Childbirth can be defined as a very important event in the life of every woman. This remarkable event has significant psychological and emotional impacts [1,2]. However, fears arise during pregnancy regarding delivery and pregnancy processes for various reasons, such as: young age of mother, the concern of genetic disorder in the baby, anxiety about the physical conditions of the hospital, negative impact on the pregnancy through posts about the pregnancy process on social networks (social media effect), fear of being alone during childbirth, unplanned pregnancy and not receiving support from spouse or family, among others.

Studies have demonstrated that Fear of Childbirth (FOC) prevalence changes across countries, its worldwide prevalence reported being 14% via a meta-analysis [3]. The prevalence of FOC changes post-partum depression and stress disorders and can even prolong labor [1].

FOC can affect the mode of delivery. This fear has led to an increasing number of cesarean deliveries worldwide [4]. Consequently, it is a major cause of a cesarean delivery preference [5]. These preferences were prevalent in women who had a history of psychological and obstetrical situations [6]. This preference is also considered a right to self-decision-making in developed countries [7]. However, nulliparous women had a higher level of FOC [7], while low-risk pregnant women who preferred vaginal deliveries had the highest level of FOC compared to alternative delivery processes [8].

Several studies have so far focused on suggesting a new cut-off value for determining FOC levels for different populations. However, no consensus could be achieved regarding the common threshold value for FOC. Currently, Wijma Delivery Expectancy/Experience Questionnaire (WDEQ-A) [9] is the most common tool for assessing FOC, and its score of 85 has been commonly used to classify the fear. Despite its common use, this cut-off value may not be useful either to detect the fearful sample or discriminate between fearless and fearful groups in different samples. Therefore, it can be considered essential to propose a cut-off value to identify the levels of FOC and to specify factors associated with FOC based on the newly suggested cut-off value that can reflect the characteristics of the population in question.

This study proposed a new cut-off value for FOC based on WDEQ-A scores of the participants and investigated the relationship between FOC and mode of delivery based on the newly suggested cut-off value for FOC.

This paper-based cross-sectional study was conducted between March 2019 to February 2020 at Gynecology and Obstetrics Policlinics of the Kirsehir Ahi Evran University Training and Research Hospital. This study was approved by Kirsehir Ahi Evran University non-interventional clinical studies ethical committee (Approval Time: 26/03/2019 - Approval Number: 2019-06/68).

The inclusion criterion was defined as 18 years or older pregnant women, while pregnant under the age of 18 years were excluded. The sample size was determined by the G*Power (3.1.9.6) program, and 330 participants were calculated with 80% power and 5% error rate to detect a small effect size, namely, Cohen’s d = 0.2. Also, the Cronbach alpha internal consistency coefficient was calculated as 0.866 in this study. Data was collected via a validated Turkish version of the 33-item WDEQ-A Scale and General Information Form that was created by the researchers by reviewing the existing literature. The form included questions regarding sociodemographic (age, educational level, and occupation), obstetric (unplanned pregnancy, fear of bleeding during delivery, and concern for disorder in the baby after delivery), and social-related (social-media effect and fear of not being able to return to former social life after delivery) determinants. This study was conducted with the participants who provided written informed consent. The original W-DEQ is a 33-item questionnaire aiming to measure the FOC during pregnancy. The items are in the form of a six-point Likert scale yielding a minimum 0 and a maximum of 165 points [9]. The reliability and validity of the Turkish version of the scale have been studied previously [10]. Some items (2, 3, 6, 7, 8, 11, 12, 15, 19, 20, 24, 25, 27, and 31) were reverse-scored for the concordance.

A total of 330 pregnant women were included in the current study. The overall mean and median ages were 27.55 and 27 years, respectively (range: 18–45 years; standard deviation: 5.743 years). The mean WDEQ-A score was 52.52 ± 25.034, (range: 4–122; median: 48.5). The baseline characteristics of the participants are presented in Table 1.

