Procedure
We adopted a cross-sectional online questionnaire through a secure survey platform (Sphinx software), with a 9-month enrolment period. The study and consent procedures were approved by the ethics committee of the University (Comité d’Ethique de la Recherche Tours-Poitiers, n°2019-09-01). Online advertisement and snowballing were used to recruit participants. All participants endorsed an online informed consent and were subsequently directed to the online questionnaire. Measures were taken between 1-month to 12-months postpartum. Male or female participants completed the online survey assessing demographics, obstetric history, psychological history, birth experience, symptoms of anxiety, depression, PTSD, perceived parental bonding, and social support. Participants were instructed to answer about their experience during the birth of their most recent baby. They were asked about potential traumatic events during (or immediately after) the birth, and if they experienced post-traumatic symptoms related to this birth. The online procedure facilitated access to the study for a larger number of participants and also allowed participations based on the availability of subjects.
The exclusion criteria comprised: women who gave birth less than 1 month or more then 1 year ago; partners of women who gave birth less than 1 month or more than 1 year ago; persons under guardianship or curatorship; persons with difficulty to understand the informed consent form; participants who did not fill out the survey completely.
Measures
Biographical data were collected using sociodemographic questions, medical, psychological, obstetrical history, and traumatic life events. A self-reported psychometric assessment allowed to measure the PTSD-FC, postnatal depression, social support, and perceived mother-infant bond.
Sample description. The information regarding obstetric and birth-related history were collected through several questions such as the number of pregnancies, the number of children, the sex of the last child, the mode of the last birth – emergency caesarean, vaginal vacuum, vaginal (no instrument) and vaginal (forceps). We also investigated the experience of distressing events during childbirth by asking “Have you experienced particularly upsetting or traumatic event during childbirth?” We also asked if the participants have lost a child before (i.e., miscarriage, stillbirth, etc.). Moreover, we explored the history of other traumatic events by asking the participants if they have experienced a traumatic event in their lifetime (e.g., accident, natural disaster, illness, unexpected death, violent attack, or sexual abuse). If yes, they were asked to give a brief description of the traumatic event.
The
City Birth Trauma Scale (CBTS) is a self-reported questionnaire developed by Ayers et al. (4) in response to the need for an instrument for assessing PTSD following childbirth [
26]. This scale includes 31 items, 29 of which correspond to the diagnostic criteria for PTSD according to the DSM-5 (including intrusion, avoidance, negative cognition and mood, and hyperarousal), as well as additional questions to assess the subjective criterion and symptoms of emotional insensitivity [
26]. Of the 31 items, 23 of them are based on a Likert-type scale (0 = not at all, to 3 = 5 or more times). The questionnaire has two subscales to evaluate general symptoms and birth-related symptoms (score range 0–69). The response scale for symptoms asks for frequency of symptoms over the last week. The highest score reveals a higher risk for PTSD. An additional questionnaire for partners was also developed. It helps identifying women and partners with PTSD-FC. The psychometric properties have been evaluated and validated by Ayers et al. (4) and replicated in Hebrew [
27], Croatian [
28] and Turkish [
29]. The CBTS has shown good psychometric properties to detect symptoms of PTSD-FC. According to the study of Ayers et al., the reliability analyses of CBTS showed high internal consistency (Cronbach’s α = .92) [
26]. In this study, a non-validated French version of CBTS was used. We also performed reliability analysis for CBTS total and subscales. Similar to previous studies, our results showed high internal consistency for CBTS total scale (α = .92), for CBTS birth related symptoms (α = .90) and CBTS global symptoms (α = .91).
The
PTSD checklist for DSM-5 (PCL-5) is a questionnaire developed by Weathers et al., according to the diagnostic criteria of the DSM-5 [
30]. The scale is composed of 20 items assessing the intensity of the 20 criteria for PTSD symptoms presented in the DSM-5. Responders were invited to respond a Likert-type scale (0 = not at all, to 4 = extremely) to evaluate the level of bother that they felt for each item during the past month [
30]. This self-questionnaire was validated for the screening and monitoring of PTSD and validated in French [
31]. Scores higher than 33 indicate provisional diagnosis of PTSD. High internal consistency was demonstrated for the French version of PCL-5 (Cronbach’s α = .94) [
31] and in the present study (α = .94).
The
Edinburgh Postnatal Depression Scale (EPDS) is a 10-item scale developed by Cox et al. [
32] translated and validated in French [
33]. Responders were invited to score each item from 0 to 3, which allows to assess depressive symptoms in the post-partum period. The scores range from 0 to 30. Score more than 10 indicates provisional diagnosis of depression. The Cronbach’s coefficient of .76 was demonstrated for the French version of EPDS [
33]. Our results showed high internal consistency (α = .886).
The
Medical Outcomes Study (MOS) is a 20 item-scale developed by Sherbourne & Stewart [
34]. The items are divided into 4 categories to evaluate different types of support (emotional, tangible, affectionate and positive social interactions). Responders were invited to score (0 = none of the time, to 5 = all of the time) for each item in order to evaluate how often these different types of supports are available. The internal consistency was found high for MOS in the study of Sherbourne & Stewart (Cronbach’s α = .91) [
34] and very high in our study (α = .958).
The
Mother-to-Infant Bonding Scale (MIBS) is an eight item-questionnaire developed by Taylor et al. [
35]. Each item refers to an adjective to evaluate the participants feelings towards their baby during the first week that follows childbirth. The items are based on Likert-type scale (0 = not at all, to 3 = very much). The scores range between 0 and 24. This scale assesses the perceived difficulties of the mother-child bond during the first weeks of the newborn’s life. It was validated in French [
36]. The scale was also used with fathers to assess their bonding scores with their child [
37]. The internal consistency was demonstrated for MIBS as acceptable in previous studies (Cronbach’s α = .71) [
36] and high in the present study (α = .824).
Statistical analyses
Descriptive statistics were calculated for sociodemographic, medical, psychological, obstetrical characteristics as well as traumatic life events. The bivariate ANOVA tests were performed to test the group differences on the scores of CBTS total and subscales for mothers and for their partners separately (dependent variables). The tests were performed between groups who were either exposed to a potential risk factor or not (i.e., child loss, past traumatic experiences) or whether the different modes of birth differed (i.e., mode of birth, distressing childbirth). Regarding post hoc tests for ANOVA, we first performed multiple comparison between independent variables. We then performed the post hoc Tukey test when the Levene’s test for homogeneity of variances was non-significant, and the post hoc Games Howell test when the Levene’s test for homogeneity of variances was significant.
We also performed Bravais-Pearson correlations to calculate the bivariate associations between the MOS, the MIBS, the EPDS, the PCL-5, the CBTS total scale and subscales for mothers and for partners. The Bravais-Pearson correlations could allow us to explore the relationship between several biopsychosocial factors (i.e., depression, perceived social support, mother-infant bonding) and CBTS total and subscales (possible PTSD-FC). Descriptive and statistical analyses were performed using SPSS 24.0 software (IBM).