| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychology, University of Bath, Bath, UK (E.K.); Goldsmiths College, London, UK (S.H., D.E.); and Imperial College London, London, UK (C.D., A.H.).
Address correspondence and reprint requests to Edmund Keogh, Department of Psychology, University of Bath, Bath, BA2 7AY, United Kingdom. E-mail: e.m.keogh{at}bath.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
Methods: Sixty-five women booked for an awake cesarean section with a regional nerve block and their birth partners were recruited. Data were collected at three time points for the mothers, before, during the cesarean section and after delivery on the postnatal ward, and at two time points for the birth partners (before and during the cesarean section).
Results: Maternal fear responses varied during the operation, in that fear was greatest at the point of administration of the nerve block. Within mothers, preoperative negative expectations were related to fear experiences during delivery, which was in turn related to their postoperative pain. Maternal anxiety sensitivity was found to mediate the relationship between negative expectations and fear, whereas birth partner’s fear mediated between maternal fear and postoperative pain. Mothers’ prenatal perceptions of control over drugs predicted their postoperative pain.
Conclusions: Maternal fear during cesarean section not only fluctuates, but may be influenced by psychosocial factors, including their birth partner. Psychosocial factors were also important predictors of postoperative experiences. Interventions that appropriately manage psychological and social factors during cesarean delivery may facilitate a more positive experience for mothers.
Key Words: cesarean section • pain • anxiety sensitivity • birth partner • social support • fear
Abbreviations: ASI = Anxiety Sensitivity Index; SF-MPQ = Short form McGill Pain Questionnaire; VAS = visual analogue scale; VerbAS = verbal analogue scale; CS = cesarean section; MPQ-S = Sensory scale of the short form McGill Pain Questionnaire; MPQ-A = Affective scale of the short form McGill Pain Questionnaire
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
One concern of many prospective mothers is the experience of pain (either from normal or medically induced stimuli), and associated fear (4). Physical and psychological techniques are available to help women with pain during and following normal or operative delivery, including interventional nerve blocks through to less invasive procedures, such as individual or partner-assisted coping strategies (5).
Professional care and support from midwives and doulas can also reduce negative experiences of childbirth, especially during labor/delivery, when fears, expectations and a range of different emotions (negative and positive) are reported (6). However, it is unclear what part birth partners, who are usually nonprofessionals, play in women’s experience of childbirth. Some report that birth partners improve birth experiences (7,8), whereas others report less positive outcomes (9–11).
Mode of delivery is considered an important factor in maternal birth experience (1). Given that most deliveries have been vaginal, research into the psychosocial factors involved in childbirth has concentrated on this mode of delivery. However, there are an increasing number of cesarean sections (CS) being performed, with up to 30% in many hospitals (12,13). A CS involves major surgery and is usually performed awake under regional anesthesia, in contrast with less invasive vaginal delivery (although it should be acknowledged that some vaginal deliveries do involve surgical interventions, e.g., episiotomy).
Although the public perception is that elective CS is an "easier option" with respect to overall pain experiences compared with labor pain, this may not be the case. Actual pain during CS delivery is usually more controlled but the whole procedure is not painless. Women report pain leading up to surgery, and the operation itself involves different physical sensations to those of vaginal birth (14). Furthermore, CS is associated with longer periods of postnatal pain and recovery, including mobility restrictions. Although pain during and following CS may have importance for a mother, it has not been extensively investigated in its own right except during clinical trials of analgesic drugs. As the experience of vaginal and cesarean delivery and their medical care is different, it cannot be assumed that the same factors that influence one will influence the other.
While the reasons for increasing CS rates have been widely debated (15,16), fear of pain during childbirth is cited as one of the contributory factors (4). Given that presurgical anxiety serves as an important predictor of postoperative pain and recovery (17–19), prenatal maternal anxiety may determine maternal pain experiences of CS. Additionally, social expectations are that birth partners accompany and support mothers during delivery; they are also likely to experience fear (20). While anxiety in the birth partner may contribute to maternal experiences of CS, very few studies have exclusively targeted the interactive role that such psychosocial factors may have on the painful experiences of mothers during elective CS.
