The relation between joint physical custody, interparental conflict, and children’s mental health

Objective: This study investigated the relationship between joint physical custody and children’s mental health, and tested whether interparental conflict moderated the association. Background: Joint physical custody is an emerging post-separation care arrangement that is expected to counter the negative effects of family dissolution on children’s overall wellbeing. There is, however, substantial disagreement about the impact that joint physical custody may have on children’s mental health when interparental conflict is high. Method: The statistical analysis was based on data from the Family Models in Germany (FAMOD) study, which was conducted in 2019. The analytical sample consisted of 1,087 post-separation families practicing either sole physical custody or joint physical custody. Linear regression models were estimated to determine the relationship between physical custody arrangements, interparental conflict, and children’s mental health problems. Results: Living in a joint physical custody arrangement was positively related to children’s mental health, whereas high levels of interparental conflict were negatively related to children’s mental health. However, when levels of interparental conflict were high children in joint physical custody arrangements displayed levels of mental health problems that were quite similar to those of children in sole physical custody arrangements. Conclusion: The findings emphasize that joint physical custody is not a “one-size-fits-all model” that is suitable for all post-separation families, and that it is only beneficial for children’s mental health when interparental conflict is low.


Introduction
Although it is still common in many Western countries for children to live either primarily or exclusively with their mother after a parental separation or divorce (Cancian et al. 2014), research has identified a recent trend towards fathers being more involved in the upbringing of their children in post-separation families (Kalmijn 2016;Parkinson 2011). This trend has been accompanied by significant changes in the distribution of parenting time among mothers and fathers (Parkinson 2011). Joint physical custody is a new parental care arrangement in which children reside alternately in the two parental households, and spend substantial amounts of time with each of their parents after family dissolution (Steinbach 2019). While the prevalence of joint physical custody is still comparatively low in Germany, with estimations suggesting that 5% of all post-separation families practice this type of care arrangement (Walper 2016), joint physical custody is far more common in northern European countries like Norway and Sweden, with respective shares of joint physical custody families on all post-separation families of 30% (Kitterød & Wiik 2017) and 40% (Fransson et al. 2018a). Thus, joint physical custody has become an accepted alternative to more traditional post-separation care arrangements in many Western societies (Bergström et al. 2015;Fransson et al. 2018b;Melli & Brown 2008;Spruijt & Duindam 2009).
Because fathers in joint physical custody arrangements are more strongly involved in their children's lives after family dissolution than fathers in sole physical custody arrangements, some researchers have argued that joint physical custody may counter the negative effects of a parental break-up, and thus contribute to children's health and wellbeing in post-separation families (Bastaits & Pasteels 2019;Bauserman 2002;Braver & Votruba 2018). However, previous studies have provided evidence that joint physical custody may not be a "one-size-fits-all model" (McIntosh & Chisholm 2008). Whereas there is general consent among researchers and practitioners that the majority of children in post-separation families will benefit from two deeply involved parents, there is substantial disagreement about the impact of frequent post-separation contact with both parents on children's health and well-being when levels of interparental conflict are high (Mahrer et al. 2018;McIntosh & Chisholm 2008;Steinbach 2019). Despite the evidence that ongoing conflict between the parents is a chronic stressor for children that can have detrimental effects on their development (Amato 1993), the findings of previous research on the relevance of interparental conflict for the relationship between post-separation care arrangements and children's health and well-being have been largely contradictory (Elam et al. 2016).
Thus, in light of the growing prevalence of joint physical custody in Western countries, it has become increasingly important for researchers to identify the conditions under which children may benefit from this new care arrangement (McIntosh 2009). This is relevant not only with regard to the legislative context, but also with respect to the guidelines that should govern court decisions on physical custody. However, despite the clear relevance of this topic, so far only a small number of empirical studies have considered the moderating role of interparental conflict when examining the association between joint physical custody and children's mental health (Kalmijn 2016). To help close this research gap, the present study investigates the potential relationship between the physical custody arrangement, interparental conflict, and children's mental health in postseparation families. Using data from the Family Models in Germany (FAMOD) study, linear regression models are estimated for 1,087 post-separation families with either joint physical custody or sole physical custody arrangements. The purpose of the analysis is to examine the association between joint physical custody and children's mental health problems, and to test whether living in a joint physical custody arrangement is beneficial for children in post-separation families that are characterized by high levels of interparental conflict.

