in the childbirth care.


Antenatal childbirth education is widely accepted as a positive approach to preparing the pregnant women for the experience of childbirth. However, the relationships among childbirth education and labour outcomes are inconclusive. The most recent meta-analysis concludes that there is insufficient evidence to determine the effects of antenatal education on psychological, physical and social adjustment (Gagnon 2004). The limited evidence on childbirth education may be explained by the methodological weakness of studies, such as small sample size and limited theoretical-grounded interventional approaches (Gagnon 2004). Nevertheless, other researchers (Lowe 2000, Spiby et al. 2003) comment on the traditional nature of childbirth education focussing on transferring information about childbirth rather than putting effort into preparing pregnant women to identify and strengthen personal coping resources and confidence for childbirth.

Self-efficacy theory (Bandura 1997) provides a framework for a specific educative intervention that enables individuals to develop their self-care behaviours. It is a dynamic cognitive process where the individual evaluates her/his capabilities to perform in designated behavioural domains. There are two conceptually independent components that comprise an individual’s perceptions of self-efficacy: outcome expectancy (OE) and efficacy expectancy (EE) (Bandura 1986). Outcome expectancy refers to the belief in the likely consequences that behaviour will produce whereas EE refers to an individual’s perceived ability to perform a behaviour. A large body of evidence has revealed that the impact of different methods of treatment on health behaviour was partly mediated through their effects on perceived self-efficacy (Lenz & Shortridge-Baggett 2002).

There are four approaches to enhance the perception of self-efficacy (Bandura 1986, 1997). First, the performance of accomplishments provides the most influential source of efficacy information, because it is based on the authentic mastery experiences. Second, the vicarious experience enhances the observers’ belief about their own capabilities when they have appreciated the accomplishment of their peer group through perseverance and determination (Gonzalez et al. 1990). Third, verbal persuasion involves exposure to the verbal judgments of others and is often used in combination with other sources (Bandura 1997). Finally, physiological status provides the fourth source of efficacy information. Persons who are equipped with skills to reduce aversive physiological reactions will evaluate their self-efficacy positively (Maddux & Lewis 1995). All strategies that reduce and control emotional tension, such as relaxation training, can enhance self-efficacy (Gattuso et al. 2001). Therefore, this final source of self-efficacy can be seen in the context of the other three sources. In fact, researchers (Keefe et al. 1990, Bandura 1997, Cheal & Clemson 2001) stated that the inclusion of a combination of the first three sources of self-efficacy information in a multi-focussed interventional programme, which included health education, skills practice and mastery experience, role modelling and self-affirming verbal persuasion, produced the best effect.

Self-efficacy has been shown to be important in influencing how a labour is perceived and coped with physically. According to Bandura (1997), a strong belief in one’s efficacy to exercise some control over one’s physical condition may serve as a psychological prognostic indicator of the probable level of health functioning. Women with a higher level of self-efficacy or confidence in their abilities to use particular behaviours to cope with the delivery report lower levels of anxiety (Lowe 2000) and decreased pain perception (Lowe 1996). That perceived efficacy makes pain easier to manage is corroborated by other studies of acute and chronic clinical pain (Holman & Lorig 1992, Resnick et al.2007). In addition, those with a firm belief that one can exercise some control over pain and one’s physical functioning report fewer pain behaviours, less mood disturbance and better psychological well-being (Green & Baston 2003).

Although several studies on childbirth have revealed the important role of self-efficacy on women’s ability to cope with labour and delivery (Lowe 2000, Slade et al. 2000), little effort has been made by health professionals to integrate self-efficacy theory into perinatal care. Based on the evidences of the effective application of self-efficacy theory in health care interventions in other health-care areas, (Lev & Owen 2000, Dennis et al. 2001), it is postulated that the efficacy-enhancing educational intervention can promote women’s self-efficacy for childbirth and thus their abilities to cope with childbirth.



