Fear of Childbirth and Depressive Symptoms Among Pregnant Women: A Cross-sectional Survey in Tanzania


 Background: Many women experience fear of childbirth (FoB) and depressive symptoms (DS) during pregnancy, but little is known about FoB among Tanzanian women. The current study aimed to assess the prevalence of FoB and DS among pregnant women and determine predictors of each and both, focusing on sociodemographic and obstetric predictors.Methods: A cross-sectional study was conducted at six health facilities in two districts in Tanzania between 2018 and 2019. In total, 694 pregnant women with gestational age between 32 and 40 weeks and expecting vaginal delivery were consecutively recruited and assessed for FoB and DS. We collected data through interviews using the Wijma Delivery Expectancy Questionnaire Version A and Edinburgh Postnatal Depression Scale. We performed multivariable logistic regression to investigate the predictors of FoB and DS. Results: The prevalence rates of FoB and DS among pregnant women were 15.1% and 17.7%, respectively. FoB and DS were more likely in women aged above 30 years (AOR 6.29, 95% CI 1.43–27.84) and in single mothers (AOR 2.57, 95% CI 1.14–5.78). Women with secondary education and above and those who had given birth previously were less likely to have FoB (AOR 0.22, 95%CI 0.05–0.99) and DS (AOR 0.27, 95% CI 0.01–0.87). Women who had previous obstetric complications, and those who did not receive any social support from male partners in previous childbirth were more likely to have FoB and DS. FoB was strongly associated with DS (AOR 3.42, 95% CI 21.20–5.53). DS only was more common in women who had inadequate income (AOR 2.35, 95% CI 1.38–3.99) or had previously experienced a perineal tear (AOR 2.32, 95% CI 1.31–4.08).Conclusions: Not having a formal education, having only primary education, being aged above 30 years, being single, being nulliparous, having experienced obstetric complications, and having a lack of social support from a male partner during previous pregnancy and childbirth were predictors of FoB and DS during pregnancy. FoB and DS were strongly associated with each other. It is vital to identify at-risk women early, to offer support during pregnancy and childbirth.


Introduction
Women's childbirth experiences are multidimensional and unique to each woman, and can be positive, negative, or a mixture of both. The experience can be in uenced by personal feelings, perceptions of personal performance [1], mental and physical health, and communication and interaction with health care providers [2,3]. Moreover, type of childbirth, childbirth preferences regarding vaginal birth versus caeserean section (C/S) [4,5], freedom in the expression of feelings related to childbirth, involvement in the birth experience, support from a partner and relatives [1,2,5], and other environmental factors, like lack of privacy during childbirth [1], can affect the childbirth experience. Some women who have a negative birth experience develop fear of having more children in the future [6][7][8].
Quantitative studies across the world have shown that factors associated with FoB include sociodemographic factors like young maternal age, lack of social support from male partners, relatives and friends, unemployment, nancial worries, and a history of abuse [22][23][24]. Obstetric factors like nulliparity and adverse obstetric events, such as a previous operative birth, may provoke childbirth fear [22,[25][26][27]. Also, not being mentally/physically prepared for giving birth, an expectation of unendurable pain, a feeling of loss of control during labour and birth, and fear of death have been deemed to predict FoB [25,26]. Inappropriate support from the birth team contributes to low childbirth con dence among women [22,26].
FoB is associated with pre-existing psychological problems like DS. An association between FoB and depression has been reported in a register study of 788,317 pregnant women in Finland, where depression was found to be the most potent risk factor for FoB, regardless of parity [34]. There are similar ndings from other studies performed around the world [35][36][37].
Fear of childbirth may have consequences in the lives of women and their babies. It is associated with poor emotional and psychological health for women in general [23,38]. During childbirth, it can lead to C/S on maternal request [6,[39][40][41], increased use of pharmacologic pain relief [38], prolonged labour [42], aggravated maternal and foetal distress [38], and increased obstetric interventions, like labour argumentation [39,40]. Maternal FoB and DS can also affect the infant's future, potentially leading to decreased birth weight, delayed psychosocial development, reduced breastfeeding duration, hindered growth, severe malnutrition, more episodes of diarrhoea, and poor compliance with immunisation plans [23,42,43]. This might create problematic mother-newborn bonding, with subsequent attachment di culties that may affect infant growth and development, including cognitive development [44][45][46].
To the best of our knowledge, no published studies have assessed FoB and its predictors among Tanzanian women, despite advocacy on integrated mental health in maternal and child health services in the health system and proven long-lasting consequences of maternal FoB and/or DS. Studies on the prevalence of FoB and DS, their predictors, effects, and management, have been performed mainly in European countries, with a few from India [47] and African countries [13,14]. In Tanzania, some studies have assessed DS during pregnancy, revealing a prevalence of 11.5% in the Kilimanjaro region, northern Tanzania [17], and 33.8% in the Mwanza region [19]. Antenatal care relates to women's care during pregnancy, with the target being a healthy mother and child at end of pregnancy/birth. It is essential to increase understanding of FoB and DS during pregnancy to enhance the chances of providing the right support. Hence, the study's primary objective was to determine the prevalence of FoB and DS among pregnant women at ³ 32 weeks of gestation. The secondary objective was to investigate the predictors of FoB and DS, focusing on sociodemographic factors and previous obstetric experiences.

