Abstract
Children's psychological adjustment following parental separation or divorce is a function of the characteristics of the custodial parent, as well as the degree of postdivorce parental cooperation. Over time, custody has shifted from fathers to mothers and currently to joint arrangements. In this retrospective chart review of family court clinic records we examined predictors of custody and visitation. Our work improves on previous studies by assessing a greater number of predictor variables. The results suggest that parental emotional instability, antisocial behavior, and low income all decrease chances of gaining custody. The findings also show that income predicts whether a father is recommended for visitation rights and access to his child or children. Furthermore, joint custody is not being awarded as a function of parental postdivorce cooperation. At issue is whether parental emotional stability, antisocial behavior, and income are appropriate markers for parenting capacity and whether visitation rights and joint custody are being decided in a way that serves the child's best interests.
After divorce, children respond rather predictably with fears about their personal security, anger over the loss of family integrity, concerns about parental well-being, and loss of parental attention and involvement.1 Divorce increases the risk of adjustment problems in children and adolescents2 and doubles the risk of serious psychological and social problems.3 The purpose of a family court clinic in a custody dispute is to provide guidance to family court judges as they decide the family arrangement that is most likely to mitigate such risk.
Over the centuries, custody preference has transitioned from fathers (who claimed the child or children as property) to mothers (with the tender years doctrine presuming the necessity of maternal bonding during a child's early years) and, more recently, to an increased frequency of joint arrangements in which mothers and fathers share custody.4 Custody can be either legal (the right to make decisions relating to the health, education, and welfare of the child) or physical (the right to have the child live with the parent). Furthermore, both of these forms of custody can be sole or joint, with joint legal custody, meaning that both parents have equal rights to make major decisions about their child, and joint physical custody, meaning that the child spends a significant amount of time with both parents. Conversely, sole legal custody leaves major decision-making in one parent's hands, and sole physical custody dictates that the child live primarily with one parent.5
The nationally accepted best-interests-of-the-child (BIC) legal standard requires judges to take the child's best interests into account when awarding custody. This act defines the child's best interests as a composite of the following five factors: the wishes of the child's parent or parents as to custody; the wishes of the child as to the custodian; the interaction and interrelationship of the child with the child's parent or parents, the child's siblings, and any other person who may significantly affect the child's best interest; the child's adjustment to home, school, and community; and the mental and physical health of all individuals involved. The BIC standard has been widely criticized for its lack of guidance for judges in making child custody determinations. In the absence of clear guidance from the law, judges have turned to mental health professionals and custody evaluations for help in discerning children's best interests and, by doing so, family courts have implicitly embraced the value that children's psychological well-being is first on the list of best interests.1
Following divorce or separation, children's adjustment has been found to be significantly influenced by characteristics of the custodial parent(s)6 and the degree of continued parental conflict and hostility,7 particularly when parents use their children to express their anger and are aggressive with one another in front of their children.8 Protective factors cited by Kelly and Emery9 include the psychological adjustment and parenting of custodial parents, the type of relationship that children have with their nonresident parents, and the extent and type of conflict between parents. With respect to characteristics of the custodial parent, prior qualitative surveys have cited parental psychological stability, income, substance use, moral character, and criminal record as influential on custody outcomes; a few quantitative studies have also been conducted to examine what parental characteristics influence custody outcomes. Our study built on prior studies by expanding the number of parent characteristics included, as well as including interparental dynamics and child factors as predictor variables of custody and visitation outcomes; it also quantified all of these relationships.
Several qualitative studies, in the form of surveys of judges and mental health professionals, have been conducted to understand which parental factors are most influential in deciding or recommending custody arrangements. The three major themes that arise in the qualitative literature involve parental mental stability, moral character (e.g., criminality), and financial and material assets. Settle and Lowery,4 in a survey of judges, found the most important parental factor to be mental stability, with each parent's moral character being the fourth most important of 20 parental factors. Felner and colleagues,10 in a survey of judges and attorneys, found the most frequently cited criteria to include emotional stability and financial resources. Keilin and Bloom,11 in a survey of mental health professionals, found the psychological stability of each parent to be one of the most important factors. Ackerman and Ackerman,12 in a survey of psychologists, found the most influential parental factors to include substance use, psychological stability, a history of psychiatric hospitalizations, a criminal record, and treatment with psychiatric medication. Wallace and Koerner,6 in a survey of judges, found parental substance use, mental health, financial resources, and criminal activity to be the most influential factors, and Bow and Quinnell13 surveyed psychologists who rated parental mental health, capacity to provide for the child financially and materially, and moral fitness as highly significant. A survey of attorneys and judges found the top reasons for child custody evaluation referrals to be parental conflict, mental instability, allegations of physical or sexual abuse, and alcohol abuse.14 Of the six surveys cited, mental stability was an important factor in all six, moral character and criminality a factor in five, and financial and material resources a factor in four. What is not revealed in these qualitative studies is the quantified effect of each of these factors on custody outcomes.
