The association between depressive symptoms during pregnancy and post-delivery fear of childbirth; a prospective study

ABSTRACT

Background

Fear of childbirth is an important reason for a caesarean section on request.

Objective

To assess the association between depressive symptoms during pregnancy and post-delivery fear of childbirth (PFOC).

Methods

We prospectively studied pregnant women from two hospitals in the Netherlands. Women completed the Edinburgh Depression Scale (EPDS), the Wijma Delivery Experience Questionnaire (W-DEQ B) and questions concerning risk factors. Depressive symptoms were assessed at baseline and six weeks post-delivery. PFOC was assessed six weeks post-delivery. Baseline characteristics and pregnancy outcomes were compared between women with and without a depression at baseline. The association between depression and PFOC was assessed with multivariable logistic regression analysis.

Results

245 women participated in this study. At baseline 11% suffered from depressive symptoms. There were no differences in pregnancy outcomes. Women with depressive symptoms more often suffered from depressive symptoms six weeks post-delivery (adjusted OR 4.9, 95% CI 1.4–17). PFOC six weeks post-delivery was present in 11%. Women with depression were at increased risk of PFOC six weeks post-delivery (adjusted OR 9.2, 95% CI 2.6–32).

Conclusion

This study shows that women with depression at baseline are at increased risk for depression and PFOC six weeks post-delivery.

KEYWORDS:

• Depression
• fear of childbirth
• pregnancy
• post-partum
• psychiatry

Introduction

A meta – analysis showed that 14.8% of pregnant women experience depressive symptoms, 12.7% of these women meet the standard for a major depressive episode (Gavin et al., 2005). A higher prevalence of depression is found in ethnic minorities with a prevalence ranging between 17.5–19.5% in South-Asian and Middle-Eastern women respectively (Shakeel et al., 2015). In the Netherlands, the prevalence of depression during pregnancy was found to be between 5–10% (Bergink et al., 2011).

In women with depression, fear of childbirth (FOC) during pregnancy is more frequently observed (Saisto et al., 2001; Soderquist et al., 2004). The prevalence of severe fear of childbirth in the literature is estimated between 7.5 and 15.6% and is equally distributed in both primi- and multiparous women (Nieminen et al., 2009; Nordeng et al., 2012).

Women with high levels of FOC during pregnancy also had high levels of FOC during and after delivery (Alehagen et al., 2006; Zar et al., 2001), which was unrelated to obstetric complications in a previous pregnancy (Sluijs et al., 2012). A recent meta-analysis on FOC showed that FOC was a major predictor for the requirement of psychiatric care after childbirth. In addition, it also predicted an increased risk for PTSD symptoms, underlining the consequences of FOC in the postpartum period (Dencker et al., 2018).

FOC was also found to influence reproductive choices that women make. Women with FOC interviewed 7–14 years after their first birth, less often gave birth for a second time as compared to women without FOC (Moller et al., 2018). FOC also influenced interpregnancy intervals (Dencker et al., 2018).

Some studies showed that FOC is associated with obstetric complications like prolonged labour, instrumental vaginal deliveries and caesarean sections (Johnson & Slade, 2003; Laursen et al., 2009; Ryding et al., 1998). Also, women with FOC more often request a caesarean section. (Coskuner Potur et al., 2017).

Depression and fear in the mother has substantial consequences for the infant as well. Previous studies described negative effects on breast-feeding, bonding, mother–infant interactions, infant temperament, sleep, mental development, health and internalising in infants and on conduct disorder in adolescents (Field, 2018). In addition infants of depressed mothers score low on social engagement and had less mature regulatory behaviours as compared to infants from non-depressed mothers (Feldman et al., 2009).

Until now, no prospective study has been carried out to examine the association between antenatal depression and postpartum fear of childbirth (PFOC). Women with depression are a vulnerable group of patients. If these women indeed have a higher risk of PFOC, for caregivers this is important to be aware of and to pay attention to this condition during pregnancy and postpartum.

We hypothesised that women with depressive symptoms during pregnancy more frequently experience post-partum fear of childbirth. We therefore performed a prospective cohort study to investigate the association between antenatal depression and PFOC post-delivery.

Methods

Design and setting

Between April 2010 and May 2012, we performed a population-based prospective cohort study. We included pregnant women attending antenatal care in two hospitals in the Netherlands. Both hospitals are teaching hospitals and have an additional outpatient clinic where pregnant women with psychiatric problems can receive special care.

Inclusion and exclusion criteria

All pregnant women attending antenatal care at the obstetric outpatient clinic with a pregnancy duration of less than 20 weeks were eligible for this study. This means that we included both women who received care as usual and some women who received extra attention because of prior psychiatric problems. The only exclusion criterion was insufficient capability to read and write the Dutch language.