This paper describes a single-blind, randomised controlled trial of a self-efficacy enhancing educational programme (SEEEP) which involved two 90-minute sessions on promoting Chinese first-time pregnant women’ perceived self-efficacy for childbirth and coping abilities for pain and anxiety during labour.

Research hypothesis

  • 1Women who receive the SEEEP will report significantly higher OE and EE from women who receive usual care;
  • 2Women who receive the SEEEP will report significantly lower level of anxiety at each of the three stages of labour (early, middle and last) from women who receive usual care;
  • 3Women who receive the SEEEP will report significantly lower level of pain intensity at each of the three stages of labour (early, middle and last) from women who receive usual care;
  • 4Women who receive the SEEEP will report significantly higher level of coping behaviour during labour from women who receive usual care.


The study was conducted from August 2003–April 2004 in Hong Kong. First-time pregnant women who sought antenatal services at out-patient clinics of a regional teaching hospital (OPD) were invited to participate in the study if they met the following criteria: (1) a Hong Kong Chinese resident, (2) at the 32nd–34th weeks’ gestation, (3) planning to have a singleton by vaginal delivery, (4) aged 18 or above, (5) able to read and understand Chinese and (6) willing to give written informed consent. Women were excluded if they (1) would not stay in Hong Kong after discharge and (2) had a planned caesarean section.

Interventions related to psychosocial and educational modalities in pregnant women often result in a medium effect size (Saisto et al. 2000). It was therefore conservative to regard SEEEP as having a medium-size effect on the outcome variables. The required sample size for the detection of a medium effect size of 0·5 at a power of 95% and two-tailed alpha of 0·05 is 50 in each group (Stevens 2002). However, in reference to the previous studies with given attrition rates ranging from 40–50% (Brugha et al. 1998, Saisto et al. 2001), we envisioned an attrition rate of approximately 48%. Hence a total sample size of 192 subjects with 96 in each comparative group was considered to be the optimal for this study.


The short form of the Chinese Childbirth Self-Efficacy Inventory

The original Childbirth Self-Efficacy Inventory (Lowe 1993), based on Bandura’s (1997) self-efficacy theory, is a psychometrically sound diagnostic tool for evaluating women’s coping ability for childbirth (Lowe 1993, Drummond & Rickwood 1997, Sinclair & O Boyle 1999). There are two types of expectancy that exert powerful influences on childbirth behaviour, namely, OE and EE (Bandura 1977). OE is the belief that enhanced childbirth experience may result from engaging in the specific behaviours, whereas EE for childbirth refers to a woman’s perceived ability to perform specific coping behaviours during labour.

The short form of the Chinese Childbirth Self-Efficacy Inventory (CBSEI-C32) (Ip et al. 2007) comprises two parallel subscales: outcome expectancy (OE-16) and efficacy expectancy (EE-16). The two subscales, which consist of the same 16 items addressing common coping behaviours for childbirth, such as breathing and relaxation exercise, distraction and positive thinking, were adopted for measuring the overall perception of self-efficacy for labour. All responses are made on a 10-point Likert-type scale from ‘1 – not at all helpful’ to ‘10 – very helpful’ for the OE-16 subscale and ‘1 – not at all sure’ to ‘10 – very sure’ for the EE-16 subscale. Each measure yields a scale score between 16–160. The two subscales of the CBSEI-C32 have been shown to be valid and reliable self report measures of women’s confidence in coping with childbirth among Chinese population with reported Cronbach’s alpha over 0·9 (Ip et al. 2005, 2007). In this study, the OE-16 and EE-16 of the CBSEI-C32 consistently reported very good reliability with Cronbach’s alpha ranging from 0·92–0·97 across the two measured time points.

