A Culturally Grounded Prenatal Coparenting Intervention: Results of a Randomized Controlled Trial with Unmarried Black Parents

Abstract

This study examines the efficacy of a prenatal intervention designed to promote healthy coparenting relationships in families where low-income, unmarried mothers and fathers were expecting a first baby together. One hundred thirty-eight Black and mixed-race mother-father dyads participated. Coparent dyads were randomly assigned to either a treatment as usual (TAU) group, receiving referrals and navigation support to existing community services (control), or to TAU plus invitation to a series of 6 dyadic Focused Coparenting Consultation (FCC) sessions led by a male-female mentor team (intervention). 71% of those prenatally assessed were later reassessed at 3 months postpartum. Both mothers and fathers contributed reports of coparenting, father engagement, physical and psychological IPV, and depressive symptoms. Intent-to-treat analyses indicated: 1) parents in the intervention group reported more positive coparenting at 3-months post-partum than did control group parents 2) mothers in the intervention group reported significantly more time spent by fathers with the child than did control group mothers; 3) parents in the intervention group had significant reductions in psychological IPV compared to parents in the control group and 4) both mothers and fathers showed reductions in self-reported depression over time, with no differential impact of group. Findings suggest that the FCC intervention may produce modest but important benefits for unmarried, low-income Black coparents in the transition to parenthood.

The transition to parenthood (TTP) is often experienced as a rewarding but challenging process introducing rapid shifts in couple relationships and new forms of stress. A generation of research has demonstrated that prenatal relationship quality is a particularly relevant risk or protective factor for post-natal family functioning. The significance of couple functioning for married adults make a case for supporting couples prior to the arrival of a new baby (Cowan & Cowan, 1992; Kuersten-Hogan & McHale, 2021). Research has also documented that for couples experiencing abnormally high levels of stress, including lower socioeconomic families and younger couples, transitions to new parenthood are fraught with risk for numerous forms of postnatal distress, including paternal disengagement, negative attributions about child behavior, and potentially abusive behaviors (Belsky, 1993; Fagan, 2014; Holden & Banaz, 1996).

Historically, most TTP studies enrolled predominantly white married mid-SES participants (Baucom, Chen, et al., 2018; Karney, Kreitz & Sweeney, 2004). Far fewer focused on Black couples. Comparing mid-SES Black couples to similar income white couples, Hobbs and Wimbush (1977) reported that while describing slightly more difficulty in adjusting to their first child, a higher percentage of Black than white mothers labelled their marriages as happier and more satisfying after their baby was born than before. Positive parenthood adjustment was later echoed in studies documenting that after a child’s birth, Black husbands and wives report more egalitarian views of gender equality than white husbands and wives, including shared responsibility for parenting (Kane, 2000; Orbuch & Eyster, 1997; Glauber & Gozjolko, 2011) with men described as “hands on” and involved with daily child rearing by their wives (Crowley & Mouzon, 2019). Blanchard and LeBaron (2019) posited that while some differential pregnancy outcomes exist for white and Black parents (i.e., gestational period length, preterm delivery), other aspects of couples’ TTP experience may be universal among middle class U.S. couples.

Such studies provided relevant guidance for understanding positive TTP adaptation in Black families who were not facing sociodemographic risks related to poverty and unmarried parenthood. Less is understood about positive TTP adaptations of Black families led by unmarried coparents – indeed, about the TTP of most families led by unmarried coparents – facing such risk (Lu et al., 2010). On the one hand, the national Fragile Families and Child Well-Being study of unmarried parents transitioning to parenthood found that over 80% of unmarried fathers were at least episodically present during the mother’s pregnancy, a signal of initial dedication to the shared future child. On the other, by the time that child entered kindergarten, only two in five unmarried fathers remained meaningfully engaged. Something significant shifts over time in unmarried men’s ambitions to be active partners in a mother-father-child family triangle, with unfavorable impacts for children (McLanahan, Tach and Schneider, 2013).

In response, new programs aspiring to promote positive couple relationships, father involvement, and coparenting appeared. A significant body of research details that coparenting relationships have more proximal effects on child adjustment than do couple relationships (Feinberg, 2003; McHale & Lindahl, 2011). Coparenting is also a particularly salient index for unmarried parents, as coparenting relationships can not only often outlive committed couple relationships but can develop and endure even when no committed couple relationship exists (McHale, Waller & Pearson, 2012). Among programs designed to support unmarried parents, most prolific have been Couple Relationship Education (CRE) interventions that bring parents together as couples to focus on bettering their own relationship (Hawkins and Erikson, 2015) and Responsible Fatherhood (RF) programs working principally with fathers to provide financially for their children as they try to build a communicative relationship with their child’s mother (Moore et al., 2018; Holmes et al., 2020). CRE and RF programs do often achieve sought-after impacts on couple functioning and/or father engagement but have not been as effective in benefitting coparenting (Hawkins and Erikson, 2015; Holmes et al., 2020; Moore et al., 2018).

