Coparenting and the Transition to Parenthood: A Framework for Prevention

Abstract

The way that parents work together in their roles as parents, the coparenting relationship, has been linked to parental adjustment, parenting, and child outcomes. The coparenting relationship offers a potentially modifiable, circumscribed risk factor that could be targeted in family-focused prevention. This paper briefly outlines an integrated and comprehensive view of coparenting, and suggests that the time around the birth of the first child is an opportune moment for coparenting intervention. To support the development of such prevention programs, an outline of the possible goals of coparenting intervention is presented with a description of the processes by which enhanced coparenting may have effects in each area. The paper discusses several issues involved in developing and disseminating effective coparenting interventions.

A basic axiom of prevention science is that the development of empirically-based prevention programs depends on the identification of modifiable risk and protective mechanisms. However, while an abundance of research has been carried out on risk mechanisms in the family, the number of effective, widely disseminated family-focused preventions has been limited (Tolan, Quintana, & Gorman-Smith, 1998). In fact, of the four family-focused prevention programs among the original 10 programs designated as Blueprints for Violence Prevention, three are actually closer to the treatment/therapy end of the intervention continuum than the prevention end. Nurse Home Visiting was the only family-focused program among the empirically-based Blueprints that is a true prevention program. However, Nurse Home Visiting and most other prenatal and early childhood interventions focus on mothers and babies—despite the fact researchers have known for 20 years that the relationship between the parents is strongly associated with parenting and child outcomes (Emery, 1982).

The overarching idea of this paper is that recent empirical work on the coparenting relationship supports an empirical and conceptual framework for potentially important advances in family-focused prevention. The coparenting relationship provides the conceptual framework for integrating two prominent areas of family-focused prevention—parenting interventions and couple relationship programs. Although there have been striking advances in the basic science underlying these two areas of study, the interventions developed for these subsystems have not been well integrated (Sanders, Nicholson, & Floyd, 1997).

Coparenting refers to the ways that parents work together in their roles as parents. This paper proposes that there are four basic components of coparenting: support versus undermining in the coparental role; differences on childrearing issues and values; division of parental labor; and management of family interactions, including exposure of children to interparental conflict. This conceptual definition, based on empirical work and previous theory, is a framework that requires further discussion, validation, and refinement. However, the framework serves to clarify the distinction made by a number of researchers between how parents share and coordinate parenting responsibilities and the rest of their instrumental and expressive relationship components (Frank, Jacobson, & Hole, 1986; Feinberg, Reiss, Hetherington, & Plomin, 2001; Gable, Belsky, & Crnic, 1995; McHale, Kuersten-Hogan, Lauretti, & Rasmussen, 2000). Although the quality of coparenting and the overall couple relationship are probably linked through reciprocal relationships (Maccoby, Depner, & Mnookin, 1990; Margolin, Gordis, & John, 2001; McHale et al., 2000), the linkage is only partial. Thus, coparenting distress is not equivalent to couple relationship distress, nor is coparenting positivity equivalent to couple intimacy. Further, coparenting and overall couple relationship dynamics have differential associations with outcomes of interest.

WHY FOCUS ON COPARENTING?

A number of studies have now demonstrated that the coparenting relationship is linked to parenting and child outcomes. A recent report, for example, indicates that supportive and undermining coparenting relations when a child is 3 years old predicts child externalizing behavior at 4 years (Schoppe, Mangelsdorf, & Frosch, 2001). Although such links are important, there are more specific reasons to focus on coparenting as a means of integrating research on couple relations and parent–child relations.

First, in practical terms, the coparenting relationship offers a more circumscribed and potentially modifiable target for intervention than the overall couple relationship. Although programs have been successful in altering aspects of the couple relationship, an intervention focused on the more limited coparenting relationship may have stronger effects.

Second, research indicates that the coparenting relationship is more powerfully and proximally related to parenting than and other aspects of the couple relationship. When the general couple relationship and coparenting are compared in the same study, coparenting often is found to be of greater significance. For married couples, Abidin and Brunner (1995) found that the parenting alliance, not marital adjustment, is significantly associated with parenting style. Bearss and Eyberg (1998) reported that the parenting alliance had a stronger relationship with child problems than did marital adjustment. Our analyses with data from the nondivorced subsample of the Nonshared Environment and Adolescent Development project has supported the findings of both of these studies (Feinberg, Neiderhiser, Reiss, Hetherington, & Simmens, 2000). Similar findings about the importance of the coparenting relationship have been obtained for divorced parents (see review in Whiteside & Becker, 2000; also see Camara & Resnick, 1989; Ihinger-Tallman, Pasley, & Beuhler, 1995).

This research suggests that there is domain specificity in the effects of coparenting relations on parenting and child outcomes. Further evidence of domain specificity was recently reported by Frosch, Mangelsdorf, and McHale (2000): observed inter-parental conflict during family play with a 6-month old infant, but not dyadic marital interaction conflict, predicted attachment security at 3 years. Such domain specificity may partly justify Margolin’s view (Margolin et al., 2001) that coparenting represents a risk mechanism, while general marital conflict or marital quality may represent a risk indicator (Rutter, 1994). In developing prevention programs, it is obviously much more important to target the risk mechanism rather than a factor that may merely be a marker of risk.

The view of coparenting as a risk mechanism is supported by evidence of coparenting as a mediator of the relationship between the couple relationship and parenting both cross-sectionally and longitudinally (Feinberg et al., 2000; Floyd, Gilliom, & Costigan, 1998; Gonzales, Pitts, Hill, & Roosa, 2000; also see Belsky & Hsieh, 1998). In a recent study, coparenting mediated the relationship between marital conflict and parenting (i.e., the association between marital conflict and parenting became nonsignificant when coparenting was added to the model; Margolin et al., 2001).

OUTLINE AND SCOPE OF THIS PAPER

The perspective offered here is that the coparenting relationship is an important and potentially modifiable influence on parenting and child outcomes, as well as a mediator of other influences (e.g., individual parent characteristics, work pressure, marital conflict). However, the outline of a cohesive, integrated conceptual model of coparenting that could guide intervention development has not been available until recently (Feinberg, 2001). Thus, the first goal of this paper is to outline a model of coparenting as four interrelated components. The integration and organization of the emerging literature on coparenting during the last decade into these four components holds promise for guiding the design of intervention. It should be noted, however, that this model is proposed based on previous research; further research is required to validate and undoubtedly modify and refine the model.

An understanding of the differential importance and trajectories of these four components during the family life cycle would contribute to intervention design. However, research on this topic has not been conducted to date in a comprehensive, longitudinal manner. Nonetheless, based on existing evidence and theory, the second aim of this paper is to propose a framework for intervening in a particular developmental phase that holds great promise. Research suggests that the transition to parenthood is perhaps the best time to intervene to enhance the coparenting relationship because of expectant and new parents’ particular openness to change (Duvall, 1977; Elliott, 2000; Pryce, Martin, & Skuse, 1995). Researchers have frequently noted the strains of the transition period for the family and frequently counseled the development of preventive interventions (Antonucci & Mikus, 1988; Belsky & Pensky, 1988; Coiro & Emery, 1998; Frosch et al., 2000; Grossman, Eichler, & Winickoff, 1980; NICHD Early Child Care Research Network, 2000), yet few have been developed. Couples entering the family formation stage require guidance, support, and skills designed specifically for the transformations and stresses of this period. Based on a brief review of the transition to parenthood period, this paper articulates a series of goals that a coparenting-targeted intervention can address during this period, as well as the processes through which enhanced coparenting may affect outcomes. Although the focus is on the transition to parenthood period, I also suggest a number of ways that an understanding of coparenting may enhance interventions during other family developmental periods.

