1 Acta Obstet Scand 1998; 77: 542-547 Fear of Childbirth during Pregnancy may Increase the Risk of Emergency Cesarean Section Elsa Lena Ryding1, M.D. Barbro Wijma2, M.D., Ph.D., associate professor Klaas Wijma2, Ph.D., associate professor Håkan Rydhström3, M.D., Ph.D. From the Departments of Obstetrics and Gynecology, 1Central Hospital, Helsingborg, 2University Hospital, Linköping, and 3University Hospital, Lund, Sweden Address for correspondence: Elsa Lena Ryding, M.D. Department of Obstetrics and Gynecology Central Hospital S - 251 87 Helsingborg Sweden Tel +46 42 10 00 00 Fax +46 42 10 22 05 Running headline: Fear adds to risk of emergency cesarean 2 Abbreviations: CS: cesarean section; W-DEQ: Wijma Delivery Expectancy Questionnaire; STAI: State-Trait Anxiety Inventory; SCI: Stress Coping Inventory; SGA: small for gestational age; OR: odds ratio; CI: confidence interval. 3 Abstract Fear of Childbirth during Pregnancy may Increase the Risk of Emergency Cesarean Section Background. The purpose of our study was to elucidate the association between fear of childbirth, general anxiety, and stress coping during the third trimester of pregnancy, and a subsequent delivery by emergency cesarean section. Methods. In a case-control study, 1,981 Swedish-speaking women completed three self-assessment questionnaires at 32 weeks' gestation. Ninety-seven of these women were delivered by emergency cesarean section. Fear of childbirth, general anxiety and the stress coping ability of these 97 cases were compared with the same features in 194 controls, matched for age and parity. Results. Women, subsequently delivered by emergency cesarean section, reported a greater anxiety and a poorer stress coping ability, and, most obviously, a greater fear of childbirth at 32 weeks' gestation. After elimination of possible confounders, the odds ratio for emergency cesarean section was examined for women whose scores were above various cut-off points according to the fear of childbirth measuring instrument. For women with a serious fear of childbirth the odds ratio was 3.0 (95% confidence interval 1.4 to 6.6), and the population attributable risk 0.167. Conclusion. Fear of childbirth during the third trimester of pregnancy may increase the risk of subsequent emergency cesarean section. Key words: fear of childbirth; anxiety; stress coping; pregnancy; emergency cesarean section 4 The evidence of that maternal stress and anxiety have a deleterious effect on the outcome of pregnancy is not unanimous (1, 2, 3). The theory that anxiety during pregnancy can interfere with the delivery process has been supported by some prospective studies (4, 5, 6, 7). Such interference can occur directly by psychophysiological pathways. Pharmacologically, epinephrine has been associated with enervated uterine contractility, and norepinephrine with intensified uterine contractility (8). Self-reported anxiety during labor was found to be correlated with the concentration of epinephrine in plasma (9). Longer duration of the second stage labor has been observed in women with higher catecholamine concentrations (8). Anxiety can also affect the birth process indirectly, by influencing maternal behaviour, e.g. by stimulating the urge to smoke. However, a more recent epidemiological investigation (10) could not demonstrate any relation between anxiety during pregnancy and non-spontaneous delivery. Women with antenatal fear of childbirth have also been found to run an increased risk of dissatisfaction with their delivery experience (11). They also tend to have an increased rate of cesarean section (CS) and operative vaginal delivery (12). According to Barlow (13), anxious apprehension is "an unpleasant affective state associated with perceptions of future unpredictability and uncontrollability". When a person with a specific anxiety has to face the feared situation, there is an increased physiological arousal. There is also an attentional narrowing with a heightened perception of threat-related stimuli. As a result the worry is felt more intensely. The person's performance is then often dysfunctional, which further aggravates the situation. Applied to obstetrics, this could mean that the specific anxiety is fear of childbirth (14) and the feared situation, labor. Increased physiological arousal may affect the uterine function and, due to attentional narrowing, the woman in labor perceives most events as threat-related stimuli. An example of the consequence of such intensified worrying, dysfunctional performance and worsened situation is the difficulty of establishing a good rapport between the obstetrical staff and very anxious parturients. It