Study design
This was a cross-sectional study investigating the prevalence and predictors of FoB and DS among pregnant women in Tanzania.

Study setting
We carried out the study in the Mkuranga and Kisarawe districts in the Pwani region, Tanzania. The Mkuranga district has 56 health facilities in total: 1 hospital, ve health centres (2 governmental and three privately owned), 50 dispensaries (38 governmental and 12 privately owned). The Kisarawe district has a total of 36 health facilities, including 1 district hospital, 3 health centres, and 31 dispensaries (26 are governmental and ve privately owned).
In each district, one district hospital and two health centres, all of which were government-owned, were selected for the study. The selected health facilities had a volume of at least ten antenatal visits per day, and the district hospitals had facilities for childbirth by both vaginal delivery and C/S. In Tanzania, antenatal care is accessible in every public health care facility at no charge. A pregnant woman is recommended to make at least eight antenatal visits at these facilities prior to delivery; at these facilities where nurse-midwives are the primary care providers. In case of any complications identi ed when providing services, the nurse-midwife refers the woman to a specialist or higher-level health facility for further management.

Study participants and procedure
We recruited and interviewed a consecutive sample of pregnant women seeking antenatal services at selected health facilities from September 2018 to March 2019. The eligibility criteria were: being pregnant in at least the 32 nd week of gestation; speaking Kiswahili; being resident in the selected district; having had no previous C/S; anticipating vaginal birth; attending the relevant health facility for antenatal care. Of the 1,130 pregnant women approached, 702 were invited to the study, 8 refused participation, and 694 provided survey data. Women who refused to participate did so due to lack of time, having other commitments after receiving antenatal care or that their male partners were waiting for them; some women felt uncomfortable taking part in the survey.
We trained six research assistants (RAs) in the data collection tools, use of a visual scale, study participant recruitment, and data collection procedures, including ethical principles in data collection. During the actual data collection, RAs recruited study participants, obtained informed consent, and performed data collection through face-to-face interviews in the selected antenatal clinics. Interviews were selected as the data collection method, to maintain data collection consistency, overcome the illiteracy challenge, and enhance comprehension of the questions and data collection scales. To decrease interview bias, participants could use a visual analogue scale to rate levels of fear for the Wijma Delivery and Expectation Questionnaire version A (W-DEQ-A). This reduced the risk of having different verbal explanations for response items from different research assistants. To enhance validity, all interviews were conducted in Kiswahili.