A few quantitative studies, in the form of retrospective court case reviews, have also been conducted. The major themes that arise from them involve the importance of parental income, education, and maturity on custody awards. Kunin et al.15 reviewed disputed child custody cases to analyze predictors of judicial physical custody decisions. They found that predictors of decisions favoring the mother included general-impression variables, such as physical appearance, social skills, social adjustment, and maturity. Predictors of decisions favoring the father included the father's maturity, a lack of drug abuse history, and the stability of his living arrangement. In a review of Wisconsin divorces, Cancian and Meyer16 found that the likelihood that a father would gain some degree of physical custody (either shared or sole) increased as his proportion of the couple's total income increased and that as the mother's income increased, there was a decreased likelihood of paternal sole custody. Christensen and colleagues17 also reviewed court cases and similarly found that fathers with higher incomes were more likely to be granted custody. Fox and Kelly18 reviewed divorce cases and found that sole paternal physical custody was less likely when the mother had a college degree.
These four quantitative studies reveal a pattern in which parental maturity, financial assets, and level of education influence custody outcomes. Lacking in these quantitative studies, but included in the current study, is the variety of predictor parental variables. This variety allows not only for an expanded examination of relevant parental factors in determining custody but also provides the ability to control for potential confounding variables. For example, it might be argued that income, education, and the general-impression variables noted in these quantitative studies merely serve as surrogate markers for other parental factors, such as mental stability, moral character, and criminality. In our study, we controlled for such potential confounding.
With respect to parental conflict as it relates to children's adjustment, the results have been mixed, with some studies reporting no association between postdivorce conflict and later adjustment in young adults19 and other studies finding that postdivorce conflict has more adverse effects than does conflict in married families.20
Hypotheses
Based on our literature review, the following hypotheses were generated:
A parent is less likely to be awarded sole physical custody or visitation when that parent has a low income, a low education level, a history of arrests, a history of involvement of family protective services (including a history of restraining orders and involvement of child protective services), a history of outpatient mental health treatment, a history of substance use, or a history of psychiatric hospital admissions.
If at least one child involved in the custody dispute carries a psychiatric diagnosis, has a chronic medical problem, or has had mental health treatment, physical custody shifts toward the mother.
Joint physical custody is more likely when there is reported communication between parents and less likely when both parents characterize their separation as hostile.
Methods
Data Collection
A chart review of the records of a family court clinic was conducted after this research was approved by the International Review Board (IRB) of Cambridge Health Alliance. The court clinic calls on doctorate-level psychologists and licensed social workers who provide recommendations to family court judges on custody and visitation arrangements. The clinic is located in a large metropolitan area in the Northeast region of the United States, serves a population of families that is less affluent than the general population, and is diverse in language, race, and ethnicity.
Each clinic chart documents the evaluation of a case in which custody and visitation were contested. As part of the clinic's protocol, intake questionnaires are given to both parties on arrival at the clinic. These questionnaires assess each parent's satisfaction with the current custody and visitation arrangement; the presence or absence of co-parenting communication; each parent's characterization of their separation and divorce; the sociodemographic, mental health, and legal histories of the caregivers, including ethnicity, education, income, health, substance use, and legal involvement; and each parent's understanding of whether the child or children have received mental health treatment or a psychiatric diagnosis or have a chronic medical condition. Each chart was reviewed for the presence of two completed intake questionnaires, one by each biological parent, with completion defined as response to at least 75 percent of the questions. After selecting those charts with two completed questionnaires, each chart was further reviewed for the presence of a clinic evaluation with recommendations on custody and visitation; only those cases with clear recommendations were included in the analysis.