Background and hypotheses
Because separation and divorce rates remain relatively high in Western countries (Härkönen 2014;Vanassche et al. 2013), the well-being of children in post-separation families has attracted a substantial amount of attention from researchers. Against this background, a large number of empirical studies have shown that, on average, children in nuclear families fare better than children in post-separation families with respect to their health and overall well-being (for overviews, see Amato 2010;Härkönen et al. 2017). Most of these studies have sought to explain the negative effects of a parental separation or divorce by concentrating on five factors: the loss of contact with the nonresidential parent, the adjustment of the residential parent, economic hardship, stressful life changes, and interparental conflict (Amato 1993;Amato 1994). One factor that has been found to be particularly crucial in accounting for the lower well-being of children in post-separation families is the lack of resources that children experience in traditional post-separation care arrangements. In sole physical custody arrangements, in which children live primarily or exclusively with one parent (in most cases with their mother) and have either no or only limited contact with the other parent (Cancian et al. 2014), children generally suffer from the significant loss of the emotional, social, and financial resources previously provided by their nonresidential parent (Steinbach 2019).
Joint physical custody is a post-separation care arrangement in which children reside alternately in the two parental households, and spend substantial amounts of time with both their mother and their father. Although there is no official definition of the term, researchers referring to joint physical custody usually mean an arrangement in which children spend at least 30% of their time with each of their parents (Steinbach 2019). Thus, the main reason why scholars have suggested that joint physical custody may counter the negative effects of a parental separation or divorce and benefit children in post-separation families is that this care arrangement implies that both parents are strongly involved in the upbringing of their children. For instance, practicing joint physical custody increases the amount of time children spend with their nonresidential parent, and thus reduces children's perception of loss, and the degree to which they worry about or feel responsible for their nonresidential parent in a sole physical custody arrangement (Turunen 2017). In addition, more frequent contact between children and their nonresidential parent is likely to improve children's economic well-being, as nonresidential parents, through their greater involvement in their children's lives, should have more incentives to provide their offspring with financial resources (Bauserman 2002;Köppen et al. 2018). Furthermore, frequent parent-child contact after family dissolution "strengthens the parent-child bond and facilitates the kind of authoritative parenting style, with high levels of support and control, that has been found to be positive for child development" (Turunen 2017: 372).
There are also reasons to assume that children may indirectly benefit from joint physical custody, as this type of care arrangement may have positive effects on the wellbeing of the residential parent, which should, in turn, be beneficial for the mental health of children. For instance, compared to their counterparts in sole physical custody arrangements, residential parents -in the majority of cases the mothers -who practice joint physical custody may experience lower levels of stress because they do not have to carry the burden of sole physical parenting, and they may find it easier to balance child care and paid employment (Bernardi et al. 2018;Steinbach 2019). As it has been shown that negative emotions and feelings, which may result from prolonged experiences of stress, can be transmitted from parents to their children (Augustijn 2020;Larson & Gillman 1999), living in a joint physical custody arrangement should indirectly increase children's mental health. Moreover, previous research has suggested that mothers with sole physical custody are less likely to repartner than mothers with other post-separation care arrangements (Vanassche et al. 2015). As the formation of a stepfamily is likely to improve children's economic situation, children would again indirectly benefit from joint physical custody.
Similarly, living in a joint physical custody arrangement may have positive indirect effects on children through enhancing the well-being of their nonresidential parent -in most cases the fathers. Research has established that a separation or divorce can negatively affect adults, especially when minor children are involved (Leopold 2018;Leopold & Kalmijn 2016). Nonresidential parents may experience higher levels of stress and depression than residential parents if they lose contact with their children or fear that they will not be able to maintain satisfying levels of contact. Limited parent-child contact may, in turn, lead to a poorer parent-child relationship and the parent's perceived inability to fulfill normative expectations regarding their parental role, all of which may negatively affect their mental health and well-being (Evenson & Simon 2005;Leopold 2018;Leopold & Kalmijn 2016).