Before describing the four-component coparenting model, it is prudent to address three issues that may have already occurred to the reader. First, the term coparenting may require clarification. Some writers promote an “equal” division of parenting work (Deutsch, 1999). Indeed, the division of labor is one of four aspects of coparenting in the model presented below. However, the perspective taken here is that high-quality coparenting involves the construction and operation of a mutually agreed upon partnership structure. Individual parents’ satisfaction with the fairness and acceptability of the division of parenting work, as well as the process by which a mutual understanding is reached, are more important than whether parents attain an objectively equal division of parenting labor or any other prescribed coparenting arrangement.

Second, the concept of coparenting entered family research through research into divorced families (Camara & Resnick, 1989; Maccoby et al., 1990). In that literature, it was found that marital and ex-marital conflict were relatively important influences on children compared with the experience of divorce itself (Amato & Keith, 1991). The importance of coparenting has subsequently found to be present in intact families as well (Belsky, Crnic, & Gable, 1995; Belsky, Putnam, & Crnic, 1996; Brody, Flor, & Neubaum, 1998; Cowan & Cowan, 2000; Gable et al., 1995; Margolin et al., 2001). The focus in this paper is on coparenting among intact families, although portions of the model may be applicable to divorced families.

Third, on a related note, the reader may be justly concerned about the generalizability of a coparenting model to the diversity of family structures. The co-parenting model presented here is based largely on available research that has frequently been limited to the traditional heterosexual, two-parent family structure. However, the coparenting relationship is considered to be generalizable to any relationship between two parental figures who share parenting responsibility. Thus, extending the framework to the mother–grandmother relationship may be useful in designing interventions with young, single, and/or unmarried mothers. Comparative research involving coparenting among diverse family types is an important direction for further work.

However, despite the importance of examining how the coparenting framework can be extended to other family configurations, it is also important to note that at least during the transition to parenthood, the current model is directly generalizable to a vast majority of U.S. families. Married couples represent the family constellation of two thirds of U.S. newborn infants (Ventura, Anderson, Martin, & Smith, 1997). Of the one third of all U.S. births in which parents are not married, parents cohabit in almost half (Carlson & McLanahan, 2001). Thus, the model is directly generalizable to almost 85% of U.S. families of new infants. In a portion of the remaining 15% of families, a boyfriend with some initial degree of commitment to the infant is present. One important factor influencing the involvement of such fathers in the rearing of the child over time is the tenor of the couple relationship, perhaps particularly the coparenting relationship (Doherty, Kouneski, & Erickson, 1998).

COPARENTING COMPONENTS AND MODEL

This section presents a model of four interrelated components of coparenting, which is drawn from several sources (e.g., Belsky, Putnam, et al., 1996; Brody & Flor, 1996; Brody et al., 1998; Cowan & Cowan, 2000; Ihinger-Tallman et al., 1995; Margolin et al., 2001; McHale, 1995). This is not meant to be an exhaustive description of coparenting, but a framework for organizing theory, research, and intervention. Further research and theoretical development of the concept will undoubtedly result in modifications (see suggestions for future research in Feinberg, 2001).

The first component of coparenting is support versus undermining in the parental role. This component of coparenting relates to each parent’s supportiveness of the other: affirmation of the other’s competency as a parent, acknowledging and respecting the other’s contributions, and upholding the other’s parenting decisions and authority (Belsky, Woodworth, & Crnic, 1996; McHale, 1995; Weissman & Cohen, 1985). The degree of support versus undermining of the other parent in the parental role has been associated with both parenting and child outcomes. For example, positive perception of the coparenting relationship has been linked to perceived parental competence and child and adolescent behavior problems (Floyd & Zmich, 1991); authoritative parenting and lower parenting stress (Abidin & Brunner, 1995); and, independently of variance accounted for by parenting, preschool boys’ inhibition (Belsky, Putnam, et al., 1996).

Childrearing disagreement, the second coparenting component, involves differences of opinion over a range of child-related topics, including moral values, discipline, educational standards and priorities, safety, peer associations, and so on. The extent of childrearing disagreement has been linked to child problems in the preschool period (Deal, Halverson, & Wampler, 1989), child behavior problems at 5 years (Block, Block, & Morrison, 1981), and psychological and behavioral adjustment at 18 (Vaughn, Block, & Block, 1988). Childrearing disagreement is a better predictor of behavior problems for young children than general marital conflict and adjustment (Jouriles, Murphy, Farris, Smith, et al., 1991; Snyder, Klein, Gdowski, Faulstich, et al., 1988; but see Ingoldsby, Shaw, Owens, & Winslow, 1999).

Childrearing disagreement per se may not be problematic. For example, parents who “agree to disagree” are able to maintain high levels of mutual coparenting support. These parents who actively and respectfully negotiate resolutions of disagreements may not experience detrimental effects from childrearing disagreement. Instead, childrearing disagreement may be important to the extent that it negatively affects other components of coparenting or family life. Issues of childrearing disagreement may play a particularly salient role in affecting consistency of discipline practices across parents.

The third component of coparenting relates to the division of duties, tasks, and responsibilities pertaining to daily routines, child care, and household tasks; financial, legal, and medical issues; and other child-related duties. Mothers report that the issue of household chores is the single most important trigger of conflict in the postpartum period (Cowan & Cowan, 1988a). Indeed, mothers’ perception of fairness in fathers’ contributions is linked to increased marital quality over the transition to parenthood, while perception of inequity is linked to decreased marital quality (Terry, McHugh, & Noller, 1991). The importance of mothers’ perceptions in this domain appear to be crucial, probably because mothers generally perform the majority of household tasks and take on ultimate responsibility for almost all child-related issues (Aldous, Mulligan, & Bjarnason, 1998; Demo, Acock, & Hurlbert, 1993; Hetherington et al., 1999; Lamb, 1995; Peterson & Gerson, 1992).

The fourth coparenting component, parents’ management of interactional patterns in the family, is comprised of three aspects: conflict, coalitions, and balance. The first aspect, interparental conflict, has been repeatedly linked to children’s externalizing behaviors (Buehler et al., 1998; Emery, 1982; Johnson & O’Leary, 1987; Jouriles, Bourg, & Farris, 1991; Rutter, 1994), but also to internalizing disorders and other problems (Holden & Ritchie, 1991; Jouriles, Barling, & O’Leary, 1987; Jouriles, Murphy, & O’Leary, 1989). The influence of interparental conflict on children begins early: Children as young as one year exhibit distress to parental hostility (Cummings, Zahn-Waxler, & Radke-Yarrow, 1981; Grych & Fincham, 1990).

Not all couple conflict represents a coparenting process. Nonchild related conflict that occurs “behind closed doors” does not fall into the coparenting domain. Exposure of children to conflict—especially frequent, unresolved, and/or physical conflict (Grych & Fincham, 1990)—is the central issue in terms of how parents jointly manage couple conflict. Owen and Cox (1997) report that prenatal and 3-month postpartum marital conflict is linked to disorganized infant attachment patterns at 1 year, independently of both parental ego development and warm, sensitive parenting.

Although the multiple mechanisms linking couple conflict with child adjustment are probably complex (Frosch et al., 2000; Goldberg & Easterbrooks, 1984), Cummings and associates have found evidence for a direct effect by documenting the immediate reaction of children to simulated adult anger (Cummings, 1987; Cummings, Ballard, & El-Sheikh, 1991; Cummings, Ballard, El-Sheikh, & Lake, 1991; Davies, Myers, & Cummings, 1996). Further, children whose parents reported a history of physical aggression in the marriage demonstrated increased levels of behavioral and emotional dysregulation during exposure to marital conflict (Cummings, Pellegrini, Notarius, & Cummings, 1989; also see Davies, Myers, Cummings, & Heindel, 1999). Thus children may become sensitized to conflict, and conflict may disrupt development of emotional or self-regulation and family-level emotional security.