Intake questionnaires provided the predictor variables and one outcome: current custody. The clinic evaluations provided one or two outcomes: recommendations on custody and visitation, depending on whether a parent was petitioning for a change in custody or a change in visitation. All predictor variables and outcomes were coded into a database by the primary author.
Charts of 1,100 custody and visitation cases from 1999 to 2009 were reviewed. Of these, 275 (25%) had two completed intake questionnaires by both biological parents. Of the 275 charts with two completed questionnaires, 202 (73%) had clear clinic recommendations on physical custody and visitation.
Predictor Variables
Predictor variables included eight parental factors, two interparental communication and hostility factors, and a single child factor. The following parent factors were recorded: race and ethnicity (white or nonwhite); income, excluding child support received (more or less than $20,000 per year, which is an approximation of the 2011 federal poverty level for a three- to four-person family. The United States Department of Health and Human Services cited $18,530 and $22,350 as the 2011 poverty-level incomes in the 48 contiguous states and the District of Columbia for three- and four-person families, respectively21); level of education (high school graduate or did not complete high school); a history of or present outpatient mental health treatment (yes, no); a history of psychiatric hospitalizations (yes, no); a history of or present substance use (yes, no); a history of arrests (yes, no); and a history of family protective service involvement, including both a history of restraining orders and involvement of child protective services (yes, no). The following two interparental dynamics were recorded: whether both parents characterized their separation or divorce as hostile and whether at least one parent reported communication with the other on at least one aspect of their child's or children's functioning. The three child factors were whether at least one child involved in the custody dispute had received mental health treatment, as reported by at least one parent; whether at least one involved child had received a psychiatric diagnosis, as reported by at least one parent; and whether at least one parent had reported that at least one child involved in the dispute had a chronic medical problem.
Outcome Measures
The outcome measures were current custody arrangement (before clinic evaluation) and the clinic's recommendations on custody and visitation. Current custody was determined by reviewing both the clinic's evaluation and the intake questionnaires, in which each parent was asked about the current custody arrangement. If there was a disagreement between the parents on their understanding of the current custody arrangement, the clinic's report was used to provide the closest impartial estimate of the actual arrangement before recommendations; of note, parents agreed on the current custody arrangement more than 90 percent of the time. We operationalized custody and visitation recommendations into the following seven outcome variables: maternal sole physical custody before the clinic's recommendation; maternal sole physical custody, as recommended by the clinic; paternal sole physical custody before the clinic's recommendation; paternal sole physical custody as recommended by the clinic; joint physical custody before the clinic's recommendations; joint physical custody as recommended by the clinic; and paternal visitation rights as recommended by the clinic. The family court clinic evaluators consider a physical arrangement to be joint when both parents have custody of the child at least 35 percent of the time (whereas a joint arrangement before the clinic evaluation has no such time quantification). Of note, maternal visitation rights, as an outcome, were studied, but the number of cases in which the clinic made recommendations on this parameter was insufficient to generate meaningful comparisons.
Data Analysis
We first described the sample in two ways: by reporting the rates of the predictor variables for each of the 202 divorced couples and by assessing both the custody arrangement before clinical evaluation and the custody and visitation recommendations made after clinical evaluation. To identify the independent relationships between the predictor variables and our outcomes of interest (current and recommended custody and recommended visitation), we estimated multivariate logistic regression models for each outcome and accounted for missing data in the regression analyses by multiple imputation. The significance and direction of the odds ratios for each predictor variable can be interpreted as the independent association between the predictor variable and the outcome of interest, after adjustment for all other measured factors.
Results
Most of the study subjects reported their ethnicity as Caucasian (Table 1). The mothers in the study population reported higher income and better education than did the fathers; incomes for the mothers and the fathers ranged from less than $5,000 to more than $50,000. The mothers were more likely than the fathers to report a personal history of mental health treatment(s) and psychiatric hospitalizations. Conversely, the fathers were more likely to report past or present substance use and a history of arrests and restraining orders, when compared with the mothers.
With respect to child mental health treatment history, in more than half of the cases (52%) in this study, at least one parent reported that at least one child involved in the custody dispute had received mental health treatment. In relation to interparental dynamics, roughly two thirds of the parents (67%) agreed that their separation or divorce was hostile, although, in more than half of the cases (55%), at least one parent reported communication with the other on some aspect of their child's functioning (emotional, academic, behavior, matters of discipline, or health).
