Aims
In the present study, we evaluated three possible pathways of action of avoiding (e.g., having a sCS) versus approaching (e.g., having a natural childbirth) the challenges of childbirth in pregnant women with high FOC. We hypothesized that following MBCP, several mechanisms of change would contribute to natural childbirth: the change from high to lower FOC (emotional pathway), the change from high to lower catastrophic beliefs about labour pain (cognitive pathway), and the change from low to higher mindful awareness (attention pathway). Additionally, we tested whether the number of completed MBCP sessions and the minutes of meditation practice at home were associated with the outcome.
Procedure and subjects
In this study, we analysed the “I’ve Changed My Mind” RCT-data in which 141 pregnant women without a priori (medical) restrictions for natural childbirth experiencing high FOC were randomized to MBCP (n = 75) or ECAU (n = 66) [6]. There were no significant pre-intervention differences between conditions for demographic predictors and outcome measures [6]. Below, a summary of the methodological details most pertinent to the current study are presented. The study’s procedure, which includes the rates of recruitment, reasons for refusal, exclusion, withdrawal and attritions, as well as the course of randomization and masking, has been published previously [6]. Inclusion criteria were an age ≥ 18 years, fluent in the Dutch or English language, between 16 and 26 weeks pregnant at baseline, and high levels of FOC as indicated by a score ≥ 66 on the Wijma-Delivery Expectation Questionnaire (W-DEQ-A) [31]. Exclusion criteria were psychotic episodes, suicidal risk, substance use and dependency, borderline personality disorder, current trauma or traumatic stress disorder, HIV infection, multiple gestations, high risk for premature labour, or participation in other MBPs in the past year.
Intervention: mindfulness-based childbirth and parenting (MBCP)
The intervention consisted of the face-to-face, group-based MBCP program for expectant parents published as the course book “Mindful Birthing” [32]. MBCP was originally designed to teach life-skills and to promote healthy pregnancy and childbirth to all expectant parents. In our trial, we adapted it for pregnant women with FOC. Adaptations were focused on facilitating participants in every session with skilful responding to anxiety- and fear-related responses in guided meditations and enquiry. The nine weekly sessions, with up to six couples in a group, lasted 3 h, and were delivered by experienced midwives certified in MBCP. Sessions included: mindfulness meditation practice (e.g., body scan, sitting and walking mediations, speaking and listening meditation on fear and happiness, yoga) and enquiry; and teachings about psychobiological processes in childbirth (e.g., physiology of labour pain, dilatation, delivery and postpartum) and in new-borns. Participants were asked to commit to daily meditation practices at home for 30 min. MBCP was free of charge, and the sessions took place at mindfulness centres in Amsterdam and The Hague, The Netherlands. MBCP feasibility and participant’s attendance are presented elsewhere [6].
Active control condition: enhanced care as usual (ECAU)
ECAU consisted of two individual fear of childbirth consultations of 1.5 h for the expectant couple. Both consultations were spread over a nine-week period (similar to MBCP) and were delivered by trained midwives. ECAU was developed specifically for anxious pregnant women by the research team to reduce FOC by gaining insight into the factors causing and maintaining fear and stress around pregnancy, birth and the postpartum period (the first consultation); and making a coping plan to deal with fears and stressors and discuss psychoeducation about fear (the second consultation). More specifically, the first consultation was based on the Biopsychosocial Model [33], and the second consultation consisted of writing the commonly used Childbirth Plan of the Royal Dutch Organization of Midwives (KNOV) [34]. ECAU was free of charge, and the consultations took place at the couple’s home.
Measures
Time
Measurements of FOC, catastrophic beliefs about labour pain, and mindful awareness were collected at pre-intervention (T1) and post-intervention before childbirth (T2). The childbirth mode including obstetric interventions used during childbirth were collected after birth (T3). Participant characteristics were collected at T1.
Pathways of action
Emotion pathway: fear of childbirth
The emotion pathway was operationalized as FOC and assessed with the 33-item W-DEQ-A [32]. The questionnaire operationalizes emotions around childbirth (e.g., ‘How do you expect you will feel during delivery; ‘lonely, strong, confident, scared, happy, proud’) as covering general fear, negative appraisal, loneliness, lack of self-efficacy, lack of positive anticipation, and concerns about the child (range 0–165). Higher scores indicate more FOC: high (W-DEQ-A ≥ 66); severe (W-DEQ-A ≥ 85); and phobic FOC (W-DEQ-A ≥ 100) [35]. The W-DEQ-A showed good reliability in an average sample of pregnant women at 16–26 weeks pregnancy (α =0.94) [6]. Cronbach’s α at T1 and T2 in the present study was 0.95.