Care should be taken in translating such findings to intervention. A distinction should be made between the active coping strategy of avoiding hostile conflict in the presence of children and the general withdrawal from interaction. Withdrawal in the couple relationship (which may represent an advanced state of relationship deterioration), has been associated with negative outcomes for children and the couple relationship (Cox, Paley, Payne, & Burchinal, 1999; Katz & Gottman, 1993). Further, not all conflict is harmful; constructive management of conflict seems to be beneficial or at least not detrimental (Cummings & Wilson, 1999; Easterbrooks, Cummings, & Emde, 1994). It is also important to note that resolution of conflict mitigates the negative effects of conflict (Cummings, Vogel, et al., 1989; Cummings, Ballard, El-Sheikh, & Lake, 1991; Cummings, Simpson, & Wilson, 1993).

A second aspect of parents’ role in managing family interactions relates to the presence of a unified coparental coalition versus triangulation of the child in an overt or covert parent–child coalition (Ihinger-Tallman et al., 1995; McHale, 1997; Minuchin, 1985). Even mildly dissatisfied couples tend to triangulate children (Lindahl, Clements, & Markman, 1998). Several family researchers have described the negative effects of triangulation and scapegoating processes within families (Christensen & Margolin, 1988; Kerig, 1995; Lindahl et al., 1998; Maccoby, Buchanan, Mnookin, & Dornbusch, 1993; Minuchin, Rosman, & Baker, 1978; Vuchinich, Emery, & Cassidy, 1988). Triangulation of children is linked to martial dissatisfaction (Margolin et al., 2001). Retrospective research demonstrated that a father–daughter alliance during childhood predicted depression, anxiety, and low self-esteem in young adult women even after controlling for current father–daughter alliance (Jacobvitz & Bush, 1996). Issues of triangulation and alliances may also involve other members of the family—for example, the meaning of certain parent–child relationships may depend on parent–sibling relationships (Feinberg et al., 2000; Feinberg, Reiss, & Hetherington, 2001) or the role played by a grandmother in a family system. Nevertheless, the central issue here is how the coparents jointly manage such family relations.

The final aspect of family interaction is balance between parents in interactions with the child. The issue of interactional balance concerns in part the relative proportion of time each parent engages with the child in triadic situations. Discrepant levels of parental involvement in triadic play predict later teacher rated anxiety, while parental hostility and competitiveness in such interactions predict more aggression (McHale & Rasmussen, 1998). Further, another study found that parenting discrepancy (triadic balance) is independent of husband or wife reports of marital quality (McHale, 1995).

An Inclusive Model

It was mentioned above that coparenting is a potential mediator of a number of influences on parenting and child outcomes. In fact, the proximal relationship of coparenting to both the couple relationship and parenting places coparenting in the center of a complicated web of associations among individual, family, and contextual influences.

Individual characteristics of both parents and children are likely influences on the coparenting relationship. Individual parent characteristics, ranging from cognitions (e.g., parental beliefs) to more affective features (e.g., depression, hostility) affect parents’ ability to cooperate in childrearing and family management (Belsky & Hsieh, 1998). It is also likely that child temperament may affect the coparenting relationship. For example, when parents are experiencing difficulty managing a child’s behavior, the consequent stress, frustration, and feelings of failure may lead each parent to blame the other’s parenting for the child’s behavior. There is evidence of the influence of child problems on the couple relationship (Mash, 1984), however a more precise investigation would likely find that child temperament and behavior are more closely linked to the coparenting relationship (due to domain specificity).

Another individual characteristic, stress, is a likely entry point for the influence of environmental, contextual factors on coparenting. In this view, stress—whether derived from pressure at work or financial difficulty, extended family relations, or other sources—typically undermines individuals’ functioning, leading to less ability to tolerate frustration and more negative interpersonal relationships (Atkinson et al., 2000; Garmezy, Masten, & Tellegen, 1984; Matheny, Aycock, Pugh, Curlette, et al., 1986). Thus, high levels of environmentally-influenced stress will tend to result in less supportive, more conflictual co-parenting. In one investigation of the relationship of stress to coparenting, stress accentuated the relationship between spousal differences and negative coparenting interactions (Belsky et al., 1995). However, environmental or contextual variables may also serve as protective factors. Extra-familial social support may be a general protective factor (Johnson & Sarason, 1978) facilitating the coping of families experiencing stress and enhancing the coparenting relationship.

To the extent that coparenting is influenced by individual, family, and environmental characteristics and in turn influences relevant outcomes, coparenting can be viewed as a mediator. This view of coparenting as a mediator is relevant to intervention, as a general prevention strategy is to target modifiable mediators of outcomes. However, an understanding of coparenting as a protective factor that moderates risk-outcome relationships is also relevant to prevention. For example, Floyd et al. (1998) suggest that positive coparenting may protect parenting quality and child adjustment from the negative effects of depression in one parent. Others have suggested that a strong coparenting relationship may diminish the effects of couple conflict on children (Abidin & Brunner, 1995; McHale, 1995): high levels of hostile couple conflict may be detrimental only when coparenting quality is low, not when it is high.

THE TRANSITION AND POTENTIAL FOR COPARENTING INTERVENTION

The transition to parenthood holds great potential for coparenting intervention for at least two reasons: First, new parents are relatively open to education and support as they create an “emergent family system” (Cowan & Cowan, 2000). Parents’ openness at this stage may in part be due to the strains and stresses of the period. Although there is a wide range of experience across the transition to parenthood, research supports the view that new parents experience a high level of stress and strain (Antonucci & Mikus, 1988; see review in Sanders et al., 1997), which continues beyond the initial months post-partum (Shapiro, Gottman, & Carrere, 2000). The Cowans have argued persuasively that the stresses and vulnerability of even “low risk” couples have been underestimated: Parents who are married, have fairly good relationships, and are well off in socioeconomic terms, experience difficult strains as they enter parenthood and create an “emergent family system” (Cowan & Cowan, 2000; also see Belsky & Pensky, 1988).

A second reason why a coparenting intervention holds promise at the transition to parenthood relates to the emergence of the coparenting relationship itself. The coparenting relationship may be more fluid and malleable during its emergence and early development than later when the coparenting patterns are more canalized and buttressed by years of history. One direction for future research that would therefore inform intervention development is a close understanding of how parents create and shape the early coparenting relationship (Cox, Paley, Burchinal, & Payne, 1999).

One reason that there has been limited research on the early development of the coparenting relationship has been a primary focus on mothers in basic and applied research. For example, most existing interventions during the family formation stage focus on mothers and babies. A growing interest in the role of fathers has developed over the last decade as research highlighted the influence of fathers’ parenting on children—especially on children’s externalizing behavior (DeKlyen, Speltz, & Greenberg, 1998; Katz & Gottman, 1994; Parke, Cassidy, Burks, Carson, & Boyum, 1992; Stocker & Youngblade, 1999). More recently, researchers have also discovered that fatherhood may have positive effects on the life trajectories of individual fathers (Eggebeen & Knoester, 2001). As researchers move to incorporate fathers into early family interventions, an understanding of the coparenting relationship during the transition to parenthood will be essential.

Much of our best data on the family formation period comes from comparison of the experiences of parents and nonparent couples, although there is some question about the validity of this comparison (Belsky & Pensky, 1988). When the comparison is made, the evidence sometimes suggests that marital deterioration is similar across the groups (MacDermid, Huston, & McHale, 1990; McHale & Huston, 1985; White & Booth, 1985), although there is also evidence that deterioration is more precipitous and pervasive for parents (Cowan & Cowan, 1988b; Kurdek, 1993; Lindahl, Malik, & Bradbury, 1997). Importantly, research suggests that the transition affects families in broadly different ways (Belsky & Hsieh, 1998). In many families, these normative stresses evoke individual, couple, and social network coping responses that allow the family to move through this transitional stressful phase with renewed strength and maturity (see Prancer, Pratt, Hunsberger, & Gallant, 2000). Other families weather this stressful period suffering transient distress, without significant long-term effects (either positive or negative) on adjustment and relationship quality. However, a substantial number of families apparently do not cope successfully with these developmentally normative stresses, and the distress experienced during this stage leads to significant individual and family difficulties (Cowan, Cowan, Heming, & Miller, 1991; Gloger-Tippelt & Huerkamp, 1998; NICHD Early Child Care Research Network, 1999).