then becomes difficult to offer help in a way which is acceptable. Thus, maternal anxiety may be correlated with uterine dysfunction, such as hypo- or hyperactive contractions, and with dysfunctional performance during labor. It can be hypothesized that anxiety and, specifically, fear of childbirth, as well as a diminished ability to cope with stress, imply an increased risk of intrapartum complications, such as prolonged labor or fetal asphyxia. An emergency CS is then sometimes considered the safest way to conclude a complicated delivery. The aim of this case-control study was to assess the association between fear of childbirth, general anxiety, and stress coping during the third trimester of 5 pregnancy, on the one hand, and a subsequent delivery by emergency CS on the other. Material and methods From 1st January, 1992 to 28th February, 1993, most women at 32 weeks' gestation, intending to give birth at Helsingborg Central Hospital, were invited to participate in the study. Practically all pregnant women living in this catchment area of about 38,000 females aged 15-44 years give birth at the same unit. The frequency of emergency CS delivery during the study period was 6.3% in Helsingborg, and the overall CS rate, 9.1%. Swedish-speaking women attending the antenatal program at all nine antenatal clinics in the area, at three of the private gynacologists in the city, or at the hospital's clinic for complicated pregnancies, were eligible for the study with the following exceptions. A few women who subsequently did not give birth at Helsingborg Hospital were excluded, as also were 3 women who received specific treatment for their fear of childbirth. Women scheduled for an elective CS were also excluded, as they were at risk for an emergency CS only during a limited time of their pregnancy. The remaining group, eligible for the study, comprised 2,361 women out of a total population of about 2,700 pregnant women during the study period. The majority of those excluded could not speak, read or write Swedish. The women completed the questionnaires at the antenatal clinics. The midwives arranged for the women to sit in seclusion for about 30 minutes, during a visit as close as possible to 32 complete gestational weeks. A total of 1,981 women (84%) completed the questionnaires. The mean age of the whole sample was 28 years (range 16-43). Altogether 380 eligible Swedish-speaking women did not participate, 206 due to organizational failure, while 174 declined. The midwives and physicians caring for the women during pregnancy and labor had no access to information contained in the questionnaires. Nineteen of the 378 drop-outs (5%) were delivered by emergency CS, as were 97 participants (4.9%). These 97 constituted the case group. Their mean age was 29 years (range 19-43); 60% of them were nulliparous. A control group of 194 women matched for age and parity (nulli- or multiparity) were selected from among the remaining participants. Obstetrical data were collected from the antenatal and hospital records. The psychological measures consisted of the following questionnaires: The Wijma Delivery Expectancy Questionnaire (W-DEQ) (15), the State-Trait Anxiety Inventory (STAI) (16), the Stress Coping Inventory (SCI) and a demographic questionnaire. 6 The W-DEQ version A measures fear of childbirth as operationalized by the cognitive appraisal of the approaching delivery. The 33-items graphic self- assessment rating scale has six scale steps per item, ranging from "not at all" to "extremely" with a minimum score of 0 and a maximum of 165. A score exceeding 84 was found for 10% of the entire sample (n=1,981) at 32 weeks' gestation. The women who scored above 84 were considered to be suffering from serious fear of childbirth. The STAI measures trait and state anxiety. The version of STAI used in this study is the part measuring trait anxiety (20 items with four scale steps per item) by means of an instruction referring to the situation during the current pregnancy. The SCI measures stress-coping. It is a recently developed research measure of the individual's appraisal of her adaptive resources for dealing with stressful situations. When filling in the SCI, the subject is instructed to rate on a six-point Lickert scale her personal perception of how often she is able to cope with 41 stressful situations. The five points are marked with "almost always", "very often", "occasionally", "rarely" and "never". Thus the minimum possible score is 41, and the maximum, 246. The higher the score, the greater is the stress-coping ability. The reliability of the SCI was assessed in a group of pregnant women (N=227) with the following results: Chronbach's alfa = 0.94; split-half correlation corrected with Spearman-Brown prophecy formula = 0.93. The correlation between SCI and STAI in the same group was 0.48. The correlation beween SCI was 0.25. Possible confounders were identified from background data. These included education, marital status, height and weight before pregnancy as well as obstetrical risk factors such as serious complications of pregnancy, twin pregnancy, history of previous emergency CS, history of infertility more than 2 years, and smoking during pregnancy. Women who smoked only on "special occasions" were classified as non- smokers, but both light and heavy smokers as smokers, since there was so few light smokers (less than 10 cigarettes a day). Educational level was reported as 1) <9 years of elementary schooling, 2) 9 years of elementary schooling, 3) senior high school, or 4) college/university education. The infants were classified as small for gestational age (SGA) if their birth weight was <mean weight -2 SD according to the Scandinavian Standard (17). Statistical analysis was performed by means of unpaired t-test for continuous variables and Chi-square test (or Fisher's exact test in the case of small numbers) for categorical variables. Statistical significance was defined as p < or = 0.05, and odds ratios (OR) with 95% confidence intervals (CI) are presented. The study was approved by the regional committee for ethics in science. 7 Results Table I shows the two groups in relation to certain demographic and obstetrical variables. Alcohol consumption was also recorded, but was invariably denied (except for drinking on rare occasions). On the basis of these findings the women in the case group could be characterized as having a higher prepregnant weight, more often having a history of infertility and previous emergency CS, and an increased rate of twin pregnancies, pre-eclampsia, and SGA babies. The cases and controls differed concerning their scores in the three psychological questionnaires (Table II), but most significantly regarding the intensity of fear of childbirth as measured by the W-DEQ. Therefore, the relative risk, estimated by OR for emergency CS was examined for women who scored above various cut-off points in the W-DEQ, in comparison with the rest. The results are shown in Fig. 1. The mean W-DEQ score for the entire sample (n=1,981) was 54.1+21.1 for nulliparous women and 57.7+22.1 for parous women. The variables characterizing the women in the case group (Table I) were regarded as possible confounding factors only if they were (i) known risk factors for delivery complications leading to emergency CS, (ii) might be correlated with fear of childbirth, and (iii) had already been present at 32 weeks of pregnancy. Prepregnant weight was considered a factor that was not likely to be related to fear of childbirth, nor was it in fact correlated to the W-DEQ score. Smoking during pregnancy is a known risk factor for childbirth complications, but since the women in the control group smoked about as often as those in the case group, it was not considered to be a confounder in the present study. Possible confounding factors present from early pregnancy were: a history of infertility, a previous emergency CS, and a twin pregnancy (diagnosed during the first 17 weeks of pregnancy). Even when those 26 women with one or more of these variables were excluded from the analysis, together with their 52 controls, the W-DEQ scores remained about the same (64.0+20.6 vs 53.9+20.2). The OR for emergency CS for women who scored above various cut-off levels in the W-DEQ in comparison with the rest, were recalculated and found to be about the same as the results of the original material. These 26 women were therefore replaced in the analysis. The records of those women who had pre-eclampsia or hemorrhage during pregnancy, false labor leading to hospitalization, or an SGA baby were then carefully scrutinized. Exclusion from the analysis was made only when a complication was already recorded at the time when the woman completed her questionnaire. Accordingly, 9 women were excluded together with their 18 controls. The W-DEQ 8 scores remained about the same (64.7+21.5 vs 54.1+20.2). The OR for emergency CS was re-examined. The results are shown in Fig. 1. Thirty women were considered to have a serious fear of childbirth, with a W-DEQ score exceeding 84. OR for emergency CS was 3.5 (95% CI 1.6 to 7.3) in comparison with women who scored less. After exclusion of those 2 women who must have known about the serious complications of pregnancy (one early pre-eclampsia and one IVF