Consequently, coparenting specific interventions began surfacing. Pilkington et al. (2019) provided an initial examination of their impact. Reviewing RCTs of interventions featuring coparenting content, including 12 that assessed coparenting outcomes, Pilkington determined that only half reported a positive intervention effect on at least one coparenting measure, with only 4 of 7 trials assessing father involvement documenting a positive effect. Results are hence mixed as to whether interventions forefronting coparenting content exert significant impact on coparenting or father engagement with children. Apropos to the current report, only one trial examined in the Pilkington et al. (2019) review was specific to Black coparents - a Protecting Strong African American Families (ProSAAF) in-home program (Barton, et al., 2018; 2020). ProSAAF was designed to promote coparenting and parent–child relationships in African American families parenting children aged 9 to 14. Though not a study of the TTP, preliminary coparenting benefits were documented (Lavner et al., 2019). There was also evidence ProSAAF had positive indirect effects on several child outcomes through promoting improvements in couple functioning and better parent–child relations (Lavner et al., 2020).

To underscore again, most coparenting specific interventions, including nearly all TTP interventions that have engaged both coparents, have catered primarily to white families. Prior to the current program of work, no dyadic coparenting intervention at the TTP had been designed specifically for or tested with low-income Black families led by unmarried parents, a sizable U.S. subpopulation for whom coparenting is already a time-honored community and family adaptation. The current study hence reports initial findings from the first RCT of a dyadic coparenting-specific intervention designed to meet the cultural context and needs of unmarried, low-income Black parents expecting their first child together.

A Culturally Grounded Prenatal Coparenting Intervention: Background Considerations

In Black communities throughout America, there has long been a wariness regarding government-sponsored programs that pledge help but ultimately disappoint and intrude too invasively into the family’s private domain (Boyd-Franklin, 1995). In the case of governmental CRE and RF programs created to strengthen families, emphasis has been given to relational and economic benefits of marriage, with responsible fatherhood linked to child support enforcement (Noyes, Vogel & Howard, 2018). While the coupling and child support emphases of federal CRE and RF programs do speak to some unmarried Black coparents, many others ascribe to culturally founded and -centered coparenting adaptations, including some that delink partnering from parenting (Florsheim et al., 2011). Unmarried Black parents’ reliance on relationship suspension and pair-bonding can sometimes serve as an adaptive strategy for dealing with structural racism, environmental and family barriers to long-term relationships; enabling child well-being and father engagement even as parents wait to formalize relationships (Roy et al., 2008).

Acknowledging the distinctive structural, environmental, and family circumstances unmarried Black coparents face in the United States, curricular design and programming for this study was steered by Black community voices. A group of 12 community elders, mentors, advocates, clergy members and healthcare professionals reviewed a dyadic intervention model called Focused Coparenting Consultation (FCC; McHale & Irace, 2010; McHale & Carter, 2012). FCC has three stages: consciousness-raising about coparenting and its impact, skill building, and guided enactments wherein parents practice new skills they have learned with interventionist support. FCC’s distinctive features include (a) delivery in a dyadic, not a group format; (b) an emphasis on interparental communications concerning the child and (c) absence of any specific content concentrating on interparental intimacy or affection. Parents progress from general exercises building consensus on perceived differences in their parenting philosophies and values to more challenging conversations about contentious issues they themselves identify as current or potential obstacles to their shared coparenting (residency and time-sharing, conflicts with extended family, presence of children from prior unions). FCC sustains persistent focus on children across all 6 dyadic sessions.

The community leaders recommended adaptations to FCC: a 1-on-1 mentor-to-parent rapport-building stage in advance of dyadic meetings to promote trust; reconceptualization of the first FCC meeting to involve a get-to-know-you meal together with the interventionists (who were called mentors); regular experiential exercises; emphasis on intergenerational legacies; and use of concrete examples, visual graphics, and explicit discussion of the status of Black children in the local community. Curricular elements are described in McHale, Stover & McKay (2021). The result was a grounded community-based coparenting program and curriculum called “Figuring It Out for the Child” (FIOC) (see Table S1 for session details).

Community leaders felt parents would not explore relationship-based programming if basic family needs were going unmet. Ooms and Wilson (2004) called for CRE programming to be embedded within a community service support structure and so relationship programming in FIOC was supplemented by access to a resource and referral (R&R) navigator who helped parents leverage already-existing community-based services. Navigation included referral to agency programs specializing in employment, housing, childcare, educational opportunity, material supports, and other essentials considered to be “Treatment as Usual” (TAU) in the community. To ensure equitable access for all families, R&R navigation to help access TAU supports was also afforded to families who did not receive the intervention (Warren, 2016).