Shifts during the postpartum transition period include changes in the household division of labor, extra-familial roles, time for the couple to be together, and sex (Cowan & Cowan, 1995). Differences between men and women’s experiences and roles are accentuated during the transition period, intensifying traditional gender role differences (Salmela-Aro, Nurmi, Saisto, & Halmesmaki, 2000; similar findings have been reported in a sample of Turkish couples: Hortacsu, 1999). Late in pregnancy, the division of labor is least sex-typed of any period (see Belsky & Pensky, 1988). However, as the roles of “father” and “mother” become larger in parents’ psychological space, the role of “husband/wife/lover” diminishes for both parents (Cowan & Cowan, 2000). Women work less after pregnancy, and their identity as “worker” and “student” diminishes, but men work more (Cowan & Cowan, 1992). Changes in men’s and women’s activities outside the home appear to impact the division of labor inside as power inside family negotiations accrues partly based on economic earning productivity (Ishii-Kuntz & Coltrane, 1992). The absolute burden on women of domestic work, even apart from child care, increases after the first child. These changes lead to increased feelings of distance, dissatisfaction with “who does what?,” and spousal conflict (Cowan & Cowan, 1992, 2000).

Expectations

Parents’ expectations regarding these issues appear to play a key role in couple adaptation (Cowan, 1988). For example, couples generally do not anticipate the traditionalization of roles (Belsky, Ward, & Rovine, 1986; Cowan et al., 1985), and violation of expectations appears to play a role in the link between marital dissatisfaction and traditionalization (Kalmuss, 1992; Kerig, Cowan, & Cowan, 1993). Cowan et al. (1991) reported that when a husband’s actual involvement in care of the baby was discrepant with a wife’s earlier prediction, the wife showed steeper declines in marital satisfaction after birth. Neither the objective division of labor, nor the perceived division of labor predicts marital adjustment, marital trajectories, or depressive symptoms (Belsky & Hsieh, 1998; Bristol, Gallagher, & Schopler, 1988). However, the discrepancy between each parent’s expectations and perceptions of responsibility for childcare support are significantly related both to depression and marital adjustment for both parents (also see Kalmuss, 1992; Voydanoff & Donnelly, 1999). When expectations are not met, a sense of unfairness and resentment may be engendered (Goodnow, 1998), leading to increased parental stress. Such feelings may lead to negative emotional arousal that interferes with warm, sensitive interaction with the child.

Much of the research on expectations has centered on involvement and division of labor; however, it may be that expectations in other coparenting domains may be important as well although the research on these other areas has not yet been conducted. For example, expectations regarding the how a partner will support instead of undermine one’s own parenting, may be important. In addition, it is likely that couples do not expect the level of conflict in their relationship to increase. Finally, expectant parents may not be aware of differences in childrearing values with their partners as these differences have not yet become salient.

The importance of expectations has direct relevance to prevention: One important focus of preventive interventions may be to help couples identify and expand the possibilities encompassed by their expectations. Fortunately, expectations and beliefs are not static. In a longitudinal study of newlyweds (Johnson & Huston, 1998), change in expectations and beliefs was common—although wives tended to change towards husbands’ viewpoints more than vice-versa. In fact, the more love reported by the wife, the greater her change towards her husband’s perspective over time. Although the pattern of change reported in that study was asymmetric, the good news in terms of intervention was that expectations do appear to be malleable.

The recognition of the importance of expectations also indicates a need for corresponding research into how expectations and early anticipation of coparenting develops. Just as the transition to parenthood begins in the prenatal period (Shapiro, Diamond, & Greenberg, 1995), so too does the coparenting relationship. Expectant parents begin to anticipate and make plans around coparenting issues, although the extent to which they explicitly address these issues with each other is not known. In addition, preparation for the baby—including preparing the living quarters, obtaining furniture and supplies, attending prenatal health visits and childbirth classes, and arranging for postpartum support—begin to set precedents for how child-related tasks are shared and managed. During this period, there is also psychological and emotional preparation regarding parenthood and salient issues for each partner begin to emerge (e.g., responsibility, dependency, loss of freedom). The manner in which expectant parents support each other as these issues emerge may have implications for the handling of issues regarding coparenting support versus undermining.

Readiness

Interventionists should be aware that couples may have varying levels of readiness to address co-parenting issues. One factor that may influence couple readiness is whether the pregnancy was planned. Intentionality of the pregnancy has been examined by a few researchers; for example, Cox, Paley, Payne, et al. (1999) found that intentionality of the pregnancy is linked to parental adjustment. The potential links between intentionality of the pregnancy and couple readiness may run in both directions. It is likely that readiness leads to a planned pregnancy: couples that have actively planned to become parents have often recognized a degree of “readiness” (emotional, financial, lifestyle) in themselves. However, there may also be a causal pathway in the other direction: Intentionality of pregnancy may influence the subjective sense of “readiness” to become a parent. Couples who planned to have a child may experience a greater level of joint “control” over the impending life change, and thus be more willing to engage in couple-level preparations including development of an explicit coparenting relationship (i.e., discussing and negotiating expectations and roles).

Even for parents who intended to have a baby, issues of readiness are important—and especially so perhaps for fathers. The expectant mother is generally more attuned to the impending birth than the father in part for biological reasons (Grossman et al., 1980), and the father’s perceived lack of readiness can have dire consequences. Fathers who report not being ready in late pregnancy report lower self-esteem, more depression, and lower marital satisfaction at 18 months postpartum (Cowan et al., 1991).

A lack of subjective “readiness” may retard the anticipation and development of the coparenting relationship components with potential postpartum effects on mutual support, sharing of responsibility, resolution of childrearing differences, and commitment to joint management of family interaction. Intervention during the transition to parenthood period, therefore, should relate to a couple’s readiness to confront coparenting issues. Couples who are not yet “ready” may need to do preliminary work. Such couples may need to address issues of readiness explicitly, to experience a sense of control in setting the pace for their preparation for parenthood and coparenthood, and to address possibly differing levels of readiness within the couple.

INTERVENTION GOALS AT THE TRANSITION AND COPARENTING

The best known of the few preventive interventions developed for couples at the family formation stage period is the intensive Becoming A Family project implemented by Cowan and Cowan (1987), a 25-session, 6-month program based largely on a discussion, support group model. Although this seminal intervention and research project demonstrated that intervention during the transition period can have positive effects (Cowan & Cowan, 1992), it has not been widely replicated. This program may have been too intensive for widespread dissemination. A few less intensive transition to parenting programs for couples are currently being developed and tested; however the degree to which they incorporate theoretically-based goals and content targeting coparenting varies. This section aims to clarify the processes by which improved coparenting may lead to attainment of several potential intervention goals.

Intervention Goal 1: Improve Parent Adjustment

Research has demonstrated a link between major life changes and increased levels of both physical and psychological problems (Dohrenwend & Dohrenwend, 1974). Parenthood, with associated and ongoing changes in roles, relationships, routines, responsibilities, identities, and task demands (Fish, Stifter, & Belsky, 1993; Grossman, 1988; Levy-Shiff, Dimitrovsky, Shulman, & Har-Even, 1998), represents a paradigmatic life change with potentially serious consequences. For example, about a third of low-risk parents report clinical levels of depression 18 months after the birth of the first child (Cowan et al., 1991). In a 5-wave study of mothers over the first 3 years of parenthood, about 38% reported clinical depression at one to three of the five waves of data collection, and an additional 8% were clinically depressed at four or all five waves (Network, N. E. C. C. R., 1999).