Instruments and measures
The participants were interviewed regarding demographic and obstetric characteristics and their feelings and expectations of childbirth, using the W-DEQ-A [48]. They were also assessed on DS using the Edinburgh Postnatal Depression Scale (EPDS) [49]. Questions on obstetrics and social support were asked in relation to the history of the most recent previous pregnancy and childbirth.
The W-DEQ-A is a 6-point Likert scale questionnaire with 33 items ranging from 0 (not at all) to 5 (extremely), yielding a minimum score of 0 and a maximum score of 165. The higher the score, the more intense the FoB. The cut-off points from the original tool are as follows: a score of £ 37 is considered a low level of fear, a score of 38-65 re ects a moderate level of fear, a score of 66-84 signi es a high level of fear, and a score of ³ 85 indicates a severe level of fear [50]. In this study, we de ned FoB as scoring ³ 66 and no FoB as scoring < 66. The original W-DEQ-A has been reported to be a reliable questionnaire with a Cronbach's alpha of 0.93. The original English questionnaire version was translated into Kiswahili and reviewed by a team of native Kiswahili speakers and professionals in midwifery, obstetrics, behavioral science, psychology, and psychiatry. Further, a pilot study and validation process were carried out and found that the translated W-DEQ-A was a reliable tool, with a Cronbach's alpha of 0.83 [51] (unpublished, submitted).
The EPDS is a tool used to identify common perinatal mental disorders and has been used to screen women at risk of developing depression during pregnancy or/and after childbirth. The EPDS is a four-point Likert scale with 10 items ranging from 0 to 3. Participants rate each item based on how they have felt in the preceding seven days. The lowest score that can be obtained on this scale is 0, while the highest is 30. The higher the score, the higher the risk of having DS. Scores ³ 10 suggest the presence of DS, while scores ³ 13 represent depressive illness of varying severity [49]. In this study, we de ned DS as scoring ³ 10 and no DS as scoring < 10. he internal consistency reliability of the original EPDS was excellent, with a Cronbach's alpha of 0.88. Our study had a Cronbach's alpha of 0.85. The EPDS has been translated into Kiswahili in Kenya and showed an acceptable Cronbach's alpha of 0.78 [18]. We adapted the EPDS from Kenya, translated by Green and colleagues [18], for data collection in Tanzania.

Data analysis
Data were analysed using the Statistical Package for the Social Sciences (SPSS) version 26. Prior to analysis, the normality test was performed for the total scores of FoB and DS using the Q-Q plot, histogram, box plot, and Kolmogorov-Smirnov. Descriptive statistics were used to summarise demographics, obstetric characteristics, and the prevalence rates of FoB and DS. A chi-squared test was performed to determine the associations between sociodemographic and obstetric variables and the prevalence rates of FoB and DS. Logistic regression was used to estimate the effects of sociodemographic and obstetric variables on the prevalence rates of FoB and DS after adjusting for potential confounders. Univariable analysis was conducted to assess the association between each sociodemographic and obstetric variable and the presence of FoB and/or DS. All the variables with p < 0.2 in the univariable analysis were subjected to several multivariable logistic regression models using a hierarchical approach. A p value of < 0.05 was considered signi cant. Effect sizes were presented using odds ratios (ORs) with 95% con dence intervals (CIs). The outcomes of interest were grouped as (1) FoB vs. no FoB and DS vs. no DS, (2) Neither FoB nor DS vs. FoB but no DS, DS but no FoB, and FoB in combination with DS.

Results
Description of the population In total, 694 eligible pregnant women agreed to participate in this study. Their age varied between 14 and 46 years, with a median age of 26 years, and the majority (73.2%) were married and had primary education (61.8%). The majority were nancially and socially supported by their male partners and had planned the pregnancy.

Prevalence of FoB and DS
The total scores for FoB ranged from 1 to 102, while those for DS were 0-27. The median scores with interquartile range (IR) for FoB and DS were 48 (24.0) and 3 (7), respectively. The overall prevalence of FoB was 15.1% and that of DS was 17.7%. The results indicated that about 37 (5.3%) of the pregnant women had both FoB and DS, 68 (9.8%) had FoB but no DS, 86 (12.4%) had DS but no FoB, and 503 (72.5%) had neither FoB nor DS.
Scores for the item in the W-DEQ-A Scores for each item in the W-DEQ-A are shown in Table 1 in descending order of prevalence. The most common fear reported by pregnant women was labour pain. The items for which most women scored low were longing for a child and childbirth being fun.   Tables 3 and 4 show different logistic regression models for both univariable and multivariable analyses of FoB and DS as outcomes. The variables associated with FoB, DS, and both FoB and DS, were entered into multivariable analysis models, one after another. Age, education, and parity were included in every model, since they had been shown to be confounders of FoB and DS predictors in previous studies.  Another strong predictor of FoB was having a low educational level and lack of social support from a male partner in a previous childbirth. The strongest predictors of DS were being single, inadequate income, lack of social support from a male partner in a previous childbirth, having ever experienced obstetric complications, and have experienced a perineal tear before the index pregnancy ( Table 3).
As shown in Table 4, the predictors of DS only, as compared with no FoB or DS were: lacking formal education; being unmarried; having inadequate income; having experienced a perineal tear in a previous delivery. The odds of having both FoB and DS as compared with no FoB and no DS were higher in older, single, and/or nulliparous women, those with a low education level, and those who had experienced obstetric complications.