The VAS for pain and anxiety during labour

Women’s self-reported anxiety and pain at three stages of labour were measured using 10 cm Visual Analogue Scales (VAS) with Chinese verbal anchors ranging from one end of the scale (0 = no anxious/no pain at all) to the other end (10 = extremely anxious/unimaginable pain). The three stages of labour were the first third, second third and last third of the duration of the first stage of labour (the time which begins with the onset of regular contractions and ends with full dilation of the cervix). The VAS had been used in previous studies and shown to be an accurate and sensitive tool for the assessment of the levels of anxiety and pain in three stages of labour as recalled by the postpartum women (Ip et al. 2005, Cheung et al. 2007).

Childbirth coping behaviour scale

A Childbirth Coping Behaviour Scale (CCB) is a self-developed scale to measure the participants’ use of the coping behaviours during labour. All 16 items which were extracted from the CBSEI-C32 (Ip et al. 2007), used a four-point Likert scale (from ‘1 = never’–’4 = often’) to indicate their frequency of non-pharmacological coping behaviours executed throughout labour. Examples of the items are ‘relax my body’, ‘use breathing during labour contractions’ and ‘focus on the person helping me in labour’. The measure yields a scale score between 16–64. The Cronbach’s alpha of CCB in the present study was 0·86.

The self-efficacy enhancing educational programme

The SEEEP aimed to enhance self-efficacy in the first-time pregnant women, in the hopes that they could successfully accomplish behaviours for coping with stress and pain in childbirth. It comprised two 90-minute interactive educational sessions. The first researcher conducted the sessions on two consecutive Saturday afternoons in the hospital clinical venues. The timing of the educational programme was chosen to facilitate working women who were not available during workdays. Each session was designed in a grouping of six people or fewer for an optimal interaction and discussion.

According to the literature on multidimensional efficacy-enhancing interventions based on Bandura’s self-efficacy theory (Keefe et al. 1990, Bandura 1997, Cheal & Clemson 2001), the protocol of SEEEP was an integration of the first three sources of self-efficacy information (performance accomplishments, vicarious experience and verbal persuasion) that included five main learning activities. First, offering health education to prepare and motivate participants to participate actively in the learning activities. The content of the information included the biopsychological phenomena of childbirth; the strategies of coping with childbirth discomfort and its relationships with self-efficacy for childbirth. Second, providing a demonstration of coping behaviours, including breathing and relaxation, distraction and cognitive restructuring of pain, to help participants control emotional tensions and pain during labour. Third, returning demonstration of the taught coping skills by the participants. Fourth, vicarious observation of role models in a VCD depicting how two Chinese pregnant mothers modelled perseverant success in reinstating control over childbirth with the use of coping behaviours. Lastly, making a verbal contract to rehearse the taught coping skills with the use of mastery aids at home between and after sessions.

To enhance performance accomplishment, the participants were given a pamphlet summarising the main coping methods for guiding self-practice and a practice log for daily entries as a means of self-monitoring of their successful practice efforts in the hope that a sense of control was built through structured practice in the exercise of control over challenging conditions (Speck & Looney 2001). Additionally, verbal persuasion through positive encouragement and giving the participant a sense of accomplishment from the results of her practice by the researcher and other participants could also help the participant to gain a sense of confidence in her self capability to cope with childbirth.

The control group received usual care, which included a regular physical check-up and attending childbirth classes on a voluntary basis. In Hong Kong, there was no standardised system of antenatal childbirth education. Participants could attend various types of childbirth classes, which were conducted by either public hospitals or community health care centres. The classes were structured in various formats, ranging from one session on a ‘walk-in’ basis to a programme of 4–6 sessions in a series.

Procedure and ethical consideration

Upon receipt of ethical approval from an institutional Ethics Committee, two part-time research assistants (RAs) were trained for recruitment of the eligible subjects and data collection. One RA was responsible for reviewing the clinic’s follow-up records of the pregnant clients and screened the potentially eligible participants. The RA would then approach the eligible women who were present for an antenatal check-up to explain the aim and procedure of the study. All of the participants were assured that their responses to the questionnaires would be kept confidential, that their participation was entirely voluntary and that they could withdraw at any time.