The proposed cut-off value was based on the distribution of the FOC scores, the fourth quartile of the scores as 69 was proposed for determining FOC in unselected Turkish population. A statistically significant difference between the prevalence of low, moderate, high, and severe FOC in terms of quantile classification and the prevalence of FOC in terms of widely accepted FOC cut-off value of 85 as proposed by Wijma et al [9] (χ2: 13.785, p-value < 0.001). However, the prevalence of FOC was observed to be similar across the delivery mode (χ2:1.281, p-value: 0.865), and maternal age groups (χ2:2.036, p-value: 0.361).

The mean FOC levels were 54.04 ± 24.957, 52.227 ± 25.724, and 51.227 ± 24.18 for the previously cesarean, nulliparous, and previously vaginal delivery groups, respectively (p=0.668). Furthermore, the highest FOC levels were observed in self-employment workers (62.722 ± 31.301), while the lowest levels were observed in women without any concern for disorder in the baby (44.77 ± 22.898). The group comparison results are presented in Table 2.

The results of four different binary logistic regression models (all sample, pregnant women with previously caesarean birth experience group, nulliparous group, and vaginal delivery groups) revealed that the concern for any disorder in the baby after the delivery was a significant risk factor for all sample groups (AOR = 0.339, 95% CI = 0.154–0.744; p = 0.007).

The social-media effect (AOR = 0.334, 95% CI = 0.127–0.881; p = 0.027) and concern for any disorder in baby after the delivery (AOR = 0.218, 95% CI = 0.058–0.817) were significant risk factors for caesarean delivery group; education level (AOR = 0.134, 95% CI = 0.024–0.747; p = 0.022 for secondary school and AOR = 0.033, 95% CI =0.003–0.338; p = 0.004 for high school) and fear of bleeding during delivery (AOR = 0.203, 95% CI = 0.055–0.754; p = 0.017) were significant risk factors in the vaginal delivery group; while unplanned pregnancy (AOR = 3.669, 95% CI = 1.180–11.405; p=0.025), social-media effect (AOR = 0.350, 95% CI = 0.130–0.942; p=0.038), fear of bleeding during delivery (AOR = 0.187, 95% CI = 0.060–0.585; p = 0.004), concern for any disorder in the baby after the delivery (AOR = 0.344, 95% CI = 0.12–0.977; p = 0.045), and fear of not being in formal social life after delivery (AOR = 0.238, 95% CI = 0.090–0.627; p = 0.004) were significant determinants for the nulliparous group.

The discriminative powers of these regression models were 0.683 (95% CI: 0.614–0.752), 0.691 (95% CI: 0.574–0.808), 0.783 (95% CI: 0.659–0.906), and 0.833 (95% CI: 0.759–0.907) for all samples, pregnant women with the former cesarean experience group, vaginal delivery group, and nulliparous group, respectively.

In this study, the sensitivity, specificity, and Cohen’s Kappa measures were calculated based on the consideration of other cut-off values that were defined in literature as gold-standard for the WDEQ-A score of FOC. The sensitivity, specificity, and Cohen’s Kappa statistics of other cut-off values in the literature were calculated as follows: (a) 50, sensitivity: 56.6%, specificity: 100%, and Cohen’s Kappa: 0.502; (b) 60, sensitivity: 70.9%, specificity: 100%, and Cohen’s Kappa: 0.752; (c) 66, sensitivity: 89.2%, specificity: 100%, and Cohen’s Kappa: 0.922; (d) 71, sensitivity: 100%, specificity: 98.8%, and Cohen’s Kappa: 0.975; (e) 81, sensitivity: 100%, specificity: 89.2%, and Cohen’s Kappa: 0.723; (f) 85, sensitivity: 100%, specificity: 85.2%, and Cohen’s Kappa: 0.577; and (g) 100, sensitivity: 100%, specificity: 78.7%, and Cohen’s Kappa: 0.252.