pregnancy with haemorrhage) the OR for emergency CS was 3.0 (95% CI 1.4 to 6.6), and the population attributable risk, 0.167. Compared with the women with a lower W-DEQ score, these 28 women also had a higher general anxiety level (mean sum score of STAI 45.5+20.3 vs 33.5+8.6, p<0,0001) and a lower stress tolerance (mean sum score of SCI 161.4+23.8 vs 177.8+20.9, p<0,0001). There was a tendency for smoking to be more common (46% vs 27%, NS) among the women with a serious fear of childbirth, but they did not differ from the rest of the subjects with respect to any of the other variables studied. Sixteen of those women with a serious fear of childbirth belonged to the case group after the exclusion of women who may have known about pregnancy complications when completing the questionnaire. Imminent fetal asphyxia was the main indication for emergency CS in 69% of those women, compared with 39% of those women without a serious fear of childbirth (p<0.05). Protracted delivery was the main indication for emergency CS in 13%, compared with 21% of the women without a serious fear of childbirth (NS). Imminent fetal asphyxia or placental abruption was the main indication for emergency CS in 48% of all those women who were smokers, compared with 47% of all the non-smokers (NS). Thirty-nine women were operated because of imminent fetal asphyxia. Within that group, 4 out of 11 women with serious fear of childbirth smoked. Eight out of 28 women without serious fear of childbirth smoked. No correlation was found between smoking and serious fear of childbirth within the asphyxia group. Discussion The results of the present investigation suggest an increased risk of emergency CS when a pregnant woman suffers from serious fear of childbirth. Study of those suffering in this way has revealed a picture of patients in need of special antenatal care. Their high level of general anxiety and low stress tolerance may inhibit their postpartum adaptation process. They are also more likely to be dissatisfied with their delivery experience (11), which can provoke posttraumatic stress reactions (18, 19) or 9 other emotional distress postpartum (20, 21, 22, 23). Imminent fetal asphyxia was the chief reason for emergency CS in this group, but whether or not this was related to anxiety, exemplified by maternal smoking, hyperventilation during labor, hyperactive uterine contractions, or some other hitherto unknown factor, was not ascertained in this study which focused on antepartum variables. The results do not support the theory that fear causes hypoactive uterine contractions or that it prolongs labor. Since there were few women with serious fear of childbirth subsequently delivered by emergency CS, these results must be regarded with some degree of caution. If, hypothetically, all cases of serious fear of childbirth could be eliminated, the results of the present study would indicate a possibility of reducing the number of emergency abdominal deliveries at our hospital by 16.7%. Thus, to diagnose and treat fear of childbirth in pregnant women ought to be an important task for antenatal programs. Increased awareness of the problem could help midwives to identify those pregnant women who need special care. Determination by means of the W-DEQ, if further developed for diagnostic purposes, might be a practicable way to identify these women. The questionnaire, which takes 5-10 minutes to complete, provides an objective estimate of the degree of women's fear of childbirth. It has been suggested that serious fear of childbirth can be treated (24, 25, 26, 27), but the effectiveness of treatment programs remains to be evaluated. In our opinion, elective CS necessitated by fear of childbirth (28) can be restricted to serious cases not amenable to treatment. Our study sample is believed to be representative of the total population of pregnant Swedish-speaking women in the Helsingborg area. To perform a case- control study instead of analysing the entire material was considered the most cost- effective course. Analysis of the entire cohort might well have added to the information. The decision to regard a W-DEQ score of >84 (the 10% level of the total sample of 1,981 women) as an indication of serious fear of childbirth was dictated by the need for a suitable number of women, belonging to a clinically relevant category, in the analysis. Areskog et al. (29) found that 6% of Swedish women during late pregnancy suffered from a disabling fear of childbirth, while 17% felt a more moderate but still troublesome fear. Objections could be raised against our way of examining the possible confounding factors. For instance, as smoking