In many ways, the approach utilized in this initiative was pioneering. Beyond taking a lead from community voices in modifying programming, the very notion of a dyadic coparenting intervention accommodating needs and issues of unmarried Black coparents at the transition to new parenthood was one that had not previously been explored. An initial pre-post evaluation pilot study of FIOC (McHale et al, 2015) had found that at 3 months postpartum, unmarried coparents showed statistically significant gains on 8 of 12 coparenting communication and relationship variables examined, with moderate to large effect sizes. However, without a control group, those improvements could not be definitively attributed to the intervention. The current study was designed to test whether there is an intervention effect of FIOC on both coparenting and father engagement, as reported by both coparents in a randomized trial.

Coparenting Interventions and Risk at the TTP: Intimate Partner Violence (IPV) and Postpartum Depression (PPD)

Postpartum intimate partner violence (IPV) is disconcertingly common (Saltzman et al., 2003), and the TTP strains intimate relationships (Cowan & Cowan, 1992). Challenges in coping with stressors increase IPV risk (Islam, 2017), and hence coparenting interventions – especially dyadic ones – carried out during the pregnancy must weigh IPV carefully. Results of prenatal coparenting intervention studies to date have been mixed with respect to IPV outcomes (Heyman et al., 2019; Florsheim et al., 2011; Kan & Feinberg (2015). Intervening dyadically can be risky, and in the U.S. some state statutes prohibit intervening dyadically if IPV has been identified (Austin & Dankwort, 2003). At the same time, there is some evidence that dyadic interventions can be of help to certain couples who stay in relationships following situational couple violence (i.e., IPV in the context of escalating arguments not marred by the power and control dynamics typical of intimate terrorism; Stith et al., 2003). Also of relevance, Florsheim and colleagues (2011) reported that in a young high-risk population, a careful dyadic coparenting intervention appeared to help prevent IPV. To date, however, dyadic interventions for unmarried coparents who report situational IPV preceding the TTP – particularly those not in committed relationships -- have not been the subject of systematic study.

Such studies are important, given that parents in relationships dotted by situational violence pre-pregnancy frequently do stay engaged after their baby’s arrival. Declines in IPV are sometimes observed during pregnancy (Saltzman et al., 2003), but IPV behaviors of both men and women frequently return post-pregnancy (Charles & Perreira, 2007). IPV is common among young, unmarried (Charles & Perreira, 2007; Saltzman et al., 2003), and low income (Cunradi et al., 2000) couples, though studies specifically examining IPV between Black coparents during the TTP have not been conducted. A range of causes that disproportionately affect Black individuals (Caetano et al., 2005, 2003) - educational disparities, economic privation, mental health, substance misuse (Dowd et al., 2005; Golden et al., 2013) and racism mediating the impact of societal and individual-level risk factors (Al’Uqdah et al., 2016; Hampton et al., 2003) - may also impact IPV. Extant data indicate that Black women are more than twice as likely to be victims of IPV (Saltzman et al., 2003), and significantly more likely to engage IPV behaviors themselves (Cunradi et al., 2000). More frequent and serious woman-initiated, and more mutual bidirectional violence among Black couples (Caetano et al, 2005) has also been documented.

Given the distinctive circumstances of unmarried expectant coparents, the risks posed by IPV, and the possibility that drops seen in IPV during the pregnancy may indicate perpetrating parents not wanting to harm their unborn child (Håland et al., 2016), serious consideration must be given to developing helpful interventions to mitigate post-partum risk. The TTP itself offers a window of opportunity to connect with unmarried parents to promote positive coparenting and potentially mitigate IPV. Positive coparenting is protective to children even when there has been IPV (Fainsilber, Katz & Low, 2004), but more importantly parenthood itself can motivate behavioral change (Stover & Farrell, 2019). A contextually grounded family support intervention strengthening communication and increasing focus on the child may be situated to help alleviate current situational IPV– assuming that careful assessment and safety planning can be carried out.

Besides IPV, another major and common risk attendant to the TTP is heightened depression. Among women, support from partners is inversely correlated with depressive symptomatology (Turner, Grindstaff, & Phillips, 1990), but for coparenting relationship programs, the impact on mothers’ PPD has been equivocal. Moreover, very few studies have attended to trajectories of depressive symptomatology among unmarried fathers (Pinquart & Teubert, 2010). Strengthening coparenting relationships of unmarried Black parents may have special merit; though post-natal depression among inner-city black mothers in enhanced usual care runs as high as 46% (Howell, Balbierz, Wang, Parides, Zlotnick & Leventhal, 2012), the FIOC pilot study of Black coparents’ TTP suggested a palliative effect of programming. In that study, mothers’ self-ratings of depression were significantly lower at 3 months postpartum than they had been prior to the prenatal intervention (Salman-Engin et al., 2015). The current RCT is poised to test for intervention effects of FIOC on both maternal and paternal depression.