In accordance with these assumptions, previous research has found significant differences in the overall adjustment of custodial and noncustodial fathers, as "[t]he presence of children appears to be a stabilizing force in the adjustment of custodial fathers, lessening detrimental effects and making them remarkably similar to their married counterparts" (Stewart & Schwebel 1986: 61-62). Moreover, noncustodial fathers have been found to fare, on average, worse than custodial parents on several health outcomes, including depression, anxiety, emotional health, and life satisfaction (Evenson & Simon 2005;Maslauskaite & Steinbach 2020;Stewart & Schwebel 1986). Similarly, in a sample of recently divorced fathers, fathers with either full physical custody or joint physical custody were found to have significantly higher levels of emotional well-being than, for example, noncustodial fathers (Bokker 2006). Against this background, it appears plausible to assume that joint physical custody can reduce the negative effects of a separation or divorce on the nonresidential parent's well-being through higher levels of parent-child contact and closer parent-child relationships, and thus limit the transmission of mental health problems between nonresidential parent and child. Taking all these considerations into account, the first hypothesis of this study is that joint physical custody reduces children's mental health problems in post-separation families (H1).
Although most intimate relationships are characterized by some form of conflict, and not all types of conflict are necessarily harmful (King & Heard 1999), the interparental conflict perspective by Amato (1993) suggests that a home environment that is characterized by high levels of interparental conflict is detrimental for children's overall development. In accordance with this perspective, several studies have found that children's exposure to interparental conflict is generally associated with a higher prevalence of internalizing and externalizing problems (Elam et al. 2016;Gerard et al. 2006), as well as with reduced levels of emotional well-being (Ayoub et al. 1999). Previous research has also shown that interparental conflict can provoke negative emotions in children, including feelings of fear, anger, and distress. Moreover, parents may draw their children into their conflicts and force them to take sides, which can cause the parent-child relationship to deteriorate. Children may also feel responsible and guilty for conflicts between their parents (Amato 1993), especially if the parents argue over child-related topics. Furthermore, there is evidence that interparental conflict that is very hostile and aggressive may be particularly harmful for children, because it can cause them to fear that their parents might separate or divorce, or that their parents may turn their anger and aggressiveness towards the children (Grych et al. 2000). By eliciting feelings in the children that they are not being adequately cared for, interparental conflict may further impair the children's sense of emotional security, and their ability to control their emotions (Vanassche et al. 2015).
Among researchers, there is a general consensus that interparental conflict is one of the most crucial factors in children's problems in post-separation families (Spruijt & Duindam 2009). Amato (1993), for instance, pointed out that ongoing conflict between the parents after separation or divorce (e.g., conflicts over child custody, parent-child contact, or child support) is a chronic stressor that can have negative long-term effects on children's mental health. It has also been suggested that high levels of interparental conflict after family dissolution may "aggravate the well-known negative effect of parental separation on children's wellbeing" (Kalmijn 2016: 68). Furthermore, conflicts between separated or divorced parents may be particularly harmful for children because these types of conflicts are likely to be more intense, more prolonged, and less well resolved than conflicts in nuclear families (Dunn et al. 2005). Accordingly, this study will test the hypothesis that high levels of interparental conflict increase children's mental health problems in post-separation families (H2).
Although it is plausible to assume that high levels of interparental conflict have more negative effects on children who have frequent contact with their nonresidential parent after family dissolution, comparatively few studies have investigated the moderating role of post-separation interparental conflict on the relationship between joint physical custody and children's health and well-being (Kalmijn 2016). Moreover, the findings of the existing studies on this topic have produced contradictory results. On the one hand, some of these studies failed to find any support for the hypothesis that high levels of interparental conflict reduce the positive effects of joint physical custody on children's health and well-being (for an overview, see Mahrer et al. 2018), or they found that factors like the parent-child relationships had a greater impact on children than their parents' conflicts (Nielsen 2017). On the other hand, some empirical studies have suggested that while children generally profit from joint physical custody, elevated interparental conflict outweighs the positive association between joint physical custody and children's wellbeing, and ultimately reduces child mental health in post-separation families. According to these authors, joint physical custody is beneficial for children only when levels of interparental conflict are low, and when the parents are able to communicate and cooperate (Kalmijn 2016;McIntosh & Chisholm 2008;Singer 2008;Spruijt & Duindam 2009;Vanassche et al. 2013).