Data collection

Women received a handheld device with the Mind2Care questionnaire (Quispel et al., 2012). This contains the Edinburgh Depression Scale (EPDS) and some additional multiple choice questions concerning socio-demographical, obstetrical, and psychiatric risk factors (supplemental Figure S1.).

The questionnaire was completed in privacy in the waiting room. Once a week the investigators checked all the handheld devices. All women with an EPDS score ≥ 12 received a referral to the perinatal psychiatrist.

Six weeks post-delivery, two questionnaires were sent to the patient’s home address, including a return envelope; W-DEQ B (PFOC) and EPDS (depression). After two weeks, a reminder was sent to non-responders. Obstetric and neonatal outcomes were extracted from the medical records.

Questionnaires

Edinburgh Depression Scale (EPDS)

The EPDS is a self-report questionnaire consisting of ten questions in which women are asked how they felt the previous seven days. The EPDS has been validated for use during and after pregnancy (Bunevicius et al., 2009; Pop et al., 1992). It is corrected for somatic symptoms in (early) pregnancy (Bunevicius et al., 2009). According to the original publication of Cox et al., we decided to use a cut-off value of 12. Above a threshold of 12 women are most likely to be suffering from a depressive illness and should be further assessed to confirm whether or not clinical depression is present (Cox et al., 1996).

Wijma Delivery Experience Questionnaire (W-DEQ B)

The W-DEQ B is a validated questionnaire that measures FOC after personally experiencing childbirth (version B). The fear experienced during delivery can be measured in retrospect post-delivery by W-DEQ B and is called postpartum fear of childbirth (PFOC) (Wijma et al., 1998). This is a self-defined term to emphasise the difference between anticipated fear of childbirth (FOC) and the fear someone has already experienced (PFOC).

It consists of 33 items, on which the participant can rate her feelings on a six-point scale. A score of 85 and higher was used to classify severe fear of childbirth (Adams et al., 2012).

Statistical analysis

The study population was divided into women with and without depression based on the EPDS scale.

Depression was defined as an EPDS score ≥12 at intake. To compare baseline characteristics and pregnancy outcomes, we calculated percentages and determined statistical significance for differences in dichotomous measures using a chi-square test and for continuous measures using the t-test.

Univariable logistic regression was used to describe the association between depression before 20 weeks and depression and PFOC six weeks after delivery. In addition, we performed multivariable logistic regression to correct for possible confounders.

We adjusted for ethnicity (Ternstrom et al., 2015), low education (Salomonsson et al., 2013), smoking and alcohol use (Raisanen et al., 2014) and psychiatric history (O’Connell et al., 2019), since these factors were significantly different between women with and without depression and are also known to be associated with fear of childbirth.

We also performed a mediation analysis to investigate whether the effect of antenatal depression on PFOC was mediated through postpartum depression by using the Sobel test. The associations are expressed as adjusted odds ratios (OR) with 95% confidence intervals (CI). A p value <0.05 was used to indicate statistical significance. Statistical analyses were performed using IBM SPSS Statistics 24.

Institutional ethical approval

The Medical Ethical Board of the Gelre Hospitals Apeldoorn location and the Amphia Hospital Breda exempted this study from institutional review board (IRB) approval, since the screening and referral were already part of routine care, including opting-out. In the Netherlands, screening is allowed under Dutch Law if opting-out is explicit (in Dutch: the WBO and WGBO Act). Post-hoc data collection of health outcomes from medical records is allowed provided anonymised analysis.

Results

Between April 2010 and May 2012, 245 women completed the PDA questionnaires.

Of the participating women, 28 (11%) had an EPDS score ≥12 at baseline (Table 1). Other characteristics of the women with and without depression are shown in Table 1. Women with depression were more often non-Caucasian compared to women without depression (21 vs 6.9% P = 0.02) and were less educated (54 vs 25% P = 0.001). Women with depression more often smoked or used alcohol during pregnancy compared to women without depression (32 vs 15% P 0.02 and 36 vs 14% P = 0.004 for smoking and alcohol use respectively). Finally, women with depression more often had a psychiatric history (43 vs 19% P = 0.005). The majority of women with a psychiatric history had anxiety (N = 20, 31%) and depressive symptoms (N = 30, 49%). Three women had a history of psychosis, two women had a history of eating disorders, and one woman had a history of substance abuse. Eight women had multiple complaints.

Table 1. Baseline characteristics of the women with and without depression.

