Introduction

Childbirth is an experience with many dimensions and is unique to each woman. It is normal to experience some apprehension during pregnancy and after childbirth. If the feelings affect women performing their daily activities, they might suffer perinatal mental health problems such as depression and fear of childbirth (FoB) [1]. Depressive symptoms (DS) and FoB are universal concerns that can influence one another and may occur before and/or after childbirth. There is no clear definition of FoB, but it is usually categorised into either primary when occurring before birth or secondary when arising after experiences of traumatic or distressing childbirth [1].

DS can be a consequence of long-term FoB [2], and can also be a risk factor for severe FoB during pregnancy and after childbirth [3–6]. Some women develop a phobic fear, leading to total avoidance of birth because of DS [1, 3, 4]. Several systematic reviews [7–12] and longitudinal studies [10, 13, 14] have reported the prevalence rates of FoB and DS before or after childbirth. Few longitudinal studies have followed the same women to assess these conditions before and after birth.

Rates of FoB and DS vary between countries, with global prevalence rates of 5–30% for antenatal FoB (A-FoB) [1, 7, 12–16] and 5–34% for antenatal DS (A-DS) [17–20], whereas postnatal FoB (P-FoB) ranges between 5 and 20% [16, 21, 22] and postnatal DS (P-DS) between 4 and 25% [23]. Most studies have reported lower FoB and DS rates during the postpartum period than during pregnancy [10, 11, 16, 22, 24].

FoB can be associated with expectations and experiences of labour such as pain or operative procedures; fear of being incapable of giving birth; perceived less childbirth knowledge; not having a voice in decision making; the outcome of labour, including the woman’s and baby’s wellbeing; previous obstetric complications; expectations of parenthood or fear of becoming a parent, low social support and experiences of a non-supportive partner and/or health care providers [22, 25–30]. Severe FoB can affect childbirth, for instance, leading to increased labour duration or a more significant number of obstetric interventions and inducing fear of subsequent pregnancy [25, 28, 31, 32]. DS is more complicated since personality resources can have an impact together with life events. Several factors may be related to DS, such as being single, divorced or separated, not having a formal education, deployment, young age, having a large family, financial instability, previous caesarean section, history of miscarriage, unplanned pregnancy, lack of preparation for childbirth, being a victim of gender-based violence, lack of social support, lack of parenting knowledge and low self-efficacy [18, 19, 33–36]. DS can interrupt women’s daily activities, mother-infant bonding and infant sleeping patterns [24, 37, 38].

Studies have explored the prevalence rates of common perinatal mental disorders (CPMDs) in Tanzania, including DS, anxiety and post-traumatic stress disorders [19, 39–43]. However, the rates differ between regions. FoB after childbirth is understudied. Some studies have indicated that FoB and DS can be present in the same woman [5, 44]. A study in Tanzania revealed an association between FoB and DS during pregnancy and investigated associations with sociodemographic and obstetric factors [40]. Screening of mental health disorders is limited during routine prenatal care. One study in Dar es Salaam, Tanzania, indicated that in women with the possibility of having mental health disorders, their symptoms were not identified and diagnosed if they didn’t raise their concerns by themselves. Probably due to inadequate knowledge and skills among healthcare providers in identifying and managing prenatal mental health problems [43]. We could not identify an informational package which is integrated into routine prenatal care guidelines to guide healthcare providers in providing women with information about CPMDs. However, one cluster-randomised controlled trial in 12 villages in rural Ifakara, Tanzania, was done to assess the effect of a community health worker-delivered intervention on maternal depressive symptoms in rural Tanzania [45]. To our knowledge, no study has compared the prevalence rates of FoB and DS before and after childbirth in the same cohort of women and any associations with other factors. Hence, this study aimed to assess the prevalence rates of FoB and DS after birth, including patterns of FoB and DS from pregnancy to postpartum. We also assessed predictors for the development and persistence of FoB and DS in women after childbirth.