The factor most consistently identified with post-partum depression and maternal adjustment generally, is lack of social support—and especially “spousal support” (Brown & Harris, 1978; Crnic & Greenberg, 1987; Gotlib, Whiffen, Wallace, & Mount, 1991; O’Hara & Swain, 1996; Quinton, Rutter, & Liddle, 1985).² “Spousal support” is operationalized differently among studies, but generally refers to aspects of both the general couple relationship and the coparenting relationship. Dunn’s review finds that marital support is associated with functioning in a number of domains, including maternal adjustment and parenting competencen (Dunn, 1988). A meta-analysis concluded that low support from the father was related to severity of postpartum depression (O’Hara & Swain, 1996). Further, low partner support is linked to social–emotional problems in the children of adolescent mothers (Sommer et al., 2000). Although much of the research in this area has focused on the effects of social support of mothers, similar evidence has been collected indicating that mothers’ support of fathers’ parenting is a key factor influencing paternal adjustment and parenting (Allen & Hawkins, 1999; Grossman et al., 1980; Jordan, 1995; Seltzer & Brandreth, 1995). Once again, based on the proposition that coparenting is a central and proximal influence on parenting, it is expected that coparenting support is a more powerful influence on parental adjustment and depression than other aspects of couple support—especially during the family formation period.

The mechanisms through which social support exerts a generally beneficial influence has not been thoroughly addressed in the literature.³ However, in an important study, Cutrona and Troutman (1986) report that social support exerts a protective function against maternal depression primarily through the mediating role of parental self-efficacy. The concept of parental self-efficacy is based on Bandura’s self-efficacy theory (Bandura, 1977). Self-efficacy, one’s sense of competence and effectiveness in a domain, is thought to mediate the influence of past experience on future performance. Low-efficacy individuals experience relatively higher self-doubt, anxiety, and generally negative emotional arousal (Jerusalem & Mittag, 1995). Low efficacy individuals also tend to perceive task demands as threatening and avoid challenge, and have lower persistence on tasks.

Coparenting support serves to bolster a parent’s sense of performing the new parenting role adequately and competently, leading to more confidence in one’s parenting (Cutrona & Troutman, 1986; Frank et al., 1986). This effect is one key to understanding why the coparenting relationship is a proximal and powerful influence: For most parents, the coparent is the individual most aware of the parent’s handling of daily (and nightly) parenting tasks. The coparent also has the most information regarding contextual factors that influence self-efficacy representations, such as the difficulty of the task at hand. For example, the coparent presumably has a high level of information regarding the child’s temperament, daily hassles, the parent’s own vulnerabilities, and other extra-familial pressures such as money or work problems that affect one’s capacity for meeting parenting task demands. Finally, the coparent generally has the greatest opportunity for input into the parent’s self-perception through daily conversations that help crystallize perceptions and interpretations (Hardin & Higgins, 1996). Thus, coparenting support is viewed as the most proximal and powerful concurrent environmental influence on perceived parental efficacy, and through this mechanism affects parental stress and depression. Through this risk pathway, the coparenting relationship operates as a causal risk mechanism and not just as a marker or correlational risk factor.

Intervention Goal 2: Improve the Couple or Marital Relationship

The elevated rate of depression in parents of young children is probably reciprocally linked to the deterioration of marital quality in most families during this period (Belsky, Spanier, & Rovine, 1983; Grossman et al., 1980; Shapiro et al., 2000; also see Kurdek, 1999). Marital distress appears to be elevated during the family formation period: at 18 months post-partum, about 20% of parents report clinical levels of marital distress—not only is this rate over and above the 20% of couples who go on to divorce within the next 4 years, but the percentage of couples reporting marital distress continues to climb for the next 4 years (Cowan et al., 1991). Not all couples experience marital deterioration, but two thirds of mothers report declines in martial satisfaction after the first child’s birth (Shapiro et al., 2000; also see McHale & Huston, 1985).

Given longitudinal research and the theory of a downward marital cascade (i.e., marriages with eroded quality tend to ever lower levels of satisfaction; Gottman, 1993), one might expect that the potential stress of parenthood would knock some couples off of a stable, relatively benign trajectory onto one with a negative slope, leading to increasing difficulty and possibly divorce or separation. Some couples may regain higher levels of positive relational quality when the stresses of parenting young children are alleviated over time. However, other couples may not make it through this period intact, or may lack the resources to repair the strains. Thus, it is not surprising that a majority of families undergoing divorce include a child under 6 (Whiteside & Becker, 2000).

The potential of a coparenting-focused intervention to enhance couple relationship quality is supported by longitudinal evidence that coparenting is associated with couple relationship outcomes (Belsky & Hsieh, 1998; Hetherington et al., 1999; Schoppe, Mangelsdorf, Frosch, & McHale, 2001). Further research is needed on the processes involved. It may be that where coparenting is positive but the couple relationship is negative (Cowan & McHale, 1996; Van Egeren, 2000), parents have successfully rallied together for the good of their children. This process may designate the coparenting sphere as a nonhostile area, which provides the basis for positive experiences and the retention of positive interactions. Parents simply taking mutual delight in their child, for example, is a strong bonding experience (see below). Parents who are able to maintain a positive coparenting relationship despite negativity in their own dyadic relationship may thus be protected from being flooded by global negativity. The maintenance of a positive co-parenting relationship may sustain an incentive for repairing dyadic negativity. Of course, it is also possible that parents who maintain positive coparenting in the face of couple negativity may possess individual or couple strengths that eventually also serve to enhance the couple relationship. Examination of the effects of a coparenting intervention may help test this alternative account.

Enhancing coparenting in order to affect the couple relationship may represent a change in the perspective of some treatment providers. For example, some family therapists view problems in the couple relationship as primary factors that must be resolved before real and lasting improvements in parenting and joint family management can occur. However, the alternative view proposed here is that targeting the co-parenting relationship may lead to effects on the couple relationship. Further, parents are often invested in perpetuating the couple relationship conflicts and negativity, and thus may be ambivalent about a goal of promoting positive relationship quality. In strategic and practical terms then, it may be easier to gain parents’ endorsement of treatment goals relating to coparenting (with the ultimate aim being the welfare of the child) instead of first focusing on improving the couple relationship.

Intervention Goal 3: Improve Parenting and Child Outcomes

Although the transition to parenthood is stressful for many parents as individuals and couples, the family relationships of these early years of a child’s life are developmentally crucial. Even in low-risk families, distress in this period is tied to later negative outcomes for children in preschool and elementary school (Cowan & Cowan, 1992). The theory offered here is that enhancing coparenting quality will improve the sensitivity, warmth, and consistency of parenting, which will increase children’s emotional health and social competence. Given recent controversy surrounding the importance of parenting for child adjustment, this section briefly describes how early parenting is important for outcomes before discussing the process by which the coparenting relationship affects parenting.

The clearest links between early family relations and later outcomes is in the area of externalizing behavior problems, a broad dimensional category which includes more specific measures of disobedience, aggression, disruptive behavior, and antisocial behavior. The development of chronic and socially disruptive behavior often begins “in the first 700 days after birth” (Tremblay, LeMarquand, & Vitaro, 1999, p. 534). Persistent antisocial behavior has been linked to early behavior problems and is termed the “early starter” variety (Moffitt, 1993; Moffit, Caspi, Dickson, Silva, & Stanton, 1996; Tremblay, Pihl, Vitaro, & Dobkin, 1994; Zoccolillo, Tremblay, & Vitaro, 1996). Early starter antisocial behavior (Moffitt, 1993; also see Loeber, 1982; Olweus, 1979; Robins & Rutter, 1990) has been linked to the development of noncompliance, disobedience, and aggression in early childhood. Evidence suggests a trajectory from noncompliance at 18 months, aggression at 24 (for boys), to externalizing problems including antisocial behaviors at 36 months (Shaw et al., 1998; Shaw & Winslow, 1997). Further, aggressive behavior beginning at about age 3 is moderately stable (Richman, Stevenson, & Graham, 1982; Shaw et al., 1998; also see Belsky, Hsieh, & Crnic, 1998), and stability has been demonstrated from age 3 to middle childhood and early adolescence (Shaw & Winslow, 1997).