Summary and Prospectus

Few coparenting-specific interventions have been designed for unmarried nonresidential coparents or taken account of life circumstances and family adaptations of lower socioeconomic Black coparents. As coparenting is a culturally rooted theme in Black families and communities, a properly centered intervention honoring Black family values and strengths might be well-suited to support coparental and father engagement during infancy. This study reports results of an RCT testing FIOC against a “Treatment as Usual” (TAU) control group. We posited that compared to TAU, parents receiving FIOC would report: 1) greater coparenting support from and teamwork with the other parent, and 2) greater engagement by fathers 3 months postpartum. With respect to secondary outcomes, FIOC parents were expected to report greater reductions than TAU parents in 3) physical and psychological IPV, and 4) depressive symptomatology across the TTP.

Method

Participants

Participants included 276 low-income, unmarried parents (138 mothers and 138 fathers), for whom the child would be their first baby together. The target population was Black, but mixed-race coparents were included if at least one parent self-identified as Black. The participants were recruited from low-income neighborhoods in a community of a Southeastern urban area of the United States. Families were excluded if either participant: 1) demonstrated evidence of psychotic symptoms or suicidal ideation; 2) recently (over the past 12 months) had been arrested and convicted for violence (assault) perpetrated against anyone including a current or former partner; or 3) were deemed high risk or in need of more intensive clinical intervention than would be available through FIOC as determined by the study’s clinical team. See Figure 1 for data summarizing recruitment, enrollment, and retention.

Figure 1.

CONSORT diagram of the FIOC for coparents randomized controlled trial

Sample demographics are detailed in Table 1. All mothers were in the second or third trimester of their pregnancy, and all parents were unmarried. One or both participants in each family were Black (89.9% fathers; 76.8% mothers), with the parent not identifying as Black most commonly White (5.8% fathers; 18.8% mothers) or of mixed-race (4.3% fathers; 4.4% mothers). Intervention and control groups were comparable at baseline on several key demographic variables, including substance use and mental health histories (Table 1). As shown in Table 2, at baseline the intervention and control groups were also comparable on two sets of variables that were examined in longitudinal analyses (IPV, depression), with no significant pre-intervention differences between groups.

Table 1.

Means, SDs, Percentages and Group Differences in Demographics at Baseline

	Intervention (n=140)		Control (n=136)
	Mother (n=70)	Father (n=70)	Mother (n=68)	Father (n=68)
Age M (SD)	24.09 (5.48)	25.91 (7.75)	24.31 (5.97)	26.93 (8.13)
Previous Children	37.1%	35.7%	45.6%	32.4%
Adverse Childhood Experiences (ACE) Score M (SD)	2.91 (2.32)	2.26 (1.98)	3.28 (2.58)	2.82 (2.23)
Drug/Alcohol Misuse History	1.4%	4.3%	5.9%	4.4%
Mental Health History/Encounters	20.0%	14.3%	20.9%	25.0%
	Family (%)		Family (%)
Currently Co-residential	51.4%		58.8%
Prior History of IPV	50.0%		63.2%
Education (Highest Grade/Certification/Degree Completed)
Junior High School	12.9%	15.7%	23.5%	20.6%
High School	47.1%	31.4%	33.8%	41.2%
GED	5.7%	4.3%	2.9%	5.9%
College	10%	11.4%	13.2%	4.4%
Some College	17.1%	28.6%	26.5%	22.1%
Trade School/Vocational School	7.1%	8.6%	-	5.9%
Income (Over the last 12 months)
Under $5,000	24.3%	15.7%	33.8%	17.6%
$5,000 to $14,999	27.1%	35.8%	16.2%	26.5%
$15,000 to $34,999	11.5%	30%	19.1%	25.1%
Greater than $35,000	7.1%	8.6%	7.3%	11.8%
Don’t know/Refused	15.7%	3.1%	13.2%	5.8%
Did no regular work	14.3%	5.7%	10.3%	13.2%

NOTE: There were no significant group differences p < .05

Table 2.

Means, SDs and Group Differences in Outcome Variables at Baseline

Pre-assessment	Intervention		Control
	Mother M(SD), n	Father M(SD), n	Mother M(SD), n	Father M(SD), n	t-score Mother	t-score Father
Psychological Aggression Self	16.64 (24.40), 70	15.74 (25.19), 65	22.22 (23.56), 68	14.37 (19.29), 62	1.36	−.34
Psychological Aggression Partner	14.93 (21.81), 70	17.31 (27.39), 65	19.25 (22.20), 68	15.53 (19.89), 62	1.15	−.41
Physical Assault Self	2.13 (4.86), 70	1.51 (4.98), 65	2.63 (6.80), 68	.43 (.95), 62	.5	−.1.67
Physical Assault Partner	1.36 (3.84), 70	5.00 (20.37), 65	.71 (2.02), 68	2.0 (4.08), 62	−1.24	−1.14
Depression	6.69 (5.20), 70	6.00 (4.89), 70	7.49 (4.28), 68	7.54 (5.04), 68	.98	1.83