Maternal physical custody was more likely both before and after the clinic's recommendation, although the rates of maternal custody decreased and the rates of paternal sole custody increased after the clinic's evaluation (Table 2). There were nine cases in which more paternal involvement was recommended by the clinic (nine maternal sole custody cases were changed: five converted to joint physical custody and four to sole paternal physical custody). The clinic recommended less joint custody than was present before their evaluation. In the case of recommended sole custody after the clinic's evaluation, noncustodial fathers were much more likely to be recommended for visitation than were noncustodial mothers, by a factor of almost two to one.
Predictors of Custody
In Table 3, we present only the significant odds ratios for predictor variables in seven separate logistic regression analyses. Each model estimated the relationship between the outcome variable of interest (custody before clinic's recommendation, clinic custody recommendation, and visitation rights) and the parental-level, interparental dynamic, and child-level factors described earlier. Analysis 1 identified that the mothers were less likely to have sole physical custody before clinic evaluation if they reported a history of psychiatric hospitalization or the involvement of family legal and protective services (including restraining orders and the involvement of child protective services) or if they reported a lower income. They were also less likely to have sole custody if at least one parent reported communication with the other on some aspect of the child's functioning (Table 3, Analysis 1). The mother was more likely to have sole custody before evaluation if the father reported a history of arrests and if both parents characterized their separation or divorce as hostile.
The clinic was significantly less likely to recommend that sole physical custody be given to the mother if she reported a history of involvement of family protective services (including child protective services and restraining orders), if she reported past or present substance use, or if the father reported a lower level of education (Table 3, Analysis 2). Maternal sole physical custody was significantly more likely to be recommended when the father reported a history of arrest and when both parents characterized their separation and divorce as hostile.
The father was significantly less likely to have sole physical custody before clinic evaluation if at least one parent reported at least one child with a history of past or present mental health treatment(s), if the father reported a history of arrests, or if the mother had a lower level of education (Table 3, Analysis 3). The father was significantly more likely to have sole custody before evaluation if the mother had a history of involvement with family protective services (including restraining orders and the involvement of child protective services) or if the mother reported a lower income.
The clinic was significantly less likely to recommend that sole physical custody be given to the father if he reported a history of arrests or if at least one parent reported at least one child with a history of past or present mental health treatment(s). Sole paternal custody was significantly more likely to be recommended when the mother reported a history of involvement with family protective services (including restraining orders and the involvement of child protective services) and if at least one parent reported the presence of psychiatric diagnosis in at least one of the involved children (Table 3, Analysis 4). Of note, the child's or children's age(s) were not recorded in this study; the literature has shown that the presence of young children reduces the probability that the father will be granted sole custody.16
A joint physical custody arrangement, before clinic evaluation, was significantly more likely if one parent reported communication with the other on some aspect of the child or children's functioning and significantly less likely if the father reported a history of arrests or if the mother reported a history of outpatient mental health treatment (Table 3, Analysis 5). The clinic was significantly more likely to recommend joint physical custody in cases in which one parent reported communication with the other on some aspect of the child or children's functioning and the clinic was significantly less likely to recommend joint physical custody when the father reported a history of arrests (Table 3, Analysis 6).
Predictors of Visitation
In cases in which maternal sole physical custody was recommended, the clinic was significantly more likely to recommend visitation rights for the father when the mother reported an income lower than $20,000 per year. Visitation rights were significantly less likely to be recommended for the father when his income was lower than $20,000 per year (Table 3, Analysis 7). The sample size for paternal sole physical custody was too small to generate meaningful estimates for assessing predictors of maternal visitation outcome.
Discussion
As found in previous studies, parental mental health, moral character and criminality, and income influenced custody recommendations. Contributions of this study to the literature include the following main findings: mental health treatment history is relevant for mothers seeking custody; moral character and criminality are relevant for fathers seeking custody; and income is relevant for fathers seeking visitation rights. In the absence of rigorous studies of the impact of the child's mental health on custody outcomes, this study shows that, when there is at least one child with mental health needs, there is a shift in custody away from the father. A joint arrangement was predicted by interparental communication and not by the nature of that communication. We found no studies on the association between parental communication and the likelihood of joint custody.