Although genetic factors as well as extra-familial influences such as peer, school, and neighborhood factors may be important, familial processes seem to be crucial in the early development of externalizing behavior problems (Loeber & Stouthamer-Loeber, 1986; Offord, Boyle, & Racine, 1989; Tolan, Cromwell, & Brasswell, 1986). For example, maternal responsiveness during the first 2 years of life predicts disruptive behavior at age 10, even after controlling for concurrent parenting and other established risk factors (Wakschlag & Hans, 1999). Patterson’s social learning theory approach has gained considerable currency: Child antisocial behavior emerges in coercive exchanges with parents (Patterson, 1982) as discipline and obedience issues become salient in very early childhood. Shaw and Bell (1993) propose an integrative model of early development of behavior problems beginning with the general negative infant reaction to the mother’s nonresponsiveness in the attachment framework (Ainsworth, Bell, & Stayton, 1991; Sroufe, 1985; Stayton, Hogan, & Ainsworth, 1971). The model posits a pathway from nonresponsive parenting and attachment problems to child noncompliance with early parental requests, such as cleaning up toys. Parental difficulty in maintaining consistent “insistence” on limits and discipline within the context of a warm parent–child relationship (Shaw & Bell, 1993), then leads to the onset of coercive cycles (Patterson, 1982; Patterson, DeBaryshe, & Ramsey, 1989; Patterson, Dishion, & Bank, 1984).

Thus, both sensitivity/warmth and competent behavioral management (Feinberg & Pettit, 2001) are important parent factors that help avoid early externalizing domain trajectories. As described above, a positive coparenting relationship, especially in the area of coparenting support, would be expected to foster parental self-efficacy, which in turn has been demonstrated to be linked to sensitivity and warmth (Teti, O’Connell, & Reiner, 1996). In fact, based on his review of existing evidence, Teti proposed that parental efficacy is the “final common pathway” to disruptions in caregiver sensitivity (Teti et al., 1996).

The role of the coparenting relationship in bolstering parental efficacy may be especially crucial when parenting is disrupted because one partner is experiencing depressive symptoms. Several studies have documented the detrimental effects of parental depression on parenting (e.g., Carter, Garrity-Rokous, Chazan-Cohen, Little, & Briggs-Gowan, 2001; Cohn & Campbell, 1992; Field, 2000; Forehand et al., 1988; Gotlib & Beach, 1995; Johnson & Jacob, 2000; Jones, Field, & Davalos, 2000; Martins & Gaffan, 2000). Research indicates that parental efficacy mediates the relationship between parental emotional distress and lack of sensitive parenting (Gondoli & Silverberg, 1997; Teti et al., 1996; Teti & Gelfand, 1991). Thus, positive coparenting support may buffer parental sensitivity from the negative effects of parental depression.

In addition to influencing parental sensitivity, positive coparenting support may also serve to foster involved, competent behavioral management through the mediating role of parental efficacy: Low parental efficacy can be expected to lead to a lower level of sustained, consistent, and competent management behavior. Parents with lower efficacy may be more susceptible to child pressure, contextual stress, and fluctuations of mood, leading to inconsistent responding and a greater tendency to yield to child demands. Thus, low-efficacy parents of ADHD children, for example, tend to withdraw from engaging with the child in a demanding task situation (Mash & Johnston, 1990).

Parents with a low sense of parental efficacy tend to have higher autonomic arousal, negative affect, and feelings of helplessness in the face of difficult child behavior. Such parents tend to use coercive discipline (Bondy & Mash, 1999), and low parental efficacy has been found to characterize abusive parents (see review in Mash & Johnston, 1990; also see Bugental, 2001). However, under certain conditions (e.g., a difficult task situation with a hyperactive child) low parental efficacy may lead not to negative parenting, but to withdrawal (Mash & Johnston, 1990). These two different behavioral consequences of low self-efficacy (i.e., harsh parenting or withdrawal) correspond to findings in a series of experimental studies: When faced with difficult or ambiguous infant signals, parents with low parental efficacy tend to demonstrate either helplessness or a high level of illusion of control (Donovan & Leavitt, 1989; Donovan, Leavitt, & Walsh, 1990). If such variability in responding is found within individual parents, as seems likely, then the contextual condition is ripe for the development of coercive cycles.

One pathway, then, from enhanced coparenting to decreased child externalizing problems may operate via increased parental efficacy and improved parenting practices (both enhanced sensitivity/warmth and behavioral management competence). A second pathway from coparenting to early child behavior problems may be through interparental consistency. Although parenting consistency is generally viewed as an important feature of a healthy family environment, there is little research on the specific effects of inconsistency between parent discipline practices. Clinical and common wisdom suggests that children attempt to find the path of least resistance by appealing to or aligning with the least restrictive parent. It is possible that the exploitation of inconsistencies between parents may lead to the kinds of negative social learning that occurs when children seek to exploit inconsistencies in one parent’s discipline practices. However, it may be difficult to distinguish empirically between the effects of inconsistency between parents and the effects of interparental conflict, childrearing disagreement, triangulation, and poor discipline practices employed by one or both parents.

Specifying the potential links from coparenting to early child internalizing problems is more difficult than for externalizing problems. First, some evidence suggests that the associations with externalizing problems may be larger than for internalizing behaviors (Buehler, Anthony, Krishnakumar, & Stone, 1997; Feinberg et al., 2000). Second, researchers’ capacity to detect early internalizing problems may be more limited than for early externalizing problems. Nonetheless, the links between coparenting and child internalizing problems deserve attention. In examining internalizing trajectories, it may be useful to include internalizing problems such as depression, anxiety, and social withdrawal, as well as domains such as school readiness and secure relationships (Ainsworth et al., 1991; Dadds & Powell, 1991; Harold & Conger, 1997; Harrist & Ainslie, 1998; Katz & Gottman, 1993; Parker, Boak, Griffin, Ripple, & Peay, 1999; Rubin, 1995).

Despite the greater difficulties in conducting research into internalizing versus externalizing problems, some relevant findings have been obtained. Interparental conflict has been linked to internalizing problems in children (Kerig, 1998), and childrearing disagreement is linked to children’s internalizing problems and cognitions such as self-blame (Grych & Fincham, 1993; Jouriles, Murphy, et al., 1991). This association has been examined longitudinally: Exposure to couple conflict at the age of 2 has been linked to depression/anxiety, withdrawal, and overall internalizing problems at age 5 (Shaw, Keenan, Vondra, Delliquadri, & Giovannelli, 1997).

The form that couple conflict takes may relate to the development of internalizing versus externalizing behavior problems. In one study of 9–15 year old children, a measure of child-reported triangulation was associated with internalizing problems and to some degree externalizing as well (Buehler et al., 1998). Overt interparental conflict however was associated with externalizing behavior only. These results support the view that overt conflict may act as a disinhibitor to children, while triangulation may place children in situations of ongoing, inner tension. Although interesting, further research is needed to determine whether it is the experience of being exposed to different kinds of interparental conflict (overt hostility vs. triangulation), or the kinds of genetic predispositions that children acquire from parents who engage in different kinds of conflict, that is responsible for the pattern of associations.