was about as common in the case group as among the controls, the smokers were not excluded, in order to retain a reasonable number of women in the analysis. However, the women most seriously afraid of childbirth did tend to smoke more often than the others, although there was no evidence that 10 smoking per se affected the rate of emergency CS, or the prevalence of imminent fetal asphyxia and placental abruption, in those women. Because so few women had a serious fear of childbirth, the effect of smoking is difficult to assess in this particular group. The causes of fear of childbirth (25) were not the subject of the present study. It is conceivable that suspected complications of pregnancy at 32 weeks' gestation may have influenced the measurements in some women who later did not receive any diagnosis of serious pregnancy complication in their records. Because such suspicions cannot have been an indication for emergency CS, they were not considered to be confounding factors. Stress coping was measured by means of a newly devised questionnaire. Such a questionnaire may be of limited value. Moreover, women less able to cope with stress, but without any specific fear of childbirth, may feel less anxious during labor, as they do not regard delivery as a particularly stressful event. This may explain why the SCI score has a weaker correlation with the emergency CS rate than has the W- DEQ score. Among the questionnaires utilized, the STAI is the best known. Previous studies using the STAI during pregnancy and intrapartum (6, 8, 30) have demonstrated positive correlations between state-and-trait anxiety and duration of labor. In the present study the association between maternal anxiety, as measured with the STAI, and subsequent emergency CS was weaker than the association between maternal fear of childbirth and subsequent emergency CS, as measured with the W-DEQ. Anxiety-prone women not specifically apprehensive about childbirth may be better prepared when facing labor and delivery. This would explain the absence in one prospective population study (10) of any effect of antenatal anxiety on non- spontaneous delivery. In that study, Perkin et al. used the General Health Questionnaire, which measures maternal anxiety in general, rather than specific fear of childbirth. Perkin's study included instrumental and breech vaginal delivery and elective CS, besides emergency CS. In conclusion, it is possible that fear of childbirth during pregnancy may increase the risk of subsequent emergency CS. To establish whether accurate diagnosis and effective treatment of fear of childbirth might reduce the emergency CS rate is an urgent topic for research. 11 Acknowledgement This study has been aided by research grants from the Stig and Ragna Gorthon Foundation and from Östergötland County Council. 12 References 1. Levin J, DeFrank R. Maternal stress and pregnancy outcomes: a review of the psychosocial literature. J Psychosom Obstet Gynaecol 1988; 9: 3-16. 2. Reading A. The influence of maternal anxiety on the course and outcome of pregnancy: a review. Health Psychol 1983; 2: 187-202. 3. Istvan J. Stress, anxiety and birth outcomes: a critical review of the evidence. Psychol Bull 1986; 100: 331-48. 4. Crandon A. Maternal anxiety and obstetric complications. J Psychosom Res 1979; 23: 109-11. 5. Standley K, Soule B, Copans S. Dimensions of prenatal anxiety and their influence on pregnancy outcome. Am J Obstet Gynecol 1979; 135: 22-6. 6. Beck N, et al. The prediction of pregnancy outcome: maternal preparation, anxiety and attitudinal sets. J Psychosom Res 1980; 24: 343-51. 7. Hodnett E, Abel S. Person-environment interaction as a determinant of labour length. Health Care for Women International 1986; 7: 345-56. 8. Lederman R, Lederman E, Work B, McCann D. The relationship of maternal anxiety, plasma catecholamines and plasma cortisol to progress in labor. Am J Obstet Gynecol 1978; 132: 495-500. 13 9. Lederman RP, Lederman E, Work B, McCann D. Anxiety and epinephrine in multiparous women in labor: relationship to duration of labor and fetal heart rate pattern. Am J Obstet Gynecol 1985; 153: 870-7. 10. Perkin M, Bland J, Anderson H. The effect of anxiety and depression during pregnancy on obstetric complications. Br J Obstet Gynaecol 1993; 100: 629-34. 11. Areskog B, Uddenberg N, Kjessler B. Experience of delivery in women with and without antenatal fear of childbirth. Gynecol Obstet Invest 1983; 16: 1-12. 12. Di Renzo G, Polito P, Volpe A, Anceschi M, Guidetti R. A multicentric study of fear of childbirth in pregnant women at term. J Psychosom Obstet Gynaecol 1984; 3: 155-63. 