These findings can be interpreted in the context of a broad body of literature on factors related to a favorable clinical outcome in children after separation and divorce. These factors include the presence of good parenting skills on the part of the custodial parent and postdivorce parental cooperation.22 One of the main resources that courts consider in making child custody determinations is information about parents and their parenting (known as parental attributes), which refers to parenting abilities and deficits. However, there is no consensus as to any one set of parental factors that an evaluator should address as reflecting parenting skills or parent attributes. As divorce and custody proceedings are governed by state law, many states have stipulated specific parental factors that should be considered, including, but not limited to, mental health, substance use, criminal history, moral fitness, and ability to provide for basic needs.5
The custody literature shows that, in fact, children's adjustment after divorce is a function of parent mental health, moral character and criminality, and income and material assets. Although the findings in this study suggest that evaluators and judges are weighing these factors in deciding custody and visitation, they also suggest that consideration of these factors may depend on whether the mother or father is being evaluated. To date, this appears to be the first study to reveal this parent-dependent dynamic.
Mental Health and Substance Use
Two decades of research have indicated that children who have a parent with mental illness are at significantly greater risk of multiple psychosocial problems.23 Children with a mentally ill parent have elevated risks of learning problems, developmental delays, attention deficit, social skills deficits, substance abuse, anxiety disorders, and somatic complaints.24 The main challenges for mentally ill parents center on their capacity and motivation to manage their mental illnesses while assuming parenting responsibilities.25 Ultimately, the literature suggests that the determination of the effect of a mental illness is based, not on a particular parental condition or diagnosis, but on how that condition affects the daily personal and parental functioning of the parent affected and on the effects of that functioning on the child.26 Mowbray and colleagues,27 for example, showed that current symptomatology and community networking are more indicative of parenting capacity than is a specific diagnosis, and Warner and colleagues28 found that the severity of symptoms, chronicity of illness, and lower adaptive functioning of parents are more closely related to poor outcomes in children than are particular diagnoses.
Most of the research on the effects of mental illness on parenting has focused on mothers. Oyserman and colleagues23 found that mothers with serious mental illness had significantly less adequate parenting skills than did those who did not have a mental illness. In considering the postdivorce period specifically, Guidubaldi and Perry29 found that children's adjustment difficulties were related to parental reports of their own maladjustments, including lethargy, frustrated dependency, depression, low self-esteem, and increased smoking and use of alcohol. According to Jenuwine and Cohler,30 the significance of parental psychiatric illness for custody decisions can be evaluated only by examining the impact of the illness on the child; for example, the disruptive effect of the repeated separations that accompany serious mental illness.
In comparing mothers and fathers, Ayoub and colleagues31 found that a positive relationship is evident between a child's emotional distress and the presence of mental illness or substance abuse in the mother, whereas there is an inverse relationship between the child's distress and mental illness or substance abuse in the father. This finding was explained by the hypothesis that, in the setting of mental illness or substance abuse, a father's contact becomes more highly supervised and limited, whereas the degree of contact between the mother and child is unaltered.
In the current study, maternal physical custody was significantly less likely before clinic evaluation if the mother reported a history of psychiatric hospitalizations (Table 3, Analysis 1). Also of note, of the nine cases in which the clinic recommended a shift in physical custody toward the father (from sole maternal custody to either sole paternal or joint custody), the mother reported a history of mental health treatment, suggesting the importance of this variable in the clinic as well. With respect to the child's best interests, a mother's report of previous psychiatric hospitalizations may be a useful predictor of decreased parenting capacity, as these hospitalization(s) may serve as a marker of psychiatric symptom severity, which has been found to be positively correlated with a child's emotional distress. In addition, such hospitalization(s) may pragmatically affect parenting capacity, given the physical separation that hospitalization requires, as suggested by Jenuwine and Cohler.30 Of note, the father's mental health was not found to be a significant predictor of custody or visitation (perhaps as a result of the increased-supervision theory put forward by Ayoub and colleagues31).