The primary objective of this study was to examine the influence of psychosocial factors on maternal experiences during CS. We focus on mothers’ and birth partners’ prior expectations, control beliefs and the role of anxiety sensitivity. We opted to study elective CS because it gave time for our measures and urgent CS may have greater potential for postnatal distress. We anticipated that women may experience pain from pregnancy disorders, associated interventions (e.g., from needles) and that a few women may experience pain during CS and require additional analgesia (14). Our study, designed for a planned delivery, allowed recruitment of birth partners and an investigation as to whether their expectations of childbirth influenced the maternal outcome. We predicted that birth partners’ fears would be associated with mothers’ fears and pain experiences, during and following surgery.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
Assessment Points
Questionnaire measures were administered in three distinct time periods: Time 1, after recruitment at week 36 and before the operation at term; Time 2, during the CS (on ward, insertion of neuraxial block, and surgical incision); and Time 3, following delivery, while recovering on ward (i.e., between 1 to 4 days postnatal (mean = 1.4 days; SD = 1.2)).
At Time 1 participants were asked to complete the measures as soon as possible and to return them completed on the day of the CS. No record was taken of the date that participants completed the measures; thus we assume that participants completed them between weeks 37 and the date of CS (mean = 39 weeks, SD = 0.9). Measures of fear and pain were gained during Times 2 and 3.
Questionnaires
Expectations of CS Questionnaire
A modified version of the Expectation and Experience of Birth Scale was used (21,22). Section A comprised of six positive items (exciting, enjoyable, satisfying, pleasant, exhilarating, challenging) and six negative items (anxiety provoking, frightening, embarrassing, exhausting, difficult, traumatic), which were accompanied by a 100 mm visual analogue scale (VAS), on which participants indicated their agreement (0 = not at all, 100 = extremely).
Section C was also included and comprised of items related to perceived control associated with pain and emotions. The revised version contained four items relating to maternal pain (How much do you expect to be able to control any pain you feel during CS; How much do you expect midwife/anesthetist/doctors will be able to control any pain you feel during CS), two items relating to maternal panic (How important is it to you to keep in control of any panicky feelings you may have during CS; How much do you expect you will be able to keep in control of any panicky feelings during CS), and three items related to medication use (How much do you expect you will be able to decide whether/when/what pain relief is used). Agreement was rated on a 100 mm VAS (0 = not at all, 100 = completely). The scale was worded to be specifically related to CS, in the context of being either the mother or the birth partner, and administered at Time 1.
Anxiety Sensitivity Index (ASI)
The ASI (23) consists of 16 items that assess fear of anxiety-related sensations on a 5-point scale (scored 0–4) to form a total score. Internal consistency is good, with Cronbach’s 
ranging from 0.82 to 0.91, and test-retest reliability reported to be 0.71 (24). This scale was administered to mothers and their birth partners at Time 1.
Fear and Pain Before, During and After CS
Current pain severity was assessed by mothers using a verbal analogue scale (VerbAS) based on the present pain index from the McGill Pain Questionnaire (25). The pain index consists of six descriptors (no pain, mild, discomforting, distressing, horrible, excruciating) scored 0 to 5, with a higher score indicating greater pain. Such measures are considered valid and reliable techniques (26). Current fear was assessed using a verbal rating fear index. Mothers and birth partners indicated how fearful they were by selecting one of five descriptors (no fear, mild, moderate, severe, worse fear ever), scored 0 to 4, with a higher score indicating greater fear.
For mothers, the pain and fear indexes were administered five times during Times 2 and 3: (i) on the ward before transfer to the operating theater, (ii) in the operating theater just after the insertion of the neuraxial block, (iii) on surgical incision, (iv) and (v) postoperatively on the postnatal ward. The fourth and fifth measures were taken at the same time, and respectively questioned mothers about their memory of the amount of fear and pain they had experienced during the cesarean, as well as their current fear and pain levels. For birth partners, the fear index was administered at i, ii and iii (i.e., up to and including the point of surgical incision).
Short form McGill Pain Questionnaire (SF-MPQ)
The SF-MPQ has been validated (27) and contains 11 sensory (e.g., throbbing) and four affective (e.g., sickening) pain descriptors, which are scored between 0 (none) and 3 (severe). The SF-MPQ was administered to the mother at Time 3, and a total score calculated.
Data Analysis
To examine the relationship between the prenatal measures Pearson’s correlations were conducted. For the investigation of change in pain scores across Times 2 and 3 in the five perioperative phases i–v Friedman’s analysis (28) was conducted due to the non-normality of distribution of some pain scores. For the fear ratings, a repeated measure ANOVA was conducted with testing phase as the within-groups factor. Effects were followed-up with simple effects tests.
