Participants
This study utilized data from the pilot RCT comparing MIL to TAU first described by Duncan and colleagues [33]. Participants were recruited through: 1) provider referral, 2) the internet and word of mouth (e.g., parenting message boards, Google advertisements, and list-serves), and 3) posted flyers targeting pregnant people with fear of childbirth. The participant sample included 30 nulliparous child-bearers with low-risk, healthy, single baby pregnancies in their third trimester who planned to give birth in a hospital and were willing to be randomized to either condition. Exclusion criteria included any previous experience in meditation or yoga (barring prenatal yoga, which did not warrant exclusion), involvement in a separate mind/body childbirth class, planned homebirth or other non-hospital setting, or a planned Cesarean delivery. The sample was 59% White (N = 17), 17% Asian (N = 5), 14% Multiracial (N = 4), 7% Black/African American (N = 2), and 3% American Indian/Alaska Native (N = 1). Regarding ethnicity, 18% of the participants were Hispanic/Latina (N = 5, missing = 1). One participant in MIL and two participants in TAU completed childbirth education classes independently (i.e., without a birth partner). Below median household income for the area (<$90k) was reported by 55% of the participants (N = 16) and 10% of the sample reported a household income of less than $10,000 a year (N = 3; See [33] for details). The sample was relatively low risk regarding mental health (see Table 1) with baseline clinical characteristics similar to those of a universal sample versus a selected population of child-bearers at higher risk for mental health issues (e.g., [30]). The current study adheres to CONSORT guidelines; see Figure 1 for the CONSORT flow chart of study recruitment and participation.
Table 1
Baseline Mental Health: Means and Clinical Cut-Offs
|
CES-D
|
STAIT
|
PSS
|
FFMQ
|
Clinical Cut-Off
|
≥16
|
≥40
|
N/A
|
N/A
|
M(SD)
|
9.66 (8.05)
|
36.07 (8.62)
|
15.34 (6.25)
|
3.47 (0.35)
|
N (%) above Clinical Cut-Off
|
6 (20%)
|
10 (34%)
|
N/A
|
N/A
|
Note. None of the baseline mean scores were above clinical cut-off for the scales that provide them (i.e., CES-D and STAIT). CES-D = Center for Epidemiologic Studies Depression; FFMQ = Five Facet Mindfulness Questionnaire; PSS = Perceived Stress Scale; STAIT = State Trait Anxiety Inventory – Trait.
Procedures
All study procedures were approved by the University of California, San Francisco (UCSF) Committee for Human Research (institutional review board), and signed informed consent was obtained from all participants. Participants were randomized to either MIL (n = 15) or TAU (n = 15) using a pre-programmed computer database. Self-report measures were completed online at four time points: time 1 (T1) was the third trimester baseline (immediately pre-intervention and pre-randomization), time 2 (T2) was the week immediately following the intervention (post-intervention but prior to birth), time 3 (T3) was the postpartum follow-up (approximately six weeks post-birth), and time 4 (T4) was one to two years post-birth. Due to the timing of received project funding for long-term follow-up, T4 assessment timing varied such that earlier cohorts completed T4 up to two years post-birth while later cohorts completed T4 at one year post-birth. Participants completed the T4 assessment on average 1.79 years post-birth (M = 93.08 weeks, SD = 0.17 years, range = 1.47-2.20 years. All eligibility screening and assessment was conducted through an online survey software (see [33] for further details of compensation and time period of data collection). The current study was submitted in fulfillment of the first author’s master’s thesis (see [49]).
Interventions
Mind in Labor (MIL): Working with Pain in Childbirth. As described above, MIL is a short, time-intensive weekend (2.5 day) childbirth education program adapted from MBCP for pregnant people and their partners that integrates mindfulness strategies for coping with pain and fear with formal mindfulness meditation for a total of 18 hours of mindfulness training. MIL was conducted by certified MBCP instructors and facilitated by the developer of MBCP. MIL teaches mindfulness strategies for coping with labor-related pain and fear through interactive, experiential exercises, alongside didactic instruction on how mindfulness may be brought to bear on childbirth preparation (e.g., birth physiology) and parenting an infant. Handouts and CDs/ mp3s of guided mindfulness meditations were provided to participants for optional practice following MIL. Participants’ birthing support partners were invited to attend MIL; most partners attended, with one woman participating in MIL independently. There was no cost to participants for the MIL program (see [33] for further details).
Treatment as usual. TAU was an active control condition in which participants were able to choose a standard childbirth education class from a list of approved community resources or a class suggested by a participant that received approval from study staff. These were typical childbirth education program options that featured no mind-body focus, mindfulness meditation, or yoga. Content was determined by reaching out to providers and inquiring about any mind/body or stress-related subject matter. Participants could also request a non-listed class in the case that a listed course was not convenient given their location/schedule, and it was evaluated using the same procedures to create the approved list. Participants were given up to $200 to cover the tuition for their approved childbirth education program.
Measures
Center for Epidemiologic Studies Depression Scale. Depression was measured by the Center for Epidemiologic Studies Depression Scale (CES-D; [50]) at T1-T4. Participants rated their experience of various depression symptoms over the past week on the widely-used, 20-item self-report measure using a scale from 0 (“Rarely or none of the time”) to 3 (“Most or all of the time”). A score of ≥16 is used to indicate clinical levels of depression. The analyses in the current study utilize CES-D scores from baseline through one-year post birth (α = .80 to .89).