Methods and materials

Study participants and procedure

We consecutively recruited and interviewed 694 pregnant women who came for routine antenatal care at the selected health facilities from September 2018 to July 2019. We computed a sample size using the two-proportion formula in Epi Info 7 StatCal, with a power of 80% and a significance level of 0.05 with two tails. A minimum sample of 616 pregnant women would need to be enrolled in this study. Inclusion criteria were: resident of the chosen study setting; attending the same health facility for antenatal and postnatal care; pregnancy of at least 32 weeks of gestation; no previous caesarean section; anticipating vaginal birth; Kiswahili speaker. After childbirth, women were followed up and interviewed between 4–6 weeks postnatally at reproductive and child health clinics, using the same approach and tools as during antenatal data collection. However, some women did not return to the same health facility after childbirth, which meant that postnatal data were collected at health facilities and within the community. Recruitment, interviews and follow-ups took place concurrently, and the final sample of women who completed participation numbered 625 (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flowchart for recruitment and follow-up of participants.

https://doi.org/10.1371/journal.pone.0277004.g001

We trained six nurse-midwives, who were not formally hired during data collection, as research assistants to collect data. An interviewer-administered questionnaire in the Kiswahili language was the primary data collection approach. We developed a visual scale to enable participants to rate their expectations and experiences in childbirth using the Wijma Delivery and Expectation Questionnaire versions A &B (W-DEQ-A&B). This would decrease the risk of having differing verbal explanations for response items from different research assistants.

Ethical approval and consent to participate

We obtained ethical clearance from the Muhimbili University of Health and Allied Sciences, Senate Research and Publication Committee, with reference number 2018-03-09/AEC/Vol.XII/91. Permission to collect data in the selected health facilities was granted by the Regional Administrative Secretary in the Pwani region, Tanzania. Before the inclusion of women in this study, they were informed that their participation was voluntary and that they were allowed to withdraw at any time without being asked about their reasons. Qualified participants provided written informed consent before their participation. Participants under 18 years of age were considered emancipated minors and allowed to sign the consent form without their parents/guardians [53]. Confidentiality and anonymity were ensured throughout data collection. Participants under 18 years were considered emancipated minors and allowed to sign the consent form without their parents/guardians [53]. Confidentiality and anonymity were ensured throughout data collection.

Results

Description of the population

Of 694 eligible pregnant women who agreed to participate, 625 were followed up in a second interview (Fig 1). In this paper, we have included the participants who provided information during pregnancy and at 4–6 weeks after childbirth to compare findings before and after and for longitudinal analysis. Sixty-nine participants (9.9%) were lost to follow-up due to poor communication (some women did not own mobile phones and others were unreachable during the follow-up), travel after birth or geographical problems in reaching them. Among those lost to follow-up, eleven (15.9%) withdrew from the study due to various reasons like infant death or lack of permission from their male partners to continue with the study. Three questionnaires for measuring FoB after childbirth were missing three responses each on how women felt when labour became most intense. The missing items were from women who underwent caesarean sections. To address this, the imputation of data was performed using the mean substitution technique.

The age of women who completed the study ranged between 14 and 46 years, with a median of 26 years and an interquartile range of 24 years. The majority (72.0%) were married and had primary education (61.4%). More than 50% of participants were multiparous, and 52.3% did not experience any obstetric complications during the index pregnancy or childbirth. More than 80% of women gave birth at health facilities, and most babies were full-term and healthy. The majority of the women (52.3%) were socially supported by their male partners. See Table 1 for further details.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sociodemographic and obstetric characteristics of the participants.

https://doi.org/10.1371/journal.pone.0277004.t001

Information on index birth and baby

Table 2 indicates that a large proportion (81.8%) of women had less than 12 hours pass between admission and birth. Participants who were lost to follow-up and hence excluded from the longitudinal and comparative analysis were younger, more often employed, married, and socially supported by male partners than those included. Also, participants who were lost to follow-up were less likely to have A-FoB and A-DS.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Information of the index baby.

https://doi.org/10.1371/journal.pone.0277004.t002

The overall scores for P-FoB varied between 0 and 96, with a median score of 47.0. The prevalence rates of A-FoB and P-FoB were 16.0% and 13.9%, respectively, and there was an insignificant decrease of P-FoB by 2.1%. A total of 40 participants (6.4%) had persistent fear during pregnancy and postpartum. On the other hand, 60 participants (9.6%) had fear during pregnancy which resolved after childbirth, while 47 (7.5%) had no fear during pregnancy but became fearful after birth (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Patterns of fear of childbirth and depressive symptoms during pregnancy and after childbirth.

https://doi.org/10.1371/journal.pone.0277004.t003

The overall scores for P-DS were 0–23. The median (IQR) for P-DS was 1 (4). The prevalence rate of A-DS was 18.2%, and that of P-DS was 8.5%. The proportion decreased significantly after childbirth (p < 0.001). Twenty-seven women (4.3%) had DS both during pregnancy and after childbirth. A total of 87 women (13.9%) had DS during pregnancy which resolved after birth, whereas 26 (4.2%) had no DS during pregnancy but developed DS after childbirth (Table 3).