Women who had signed the consent form were asked to complete a CBSEI-C32 and a socio-demographic questionnaire including questions on attendance at antenatal classes, preferred mode of delivery and medical history. On completion of the pretest questionnaires as a baseline measure, the women were randomly assigned to either the experimental or the control group by a computer-generated random-number table. For the women in the experimental group, the RA would further explain the purpose and arrangement of the SEEEP and encouraged their participation by arranging convenient times before their 36th of gestations for the two-day programme. Before leaving the participants, the RA gave a copy of CBSEI-C32 as a posttest measure with a stamped self-addressed envelop to the participants. The participants were instructed to complete it at their 37th weeks of gestation and to mail it back to the researcher.

To follow up the postnatal women another RA, who was blinded to the groups’ assignment, visited the women within 24–48 hours after their delivery. The women were asked to recall their pain and anxiety levels at three time points of their labour: early, middle and late stages, and their performance of coping behaviours during labour by completing the VASs and CCB respectively.

Data analysis

Data analysis was carried out using spss version 13 (SPSS Inc., Chicago, IL, USA). Descriptive statistics, including the frequency distribution, mean and standard deviation, were used for sample description. Inferential statistics were used to assess the homogeneity of those who completed and those who discontinued the study on the socio-demographic variables and baseline measures. Following the recommendations of Vickers and Altman (2001) for the analysis of controlled trials with baseline and follow-up measures we used multiple linear regression models to test hypothesis 1. Separate models were fitted, one with posttest OE-16 as the outcome and pretest OE-16 and an indicator variable for group membership as covariates, and one with posttest EE-16 as the outcome and pretest EE-16 and the group indicator as covariates. A doubly multivariate repeated measures analysis (Stevens 2002) was used to test research hypotheses 2 and 3. This was mainly because the participants were measured retrospectively on two variables (pain and anxiety) at each of the three stages of labour for each hypothesis. Follow-up univariate analyses and planned comparisons were conducted. The independent samples t-test was used to test hypothesis 4; testing for a difference between the two comparison groups in the performance of coping behaviour during labour. For all analyses, a two-sided p-value less than 0·05 was considered statistically significant.

The aim of this study was to focus on those participants who completed the entire programme. Therefore, no further analysis was conducted on the data of those who left the study prematurely. Any participants who did not attend the day 1 and/or day 2 sessions in full were excluded from the data analysis to ensure the variability of interventional dose effects (Stevens 2002).



Three hundred and sixty-six women who met the inclusion criteria were invited to participate in the study. One hundred and ninety-two women (52·5%) agreed to participate and signed the consent forms. Sixty women in the experimental group completed the study, while 36 withdrew, and 73 in the control group completed the study, while 23 withdrew (Fig. 1). Hence 133 completed questionnaires were obtained for data analysis, giving a response rate of 69·3%.

Figure 1.

 Study flowchart mapping recruitment and randomisation of participating pregnant women to experimental and control groups.

There were no significant differences in the selected baseline measures (p > 0·05) between the participants who completed the study and those who withdrew. Table 1 outlines the sample characteristics and their baseline data and the significance values associated with chi-squared or t-tests that evaluated potential differences between the experimental and control groups. The groups did not differ significantly in any of the baseline measures, including age, educational level, occupational status, family income, marital status, gestational age and antenatal class attendance, indicating that the sample randomisation procedures used were effective.

Table 1.   Differences in sample characteristics and baseline measures between experimental and control groups (n = 133)
 Experimental (n = 60)Control (n = 73)Total (n = 133)p
Marital status
Educational level
 Secondary form < 3711·101622·602317·29 
 Secondary form 3–74474·104561·308966·920·38
Family income in US dollars
Antenatal class attendance
Maternal age27·885·0727·815·0927·845·060·14
Gestational age278·638·35276·387·77257·7910·710·46

As a whole, the samples were composed primarily of young married Chinese with a good health history and secondary education qualifications. Approximately one-third of them were housewives and half of the respondents were employed in clerical work, with a monthly family income of less than US$ 2564. Regarding the women’s preferred mode of labour, almost all women (n = 130, 97·74%) favoured a vaginal delivery, with only 3 (2·26%) favouring caesarean section. The majority of women (n = 98, 73·68%) had attended antenatal class before joining the study and the findings on attendance were comparable to those reported by the study obstetric unit.