	Depression (EPDS ≥12, N = 28)		No depression (EPDS <12, N = 217)
	N	%	N	%	P-value
Non-Caucasian	6	21	15	6.9	0.02
Nulliparous	7	25	35	16	0.24
Low education	15	54	54	25	0.001
No partner	1	3.6	1	0.46	0.085
Smoking	9	32	32	15	0.02
Alcohol	10	36	31	14	0.004
Drugs use	0	0	2	0.92	0.61
Twin pregnancy	1	3.6	12	5,5	0.55
Psychiatric history	12	43	42	19	0.005

Pregnancy outcomes of the cohort are shown in Table 2. Pregnancy outcomes did not differ statistically between women with and without a depression. There was a small but nonsignificant difference in the number of elective Caesarean sections (21 vs 17% P = 0.52),in the number of preterm deliveries (18 vs 12% P = 0.34) and in the number of neonates admitted post-partum (18 vs 16% P = 0.81) between women with and without a depression.

Table 2. Pregnancy outcomes of women with and without depression.

	Depression (EPDS ≥12, N = 28)		No Depression (EPDS <12, N = 217)
	N	%	N	%	P-value
Start of delivery					0.52
Spontaneous	11	39	109	50
Induction of labour	11	39	74	34
Elective Caesarean section	6	21	36	17
Analgesia during labour					0.67
Epidural	8	29	64	29
Remifentanyl	5	18	43	20
Other*	0	0	8	3.7
None	15	54	102	47
Mode delivery					0.79
Spontaneous	18	64	136	62
Operative vaginal delivery	2	7	13	6
Elective Caesarean section	6	21	36	17
Emergency Caesarean section	2	7.1	32	15
Preterm delivery<37 weeks	5	18	25	12	0.34
5 minute apgar≤7	2	7.1	8	3.7	0.38
Male gender	16	57	118	54	0.78
Admission of the neonate	5	18	35	16	0.81

* Sedative or opiods during labour, spinal or general anaesthesia during Caesarean was not counted

Depressed women more frequently reported an EPDS score ≥12 six weeks post-delivery, as compared with the women without a depression during pregnancy (25 vs 6% P = 0.001, Figure 1).

Figure 1. Change in EDS score during and after pregnancy and association with PFOC.

A total of 28 women (11%) reported PFOC. More women with a depression reported PFOC (N = 10, 35%) than the women without a depression (N = 18, 8.3% P < 0.0001 Figure 1). Among the women without depressive symptoms at baseline, those who also reported an EPDS score ≥12 six weeks post-delivery, 23% experienced PFOC, whereas this was 7.4% among women with an EPDS score<12 six weeks post-delivery (P = 0.04 Figure 1).

Table 3 shows the association between depression at baseline and depression and PFOC six weeks post-delivery. Women with a depression before 20 weeks are at increased risk of depression six weeks post-delivery (adjusted OR 4.9 95% CI 1.4–17, Table 3). Subsequently, after adjustment for possible confounding factors, women with antepartuml depression are at increased risk of PFOC six weeks post-delivery (adjusted OR 9.2, 95% CI 2.6–32, Table 3). Mediation analysis showed that the effect of antepartum depression on PFOC was not significantly mediated by postpartum depression (Sobel test P = 0.36). In addition, in women with antepartum depression but no depression at the time of PFOC, 35% had PFOC. In women without antepartum depression only 8% had PFOC. This shows that even if the depression resolved, the increased risk of PFOC is still present.

Table 3. Association between depression before 20 weeks and depression and PFOC six weeks post-delivery.

	Depression (EPDS ≥12, N = 28)		No Depression (EPDS <12, N = 217)
	N	%	N	%	OR	95% CI	adjusted OR	95% CI
EPDS ≥12 six weeks post-delivery	7	25	13	6.0	5.2	1.8–15	4.9	1.4–17
PFOC six weeks post-delivery	10	39	18	8.9	6.4	2.5–16	9.2	2.6–32

*Adjusted for ethnicity, education, smoking, alcohol use and psychiatric history.

Discussion

Main findings

This study demonstrates that in a Dutch cohort of pregnant women in two teaching hospitals, 11% report depressive symptoms before 20 weeks of gestation. Among all women, 11% had PFOC six weeks post-delivery. Our study shows an association between depressive symptoms before 20 weeks of gestation and PFOC six weeks post-delivery, even after correction for known confounding factors.

Strength and limitations

The strength of our study is the prospective design and the use of validated questionnaires. Although we have interesting observations, there are also some limitations of our study. Firstly, our hospitals have a special focus on perinatal psychiatry with easy access to a perinatal psychiatrist. This may attract a different pregnant population than hospitals without this type of care and might have led to selection bias. This may have given an overestimation of the number of women with depressive symptoms. Secondly, we did not include a prenatal fear of childbirth questionnaire (W-DEQ A) at baseline, so the fear of childbirth before childbirth was not known. On the hindsight, it would have been wise to measure prenatal fear of childbirth, in order to allow for comparison of this measurement with the outcomes six weeks post-delivery.

Thirdly, the sample size is too small to make statements about differences in perinatal outcomes.

Additionally, although we corrected for differences at baseline between women with and without depression, still unknown/unreported confounding factors could have affected the observed association between depression and PFOC.