NOTE: There were no significant group differences p < .05

Procedure

Families were referred by community agencies (e.g., federal and state Healthy Start and Healthy Families programs, faith-based community programs, WIC), and self-referred after seeing advertisement flyers distributed throughout the community to merchants and service providers (hair salons, barbers) serving Black men and women. Initial inclusion criteria were that 1) at least one parent self-identified as Black; 2) the baby was the parents’ first together; 3) they were not married; and 4) both parents were willing to participate. Interested participants came to an in-person meeting to complete informed consent followed by baseline assessments. Mothers met with a female staff member and fathers with a male staff member, separately and privately. During this assessment IPV screening was conducted using a 12-item IPV Screen and the Revised-Conflict Tactics Scale (CTS2; Straus et al., 1996); the Screen includes items such as: “Have you been afraid the father would hurt you or others during an argument?”; “Has the father been excessively jealous and possessive of you?” If mother disclosed physical IPV, indicated concern the father could hurt her, or felt an intervention might increase IPV, the Danger Assessment Scale (DAS; Campbell et al., 2003) was administered to assess dangerousness.

If certain indicators appeared in mother’s IPV screen/DAS and/or CTS2 at baseline (i.e., a sum of 9 or more yes items on the DAS; reports of highly controlling behavior, threats with a weapon, severe violence like kicking, burning, scalding, attempted strangulation), the IPV screen/CTS2 was not administered to the father and the family was excluded from the study. Assessors completed safety plans with mothers and connected her with local IPV services for advocacy, safety, and other resources, and families received other community resource referrals. As an extra precaution, scores and reports from any parent who disclosed even a single item related to IPV were reviewed by a licensed clinical psychologist with over 15 years’ experience with IPV to determine appropriateness of that couple participating further in the study. Only upon close review for IPV and other study exclusion criteria were eligible families randomized to a TAU group (Control) or TAU with FIOC (Intervention). Urn randomization was used to maximize the likelihood treatment groups were balanced on presence of children from prior unions, presence of physical IPV, and paternal participation in any fatherhood programming.

Families were notified by phone whether they had been selected to continue in the study within 48 hours of baseline and told their group assignment (intervention or control). They were visited by the Resource and Referral Navigator who provided a R&R listing of existing community services, inviting parents to contact her any time aid with referrals to county services for pregnant or parenting families was needed. She also delivered gift cards for completing the intake. For those in the intervention, the assigned male and female mentor team was introduced at this time, and mentors arranged two initial 1-on-1 mentorship sessions with the parents, as per the FIOC protocol. The intervention (Table S1) then commenced. At 1-month post-partum, all families were sent a “Congratulations on the birth” card, with episodic newsletters and reminder texts sent at staged intervals to stay in touch. Families were then seen at 3 months postpartum to assess for coparenting, father engagement, IPV and depression. All study procedures were reviewed and approved by the University of South Florida’s Institutional Review Board.

Intervention Fidelity

Fidelity was monitored continuously throughout the study. A team of four raters monitored two randomly selected intervention sessions per family by listening to audiotapes and reviewing verbatim transcripts. Each rated session was evaluated by two raters. Ratings always fell within one point of each other and in cases of discrepancy, the average score was used. Competence (skill of intervention delivery) was rated from 1 (needs work) to 9 (good work). Mean Competence averaged across female Mentors was 7.48 and across male Mentors, 7.50. Adherence (extent to which a Mentor team conformed to the intervention protocol), was rated from 1 (not accomplished) to 3 (fully accomplished). Mean team Adherence averaged across Mentors was 2.89. Mentors’ fidelity to the curriculum hence met or exceeded established standards. In any unusual case where session objectives were not fully accomplished, feedback was given to the clinical supervisor who discussed the issues in supervision prior to the next session.

Measures

Coparenting.

At 3 months post-partum, both mothers and fathers rated Respect by the coparent and coparenting Communication and Teamwork on a 20-item Parenting Alliance Measure (PAM; Abidin & Konold, 1999). The PAM estimates how cooperative and respectful parents perceive their partner to have been in coparenting their baby. Sample items include “My child’s other parent believes I am a good parent’” (Respect), and “My child’s other parent and I are a good team” (Communication and Teamwork). Statements are endorsed on 5-point Likert scales from strongly disagree to strongly agree, with higher scores indicating a more positive perceived alliance.

Father Engagement.

Also at 3 months post-partum, parents completed Coley and Morris’ (2002) Father Involvement Scale to describe father’s engagement with the child. This six-item scale yields a total score estimating father responsibility (e.g., “How much responsibility do you (does father) take for raising the child?), accessibility (e.g., “How often do you (does father) see or visit with the child”), and engagement (e.g., How many hours per week do you (does father) take care of the child?). Coley and Morris (2002) developed their Scale so that it would be appropriate for use with both residential and nonresidential fathers. Scoring conventions for the three scales followed guidelines of Coley and Morris (2002) and Hernandez & Coley (2007and composite scores for mother and father reports, summing the respondent’s scores on all items, were used. Higher scores signify greater paternal involvement. The composite measure was validated on an ethnically diverse low-income sample of predominantly Latino and Black fathers (Hernandez and Coley, 2007). Cronbach’s alpha for paternal and maternal report was .78 for fathers and .86 for mothers in the current sample.