There are several reasons why the high levels of contact between children and their nonresidential parent that are typical of joint physical custody arrangements may be harmful for children's mental health when interparental conflict is frequent. For instance, joint physical custody and the associated closer relationship between children and their nonresidential parent may cause children to feel caught in the middle of their parents' conflicts (Elam et al. 2016;Kalmijn 2016;Sobolewski & Amato 2007;Vanassche et al. 2013). Other factors that may reduce children's mental health are recurrent "tension ridden change-overs between the two houses, exposure to expressed acrimony, ongoing denigration of one parent by another, and insidious embroilment of the children in supporting the separate views of each parent" (McIntosh & Chisholm 2008: 39). Although these factors may also negatively affect the mental health of children living in sole physical custody arrangements, their impact should be stronger on children living in joint physical custody arrangements, due to the closer and more intimate parent-child relationship. As children in joint physical custody arrangements live with both of their parents after family dissolution, it is also plausible to assume that this arrangement requires the parents to have more contact with each other because, for example, they have a greater need for coordination. These high contact rates between the parents may further escalate existing conflicts, or even create new conflicts that may negatively affect the children's mental health. Therefore, the third hypothesis is that joint physical custody reduces children's mental health problems in post-separation families only if the levels of interparental conflict are low (H3).

Data and analytical sample
The present study uses data from the Family Models in Germany (FAMOD) study (https://search.gesis.org/research_data/ZA6849), a national convenience sample that was conducted in 2019 and whose main objective was to explore the well-being of family members in post-separation families across Germany, with a special focus on joint physical custody arrangements. The FAMOD sample was stratified by family model (nuclear, sole physical custody, and joint physical custody families) and age of a selected target child (0-6 years and 7-14 years). In total, the survey provides researchers with information about the family life of 1,554 nuclear and post-separation families with at least one child under the age of 15 who has contact with both of his or her biological parents.
Because FAMOD includes a substantial number of families practicing joint physical custody through means of oversampling (due to the low prevalence of joint physical custody in Germany and the severe restrictions regarding the identification of these families through official statistics), it is the first study conducted in Germany that has generated data that can be used to investigate this new post-separation care arrangement. Another feature of this survey is its use of a residential calendar through which parents in post-separation families could give detailed information about the amounts of time a selected target child was living with each biological parent after family dissolution (Sodermans et al. 2014). One major advantage associated with the use of such a residential calendar is that it allows for clear distinctions to be made between sole physical custody and joint physical custody arrangements (for a detailed description of FAMOD, see Kantar Public 2020; Steinbach et al. 2020).
Although FAMOD was designed as a multi-actor study that collected data from a residential parent (anchor, i.e., the parent to whom a selected target child was officially registered), a child between the ages of 7 and 14 (target child), the anchor's current partner (partner), and the target child's other biological parent in post-separation families (expartner), all measures used in the statistical analysis of this study are based on information provided by the anchor respondents. This approach was chosen to increase the age range of the observed children, as the items measuring the children's mental health in the child questionnaire were only administered to children above the age of 10.
Because the focus of the present study is on children's mental health problems in post-separation families, in a first step, all nuclear families were deleted from the analytical sample (n = 321). Moreover, all families with children below the age of 2 had to be excluded from the sample (n = 8), as the items measuring the children's mental health problems were only administered to parents with a target child who was at least 2 years old. Furthermore, because this study examines interparental conflict after family dissolution, all families were deleted from the analytical sample if the residential parent indicated that he or she had no contact with the target child's other biological parent (n = 47). In a final step, all cases with missing values were excluded from the analysis (n = 91), leaving a final analytical sample that consisted of 1,087 post-separation families.