Self-efficacy for childbirth

The pretest data of the two subscales of CBSEI-C32 in Table 1 demonstrated that the participants had just above average level of self-efficacy for childbirth and there were no statistically significant differences between the groups (p = 0·92 for OE-16 and p = 0·64 for EE-16). The mean (SD) posttest OE-16 at 37th week were 121·70 (22·80) and 104·93 (19·79) for the experimental and control groups respectively whereas the mean (SD) posttest EE-16 at 37th week were 119·22 (23·64) for the experimental and 98·46(23·36) for the control group. Results of the linear regression analysis indicated that the mean adjusted difference in posttest OE between the experimental group and the control group was 17·4 (95% CI = 10·2–24·6, p < 0·0001) while the mean adjusted difference in posttest EE was 22·3 (95% CI = 14·2–30·3, p < 0·0001). Further adjustment for marital status, family income, gestational age and maternal education, age and occupation in the regression analysis resulted in only negligible changes in the adjusted differences.

Pain and anxiety during labour

A doubly multivariate repeated measures analysis was performed to investigate the group, time and group by time effects for two dependent variables: anxiety and pain, over the three time periods of labour. Results of the evaluation of assumptions of the analysis were satisfactory. Bartlett’s test for sphericity indicated a significant degree of intercorrelation among dependent variables (df = 1, p < 0·001). The pooled within-cell correlation between anxiety and pain was 0·37, which is in excess of 0·30 (Tabachnick & Fidell 2001). The homogeneity of the variance–covariance matrix was confirmed by Box’s M-test (F [21,58247] = 1·99, p < 0·01).

With the use of Wilks’ criterion, the combined dependent variables (anxiety and pain) were significantly affected by group (F [2,130] = 4·40, p < 0·05), time (F [4,128] = 0·37, p < 0·001) and their interaction (F [6,128] = 3·52, p < 0·01). The results reflected moderate to large effects of the two main independent variables and their interactions upon combined dependent variables; all η2 > 0·06 (Cohen 1992). Since there were significant interactions between time and group on the combined outcome variables, the intervention effects on each of the individual outcome variables were examined by repeated measure univariate analysis of variance.

The profile plots of Figs. 2 and 3 show that the difference between groups in both anxiety and perceived pain intensity were not constant across the three stages of labour. This was reflected in the fact that the group × time interaction for anxiety and pain was significant in both univariate F ([1,131] = 8·18, p < 0·01, η2 = 0·06), and F ([1,131] = 4·45, p < 0·025, η2 = 0·03) respectively. The follow-up pairwise comparisons between the comparative groups at three individual stages of labour revealed significant differences in the perceived levels of both anxiety and pain intensity in the early and middle stages of labour but not in the last stage (Table 2). The results suggested that the intervention effectively brought relief in both pain and anxiety that are present at the early and middle stages of labour but attenuated by the last stage of labour. Therefore, both hypothesis 2 and hypothesis 3 were partly confirmed.

Figure 2.

 Profile plot of anxiety across three study time points. T1 = Early stage of labour, T2 = Middle stage of labour, T3 = Last stage of labour.

Figure 3.

 Profile plot of pain across three study time points. T1 = Early stage of labour, T2 = Middle stage of labour, T3 = Last stage of labour.

Table 2.   Pairwise comparisons of mean anxiety and pain across the three stages of labour
  Experimental group (n = 60) Control group (n = 73)Mean differencep-value95% confidence interval of the difference
Mean (SD)Mean (SD)LowerUpper
  1. *Range of score.