In addition to interparental conflict, other aspects of coparenting may be linked to child internalizing problems. For example, it is likely that coparenting is linked to child internalizing problems through parental depression. Although genetic transmission of vulnerability to depression represents one influence on child depression, it is also likely that parenting behaviors disrupted by parent depression and stress are also linked to child depression (Field, 2000). To the extent that coparenting support and satisfactory resolution of division of labor issues, for example, influence parental stress and depression, then coparenting can be seen as potentially influencing child depression.

Another pathway between coparenting and child internalizing problems may involve parental efficacy: In one study, mothers of aggressive school-age children were rated as behaviorally undercontrolling, however mothers of withdrawn and internalizing children were psychologically and behaviorally overcontrolling (Mills & Rubin, 1998). One is tempted to speculate that the degree of parental efficacy is related to absolute deviation of parental control from the norm, with low efficacy parents either displaying helplessness (undercontrolling) or an illusion of control (over-controlling). Thus, the coparenting relationship may influence parental efficacy, which may then lead to different parent behaviors (undercontrolling vs. over-controlling) with different consequences for children (internalizing vs. externalizing).

In sum, there are several pathways through which coparenting relationships may influence a broad range of child adjustment. To the extent that preventing child behavior problems is a goal of coparenting intervention, such intervention should take place early. Early coparenting preventive interventions (even in the prenatal period) are indicated because the early sensitive years of a child’s life overlap with the stresses and strains of the family formation period.

Intervention Goal 4: Improve Coparenting in Divorced Families

In addition to improving coparenting in intact families to reduce levels of child behavior problems, early coparenting interventions for new parents may also serve as a preventive measure for those couples whose marriages (or relationships) will end in divorce (or separation). Introducing coparenting concepts and promoting enhanced coparenting quality early may ameliorate, in some families, the consequences of conflictual and destructive coparenting relations in a later postdivorce period.

There is an empirical need for coparenting skills in the divorcing population: In placing ex-spouses in a 2 ○ 2 matrix of high/low cooperative communication and high/low discord after divorce, Maccoby et al. (1990) found that the largest single group were couples low in cooperation and high in discord. Only 25% of divorced couples with children were successfully in achieving low discord and high cooperation.

A limited number of postdivorce programs have been developed to address this need (Leek, 1992; McKinnon & Wallerstein, 1988), and some local jurisdictions and court systems mandate classes for parents undergoing divorce. Yet the modal number of classes in such programs is one (Braver, Salem, Pearson, & DeLuse, 1996). Although more intensive programs for divorcing parents exist and may in fact be helpful, there are two obstacles to introducing coparenting concepts at the time of divorce. First, there has usually been a long period of conflict and/or withdrawal by the time of divorce, and this period has affected children. Second, the conflict and family relational patterns among parents are often entrenched at the point of divorce, meaning that intervention is more difficult at that point (Boyojko, 2000).

It is not likely that the promotion of positive co-parenting in the postdivorce relationship would be an explicit goal of interventionists or families at the transition to parenthood. Still long-term follow-up studies of intervention effects might include this outcome. The rationale would be that even if the couple relationship was not sustained, an early introduction to the concepts of coparenting may yield later beneficial effects. In this vein, it might be helpful to introduce and stress for new parents the distinction between the coparenting relationship and the general couple relationship. Creating, emphasizing, and reinforcing this distinction would be an important component of an intervention designed to buffer the coparenting relationship from potential negativity in the couple relationship during marriage (or cohabitation) and after separation.

COPARENTING INTERVENTION

A second basic axiom of prevention science is that, although an intervention may target a modifiable risk mechanism, the success of the intervention in affecting public health is additionally linked to a design that maximizes appeal to participants and can be widely disseminated.

A question that naturally follows from this axiom is whether program designers have recognized participation barriers and planned effective dissemination strategies (Berger & Hannah, 1999; also see Heller, 1996)? For example, a 1993 NIMH report noted the importance of prevention programs for couples (Coie, Watt, West, Hawkins, et al., 1993) and many programs have been developed to enhance couple relations, communication, and satisfaction (Berger & Hannah, 1999). Yet these programs frequently fall short of providing adequate preventive services. In part, this may be linked to the reluctance of individuals and couples to engage in activities seen as counseling, therapy, or emotional support: Even though marital distress is closely linked to difficulties such as depression (Hops, Perry, & Davis, 1997) and alcohol abuse (Schaap, Schellekens, & Schippers, 1991), most distressed couples do not seek out professional help (Bradbury & Fincham, 1990).

A second question is whether recruitment and participation patterns reflect an optimal pattern. Premarital counseling programs do not seem to recruit successfully among the population of high-risk couples (Sullivan & Bradbury, 1997). Several researchers and observers have noted that couple-oriented programs, and the research they are based on (Depner, Leino, & Chun, 1992), fail to involve many participants of lower socioeconomic status or couples from minority or ethnically diverse populations (Berger & DeMaria, 1999; Duncan & Markman, 1988; [Holtzworth-Munroe, 1995 #151]; Sanders et al., 1997).

In order to address these issues, we are developing a coparenting intervention to be delivered through childbirth education departments at local hospitals and other agencies. Given the near universal appeal of childbirth education, this strategy is aimed at locating the prevention within a stable local, nonstigmatizing institution. Through group sessions during the prenatal and postnatal periods, first-time parents will be introduced to the concepts and importance of coparenting, develop and practice communication and joint problem solving skills, and learn to identify and resolve coparenting issues early. In initial pilot presentations to childbirth classes, feedback has been positive. Not only have couples responded that they would be open to engaging in such a program, but childbirth educators have been enthusiastic about taking on the coparenting-focused curriculum. This congruence with the professional identity and mission of these educators is crucial for promoting a sustained, high-quality program.

The development of this program is informed by a needs assessment we conducting by telephone with new parents described in more detail elsewhere (Feinberg, 2002). The conclusions we have drawn from this needs assessment regarding parents’ core concerns has implications not just for the prevention model we are developing, but also for the development and delivery of other coparenting and parenting programs. The first conclusion we have drawn from the needs assessment is that parents are centrally concerned with the care and development of their child. Thus, attracting parents to a coparenting intervention may be facilitated if the welfare and development of the child are seen as the central goal. These results suggest that the importance of the coparenting relationship for the child’s development may be the key selling point for most parents, while an emphasis on family harmony, parental well-being, and the overall couple relationship may be significant but secondary.

A second conclusion from this study is that a universal intervention should be framed within an educational approach. Although some parents recognize a need for support, a greater number would like information and advice. Approaches that appear too psychological or that involve counseling, a support group modality, or personal revelation may not appeal to all parents. This conclusion does not mean that elements of support and personal (or couple) change should not be incorporated into a program, but that the emphasis of the curriculum and the recruitment material should be on education and information.

In general, intervention should aim to enhance protective factors as well as addressing risks. Thus, the identification of positive aspects of the coparenting relationship may indicate areas that interventions can bolster and reinforce, especially for families experiencing difficulty. For example, the observation of one’s partner in the parenting role was cited as positive by a number of parents in our needs assessment survey. Parents experiencing stress and strain may frequently utilize the witnessing of the partner in the parenting role as an opportunity to be critical and undermining. Helping parents experience this potentially positive but perhaps easily sabotaged phenomena may be a way of reinforcing the positive aspects of coparenting. An intervention might, for example, help parents conduct exercises in which they are directed to write down the positive aspects they observe in their partner’s parenting.

NEXT STEPS AND FURTHER ISSUES

The development and empirical test of coparenting interventions will provide an important means of assessing whether in fact coparenting is a central component of risk mechanisms that lead to negative parenting and child outcomes. In moving forward in this area, there are several issues to which researchers should be sensitive.