In order to predict pain and fear, regression analysis was conducted. Pain and fear scores taken during the operation were averaged. For pain, average scores taken during the operation and memory for pain were converted into new categorical variables (pain versus no-pain) and analyzed using logistic regression. For postoperative pain and all fear measures (average during CS, memory for CS, and postoperatively), standard regression was conducted. For both the logistic regression and standard multiple regressions it was necessary to reduce the number of predictors due to the sample size. Preliminary analysis was conducted for all outcomes variables to determine items for inclusion in the final model. A forward inclusion procedure was used for model selection, as it allowed the predictor with the largest simple correlation into the model, followed by subsequent predictors with significant semi-partial correlations (29,30). Separate preliminary analysis was conducted for maternal and birth partner responses and only predictors where p < .10 were selected for inclusion. In the final models maternal parity and birth partners’ previous experience of childbirth were also included, but since they did not significantly contribute to the final solution are not reported. All predictors were initially entered at the same time. If no significant results were found, stepwise regression was adopted to counter any potential overlap between variables. Mediation analysis (31,32) was conducted to determine the theoretically derived mechanisms (anxiety sensitivity, birth partner) underpinning the relationship between mother’s expectations and experiences.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
Demographics revealed that prospective mothers were aged between 20 and 43 (mean = 35 years, SD = 5), married or living with a partner (95%), classified themselves as White (89%), and possessed educational qualifications up to and including A levels (89%). Of these 37% were expecting their first child, and for 43% this was their first experience of CS (40% had at least one elective cesarean and 35% reported at least one emergency cesarean). Birth partners were aged between 20 and 45 years (mean = 35 years, SD = 5), were predominantly male (94%), married or living with a partner (95%), White (89%), and similarly educated (78% had A levels). Ninety-seven percent were the husband/partner of the mother, and the others sisters of the mother. Of the birth partners, 60% had been present at a previous delivery.
Correlations Between Prenatal Measures
Only one significant correlation was found between the birth partner prenatal measures; perceptions of control over pain was associated with perceptions of control over administration of drugs (r = 0.25, p < .05). However, when the effect of previous experience of childbirth was controlled for, this correlation became nonsignificant. For mothers, however, a different pattern of results was found that was not affected by parity. Specifically, the three perception of control scales were interrelated (r’s ranged 0.30 to 0.39). Negative expectations were negatively related to perceptions of control over pain (r = –0.32, p < .05), control over panic (r = –0.35, p < .01) and control over the administration of drugs (r = –0.46, p < .001), and positively related to mothers’ anxiety sensitivity score (r = 0.44, p < .001). Anxiety sensitivity was also related to control over panic (r = –0.25, p < .05) and perceptions of control over the administration of drugs (r = –0.31, p < .05), but not control over pain (r = –0.15, p > .05). When these variables (controlling for parity) were entered as predictors of negative birth expectations (R2 = 0.43, F(5, 56) = 8.32, p < .001), anxiety sensitivity was the strongest predictor (ß = 0.39, t = 3.52, p < .001), followed by control over drug administration (ß = –0.24, t = –2.09, p < .05).
Investigation of the relationship between mother and birth partner’s prenatal measures revealed few relationships. Negative birth expectations of mothers and birth partners were positively related (r = 0.32, p < .05), and mother’s positive birth expectations were negatively related to birth partner’s perception that pain could be controlled during CS (r = –0.27, p < .05).
Perioperative Fear
Change in Fear Across the CS
A main effect for time of testing was found (F(4, 244) = 53.69, p < .001). Figure 1 presents the mean fear (and pain) responses across the different time points. There was a significant increase in fear from before the cesarean began to when the neuroaxial block was administered, but that this significantly reduced by the time of surgical incision (p < .001). There were no significant differences between measures taken during the event and mothers’ memory of fear. The lowest fear level was postoperatively following the CS.
|
Predicting Fear During the CS
Preliminary analysis revealed that mother and birth partner’s negative expectations and mother’s anxiety sensitivity should be included in the final model. Mother’s pain during the event (pain versus no pain) and birth partner’s average fear were also included. The results of this and all subsequent regression analyses are presented in Tables 1 to 3. Maternal negative expectations and maternal anxiety sensitivity significantly predicted maternal fear (R2 = 0.44; F(5, 56) = 8.72, p < .001).
|
|
|
|
Predicting Memory for Fear During the CS
For maternal memory of fear screening resulted in mother anxiety sensitivity and birth partner perceptions of control over pain being included (Table 1). Additional variables were average fear during the CS (in both mother and birth partner), as well as pain experienced by mothers during the event (pain versus no-pain) and her memory for such pain experiences. Maternal fear during the event and her memory for pain predicted postoperative memory of fear during the CS (R2 = 0.51; F(7, 52) = 7.78, p < .001).
Predicting Postoperative Fear
For maternal postoperative fear none of the prenatal measures were eligible for inclusion, and so maternal and birth partner’s average fear scores during the CS, maternal pain experiences during the CS (pain versus no-pain) and postoperative pain (SF-MPQ) were entered in the final model (see Table 1). Only maternal postoperative pain significantly and positively predicted postoperative fear responses (R2 = 0.25; F(4, 53) = 4.46, p < .005).