Discussion
We aimed to determine the prevalence rates and predictors of FoB and DS among pregnant women attending antenatal clinics at selected public health facilities. We found that the prevalence rates of FoB and DS were in a similar range: 15.1% and 17.7%, respectively. Some women had both FoB and DS, some had FoB but no DS or DS but no FoB, and others had neither FoB nor DS. Predictors of FoB and DS were lack of formal education, age above 30 years, never having given birth, being single, having had obstetric complications in the past, and having experienced a lack of male social support at a previous childbirth. Predictors of FoB were not having a formal education, while DS was seen in women without formal education, with inadequate income, who were single, and/or who had experienced a tear/episiotomy in a previous childbirth. Another key nding was that FoB and DS were strongly associated.
The prevalence rates of FoB and DS among Tanzania pregnant women in our study are within global ranges when the same validated tools are used for analysis [10, 14, 15, 17-19, 47, 52]. The similar results across the globe might indicate that most women fear labour pain, being alone, and losing control of their body during labour and childbirth. Tanzanian women, like other women worldwide, have positive expectations on health facility childbirth.
Most women who prefer health facility childbirth seek life-saving technology in case of obstetric complications.
Some wish to have family members nearby to provide care and affection [53]. However, Tanzanian women's expectations may be hard to meet due to the high patient-health care provider ratio at most public health facilities. This could contribute to the high prevalence rates of FoB and DS. Also, disrespectful and abusive treatment from health providers during childbirth has been reported in Tanzania. For instance, women report being neglected/ignored, receiving physical and verbal abuse [54], being shouted at, threatened, slapped/pinched, or being left alone and forced to deliver by themselves [55]. All of these events could cause fear of childbirth.
Being above 30 years of age, not having a formal education or having primary education only, being single, being nulliparous, and having previous experience of obstetric complications were common among women with FoB in combination with DS. Surprisingly, we found that older women were more fearful and had more DS than younger ones. The ndings were in line with those of a study done in Norway, assessing childbirth experiences in rst-time mothers of advanced age [56]. The reason could be that older women have heard of more complications, which might intensify fear ahead of childbirth.
We found contradictory ndings, showing that younger women/teenagers were more likely to develop DS and FoB than older women [22,24,29]. The discrepancy could be due to most teenagers being nulliparous with limited childbirth experience, which could manifest as DS and FoB. Health care providers should be attentive to all women and their differing childbirth expectations. FoB and DS were more likely in single women than married women. These ndings were similar to those of previous studies [13,16,28,43].
Further, women who had never given birth previously had a higher risk of having both FoB and DS. This was in line with other studies, which reported that nulliparous women presented with FoB more often than multiparous women [6,25,27]. However, we could not nd any study assessing predictors of FoB in combination with DS for comparison.
Previous obstetric complications were strong predictors of both FoB and DS. Similar ndings have been made in several other studies: a previous negative experience predicts FoB and DS in subsequent childbirth [6,26,57,58]. Negative birth experiences can lead to hesitance at becoming pregnant or giving birth in the future, resulting in delaying subsequent pregnancies or total avoidance of later pregnancies [6-8, 10, 11, 59, 60]. This might impact on women's future reproductive lives, likely diminishing trust in the ability to give birth and trust in maternity services. Such mistrust might lead to women not seeking maternity services from health facilities or opting for a C/S rather than vaginal birth [61,62]. It is crucial to raise nurse-midwives awareness on the possible causes of negative birth experiences and to discuss how to support these women during subsequent pregnancies and childbirths. Women who have negative experiences should be identi ed during antenatal care for psychological support during pregnancy and childbirth. The same applies for women at risk of experiencing obstetric complications. These women will need to be empowered through provision of antenatal education on what to expect throughout the perinatal period.
In this study, women who received social support from their male partners during previous childbirths were less likely to report FoB and DS. This is in line with previous studies which reported that male partner support during childbirth is essential in alleviating fear related to childbirth [24,61] and preventing DS [63]. This highlights the importance of partner companionhip before, during, and after childbirth in a country where a male companion is not yet allowed or standard procedure during health facility childbirth.
Our study showed that FoB and DS were strongly associated. DS during pregnancy predicted having FoB. These ndings are in line with ndings from most systematic reviews and published studies [59,64,65].
Depressive disorders are an important health problem globally. Among the participating women, DS was more likely to occur in women without formal education, who were single, who had an inadequate income and/or who had experienced tearing and/or an episiotomy in a previous delivery. Our ndings were in line with previous studies showing that single mothers [16,28,66] and those with inadequate income [8,19,28] had increased risks of DS. Further, several previous studies have found that less educated people are more likely to be depressed than more educated ones [28,67]. There might be a pattern in how people of a similar educational background perceive the childbirth process, child upbringing, and associated resources. Another explanation could be that the more educated people are, the more they seek information from different sources, which could prevent them from emotional distress, manifesting as DS. Our ndings were also consistent with another study showing that perineal wounds due to either tearing or episiotomy were associated with DS [47].