13. Barlow D. Anxiety and its Disorders. New York: The Guilford Press, 1988. 14. Wijma K, Wijma B. Changes in anxiety during pregnancy and after childbirth. In: Wijma K, von Schoulz B, eds. Reproductive life. Advances in research in psychosomatic obstetrics and gynaecology. Lancaster and New Jersey: The Parthenon Publishing Group, 1992: 81-8. 15. Wijma K, Wijma B, Zar M. Psychometric aspects of the W-DEQ; a new questionnaire for the measurement of fear of childbirth. J Psychosom Obstet Gynecol. In press. 16. Spielberger C, Gorsuch R, Lushene R, Vagg P, Jacobs G. Manual for the State- Trait Anxiety Inventory. Palo Alto: Consulting Psychologists Press, 1970. 14 17. Marsál K, Persson P-H, Larsen T, Lilja H, Selbing A, Sultan, B. Intrauterine growth curves based on ultrasonically estimated fetal weights. Acta Paediatr 1996; 85: 843-8. 18. Ballard C, Stanley A, Brockington I. Post-traumatic stress disorder (PTSD) after childbirth. Br J Psychiatry 1995; 166: 525-8. 19. Wijma K, Söderquist J, Wijma B. Posttraumatic stress disorder after childbirth. a cross- sectional study. J Anxiety. In press. 20. Ryding EL, Wijma B, Wijma K. Posttraumatic stress reactions after emergency cesarean section. Acta Obstet Gynecol Scand 1997; 96: 856-61. 21. Thune-Larsen K-B, Möller-Pedersen K. Childbirth experience and postpartum emotional disturbance. Reproductive and Infant Psychology 1988; 6: 229-40. 22. Green J. "Who is unhappy after childbirth?": Antenatal and intrapartum correlates from a prospective study. Reproductive and Infant Psychology 1990; 8: 175-83. 23. Edwards D, Porter S-A, Stein G. A pilot study of postnatal depression following caesarean section using two retrospective self-rating instruments. J Psychosom Res 1994; 38: 111-7. 24. Ryding EL. Investigation of 33 women who demanded a cesarean section for personal reasons. Acta Obstet Gynecol Scand 1993; 72: 280-5. 25. Lukesch H, Kockenstein P, Holz Z. Therapeutische Interventionen bei Erstgebärende mit Geburtsängsten. Z Geburtsh u Perinat 1980; 184: 303-9. 15 26. Areskog-Wijma B. Treatment of fear of childbirth. In: Leysen B, Nijs P, Richter D, eds. Research in Psychosomatic Obstetrics and Gynaecology. Leuven: Acco, 1985. 27. Linderoth B. Treatment of fear of childbirth. Linköping: Department of Education and Psychology, Linköping University, 1991. 28. Ryding EL. Psychosocial indications for cesarean section. A retrospective study of 43 cases. Acta Obstet Gynecol Scand 1991; 70: 47-9. 29. Areskog B, Uddenberg N, Kjessler B. Fear of childbirth in late pregnancy. Gynecol Obstet Invest 1981; 12: 262-6. 30. Lederman R, Lederman E, Work BJ, McCann D. Relationship of psychological factors in pregnancy to progress in labor. Nurs Res 1979; 28: 94-7. 16 Table I. Socioeconomic and obstetric characteristics of women delivered by emergency cesarean section (cases) and a control group of women matched for age and parity. ________________________________________________________ Variable Cases Controls (n=97) (n=194) ________________________________________________________ Prepregnant weight (kg) mean +SD 66.1+10.6 62.9+9.4 p<0.01 Mother's height (cm) mean +SD 165.0+6.4 167.1+6.2 NS Gestational age at delivery, mean +SD 39.1+2.0 39.3+1.3 NS Education level, mean +SD 3.1+0.7 3.2+0.6 NS Smoking during pregnancy, n (%) 29 (30%) 52 (27%) NS Living alone, n (%) 4 (4%) 7 (4%) NS Infertility >2 years, n (%) 11 (11%) 10 (5%) NS Emergency CS previously, n (%) 13 (13%) 6 (3%) p<0.01 Twin pregnancy, n (%) 5 (5%) 0 p<0.01 Pre-eclampsia, n (%) 16 (17%) 2 (1%) p<0.0001 Haemorrhage, pregnancy n (%) 3 (3%) 2 (1%) NS False labor, n (%) 3 (3%) 2 (1%) NS SGA baby, n (%) 13 (13%) 7 (4%) p=0.01 ________________________________________________________ Level of education, see Material and Methods 17 Table II. Group sum scores in W-DEQ, STAI and SCI at 32 weeks' gestation of women subsequently delivered by emergency cesarean section (cases) and a control group matched for age and parity Cases Controls Mean CI p-values n=97 n=194 difference ___________________________________________________________________________ W-DEQ 64.6 +22.2 54.3 +19.8 10.3 5.3 - 15.3 p< 0.0001 mean+SD STAI 36.7 +13.6 34.0 +9.1 2.7 0.1 - 5.3 p< 0.05 mean+SD SCI 172.4+21.5 177.4+21.6 5.0 -0.3 - 10.3 p= 0.05 mean+SD ___________________________________________________________________________ 18 Fig. 1. OR for emergency cesarean section for women who scored above vis-á-vis below various cut-off levels in the W-DEQ at 32 weeks of pregnancy, with 95% CI. -------- = total material. -------- = after exclusion of those 9 women (and their 18 controls) who may have known about complications of pregnancy already when completing the questionnaire.