Perioperative Pain
Change in Pain Across the CS
Mothers’ pain levels varied across times (i) to (v) (
2 (4) = 142.32, p < .001; Figure 1). Although there were no differences in pain during the CS, memory for pain and postoperative pain were significantly different from these measures (2 (3) = 74.35, p < .05 and 2 (3) = 124.96, p < .001, respectively) and from each other (2 (1) = 13.09, p < .001). This analysis was repeated separately by maternal parity (primiparas versus multiparas) with little change in effects found.
Predicting Pain During the CS
Preliminary analysis using logistic regression revealed that none of the prenatal measures were significantly related to pain experience (pain versus no pain) during the CS. Four variables came close to reaching significance (p < .10; mother’s and birth partner’s positive expectations and control over pain) and so were included alongside average fear ratings. No significant relationships were found, 2 (6) = 8.78, p > .05 (Table 2). Stepwise entry did not alter this.
Predicting Memory for Pain During the CS
For mother’s recall of pain during the CS (pain versus no pain), preliminary analysis revealed that birth partner’s anxiety sensitivity was the only significant predictor. Since maternal control over pain bordered on significance (p < .06), it was also included, alongside pain experienced during the CS (pain versus no pain), average fear experiences (birth partner and mother) and memory for fear. No significant change between blocks was found when entering all variables together, although birth partner anxiety sensitivity was found to be close to significance (Table 2). However, if forward entry was specified then the model was significantly different from the constant only model (2 (1) = 5.99, p < .05), suggesting that the predictors distinguished between women who experienced pain and those who did not. Birth partner anxiety sensitivity was the only significant predictor (B = –0.11, Wald Test = 4.90, p < .05), indicating that higher birth partner anxiety sensitivity reduced the likelihood of mothers remembering pain during the CS (OR = 0.90; 95% CI lower = 0.82, upper = 0.99).
Predicting Postoperative Pain
There were two measures of postoperative pain (VerbAS, SF-MPQ). For the VerbAS preliminary analysis resulted in maternal control over drugs for pain and perceived ability to control pain (p < .053) being included in the final model. Average maternal and birth partner’s fear were also included as predictors, as were maternal pain experiences during CS (pain versus no pain) and postoperative fear. Maternal control over drugs was found to be the only significant predictor of postoperative pain as measured on the VerbAS (R2 = 0.30; F(6, 55) = 3.83, p < .005; Table 3). Higher perception control over drugs reduced the likelihood of mothers reporting postoperative pain.
For the SF-MPQ, screening suggested that only mother’s negative expectations should be included, alongside pain (pain versus no pain), average maternal and birth partner fear during the CS, as well as postoperative maternal fear. Regression indicated that average birth partner fear during the CS and postnatal mother fear were significant predictors of mother’s postnatal pain experiences (R2 = 0.33; F(5, 51) = 5.00, p < .001; Table 3).
Initial screening had revealed that mothers’ fear during the operation was related to postoperative pain (SF-MPQ). Subsequent regressions indicated this became nonsignificant when birth partners’ fear was included, pointing toward a potential mediation effect (31,32). We therefore sought to test our prediction that birth partner’s fear mediates the relationship between mother’s fear during the operation and her postoperative pain. Step 1 revealed that mother’s average fear during the CS was related to her postoperative pain (ß = 0.33, t = 2.62, p < .05). We also found that mother’s fear was related to birth partner’s fear (ß = 0.30, t = 2.53, p < .05). Finally, when mothers’ and birth partners’ fears were entered together, we found that birth partner’s fear was related to mother’s pain (ß = 0.31, t = 2.46, p < .05), but that the relationship between mother’s fear and her pain was not significant (ß = 0.25, t = 1.97, p > .05). This suggests that birth partner’s fear is a mechanism by which mother’s fear is related to her postoperative pain experiences. Figure 2 presents a conceptual representation of this relationship.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
Prenatal Expectations
We consider the main finding from the current study was that prenatal expectations within the mother served as an important predictor of her fear responses during the operation. Since perioperative fear responses were subsequently found to be related to mother’s postoperative pain, it seems inevitable that greater consideration of psychological factors that impact on women’s experiences of a CS could have clinical benefit.
We were not only able to show the nature of the relationship between prenatal expectations, perioperative fear and pain, but also able to isolate the potential psychological mechanisms that help explain such relationships. One reason why mother’s negative birth expectations were related to fear during the operation was partially due to her anxiety sensitivity levels. This is consistent with the notion that anxiety sensitivity is important in both pain perception (33) and childbirth (34). Additionally, we found that the reason why mother’s fear during the operation was related to her postoperative pain was to due to the fear experiences of her birth partner. This result is important because it implies that how birth partners act during the operation determines her subsequent pain.