Anxiety (0–10)*
 Early stage of labour3·26 (2·97)4·72 (2·89)−1·72<0·001−2·82−6·10
 Middle stage of labour5·16 (2·79)6·40 (2·68)−1·250·02−2·30−0·20
 Last stage of labour6·21 (3·48)5·58 (3·57)−0·890·19−0·442·23
Pain (0–10)*
 Early stage of labour2·72 (2·00)4·21 (2·62)−1·69<0·01−2·54−0·83
 Middle stage of labour5·87 (2·00)6·84 (2·21)−1·070·01−1·86−0·28
 Last stage of labour7·99 (2·51)7·99 (3·03)  2·900E-020·96−1·00  1·06

Coping behaviours during labour

The independent samples t-test showed that the experimental group reported a significantly higher level of coping behaviour during labour than the control group [mean for experimental group was 48·94 (SD 7·9) vs. 44·99 (SD 0·91) for control group; t (133) = 2·92, CI: 1·27–6·65, p < 0·01]. The magnitude of the difference in the means was moderate [η2 = T2/T + (N1 + N2 − 2) = 0·07].


The findings show that the two-session educational programme based on Bandura’s self-efficacy theory was beneficial in promoting first-time pregnant women’s self-efficacy for childbirth. The programme also promoted the women’s coping behaviour and decreased their perceived pain and anxiety in the first two stages of labour. In interpreting these findings, it is important to note that this study employed a selected sample of Hong Kong Chinese women who were married, being pregnant for the first time and expected to have a normal vaginal delivery. The majority had attained secondary education qualifications and came from the lower middle class. The homogeneity of the sample limits the conclusions that can legitimately be drawn from the study concerning the relevance of parity, socioeconomic status and preference for labour.

In this study, the two-session weekly childbirth educational intervention, based on Bandura’s self-efficacy theory was found effective in raising women’s confidence in their capabilities to cope with childbirth which was a determinant for the subsequent coping behaviours in labour. It was expected that the women who believed they were able to execute the taught coping behaviour would be able to generate the motivation necessary to practice it and carry it out in actual birth (Bandura 1977). However, the level of coping skills mastered after the educational programme might vary among the participants. Confounding factors such as partner’s support and how this affected the individual development of self-efficacy was not known. The availability of these data may increase the understanding between women’s sense of self-efficacy and the development of skills on how to influence their own motivation and behaviour (Bandura 1991).

The finding that the experimental group demonstrated lower levels of pain intensity and anxiety than the control group in the first two stages of labour was supported by previous studies. According to McCrea and Wright (1999), being involved actively in the labour process, as indicated by an increase in coping activities during labour, may help to promote women’s feelings of ‘being in control’ over a painful stimulus, which, in turn, may result in a higher tolerance of pain. According to Willmuth (1975)‘Being in control in this sense referred to the feeling of being an active participant in labour and delivery rather than a passive object of care’ (p. 30). Lowe (1996) has reported that maternal self-efficacy in childbirth was associated with lower perceived pain intensity in labour among American populations. A local study of 85 Chinese women in labour also found that women who had a higher sense of control over themselves during the process of labour were reported to have less perceived pain in childbirth (Law 2003). Similarly, the sense of feeling in control was also found to be helpful in decreasing maternal anxiety during labour both in the Western (Ayers & Pickering 2005) and the local study (Cheung 2004). Although the experimental group in this study demonstrated lower pain and anxiety levels in the first and middle stages of labour, there was no difference from the control group in the last stage of labour.