Intimate Partner Violence.

IPV was assessed prenatally and at 3 months post-partum using the Psychological Aggression and Physical Assault scales of the Revised-Conflict Tactics Scale (CTS2; Straus et al., 1996). The CTS-2 is the most widely used IPV self-report estimating violence between partners over the preceding 12 months. CTS-2 response categories gauge frequency of acts during conflict with a partner in the past year on a Likert scale and include options of “Never in the last year, but it did happen before that,” and “This has never happened.” Acts include twisting a partner’s arm, partner receiving a cut/bruise in fight, pushing/shoving, hitting/holding down, punching, slamming, grabbing, slapping, partner feeling pain, using a weapon, partner needing medical attention, and/or partner having a sprain or broken bone. CTS2 scoring involves adding the midpoints for the response categories chosen by the participant. The midpoints are the same as the response category numbers for Categories 0, 1, and 2. For Category 3 (3–5 times) the midpoint is 4, for Category 4 (6–10 times) it is 8, for Category 5 (11–20 times) it is 15, and for Category 6 (More than 20 times in the past year) it is 25.

The CTS2 shows good validity and internal consistency. In this sample, internal consistencies of physical and psychological IPV scales were similarly satisfactory: Baseline-Psychological aggression .76 (mothers), and .77 (fathers); Physical aggression .83 (mothers), and .87 (fathers); 3 months postpartum - Psychological aggression .74 (mothers), and .73 (fathers); Physical aggression .90 (mothers), and .81 (fathers).

Depression.

Mothers and fathers reported depression prenatally and at 3 months postpartum on the Edinburgh Depression Scale (Cox et al., 1987), a 10-item measure of common depression symptoms. Each statement is scored from 0 (never) to 3 (always), yielding a total range of 0–30. Reference is to the past week (e.g., “I have been anxious or worried for no good reason”; “I have been so unhappy that I have had difficulty sleeping”). 13 is the cut off score indicating major depression (Cox et al., 1987). The EDS correlates significantly with the Beck Depression Inventory (Pop et al., 1992). In this sample, the EDS had adequate internal consistency for both mothers and fathers (mother baseline α = .83; father baseline α = .78; mother 3-month postpartum α = .89; father 3-month postpartum α = .81).

Results

All tests of intervention effects were conducted as intent-to-treat analyses; data from all parents who completed the follow-up were included regardless of level of program participation. For parallel behaviors by mothers and fathers, analyses were conducted as multilevel regression (linear mixed effects) using R and the lmer package with custom scripts. This accounted for within-family dependency while yielding separate estimates for mothers and fathers. Each model includes all possible main effects and interactions of predictors. Analyses for each dependent variable (DV) and report of each given parent on that DV draw on the full set of data in which the mother or father completed the pertinent measure in each within-subject (timepoint) condition. Given constrictions on DV completion, some variation occurred in sample size for different analyses and DVs; we hence report sample sizes for each condition of each analysis rather than restrict results only to “complete” families in which mother and father completed all DVs at every timepoint. To streamline presentation, we report results and sample sizes separately, by each DV and associated research question.

Coparenting

For the coparenting outcomes, analyses involved data from 54 mothers and 48 fathers (control), and 42 mothers and 40 fathers (intervention). PAM Communication subscale means for mothers were m_c = 53.41 and m_I = 68.41 for control and intervention, respectively, and for fathers m_c = 53.43 and m_I = 68.87. A multilevel regression model incorporating both parents and including random intercepts for families detected a significant intervention group effect β = 15.44, t(139.13) = 2.68, p < .01.

For PAM Respect, subscale means for mothers were m_c = 10.13 and m_I = 12.32 for control and intervention, respectively, and for fathers m_c = 9.28 and m_I = 12.22. The multilevel model detected a marginal main effect of parent, β = .85, t(112) = 1.88, p = .06 and again, a significant intervention group effect, β = 2.94, t(141.25) = 2.82, p < .01. No interactions reached significance.