Measures
Dependent variable. The dependent variable is the child's mental health problems, which was measured using the Strengths and Difficulties Questionnaire (SDQ), a screening questionnaire that is frequently used to assess mental health in children (Goodman & Goodman 2009;Goodman et al. 1998). The SDQ consists of 25 items that cover five different subscales: conduct problems, hyperactivity, emotional symptoms, peer problems, and prosocial behavior (Goodman et al. 1998). The three response categories for these items were not true (0), somewhat true (1), and certainly true (2). For the present study, the analysis drew on information provided by the child's residential parent through the anchor questionnaire. This informant-rated version of the SDQ covers target children between the ages of 2 and 14 and consists of items such as "My child is restless, overactive, cannot stay still for long"; "My child has many worries or often seems worried"; or "My child is rather solitary, prefers to play alone". Following the general instructions for scoring the SDQ, all subscales-except for the prosocial behavior scalewere summed to generate a total difficulties score that ranged between 0 and 40, with higher scores indicating more mental health problems (for information on how to score the SDQ, see https://www.sdqinfo.org). Because the resulting scale (Cronbach's alpha = 0.82) was not normally distributed, but was skewed to the right the natural log was taken.
Independent variable. To determine which physical custody arrangement a given postseparation family was practicing, information provided by the residential parent through the residential calendar was used. In a first step, the respondent was asked to indicate how many days and nights the target child was living with either him or her or with the target child's other biological parent during the first two weeks of a typical month. If those two weeks were not representative of the last two weeks of the month, the respondent received a second calendar that displayed two additional weeks. Drawing on this information, the proportions of time that the child was living with each of his or her biological parents could be calculated. If the child was living between 0% and 29% of the time with one of the parents, the family was identified as practicing sole physical custody (0). Correspondingly, if the child was living between 30% and 50% of the time with each parent, the family was identified as practicing joint physical custody (1).
Moderator variable. To assess levels of post-separation interparental conflict, the study drew on the residential parent's responses to the question: "How often do the following things occur between you and the biological father [mother] of [target child] today?" Levels of interparental conflict were measured using the following five items: "There are tensions or differences of opinion between you and the other biological parent"; "There are heated discussions between you and the other biological parent"; "One of you strongly blames the other"; "You don't want to talk with each other for a while"; and "Arguments get out of hand". All items had response categories ranging from never (1) to very often (5), indicating increasing levels of conflict. Because the resulting mean scale was not normally distributed, the sample was split into two groups based on the scale's mean value: below average (0) and above average (1).
Control variables. To determine the child's gender, each child was identified as either male (0) or female (1). The child's age ranged between 2 and 14 years, and for the statistical analysis, the sample was split into two groups: 2-6 years (0) and 7-14 years (1). To measure both the mother's and the father's educational level, the analysis drew on information about the parents' general school-leaving certificates. Based on the respondents' answers and their gender, the sample was split into three respective groups for mothers and fathers: low educational level (0), medium educational level (1), and high educational level (2). The low educational level group was comprised of respondents who had either no schoolleaving certificate, or who had attained the lowest formal qualification of Germany's tripartite secondary school system. The medium educational level group consisted of respondents who had earned an intermediary secondary qualification. Finally, the high educational level group was made up of respondents who had, at minimum, earned a certificate fulfilling the entrance requirements for studying at a university of applied sciences. To measure the number of years since the family dissolution, the study drew on information about the year in which the survey was conducted and the year in which the parental relationship ended. According to this information, three categories were established: less than 2 years (0), 2-5 years (1), and more than 5 years (2). The quality of the mother-child relationship and the quality of the father-child relationship were assessed using the questions: "How would you generally describe your relationship with [target child]?" and "How would you generally describe the relationship between the biological father [mother] and [target child]?" The response categories for both items ranged from very poor (1) to excellent (10). Because the answers to both questions were mostly positive, and were, therefore, not normally distributed, the respective mean value was calculated and the sample was divided into two groups, considering the parent's gender: worse than average (0) and better than average (1). The descriptive statistics for all variables are displayed in Table 1.