These results demonstrated that the newly proposed threshold value has good agreement with the previously proposed cut-off values.

In this study, we aimed to analyze the relationship between FOC and mode of delivery and to propose a new cut-off value of FOC for unselected Turkish pregnant population.

The prevalence of FOC was 28.7, 24.1, and 21.6% in pregnant women with previous cesarean experience, nulliparous, and vaginal delivery groups, respectively. Moreover, the previously cesarean-experienced group had slightly higher FOC scores compared to the other two groups. Our findings showed either similarities [11,12] or conflicts [8,13] in this regard. In this study, we proposed a new cut-off value of FOC for the pregnant population based on the quartiles of the WDEQ-A scores of the participants. To the best of our knowledge, this is the first study to propose a threshold value of FOC for unselected Turkish-pregnant population. Previous studies have proposed cut-off values of FOC based on either the distribution of WDEQ-A scores [14-19] or their clinical representation [20]. Moreover, we calculated the sensitivity, specificity, and Cohen’s Kappa statistics of the newly proposed cut-off value as calculated by Calderani et al. [20] to determine its concordance with the other threshold values described in the literature.

Several other research proposed different cut-off values to categorize FOC previously. Threshold values as 66 and 71 were proposed by Zar et al [21] and Fenwick et al [18] for Swedish and Australian populations respectively, using the same method as our study i.e. fourth quartile of the distribution of WDEQ-A Scores. Moreover, Rouhe et al [22] took the highest fifth percentile and proposed 100 as a new threshold value for the Finnish population. Ryding et al. [23] proposed 85 as a cut-off value for the Swedish population, by taking the highest 10th percentile of scores of participants. Even though there are various threshold values defined in the literature, the optimal cut-off value was proposed as 85 for different populations in various studies [14,16-17].

The proposed cut-off value as 69 FOC score in the current study was found lower compared to other cut-off values defined in the literature. This result also shows that the threshold of fear of the Turkish pregnant population was lower, suggesting they are less fearful about childbirth. The reason for this lower threshold and less fearfulness might be the results of having previous birth experiences, participating in psychotherapy programs for the preparation of delivery, etc.

In the current study, no association was observed between FOC levels and sociodemographic characteristics of the pregnant population, such as maternal age, educational level, occupation, and gestational age. Similar results have been published previously [12,15,24,25]. In contrast, it was found that women who have a low education level and those who are unemployed are likely to have high levels of FOC, while maternal age was not associated with FOC [26]. The high levels of FOC were reported to be associated with educational levels and unemployment in several studies [27,28]. Unemployment was defined as a sociodemographic risk factor related to FOC by Saisto et al. [29]. Employed women were reported to be likely to have a higher level of FOC by Toohill et al. 2014 [25]. It was revealed that higher maternal age was associated with higher levels of FOC [30]. However, educational level and occupation were not reported to be significant risk factors for maternal cesarean preference, while age was a significant factor in this regard [31]. In the present study, effects of economic or socioeconomic status on FOC were not investigated directly, occupation and educational levels and their relationship with FOC were analyzed instead. Previous studies identified their relationship with delivery mode preferences. Buyukbayrak et al. reported that higher monthly income was observed to be related to a cesarean delivery preference [31], while high or unspecified socioeconomic status was reported to be associated with first-time mothers (the nulliparous group) by Räisänen et al. [32].