Coparenting Across Development

In addition to the coparenting prevention model we are developing described briefly above, the coparenting framework discussed in this paper may also inform other preventive interventions, both during the transition to parenthood phase and during other periods. For example, prenatal home visiting programs frequently focus on young mothers. In such “fragile” families, mothers and fathers’ romantic involvement and living arrangements vary, both across families and within families over time (Carlson & McLanahan, 2001). These fathers can play important roles and efforts to include them should be made in future prevention models. Further, given the emerging research on mother–grandmother relations in these families (Apfel & Seitz, 1991; Kalil, Spencer, Spieker, & Gilchrist, 1998; Szinovacz & Roberts, 1998), it may be appropriate to develop or adapt prevention models that focus on mothers and grandmothers as coparents.

Other developmental periods that may be important to target include times when problems and difficulties either emerge, or times of family transition. Thus, coparenting intervention may be useful for families during the toddler and early preschool period when behavior problems emerge in some children. Prevention at these periods may help parents’ to avoid allowing their initial reactions to the stresses engendered by early child behavior problems from becoming crystallized into chronic negative family interaction patterns. In the pre/postnatal intervention program we are developing, booster sessions will be timed to coincide with the 18–24 month period, and then again around 36 months. In addition, other intervention programs with the capacity for early prevention (e.g., Triple-P: Sanders, 1999; The Wonderful Years: Webster-Stratton & Hammond, 1990) might potentially be modified to incorporate enhanced information and training on coparenting issues.

Another period that may be ripe for the development of coparenting prevention models is the preadolescent period. Established and effective prevention programs that aim to enhance family communication in early adolescence (e.g., Preparing for the Drug Free Years: Kosterman, Hawkins, Spoth, Haggerty, et al., 1997; Adolescent Transition Program: Eron et al., 2002) may be modified to include a coparenting dimension when appropriate. The coparenting issues that arise when a child is a preadolescent may be somewhat different than those that arise at earlier periods. For example, issues of how parents manage monitoring and supervision of the preadolescent’s leisure activities may be important. To the extent that effective monitoring depends on a close parent–child relationship that fosters disclosure (Kerr & Stattin, 2000), an issue for parents to confront is how to handle information disclosed by an adolescent to one parent in the context of that parent–child relationship: Is all information to be shared openly by the parents, or would that undermine the security of the parent–child relationship and inhibit further disclosure? When is information disclosed to one parent a natural and appropriate consequence of a child’s separate relationships with parents, and when is it a symptom of, or contributor to, troublesome alliances? These are the types of questions that researchers should consider before moving coparenting intervention into the adolescent period. It may be that parents should decide these issues for themselves; however, it may also be that preventionists can provide a helpful framework or empirically-based guidance.

Integrating Parenting and Coparenting Intervention

Prevention efficacy trials serve several masters. One aim of a prevention trial is to assess whether a given intervention is effective in modifying risk factors or outcomes. In order to achieve this aim, the intervention should be delivered in a manner that is maximally effective. For example, in order to improve parenting, a coparenting intervention should also include content on positive parenting. It is reasonable to expect that the combination of coparenting and parenting content will enhance parental functioning to a greater extent than just coparenting material alone. To the extent that coparenting enhances parental efficacy and well-being, for example, parents will demonstrate enhanced levels of sensitivity and appropriate limit-setting. However, even with maximum coparental support and positive interaction, some parents may be limited by inaccurate expectations and cognitions, poor parenting models in their past, or a limited repertoire of behaviors and skills. Extending parents’ capacity for positive parenting would require the kind of education and parenting skill development that has been incorporated in several well-developed programs (e.g., Kosterman et al., 1997; Olds et al., 1998; Webster-Stratton, 1994).

An issue that naturally emerges is how to integrate coparenting and parenting content. One reasonable approach is to add a coparenting component to such programs and test the effectiveness of the standard versus modified approaches. However, it may be more effective to integrate coparenting issues into each of the modules addressed by the program, rather than adding a discrete, stand-alone coparenting component. Thus, coparenting issues and skills may be addressed when parents learn about effective limit setting, or about enhancing communication with adolescents.

A second master served by prevention trials is the attempt to determine whether underlying theory regarding risk mechanisms is correct. In order to assess, for example, the validity of viewing coparenting as central to risk mechanisms affecting parenting, it would be important to design an intervention that selectively targets coparenting. This approach may at times be in direct conflict with the goal of designing a maximally effective program. Bolstering the strength of a program by including both coparenting and parenting material would unfortunately make it more difficult to determine whether it was the coparenting content, the parenting content, or both, that affected risk mechanisms. Serving both the effectiveness and theory-testing masters requires the development of multiple versions of the program and testing each version. The cost of this more comprehensive but also more expensive approach may be difficult to justify.

Some parenting programs may already incorporate elements of coparenting in the program (e.g., Triple-P; Sanders, 1999). However, many parenting programs are designed for individual parents, not couples. Focusing on the individual parent is justified of course as it is frequently difficult to obtain participation of both parents. However, the potential efficacy of a coparenting intervention is limited if delivery is only through one parent.

Some may argue that focusing directly on parenting issues is a more direct route to changing parenting and child outcomes than focusing on coparenting; that a coparenting modification would diffuse the parenting intervention; and that enhanced coparenting is often a byproduct of parenting intervention anyway. These arguments may be valid to some extent. As parents develop a range of skills, as their parenting becomes more effective and reduces ongoing parent–child conflict, and as parents adopt standards as to what constitutes positive parenting, the factors leading to coparental distress may diminish. However, the research and framework presented above suggest that coparenting is an important and distinct sphere from parenting, and that it influences parenting. Empirical assessment of the degree to which standard parenting programs enhance coparenting is indicated.

Diversity

A final issue for researchers to consider is how to design interventions that are effective with a range of diverse families. Family diversity ranges across structural composition (e.g., divorced, step-families, unmarried parents, gay and lesbian parents), culture and race/ethnicity, and socioeconomic status. It should be emphasized that the coparent role is not restricted to married, biological parents, but can include boyfriends, grandmothers, stepparents, and others. While I have focused on intact, married couples in this paper, this focus is only a starting point. As research and intervention continues to mature, we should assess whether coparenting interventions can successfully include traditional married couples, mother–boyfriend pairs, and mother–grandmother coparenting teams in the same group? Would coparents in such a mixed group recognize universal coparenting issues in their own relationship, or should intervention groups target different types of families in order to focus on distinct issues?

In the area of cultural and racial/ethnic diversity, the research on coparenting is somewhat limited to date (but see Brody et al., 1998; Crohan, 1996; Hortacsu, 1999). This limitation restricts our ability to identify the degree to which coparenting issues differ across groups and thus the extent to which interventions require speicific modifications. There is some indication that there are broad similarities across groups in family relations and coparenting relationships (Flannery, Vazsonyi, & Rowe, 1996; Gutman & Eccles, 1999; McLoyd, Cauce, Takeuchi, & Wilson, 2000; McLoyd & Steinberg, 1998; Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000; Rowe, Vazsonyi, & Flannery, 1994, 1995), and that differences that have been found are in the levels of variables examined and not in the form of the associations among variables (Georgas et al., 2001; McHale, Lauretti, Talbot, & Pouquette, 2002). Nonetheless, we need to pay attention to cultural and ethnic issues during the design of interventions—if not because of major differences in the role of coparenting across groups, then because different intervention delivery styles and motivating frameworks may be effective with different groups. For example, it may be that emphasizing the effects of enhanced coparenting on family cohesion may be more motivating for some groups, while other groups may be more motivated by focusing to a greater extent on individual child well-being.

SUMMARY

This paper employed a recently developed conceptual and theoretical model of coparenting to indicate new territory that might be profitably explored by prevention scientists. I have suggested that interventions developed for families in the formation stage may have the greatest potential to effect change. The processes by which intervention-induced enhancement in the coparenting relationship might relate to a series of important parent and child outcomes were described. Finally, issues regarding the delivery and design of coparenting interventions were discussed. It is hoped that the perspective and information offered here is useful for those developing new lines of basic research, as well as those designing interventions to support families both at the transition to parenthood and at later points in the family life cycle.