Spielberger State-Trait Anxiety Inventory – Trait. Anxiety was measured by the Spielberger State-Trait Anxiety Inventory – Trait (STAIT; [51]) at T1-T4. The scale prompted participants to rate “how you generally feel” for 20 anxiety symptoms on a scale from 1 (“Almost never”) to 4 (“Almost always”). A score of ≥40 is used to indicate clinical levels of trait anxiety. The data from pre-intervention through one year postpartum reflected good internal consistency (α = .90 to .95).
Perceived Stress Scale. Perceived stress was measured by the Perceived Stress Scale (PSS; [52]) at T1-T4. The PSS is a widely used measure of global stress perception—i.e., the degree to which a person perceives life events to be stressful. Participants reported the frequency of 10 stress-related thoughts and feelings in the past month on a scale from 0 (“Never”) to 4 (“Very Often”). This scale is used as a continuous variable, with no set clinical cut-off, where higher scores indicate higher levels of stress. Participants’ scores on the PSS were collected at baseline through one year postpartum (α = .87 to .94).
Five Facet Mindfulness Questionnaire. Mindfulness was measured by the Five Facet Mindfulness Questionnaire (FFMQ; [53]) at T1. The FFMQ is a 39-item measure that assesses five dimensions of mindfulness identified through a factor analysis of existing mindfulness scales, which include observing, describing, acting with awareness, nonjudging of inner experience, and nonreactivity to inner experience. Items are rated on a scale from 1 (“Never or very rarely true”) to 5 (“Very often or always true”) and were averaged to obtain a total mindfulness score (α = .88); to test proposed moderated effects, only the baseline (T1) score was used in analyses.
Data Analysis Plan
We employed an intent-to-treat analysis in which all eligible and randomized participants were included in the analyses with the sample of N = 29 (N = 14 TAU and N = 15 MIL) regardless of their degree of participation in the study. One participant in the control condition was excluded from analyses due to incorrect report of parity in the eligibility assessment (see Figure 1). To address effects on child-bearers’ total distress, a distress composite score of the STAIT, CES-D, and PSS was created at each wave. This combination was found to be justified by correlations among the measures (r = .56-.90; see Table 2) and internal reliability of the composite score, which was created by standardizing and then averaging the scores (similar to composites made for other investigations of internalizing symptoms in the perinatal period; [53]).
Table 2
Correlations of Distress Components and Mindfulness
|
STAIT_T2
|
STAIT_T3
|
STAIT_T4
|
PSS_T1
|
PSS_T2
|
PSS_T3
|
PSS_T4
|
CESD_T1
|
CESD_T2
|
CESD_T3
|
CESD_T4
|
FFMQ_T1
|
STAIT_T1
|
0.631***
|
0.334
|
0.503**
|
0.676***
|
0.356
|
0.069
|
0.385
|
0.565**
|
0.332
|
-0.095
|
0.330
|
-0.678***
|
STAIT_T2
|
|
0.577**
|
0.794***
|
0.427*
|
0.759***
|
0.186
|
0.654***
|
0.413*
|
0.695***
|
0.259
|
0.550**
|
-0.404*
|
STAIT_T3
|
|
|
0.714***
|
0.140
|
0.750***
|
0.643***
|
0.745***
|
0.266
|
0.652***
|
0.712***
|
0.625***
|
-0.183
|
STAIT_T4
|
|
|
|
0.227
|
0.692***
|
0.182
|
0.904***
|
0.266
|
0.714***
|
0.426*
|
0.836***
|
-0.367
|
PSS_T1
|
|
|
|
|
0.397*
|
-0.185
|
0.127
|
0.820***
|
0.303
|
-0.202
|
0.119
|
-0.631***
|
PSS_T2
|
|
|
|
|
|
0.454*
|
0.668***
|
0.459*
|
0.806***
|
0.593**
|
0.578**
|
-0.353
|
PSS_T3
|
|
|
|
|
|
|
0.363
|
-0.089
|
0.289
|
0.750***
|
0.174
|
0.002
|
PSS_T4
|
|
|
|
|
|
|
|
0.259
|
0.710***
|
0.579**
|
0.850***
|
-0.266
|
CESD_T1
|
|
|
|
|
|
|
|
|
0.510**
|
0.047
|
0.298
|
-0.567**
|
CESD_T2
|
|
|
|
|
|
|
|
|
|
0.493**
|
0.716***
|
-0.181
|
CESD_T3
|
|
|
|
|
|
|
|
|
|
|
0.487*
|
0.061
|
CESD_T4
|
|
|
|
|
|
|
|
|
|
|
|
-0.267
|
FFMQ_T1
|
|
|
|
|
|
|
|
|
|
|
|
|
Computed correlation used pearson-method with pairwise-deletion.
|
Note. CESD = Center for Epidemiologic Studies Depression; FFMQ = Five Facet Mindfulness Questionnaire; PSS = Perceived Stress Scale; STAIT = State Trait Anxiety Inventory – Trait.
We used multilevel modeling in HLM to examine trajectories of pregnant people’s distress across time and to test proposed differences by intervention group. As outlined above, distress outcomes were operationalized by the composite score for primary analyses and broken down by specific scales (i.e., CES-D, STAIT, PSS) in secondary analyses. Level 1 modeled each woman’s distress trajectory with an intercept and linear slope; the latter was centered at the final (T4) assessment so that intercepts represented final levels of distress at 12- to 24-months postpartum. Level 2 modeled between-woman differences in distress trajectories that could be explained by intervention condition (testing study question/hypothesis 1), as well as interactions of intervention condition with baseline levels of symptoms or mindfulness (testing study question/hypothesis 2).