Father Involvement

For father involvement, data from 54 mothers and 48 fathers (control), and 42 mothers and 40 fathers (intervention) were analyzed. Analyses indicated that fathers in the intervention group were reported to be more highly engaged with infants than were fathers in the control group. Means for mothers’ reports of father involvement were m_c = 18.15 and m_I = 20.86 for control and intervention, respectively, and for fathers’ reports of their own involvement they were m_c = 21.25 and m_I = 22.05. A multilevel regression model incorporating both parents and including random intercepts for families detected a significant main effect for condition, β = 2.52, t(130.33) = 3.37, p < .001 as well as for parent, β = 3.05, t(111.71) = 4.32, p < .001. The interaction between these two variables fell short of significance, β = −1.89, t(110.28) = −1.80, p = .07, but a trend suggested that it was mothers in the intervention condition who discerned greater father involvement at 3 months post-partum than did mothers in the control condition.¹

Intimate Partner Violence

For IPV analyses, the same number of mothers (54 control, 42 intervention) contributed data, but 5 fewer fathers per group (43 control, 35 intervention) contributed data. This resulted because fathers were not administered IPV measures either at baseline or, later, at 3 months if mothers endorsed one or more critical IPV items (highlighted earlier in the Method section) during their own baseline assessment. Further, because families raising graver concerns about more severe forms of IPV did not continue in the study at all following the intake, variability in Physical Assault scores for the study sample was necessarily constrained. Psychological Aggression and Physical Assault were hence examined separately to assess whether the intervention had differential impact on the two forms of IPV.

For Psychological Aggression, relevant subscale means at time 1 for mothers’ reports on fathers were m_c = 19.25 and m_I = 15.30 for control and intervention, respectively. Subscale means for Psychological Aggression at time 2 were m_c = 17.22 and m_I = 11.12. For fathers’ reports on mothers the relevant means at time 1 were m_c = 15.53 and m_I = 17.67; at time 2 they were m_c = 15.16 and m_I = 6.67². A multilevel model of parent reports on the Psychological Aggression subscale revealed no main effects of parent, condition, or timepoint. A significant interaction between condition and timepoint was detected: β = −12.18, t(289.47) = −2.32, p < .05 such that a significant decline in Psychological Aggression was seen at 3 months for intervention but not control group parents. No other main effects or interactions reached significance.

In the IPV analyses, the DV of principal interest is not self-report, but other-report, and it varies with parent (father’s reports on the mother; mother’s reports on the father). There are hence arguably two DVs, one to each coparent, and it may technically not be accurate to combine both in a multilevel model. For this reason, we also examined and analyzed fathers’ and mothers’ data separately. A 2 (Condition, Between-S) x 2 (Timepoint, Within-S) mixed ANOVA with Greenhouse-Geisser correction was carried out once with only the fathers’ reports, and once with only the mothers’ reports. For fathers’ reports of mothers’ Psychological Aggression, this analysis detected a main effect of timepoint, F(1,72) = 6.724, p < .05, η²_G = .02, and an interaction between condition and timepoint, F(1,72) = 5.901, p < .05, η²_G = .02. Parallel effects were not significant for mothers’ reports on fathers’ Psychological Aggression. Results plotted in Supplemental Figure S1 underscore that intervention fathers reported being recipients of less Psychological Aggression from mothers over time, with no comparable change reported by control fathers. Plots of mother data did not reflect a comparable time by condition effect.

Results of this secondary analyses imply that it was intervention group fathers who reported the greatest cross-time declines in their coparent’s Psychological Aggression; the interaction effect reported for the multilevel model hence seemingly appeared driven by the clear pattern in the father report data. However, that effect apparently did not differ enough in magnitude from a qualitatively similar, but far smaller, interaction seen in mother’s reports (Figure S1) to produce a significant 3-way interaction in the full multi-level model, which combined correlated and heterogenous variances across the parents.

Last, we note that analyses completed for the Physical Assault subscale uncovered no effects or interactions approaching significance for any analysis involving this variable (all p > .05). It is important nonetheless to highlight that data did not show that engagement in a dyadic coparenting intervention heightened parents’ risk for increases in post-natal physical assault.

Depression

For father involvement, data from 54 mothers and 48 fathers (control), and 42 mothers and 40 fathers (intervention) were analyzed. Mothers’ and fathers’ reports of depression were examined in a single multilevel model, with random intercepts for families and fixed effects of parent, timepoint, and condition and their interactions. As shown in Figure S2, only a main effect of timepoint was obtained β = −2.41, t(313.61) = 2.41, p < .05. Data revealed that significant declines in depressive symptomatology were seen for both mothers and fathers in both groups.

Discussion

This study’s aims were to determine whether a novel prenatal intervention developed to support unmarried, lower income Black parents transitioning to new parenthood could stimulate greater coparenting support and teamwork and greater engagement by fathers with their infant children than seen among a comparable control group of TAU parents. The study also tested whether the coparenting intervention would exert related beneficial effects across the TTP on secondary adjustment outcomes including reductions in IPV and depressive symptomatology, compared with control parents. Intent-to-treat analyses documented several beneficial effects of the intervention. First, intervention group parents reported greater coparenting communication and teamwork with, and greater coparenting respect from, their coparenting partner than did parents in the control group. Relatedly, with respect to father engagement, parents in the FIOC condition reported greater father involvement at 3 months post-partum than did those in the control condition. Secondary analyses suggested these findings were carried especially clearly by mother-reported data, of note because mothers reliably report lower levels of involvement than do fathers (Coley & Morris, 2002). With all the challenges fathers in unmarried families face in trying to develop relationships with their infant children, it seems plausible that FIOC may have helped parents cultivate a better relationship platform from which to navigate early paternal involvement. Data hence suggest that a targeted intervention designed to enhance child-related coparental communication, collaboration and problem-solving may also help engender greater paternal engagement with infants during the early postpartum months.