Predictors of P-FoB and P-DS

Tables 4 and 5 display different logistic regression models for univariable and multivariable analyses on P-FoB and P-DS as outcomes when the groups were treated as independent and dependent, respectively. The variables associated with FoB and DS were entered into the multivariable analysis models, one after another.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Relationships between fear after childbirth and sociodemographic and obstetric factors.

https://doi.org/10.1371/journal.pone.0277004.t004

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Relationships between depressive symptoms after childbirth and sociodemographic and obstetric factors.

https://doi.org/10.1371/journal.pone.0277004.t005

Several predictors were significantly associated with P-FoB. Having A-FoB, P-DS were the strongest predictors of P-FoB. Other predictors were giving birth by caesarean section and having experienced obstetric complications in the index pregnancy. Detailed findings are provided in Table 4.

Discussion

This study described patterns of FoB and DS among women who were followed up during their postpartum period in Tanzania. Findings showed that some women experienced FoB (6.4%) and DS (4.3) during pregnancy and after childbirth, while some did not experience it while pregnant but developed after childbirth. The odds of women developing P-FoB were higher among women with more than twelve hours passed between admission and childbirth. The odds of developing P-DS were higher for women with P-FoB while being single and having experienced a lack of parent social support in previous childbirth were associated with increased odds of persisting DS. Having experienced obstetric complications in previous childbirth and P-DS were cross-cutting predictors in women for developing FoB and having P-FoB or persisting FoB. Giving birth to an ill infant predicted developing DS and having P-DS.

The current study determined that P-FoB and P-DS prevalence rates among Tanzanian women did not differ from those in other parts of the world. Compared with other studies in Tanzania, A-DS prevalence rates were lower than those reported in Mwanza [39], while P-DS rates were almost the same as those reported from Kilimanjaro [38]. The FoB variation was within global ranges, despite global differences in income level, cultural context, birth rates, health systems and other factors [16, 17, 21, 22]. Childbirth is a natural process accompanied by experiences such as labour pain [54, 55], loss of body control [56] and other unforeseeable events, which could explain why women from different backgrounds and nationalities can share similar rates of FoB and DS.

The decrease of FoB and DS over time from pregnancy to postpartum is significant when dealing with women seeking antenatal or postnatal care. The decline may follow a natural pathway, as some women may have many emotional disturbances and expectations before birth, including fear of the unknown. Generally, previous studies have indicated that FoB [16, 22, 57] and DS [11, 24, 58] are more prevalent during pregnancy than postpartum. However, a study by Fenwick and colleagues in Australia [59] reported that women delivered through emergency caesarean sections had higher levels of P-FoB than A-FoB. This could support our findings since most women in this study had a vaginal birth. Despite the decrease, FoB did not resolve entirely in some women.

Similarly, studies in other parts have shown that women with A-FoB and A-DS are likely to have FoB [22, 60] and DS [17, 61] after childbirth. This suggests that pregnancy monitoring and addressing the most severe emotional, psychological and physical issues, which could aggravate FoB and DS at both timepoints, are essential. We found that not all women with A-DS and A-FoB presented with DS and FoB after childbirth. On the other hand, some women developed these conditions only after childbirth. This indicates that DS and FoB can occur at any time from birth to the postpartum period. These findings might interest healthcare providers, particularly midwives, in raising awareness that FoB and DS can change over time. Thorough assessment and care throughout pregnancy and the postpartum period are crucial for alleviating potential adverse childbirth outcomes. This was in line with a systematic review report that revealed that DS could develop throughout the antepartum and postpartum periods [8].

The finding that FoB is associated with DS reminds healthcare providers that these conditions can coexist. Having P-DS increased the risks of P-FoB and the other way around. Thus, a thorough assessment of both conditions during pregnancy and after childbirth is paramount for early identification and timely intervention. The findings were similar to those in previous longitudinal studies, which suggested that FoB is among the strongest predictors of P-DS [2, 62, 63]. Contrary to a study from Iran assessing anxiety and FoB as predictors of postnatal depression in nulliparous women, A-FoB was not a significant predictor of P-DS [64]. The difference might be explained by the timing of assessing prenatal and postpartum FoB and DS and/or contextual factors.