Results
To investigate the relationship between the physical custody arrangement, the level of post-separation interparental conflict, and children's mental health problems, linear regression models were estimated. The results of the regression models are presented in Table 2. The first model displays the relationship between the physical custody arrangement and children's mental health problems (Model 1). The second model shows the full model that includes the moderator variable and all relevant control variables, which previous research has shown to be important for the investigation of physical custody arrangements (Model 2). Finally, the third model adds an interaction term to the regression to test whether the association between the physical custody arrangement and the children's mental health problems was moderated by post-separation interparental conflict (Model 3). The VIF scores ranged between 1.01 and 3.20, which indicates that multicollinearity did not pose a problem in the multivariate analyses.  The results in Model 1 show that children's mental health differed significantly across physical custody arrangements, with children in joint physical custody experiencing fewer mental health problems than children in sole physical custody (β = -0.16, p<0.001). This relationship remained significant even after all control variables were added to the regression in Model 2 (β = -0.10, p<0.01). As explained above, children's mental health problems were operationalized using the Strengths and Difficulties Questionnaire (SDQ). Accordingly, the findings displayed in Table 2 indicate that, on average, children living in joint physical custody arrangements had lower levels of conduct problems, hyperactivity, emotional symptoms, and peer problems than children living in sole physical custody arrangements. Based on these findings, the first hypothesis of this study, which stated that joint physical custody reduces children's mental health problems in post-separation families (H1), was confirmed.
With regard to post-separation interparental conflict, the results show that levels of interparental conflict were significantly related to children's mental health. Compared to children from low-conflict families, children who experienced above-average levels of interparental conflict were found to suffer from significantly higher levels of mental health problems (β = 0.18, p<0.001). Thus, the results of the regression models indicate that experiencing high levels of interparental conflict is indeed harmful for children's mental health. Accordingly, this study's second hypothesis could also be confirmed: i.e., high levels of interparental conflict increase children's mental health problems in postseparation families (H2).
To test whether post-separation interparental conflict moderates the positive relationship between joint physical custody and children's mental health, the interaction term in Model 3 provides the relevant information. The results show that the interaction between the physical custody arrangement and interparental conflict was positive and statistically significant (β = 0.10, p<0.05). To help explain how these results should be interpreted, Figure 1 displays the margins plot for the interaction between the physical custody arrangement and interparental conflict. As Figure 1 shows, children living in joint physical custody arrangements with low levels of interparental conflict had noticeably fewer mental health problems than the comparison group of children in sole physical custody arrangements. In contrast, when levels of interparental conflict were above average, children in joint physical custody arrangements showed approximately the same levels of mental health problems as children in sole physical custody arrangements with similar levels of conflict between the parents. Figure 1 also indicates that the existing differences between the children in the two physical custody arrangements were comparatively small when levels of interparental conflict were high, with children from joint physical custody families having only a negligible advantage in terms of mental health.
Moreover, the conditional effects for the physical custody arrangement and the level of interparental conflict in Model 3 suggest that children living in joint physical custody arrangements that were characterized by comparatively low levels of interparental conflict (β = -0.15, p<0.001) fared significantly better in terms of mental health than children living in joint physical custody arrangements that were characterized by higher levels of interparental conflict. In addition, children from sole physical custody families had more mental health problems when the levels of interparental conflict were above average (β = 0.12, p<0.01), compared to their counterparts from sole physical custody families in which levels of interparental conflict were lower. Taking all these findings into account, the third hypothesis, which stated that joint physical custody reduces children's mental health problems in post-separation families only if the levels of interparental conflict are low (H3), was also confirmed.