The non-significant findings on anxiety and pain in the last stage of labour can be explained by this unique feature of childbirth: the last stage of labour is the most intensely painful period of childbirth. Even though the labouring women are trying to cope with their labour pain by using coping skills, they may be overwhelmed by the intensity of the pain towards the end of labour (Leifer 2003). In Niven and Gijsbers (1984) study, many postpartum women reported that the intensity of pain affected their strategy. An increase in pain intensity, such as that which accompanies progress from the early stage to the last stage of labour, could render an effective strategy useless and force a change in strategy or a resort to pharmacological analgesics. In interviews with 100 childbearing women from different sociocultural contexts, Callister et al. (2003) found that women’s abilities to cope with labour pain vary in different stages of labour. As time passed and the intensity of the labour experience increased, the labouring women described themselves as exhausted and vulnerable and needed help and support from others to master their birth experience. This implies that the use of coping strategies may have a correlation with low levels of pain, but not with high levels (Niven & Gijsbers 1984). Further studies are required to establish the effectiveness of coping strategies in modulating the pain of childbirth in the last stage of labour.

Another explanation may be that the quality of the coping behaviour is more important than the quantity of coping effort. Women with high self-efficacy may think that they can use coping behaviour such as breathing and relaxation exercises to deal with the pain of labour. However, when contractions become more intense and painful, they may practice the coping behaviour ineffectively, with no reduction in the perception of pain. The findings suggest that there is a need for a more comprehensive view of self-efficacy for childbirth in considering the range of behavioural responses perceived by women as critical to coping with labour pain, particularly at the time near the delivery. Moreover, the ability to relax, breathe through contractions, concentrate, maintain emotional control and listen to instructions from others may be differently affected by the frequency of uterine contractions, fatigue, nausea, presence of others and medical interventions. Most importantly, the effectiveness and appropriateness of using these coping skills with or without pharmacological pain relieving measures to relieve the pain of labour was not measured in this study. Observational studies on women’s use of coping behaviour during labour as well as a qualitative analysis of women’s feelings towards the use of the coping behaviour are, therefore, recommended.

The attrition rate from this study was not atypical for this type of longitudinal research. The attrition rate of 30·7% for this study was comparable to the previous international (Saisto et al. 2000, 2001) and the local (Lee et al. 2003) longitudinal obstetric studies. Analyses of the participants lost to follow-up indicate that there was no significant systematic bias in attrition from the study. According to the Census and Statistics Department, Hong Kong (2004), the mean age of the women who delivered their first-baby was 29 in 2003, and 54·5% of the women in this age group were employed in clerical work. Therefore, the socio-demographic characteristics of the sample in this study were comparable to those of the general population at the time of the study.


Caution should be taken in generalising the findings in view of the following limitations. First, no inference can be concluded about the experiences of women who did not attend the educational programme. Second, the sample was recruited from the only one regional hospital in Hong Kong. Therefore, generalisability of the results to other samples of pregnant women from other geographical areas cannot be guaranteed. Last but not the least, the ratings of the anxiety and pain during labour were recalled retrospectively. It may be possible that women’s feelings towards childbirth may be influenced by experiences during the labour period, but the time constraints and available resources in hand did not allow the study to further explore this aspect.


This study reports a preliminary investigation into the efficacy of a two-session educational intervention based on Bandura’s self-efficacy theory in promoting pregnant women’s ability to cope in labour. The findings indicate that the short-term theory-based educational intervention demonstrated a significant impact on women’s ability to cope with childbirth. The results also support Bandura’s theory that self-efficacy might be an important prerequisite to the adoption and maintenance of an effective coping behaviour for childbirth. The effect of the self-efficacy enhancing educational programme was evident in reducing the levels of pain and anxiety in the labour process except in the last stage of labour. Strategies that promote labouring women’s abilities to cope with labour pain and anxiety in the last stage of labour are needed. Further studies on the effect of self-efficacy-based educational intervention on physical childbirth outcomes are warranted. Finally, from a clinical perspective, it is important to identify interventions that can be directed towards those who chose not to attend educational programmes.


The research was supported by a Departmental Research Grant of the Nethersole School of Nursing, The Chinese University of Hong Kong. We would also like to thanks the midwives of the Prince of Wales Hospital, Hong Kong in facilitating the process of the data collection.


Study design: WYI, TC; data collection and analysis: WYI, WG and manuscript preparation: WYI, TC, WG.