In the current study, FOC levels were found to be associated with social (social-media effect) and obstetric-related (concern for disorder in the baby) risk factors for the formerly cesarean group; sociodemographic (educational level) and obstetric-related (fear of bleeding during delivery) risk factors for the vaginal delivery group; while social (social-media effect and fear of not able to return to the former social life) and obstetric-related (unplanned pregnancy, concern for disorder in baby after the delivery, and fear of bleeding during delivery) risk factors were the determinants of the FOC levels for the nulliparous group. In contrast, cesarean delivery preference was found to be associated with FOC and educational level between 10 and 13 years in the primiparous and multiparous groups; age over 35 years and depressive symptoms in the primiparous group; and previous cesarean delivery and negative birth experience in the multiparous group by Løvåsmoen et al. [33]. Similarly, regardless of mode of delivery groups, Konar et al. have found parity and a concern of disorder in the baby after delivery as the significant factors for FOC for unselected Turkish pregnant sample [34]. In addition, medical risk factors and previous negative birth experiences were previously recorded to be associated with elective cesarean delivery [1]. Furthermore, increasing parity was reported to be related to cesarean delivery preference in an earlier study [31]. On the other hand, having depression was found to be related to first-time mothers (the nulliparous group) by Räisänen et al. [32]. Undergoing training programs regarding pregnancy and delivery processes were not found to affect FOC levels in our study. Contrary to our results, women who did not receive training regarding pregnancy and delivery processes had two times higher FOC scores than the women who received this training [24]. In a previous study, nulliparous women were reported to be likely to have a higher level of FOC compared to multiparous women [25]; however, in contrast, we did not find any such association for the entire study population in our study.

Numerous studies have investigated the relationship between FOC and delivery mode preferences [2,5]. Moreover, several studies have contributed to the literature by analyzing the effects of FOC on cesarean delivery [1,15,17]. This study differs from the previous studies by constructing study groups with actual mode of delivery rather than the preferred one. Thus, this study is unique since risk factors were determined based on the actual mode of delivery for each study group.

This study has some limitations. Firstly, part of this study was questionnaire-based; therefore, bias arising from this kind of study design cannot be ignored. Secondly, it was a single-center study; hence, these results cannot be generalized to the entire pregnant population of Turkey. Moreover, a limited discussion could be made for this current research since the risk factors studied are different compared to the existing literature. Therefore, the most similar risk factors were reviewed, matched with the current risk factors, and attempted to be discussed.

In the current study, the focus was on the statistical evaluation of newly-proposed cut-off value rather than its clinical relevance. Therefore, other tools for measuring FOC, such as the Fear of Birth Scale (FOBS), Tilburg Pregnancy Distress Scale, etc. were not utilized for comparison and determination of clinic usefulness. A new threshold value was proposed via the most common tool for FOC, W-DEQ, and its statistical concordance was assessed primarily. Hence, this present study should be considered as having a mainly statistics-based approach.

In conclusion, a comprehensive study, including the research of a wide range of multi-level factors of FOC, was conducted for evaluating the relationship between FOC and mode of delivery. Both sociodemographic and obstetric-related risk factors were observed to be associated with each study group. Furthermore, having the information regarding the highest prevalence of severe FOC in the previously cesarean groups training programs regarding delivery and pregnancy process are recommended to be organized to lower the FOC and increase the knowledge regarding these processes for the entire pregnant population, especially for pregnant who prefer cesarean delivery. It’s also planned to apply this scale for each unselected pregnant woman who was admitted to polyclinic right before the medical examination and specify the most fearful group via the newly proposed cut-off value for the Turkish population. Other than training programs, the most fearful group is being aimed directed to some therapy programs which are planned to be provided in the collaboration with the Psychiatry Department and psychological counselor of the hospital to help reduce the fear. Even online therapy programs and telephone-based counseling are considered to be conducted under pandemic conditions. Furthermore, couple-based therapy programs could also be organized in an attempt to reduce fear with the father's support. Moreover, we recommend the proposed cut-off value to be utilized for the Turkish population as it yielded effective diagnostic and concordance values with the previously proposed thresholds. Additionally, multi-center studies must determine the relationship between mode of delivery and FOC more precisely. Moreover, only Turkish pregnant women were included in this study. The inclusion of pregnant women of other nationalities would enable us to analyze the differences in various populations. Therefore, it can be seen as a future research option to investigate such a comparison of different populations.