Figure 1:
Interaction between post-separation interparental conflict and the physical custody arrangements in predicting children's mental health problems (95% confidence intervals) Note: Family Models in Germany (FAMOD) With regard to the control variables, the results in Models 2 suggest that the time since the family dissolution was significantly related to children's levels of mental health. Children whose parents separated between two and five years ago (β = -0.12, p<0.01) and children whose parents separated more than five years ago (β = -0.09, p<0.05) were shown to have fewer mental health problems than children whose parents separated more recently, that is, less than two years ago. Furthermore, both the quality of the mother-child relationship (β = -0.26, p<0.001) and the quality of the father-child relationship (β = -0.11, p<0.01) were positively related to children's mental health, with better than average parent-child relationships predicting significantly fewer mental health problems in the younger generation. Finally, it should be noted that no significant associations were found between children's mental health and the children's gender, their ages, and their parents' respective educational levels.

Discussion
Joint physical custody is an emerging post-separation care arrangement in Western countries in which children live about equally with both of their parents after family dissolution. Because fathers in joint physical custody are much more involved in their children's upbringing than fathers in more traditional physical custody arrangements, it is often argued that joint physical custody may counter the negative effects of a parental separation or divorce on children's health and well-being. Against this background, the purpose of the present study has been to shed more light on the conditions under which children may profit from living in a joint physical custody arrangement by investigating the association between the physical custody arrangement, the level of interparental conflict, and children's mental health in post-separation families. Because even though previous research has shown that children's exposure to interparental conflict can have detrimental consequences for their overall well-being (Amato 1993;Ayoub et al. 1999;Elam et al. 2016;Gerard et al. 2006), particularly when the conflict is between separated or divorced parents (Dunn et al. 2005), the findings of earlier studies on the relationship between joint physical custody and children's well-being in the context of high levels of post-separation interparental conflict have been contradictory (Elam et al. 2016). Therefore, the present study considered the moderating role of interparental conflict on the relationship between the physical custody arrangement and children's mental health, thus testing whether living in a joint physical custody arrangement is less beneficial for children in the context of high interparental conflict.
The results of the statistical analysis have revealed a positive association between joint physical custody and children's mental health, with children living in joint physical custody arrangements experiencing significantly fewer mental health problems than children living in sole physical custody arrangements. Furthermore, this study corroborated the findings of earlier research that indicated that high levels of interparental conflict in post-separation families have detrimental effects on children's mental health by showing that high levels of interparental conflict were significantly related to children's mental health problems. In addition, when controlling for an interaction between the physical custody arrangement and interparental conflict, the results suggested that practicing joint physical custody is not beneficial for children when conflicts between the separated or divorced parents occur frequently. When levels of interparental conflict were above average the mental health of children living in joint physical custody and sole physical custody arrangements converged noticeably, with only minor differences between the two care arrangements in terms of children's mental health problems.
Accordingly, this study has demonstrated that the apparent positive influence of joint physical custody on children's mental health is closely linked to other family-related factors and living conditions in post-separation families-in this particular case, to the level of post-separation interparental conflict. Although the statistical analysis has shown that, on average, children in joint physical custody arrangements fare significantly better than children in sole physical custody arrangements if levels of interparental conflict are low, it also appears that growing up in a post-separation family characterized by a high level of interparental conflict can exacerbate children's mental health problems, especially in the context of a joint physical custody arrangement.
This study has a number of strengths, including the use of data from the Family Models in Germany (FAMOD) study, a social survey that provides researchers with a sufficiently high number of post-separation families practicing joint physical custody, as well as detailed information about the general living conditions of the families investigated. Moreover, because FAMOD employed a residential calendar, the statistical analysis did not have to rely on the parents' subjective assessment of the time the target children spent at their mother's and their father's households. Instead, the information from the residential calendar was used to estimate the exact proportions of time children in post-separation families lived with each parent.
However, this study also has some limitations. One limitation is that the FAMOD survey was designed as a convenience sample. Thus, the findings of the statistical analysis are not representative of all post-separation families in Germany. Nevertheless, a comparison of several of the respondents' key socio-demographic characteristics (e.g., their age and educational levels) between FAMOD and other surveys that are representative of the German population (e.g., the German General Social Survey (ALLBUS)) revealed that the distribution of these characteristics was quite similar . A related limitation is that the analytical sample consisted mainly of post-separation families with comparatively low levels of interparental conflict. As a result, the variable that measured post-separation interparental conflict was dichotomized at the mean to compare families with relatively low and relatively high levels of conflict. Consequently, this study may underestimate the true impact that interparental conflict can have on the mental health of children, as the families included in the FAMOD study may be a positively selected group, particularly in terms of the quality of the interparental relationship, due to the survey's conception as a convenience sample.