Cognition pathway: catastrophic beliefs
The cognition pathway was operationalized as catastrophic beliefs about labour pain, and it was assessed by the 12-item Catastrophizing Labour Pain (CLP; range 0–60). This subscale is derived from the Labour Pain Cognitions and Coping List (LPCCL) [18]. A higher score on the CLP represents more catastrophizing of labour pain (e.g., “The pain of childbirth will be overpowering”). In the aforementioned study, the CLP showed good reliability in an average sample of pregnant women (30–34 weeks pregnant) with a Cronbach’s α of 0.84. Cronbach’s α in the present study at T1 was 0.88 and at T2 was 0.92.
Attention pathway: mindful awareness
The attention pathway in our model was operationalized as mindful awareness. Mindful awareness was assessed with the Dutch version of the 24-item Five Facet Mindfulness Questionnaire (FFMQ; range 24–120) [36]. The FFMQ consists of five subscales: Observing (e.g., “When I’m walking, I deliberately notice the sensations of my body moving”); Describing (e.g., “I can easily put my beliefs, opinions, and expectations into words”); Acting with awareness (e.g., “When I take a shower or bath, I stay alert to the sensations of water on my body”); Nonjudging of inner experience (e.g. “I tell myself that I shouldn’t be thinking the way I’m thinking”); and Nonreactivity to inner experience (e.g., “I watch my feelings without getting lost in them”). Higher scores indicate greater mindful awareness. Cronbach’s α in the present study at T1 was 0.73 and at T2 was 0.79.
Intervention outcome: gradient of childbirth mode
Gradient of childbirth mode was operationalized into an ordinal scale consisting of five categories, with higher scores indicating childbirth with more advanced obstetric interventions: 0 = natural childbirth as birth without any obstetric interventions; 1 = spontaneous childbirth with some obstetric intervention (e.g., augmentation with oxytocin or assisted delivery) not including EA; 2 = spontaneous childbirth with EA; 3 = childbirth with obstetric indication for CS made during childbirth; and 4 = childbirth by sCS.
Attendance and practice
The minutes of meditation practice at home were registered by the participants. This data and the presence of participants at MBCP sessions were collected at each of the nine weekly intervention sessions by a MBCP trainer.
Statistical analysis
The primary analysis was performed using the completers data. The allocation process was concealed from the independent outcome assessor. To test our hypotheses of the three pathways of action in the theoretical model of avoiding versus approaching the challenges of childbirth [8], we ran (1) a parallel, multiple mediation model with our hypothesized mediators and (2) single mediation models to further delineate indirect effects. Each variable was transformed to account for the difference between T1 and T2 (i.e., the change scores (Δ), T2-T1). Our independent variable was dichotomous (i.e., ECAU denoted with 0 and MBCP as 1). According to Hayes (2018), utilizing an ordinal variable as a continuous variable (as we did with our outcome variable) in a statistical mediation model is acceptable [37]. No additional covariates were added to the models since randomization of condition assignment was successful [6].
We conducted mediation analyses using the SPSSv25 PROCESSv3.3 macro [38] to test the hypothesized mediators’ effects of the type of intervention (i.e., MBCP or ECAU) on the gradient of childbirth mode. One model with the following mediators was run: (Δ) W-DEQ-A for FOC; (Δ) CLP for catastrophic beliefs about labour pain; and (Δ) FFMQ for mindful awareness. If a significant indirect effect was found, the effect size for each mediator was estimated using the bootstrapping procedure recommended by Hayes (2018). It accounts for a nonparametric distribution and retains power in the model. We tested whether the specific indirect effect was significantly different from zero by constructing 95% confidence intervals using 10.000 bootstrap samples. If zero is contained in the interval, then the indirect effect is non-significant, suggesting the data do not support the proposed indirect effect. Coefficients, standard errors, and p-values were generated [see Additional file 1]. Their corresponding bootstrap confidence intervals were calculated and are documented in Table 3. Note that coefficients are unstandardized, but the bootstrap confidence intervals are standardized [37]. Additionally, indirect paths are partially standardized, which signify the number of standard deviations by which the gradient of childbirth mode is expected to increase/decrease per a change in mediator of size unstandardized coefficient (a) [39]. In addition, to assess the relationship between outcome measure and number of completed MBCP sessions and quantity of meditation practice at home per week, Spearman’s rank-order correlations were calculated.