Strengths And Limitations
The study's major strength was the use of standardised, validated, and widely used tools for measuring FoB (W-DEQ-A) and DS (EPDS), increasing credibility. The use of a visual aid scale was an added advantage that ensured clarity when using the W-DEQ Likert scale in the W-DEQ tool. Further, the sample was large enough to determine the prevalence rates and predictors of the outcome variables. Moreover, the quality of collected data was ensured as data collection was carried out under the direct supervision of the rst author and general guidance from all team members. While many other similar studies have used self-administered tools, we used the intervieweradministered questionnaire technique, eliminating the hurdle for participants struggling to ll in responses on their own.
Using interviewers for data collection was a strength, but could also entail some limitations. If women adjusted their responses to please the interviewer, a social desirability bias might have occurred. This could lead to the prevalence rates of FoB and DS being slightly underestimated. Another limitation was that we did not know whether women in our study who were categorised as having FoB and DS had received any treatment. We were also unable to offer any support to women identi ed as having any problems. This is because the collected data were analysed and interpreted weeks later, meaning that women could not be referred for further management.
Furthermore, the tools were only for screening purposes and therefore could not be used for clinical diagnosis. Our ndings cannot be generalised to the rst and second trimester of pregnancy, as data were collected during the third trimester. Comparisons with results from other studies may be impaired because of different data collection tools and cut-off points used to measure FoB and DS. However, we compared our ndings with other studies that used only W-DEQ and EPDS tools to de ne FoB and DS [50] and DS [49], leaving out studies that used other tools to measure FoB and DS. W-DEQ-A might have been affected by being translated into Kiswahili, due to cultural diversity and differences in how women de ne and perceive childbirth in different countries. Furthermore, use of an adapted version of the EPDS validated in Kenya, where Kiswahili is also a national language, might be a limitation, as some words may have a different meaning when used in Tanzania.

Conclusions
Previous obstetric complications were the strongest predictor of FoB and DS, and FoB and DS were strongly associated with each other. Lack of social support from a male partner was also a predictor of FoB and DS. Not having a formal education was a predictor of FoB. Being single, having an inadequate income and having experienced a perineal tear were strong predictors of DS. Having FoB in combination with DS was more common in women aged above 30 years, without formal education, who were nulliparous and/or single. Understanding why some women are more prone to FoB and DS is vital in developing effective prevention and timely intervention to restore their mental health and psychological well-being throughout pregnancy, delivery, and after childbirth. This was granted by the Regional Administrative Secretary in the Pwani region, Tanzania. We obtained written informed consent from the women who agreed to participate in the study. The study participants who were under 18 years of age were considered as emancipated minors and allowed to sign the consent form in the absence of their parents/guardians [68].

Methodological consideration
All methods were carried out in accordance with relevant guidelines and regulations.
Consent to publication