An important finding from this RCT was that participation in the dyadic coparenting intervention did not elevate risk for IPV. This is notable, as 50% of intervention group parents disclosed at least one physically aggressive exchange over the past year at baseline on the CTS2. While some in the field maintain dyadic work is dangerous, raising risk if there is IPV, this was not found in our study. Some studies find physical IPV can emerge or become more severe in the postpartum, but we saw no such patterns, perhaps because both groups were afforded access to resources and referrals. Furthermore, FIOC had a salutary effect on psychological aggression. There was no parent-by-condition-by time interaction, but closer examination of scores prompted by data skew revealed particularly clear impact in father reports of mothers’ psychological aggression. Most IPV literature has rightfully addressed father-to-mother aggression, but mother-to-father psychological aggression is also an important within-family process. As any aggression within families can disrupt emotion regulation of all family members, declines seen over the early months of the baby’s life in intervention group families are significant.

Finally, data indicated that depression declined significantly not just for mothers but for fathers in both groups. In the pilot study of FIOC, improvements in depression across the TTP were likewise found for mothers at 3 months postpartum, trending positively for fathers (Salman-Engin et al., 2016). This finding is of particular consequence since when compared to White women, Black women are more likely to experience postpartum depression (Segre et al. 2006), with lack of social support related to depression symptom severity (Logsdon et al. 2000). It is conceivable that FIOC’s affording of resources and referrals to assist with basic needs and connect with local services may have offered some help to parents in both conditions.

Overall, findings were consistent with projected benefits of FIOC with more favorable views of coparenting, greater father engagement and reductions in psychological aggression differentially seen in the intervention group. The finding that parents in both groups reported significantly less depression in the postpartum than they had at intake provides additional intriguing suggestions that perhaps something about the supportive atmosphere of the project initiative itself, including universal access to resources and referrals, may have conferred some benefit. In all cases, effect sizes were modest, and so the patterning of results should be viewed principally as having provided some promising leads for future research and programming.

Because initiatives such as FIOC have been rare, they harken important questions about interest among parents or fit within the community served. The opportunity to receive assistance with needed community referrals and to participate in a coparenting enhancement program was a sufficiently attractive draw for 157 of 277 initially eligible families who could be successfully reached (57%), with both father and mother consenting to join the study. All 157 understood they were agreeing to participate in a year-long initiative in which they might be invited to take part in the multi-week dyadic intervention with their baby’s other parent. Among the remaining 120 cases who were successfully reached but chose not to enroll, most (88) involved mothers who declined to take part in screens after being contacted by a recruitment staffer. For these cases, the palatability of the intervention for mother and father together could not be adequately assessed since father was never contacted. For the remaining 32 families where mother and father were both made aware of the opportunity to take part in an initiative offering a dyadic intervention, one or both members of the couple chose not to enroll after some deliberation.

From this vantage, a prenatal dyadic intervention that offered to help unmarried parents figure things out together for their shared child appealed at least initially to lower socioeconomic Black fathers. It seems plausible to infer they, like their baby’s mother, desired to pursue a coparenting alliance after the baby came. Upon learning they were chosen for the intervention, 83% elected to attend FIOC sessions. Finally, of coparents who came to a first session, 87% completed at least 4 of the 6 prenatal sessions. These data offer useful preliminary estimates of the palatability of a culturally grounded, coparenting-specific program when offered to unmarried Black fathers and mothers expecting a first child together.

Limitations

There are several limitations worth mention. First, an intensive intake, probing about multiple realms of adversity and conflict before sound rapport could be built with parents may have inhibited reporting. Despite spread seen in IPV and depression scores, it is conceivable IPV and depression may have been under-reported. Relatedly, the ability to detect intervention effects on IPV was curtailed by the relatively low number of randomized cases in which significant aggression was reported and the skew that characterized the data. In part, this skew was a result of the screening and intake process, which weeded out cases where there were substantial concerns about safety of the fetus and mother. Still, CTS data did reveal variability in psychological aggression, and improvements were found for this variable. The small overall sample size too may have limited power to detect differences.

Conclusion

In summary, study data suggest promise of delivering a community-based Focused Coparenting Consultation intervention for unmarried expectant Black parents focused on building support and teamwork and “figuring things out for the child”. We believe there value in continuing to investigate dyadically delivered coparenting-only intervention options at the TTP for Black parents-to-be in efforts to improve coparenting and early father involvement.