Our study revealed a history of obstetric complications in the most recent previous pregnancy and childbirth as a common cross-cutting predictor for developing P-FoB and having FoB at both timepoints. Childbirth complications can be unpredictable and vary from one woman to another. Similarly, childbirth expectations during pregnancy differ and can lead to negative childbirth experiences if they are not met. Negative experiences at health facilities can lead to frustration, confusion, anger and lack of physical and emotional control, affecting women’s decisions on future childbirth [22, 65–67]. If a woman’s actual childbirth experience does not match her expectations, it can lead to the development of FoB [31, 68, 69]. Health providers have to be aware of women’s wishes and work to meet their expectations. Timely interventions such as psychological support to prevent complications during pregnancy, childbirth and postpartum will promote positive childbirth experiences.

More than twelve hours pass between admission and childbirth at health facilities could contribute to developing P-FoB. This might be due to an overly long labour time, long waiting and thinking about childbirth outcomes. The findings align with an Indian study showing that women who spent six or more hours at a health facility before childbirth more often had P-FoB [21]. Other studies have found that FoB was a predictor of prolonged labour [31, 59]. This indicates that FoB and duration of labour predict each other. Therefore, it is essential for health care providers to ensure FoB and duration of labour are considered when providing care to women during labour and childbirth.

An ill newborn, a stillbirth or an early neonatal death contributed to developing P-DS, probably because most pregnant women expect healthy babies. This highlights the importance of healthcare providers offering care that meets women’s and their families childbirth expectations. Moreover, our study findings showed that persisting DS appeared to be associated with being single and lacking social support from parents in a previous childbirth. That being a single mother was a predictor of both A-DS and P-DS is not surprising and could be due to a lack of social, financial and psychological support from a male partner. Based on the Tanzanian cultural context, single parenting can also be misinterpreted as an indicator of low status by the surrounding community, accompanied by stigma. However, there is limited literature to support or contradict our findings. Few studies have followed the same cohort of women from pregnancy through birth to postpartum to assess patterns and predictors of developing DS or having persisting DS. We have found studies that determine predictors of A-DS and P-DS independently in various groups of women. For instance, a systematic review [70] and a prospective cohort study [35] reported that single mothers had an increased likelihood of A-DS.

Clinical implications of the findings

Several factors contribute to FoB and DS before and after childbirth in Tanzania–and possibly other parts of Africa. This reflects the multiple challenges and approaches to meeting women’s expectations and promoting positive childbirth experiences. The findings have provided insight into the magnitude of the problem, influencing factors and the importance of improving the care provided to Tanzanian women. During the antenatal period, FoB and DS screening is paramount in identifying and supporting women at risk of developing mental illnesses and those who have experienced complications during previous pregnancies and childbirth.

The findings also highlight the importance of integrating mental health components into routine antenatal and postnatal care.

Strengths and limitations

To our knowledge, this is the first study to assess FoB and its relationship to DS in Tanzania. A follow-up study from the third trimester of pregnancy to 6 weeks after childbirth gave us information. It revealed patterns on how different factors can predict FoB and DS at varying timepoints in the same cohort. This could guide the appropriate timing for interventions. Furthermore, the loss to follow-up was small (9.9%). Those omitted from the longitudinal analysis were not significantly different from the original sample recruited during the antenatal period, indicating limited selection bias. We used interviewer-administered questionnaires instead of a self-administered approach which other studies have broadly used to lessen the participants’ struggle to fill in answers by themselves.

Conducting interviews might have biased our findings if the presence of interviewers influenced women’s responses. Furthermore, not including high-risk women during recruitment, such as those who had experienced caesarean childbirth or did not expect vaginal birth, was another limitation of our study. This could underestimate the prevalence rates of FoB and DS. This might also limit the generalisability among women with a previous caesarean section.

Conclusions

FoB and DS during pregnancy and postpartum were strongly associated with each other. The prevalence rates of FoB and DS were higher among pregnant women than among postnatal mothers. Obstetric complications predicted developing P-FoB and persisting FoB. Having more than 12 hours pass at a health facility before childbirth was associated with higher odds of developing P-FoB; being unemployed was associated with increased odds of having FoB at both timepoints. Given birth to an ill child or experiencing stillbirth was a predictor for developing P-DS while being single and having a lack of social support from parents predicted having DS at both timepoints. Knowing why some women are more likely to develop FoB and DS at different timepoints is essential for timely assessment, early identification and appropriate care to improve women’s health and well-being before and after childbirth. To overcome the prevailing problems, effective interventions need to start during antenatal.

Supporting information

Acknowledgments

We acknowledge, with thanks, the assistance of the Pwani region administrative team for granting permission to carry out the study. We also thank the health care providers on-site for their support during the entire time of data collection. We extend our gratitude to the research assistants and the participants for participating in this research.