Another limitation of this study arises from the use of cross-sectional data in the statistical analysis, because the causal relationship between the physical custody arrangement and the children's mental health cannot be determined based on crosssectional data. Finally, this study had to rely on proxy information provided by the children's residential parent (on, e.g., the children's mental health problems) in order to include children from a wide age range in the statistical analysis. This approach, however, can be problematic considering that the parents' assessment of their children's mental health may differ significantly from the children's own assessment (see, for example, Davis et al. 2007). On the other hand, because "children may lack the necessary language skills, the cognitive abilities to interpret the questions and a long-term view of events" (Theunissen et al. 1998: 387), it is not unusual for social science research to be based on parental proxy information.
In summary, the present study has provided convincing evidence for the assumption that joint physical custody is not a "one-size-fits-all model" that is suitable for all postseparation families. In public and scientific debates, it is often argued that maintaining a high level of contact with both parents after separation or divorce is in the best interests of the child (Bastaits & Pasteels 2019;Bender 1994;Fabricius et al. 2018). Whereas this is likely to be the case for a large proportion of children living in post-separation families, researchers and practitioners should not forget that the potentially positive effects of joint physical custody greatly depend on a variety of other living conditions that influence children's health and well-being. As the prevalence of joint physical custody increases throughout Western countries, the need for caution is becoming more pressing.
And, although the present study focused exclusively on the positive effects of joint physical custody on children's mental health, researchers and practitioners should also acknowledge that joint physical custody is associated with several characteristics that may negatively impact children's mental health. These factors include feelings of instability, frequent transitions between the parental households and the associated logistical difficulties, the need to adjust to different parental households, and increased family complexity (for summaries, see, for example , Bauserman 2002;Turunen 2017;Vanassche et al. 2013). Moreover, when investigating the relationship between different physical custody arrangements and children's health and well-being, researchers should always consider the possibility that positive associations may be the result of selection processes among post-separation families (Cancian et al. 2014;Juby et al. 2005;Kitterød & Lyngstad 2012;McIntosh & Chisholm 2008;Sodermans et al. 2013;Turunen 2017). The present study, for instance, found that parents in joint physical custody families reported lower levels of interparental conflict after separation or divorce, and that they had, on average, higher educational levels than parents practicing sole physical custody.
Although joint physical custody appears to be a viable alternative to more traditional physical custody arrangements, and to be in alignment with changing social norms, it is a demanding care arrangement not just for parents, but for children as well (Marshall 2017). Thus, for children to benefit from frequent contact with both of their parents after family dissolution, a certain set of framework conditions must be met. This study has shown that, on average, children in joint physical custody families with comparatively low levels of interparental conflict have significantly fewer mental health problems than children in sole physical custody families. However, this study has also demonstrated that one key prerequisite for practicing this type of care arrangement is the parents' ability to communicate and to cooperate in a way that does not harm their children. If the parents do not have this ability, living in a joint physical custody arrangement is not more beneficial for children's mental health than living in a sole physical custody arrangement.
Mothers and fathers sharing parenting time equally after separation or divorce may appear to be an appropriate arrangement given the swiftly changing gender roles in Western countries. However, this type of physical custody arrangement may not be in the best interest of the child if the level of interparental conflict is high and overt. If parents are unable to manage their conflicts, but still want to practice joint physical custody, it seems advisable for them to seek help from professionals, such as youth welfare officers or mediators, to achieve the basic level of cooperation that is necessary for joint physical custody to contribute to children's health and well-being. However, more research is needed to critically investigate other living conditions and factors that may moderate the relationship between joint physical custody and various forms of children's outcomes, not least to inform and support policy-makers and decision-makers in the legal context. Among these factors are the parents' involvement in their children's upbringing prior to separation or divorce, their overall parenting skills, and the children's living conditions in the two parental households.