Interpretation

The prevalence of EPDS score ≥12 before 20 weeks of gestation in our population is higher than the results from a Dutch study among pregnant women from community midwife practices. In that study a prevalence of 5.6% was found at 12 weeks of gestation (Bergink et al., 2011) compared to 11% in our study. The difference may be explained by the fact that the pregnant population of community midwife practices is a different one, with a lower risk of complications. Women in outpatient clinics (secondary care) as compared with community midwifery care (primary care) have a different medical history or experienced a complicated pregnancy. In addition, women with a severe psychiatric history will mostly be cared for in secondary care, resulting in a higher prevalence of depressive symptoms in this group.

Our findings are consistent with previous studies. The majority of the studies covering fear of childbirth have been performed in Scandinavian populations. The reported prevalence’s of FOC vary between 5–20% (Rouhe et al., 2009; Saisto & Halmesmaki, 2003). In an Australian population Fenwick et al. showed that fear of childbirth (FOC) measured in the same group before (FOC) and after childbirth (PFOC), has a declining trend. They found a prevalence of 22% of PFOC (six weeks post-delivery, defined as W-DEQ B score>71) (Fenwick et al., 2009). This percentage seems to be higher than our results (11% at six weeks post-delivery), but this may be largely explained by the lower cut off value for PFOC (71 instead of 85). If we would have used a cut off value of 71, the prevalence of PFOC six weeks post-delivery in our population would have been 21,4% and thereby comparable with the Fenwick study. Internationally, the accepted cut off value for severe FOC is 85 (Adams et al., 2012; Wijma et al., 1998).

The association between antenatal depression and FOC has been shown earlier (Saisto et al., 2001; Soderquist et al., 2004; Zaers et al., 2008). But FOC has mainly been investigated before delivery. For example, a Scandinavian group studied the association between depression or anxiety and the prevalence of FOC, measured during pregnancy at 32 weeks of gestation. They found that FOC at 32 weeks of gestation was predicted by depression (OR 2.4, 95% confidence interval 1.1–5.2) and anxiety (OR 8.4, 95% confidence interval 4.8–14.7) (Storksen et al., 2012). These findings are consistent with our findings. To the authors knowledge, this is one of the studies on the association between depression and PFOC.

Further research

FOC and depression are areas that currently are not widely investigated and remain behind in the care for pregnant women. Since we know that this has a profound consequence on reproductive choices and even the request for a caesarean section in the next pregnancy, more attention should be paid to FOC. Further research should focus on what women actually need in their pregnancy care. We should focus on both the psychological aspects, as on the physical aspects.

As already described in this discussion, on hindsight in our study it would have been wise to measure prenatal fear of childbirth, in order to allow for comparison of this measurement with the outcomes six weeks post-delivery. This is something to further investigate in the future.

In addition, future research should focus on possible treatment options for depressive symptoms that also take into account FOC since PFOC persisted also in women whose depressive symptoms had resolved. Cognitive therapy and hypnosis-based intervention were shown to be effective in reducing fear of childbirth (Moghaddam Hosseini et al., 2018). Perhaps women with depressive symptoms during pregnancy can already benefit from this, decreasing their risk of PFOC before the actual delivery.

Conclusion and recommendations

Our study shows that depressive symptoms during pregnancy are associated with PFOC.

We don’t know whether our observed association is solely due to depressive symptoms or due to (un)known other confounding factors but the implications for management are the same. Whatever the reason, our study shows an association between depressive symptoms and fear of childbirth. Therefore, we think that screening for depressive symptoms during pregnancy could help highlight women at high risk for PFOC and thereby improve our ability to care for women on an individual basis.

Contribution to authorship

MB: coordinating and planning of collecting data in the Amphia hospital Breda, analysing data, writing.

BMK: coordinating and planning of collecting data in Gelre hospital Apeldoorn, analysing data, writing.

DNMP: accompany of study in Amphia hospital, reviewing and criticizing the manuscript

HJZL: statistical assistance

KMP: accompany of study in Gelre hospital Apeldoorn, reviewing and criticizing the manuscript

Details of ethics approval

The Medical Ethical Board of the Gelre Hospitals Apeldoorn location and the Amphia Hospital Breda exempted this study from institutional review board (IRB) approval, since the screening and referral were already part of routine care, now conducted by the Mind2Care instrument, including opting-out. In the Netherlands, screening is allowed under Dutch Law if opting-out is explicit (in Dutch: the WBO and WGBO Act). De-identified data from patients’ medical records were used in post-hoc analyses.

Geolocation information

Our study was set in Apeldoorn and Breda, two cities in the Netherlands

Acknowledgments

We thank Linda Meurs, Joey de Vogel, Joyce Cantinau and obstetric counselling nurses of Gelre Hospitals Apeldoorn location for their efforts and help.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.