These results are consistent with previous research that has examined psychosocial factors in vaginal or instrumental delivery. Negative birth expectations, fear and the birth partner are known to influence experiences of childbirth (9,21,33). However, the current study is unique by focusing solely on planned operative delivery—to examine the role of birth partners and associated psychosocial influences, and the impact they may have on operative birth experiences. The results reported here confirm that social support systems may influence maternal experiences during this emotive time.
Role of Social Support
Although the value of social support during childbirth has been recognized, little attention has been given to the specific type of support, and whether it influences maternal experiences. Our current study provides original evidence that fears within birth partners are involved in negative maternal experiences during and after CS. If the effect of birth partners is reliable, then the removal of such maternal stress may mitigate the effects on mothers’ recovery. However, rather than remove birth partners a suitable alternative would be to target the emotional wellbeing of birth partners, plus offering psychological assessment and support, and informing parental choice. Indeed, there is evidence birth partners have potentially beneficial effects on maternal birth experiences (7,8). Anecdotally a number of birth partners admitted to our researchers that they had little choice in attending and felt ill prepared. If birth partner’s fears can be ameliorated, the patient’s experience could be facilitated. Further research should consider investigating fear-reduction interventions for birth partners to establish whether this improves maternal pain experiences.
Measurement Issues
When pain was assessed using the VerbAS, a different pattern of effects was found to the SF-MPQ. Prenatal perceptions of control over drug administration predicted VerbAS pain, but not SF-MPQ scores. One view is that the current results are unreliable and that caution is required with interpretation. We feel this is less likely given that we used standard measures, and so instead believe that the methods used to examine pain are important and further investigation of measurement effects would be valuable. The fact that such prenatal perceptions of control predicted postnatal pain is not only consistent with the view that sense of control facilitates good quality care during delivery (35), but is also a variable that may be easy to manipulate, especially by anesthetists with an awake, responsive mother.
Another measurement issue relates to the finding that pain did not change during the operation, although changes in maternal fear were reported; greatest maternal fear occurred when the neuraxial block was prepared and administered. The lower fear levels at surgical incision were unexpected since we anticipated that fear would be greatest at incision. This suggests that an anesthetist could be pivotal in controlling maternal fear, although the background sedative effect of a neuraxial block due to decreased afferent stimulation may have contributed (36). We also examined retrospective fear and pain estimations and found that while memory of fear was similar to those reported during the operation, pain memories were higher than those experienced. One possible reason for this could be linked to the impact of hormones such as oxytocin, which is usually raised during and following delivery, and thought to impair memory and learning (37). This adds to discussions on the reliability of memory around delivery (38,39), and if consistent may have implications for risk management, especially in a climate of increased complaints and legal claims.
Future Research
This study was limited to elective CS and so replication of procedures for other forms of delivery such as nonelective CS, when neither mother or birth partner have time for adequate preparation, is desirable. Different patterns may be expected during the stress of urgent surgery or more prolonged stages of vaginal delivery. The current study focused on the role of birth partners, and further work is needed to directly compare different types of social support during delivery. There is evidence that professionals (eg, midwives) can provide continuous support (6), and so it would be interesting to see whether different types of support work better in certain environments eg, do birth partners provide better support during home deliveries?
Implications
If our findings replicate, one implication would be to consider interventions to reduce fear. Targeted stress management could be included during antenatal classes focusing on the fears of mothers and partners (40). Research may also consider the potential impact that fear and anxiety have on postoperative recovery. In particular, the role that the birth partner’s fear has in exacerbating the fears of mothers during the operation, and the subsequent role it may have on postoperative analgesic requirements be considered. By targeting birth partner’s fears during antenatal classes, it may be possible to reduce the amount of post-natal pain experienced by mothers. Finally, since anxiety can increase recovery times (41), within the context of CS it would be opportune to study whether increased maternal fear during the procedure has an impact on mothers’ longer-term recovery from surgery, as well as other related factors such as breastfeeding and parent-child bonding. One question to ask is whether maternal experiences of breastfeeding, which can be a painful experience (42,43), are worse in those with greater fear of childbirth?
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
We thank the staff on labor ward at the Chelsea and Westminster Hospital, London, for their help.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
This work was supported by a research grant awarded to Dr. Keogh by the Department of Psychology, Goldsmiths College.
DOI:10.1097/01.psy.0000197742.50988.9e
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSNOTESREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