In custody arrangements before clinic evaluation and in arrangements as recommended by the clinic, at least one parent reporting at least one child with a history of mental health treatment predicted a shift in custody toward the mother. Although no studies were found by the authors on the significance of the child or children's mental health on judicial or evaluator custody decision or preference, it might be speculated that the overall preference for maternal custody when a child has mental health needs, as observed in this study, represents a persistence of the tender years doctrine and the perception that children with mental health needs should be in the mother's custody.
Regarding substance use, many studies have shown associations of externalizing (acting out and disruptive behavior) and internalizing (anxiety and depression) problems in children with the presence of parental substance abuse.32,33 Some literature suggests that parental substance use can have a direct adverse effect on attachment, family dynamics, and relationships and can be associated with neglect and emotional, sexual, and physical abuse.34 Other studies suggest that there are few or no direct effects of a parent's substance use on child outcomes, as the contexts of use often contain multiple and related stressors, such as co-occurring psychological and environmental problems (e.g., divorce or financial difficulties) which, it is speculated, have a direct influence on adverse child outcomes.35
In this study, the clinic was significantly less likely to award the mother physical custody if she reported a history of past or present substance use (Table 3, Analysis 2). The literature suggests that substance use is likely to reduce the probability of parental fitness in the mother, whether by direct effect or by other variables associated with both substance use and adverse child outcomes (e.g., divorce and financial difficulties as noted earlier). As with psychiatric hospitalization, substance use in the father was not found to be predictive of custody or visitation, and increased supervision with visitation may be explanatory of the latter outcome, as suggested by Ayoub and colleagues.31
Criminal History and Moral Fitness
In this study, maternal physical custody, both before and after the clinic evaluation, was significantly more likely when the father reported a history of arrests. Likewise, both before and after the clinic evaluation, paternal sole custody was significantly less likely with a history of arrests. If mothers reported the involvement of family protective services (including both child protective services and restraining orders), they were significantly less likely to have custody before the clinic evaluation and significantly less likely to be recommended custody by the clinic. Likewise, such reported involvement of mothers significantly predicted the likelihood of paternal sole physical custody before the clinic evaluation and the likelihood of a postevaluation recommendation of paternal sole physical custody by the clinic.
Criminal behavior by the father or by both parents has been found to be significantly associated with delinquency in boys.36 Otto and colleagues37 reported that one of the most consistent findings is that parents who engage in antisocial behavior tend to have children who exhibit behavioral problems that include aggression, delinquency, and other externalizing actions.
In this study, a father's reporting arrests shifted custody toward the mother. Regarding mothers, the involvement of family protective services (which, as seen in Table 1, tended to involve child protective services more than restraining orders) shifted custody toward the father and, by definition, such involvement of child protective services related to parenting capacity and the best interests of the child.
Ability to Provide Basic Needs and Income
In this study, maternal physical custody before the clinic evaluation was significantly less likely when the mother reported a lower income (Table 3, Analysis 1). Also, before the clinic evaluation, paternal sole custody was significantly more likely when the mother reported a lower income (Table 3, Analysis 3). As suggested by the literature, a lower income for the custodial parent (usually the mother) may in fact predict poorer adjustment for the child after divorce. However, also as suggested by the literature, this dynamic is likely to be a function of the particular needs of the child and how the custodial parent is able, both pragmatically and emotionally, to negotiate financial strain, if present.
The decline in standard of living after separation and divorce is greater for divorced women than for divorced men, as women typically have lower incomes and the extra expense of childrearing, with some research showing that economic stability is an important predictor of postdivorce child functioning.7
Of the six surveys of judges and mental health professionals described earlier, four showed that income and ability to provide materially for the child were important factors in custody awards. Previous reviews of the literature dealing with the effects of the father's absence on children have argued that delinquency and cognitive impairment may be primarily attributable to the depressed socioeconomic status (SES) in families headed by single parents, particularly mothers. Children can sense economic difficulty and may be emotionally affected by it, with concerns about availability of physical necessities such as food and clothing, or the feasibility of college for adolescents.38
Pearson and Thoennes39 found that children's adjustment (in terms of depression, aggression, delinquency, social withdrawal, and somatic complaints) was affected by five variables, with financial stress being the third most influential. However, Kalter and colleagues40 found that SES was negatively related, but not significantly so, to children's adjustment. According to Sales and colleagues:
The fact that a potential custodian has access to greater financial resources does not imply that a custodial award in his or her favor will necessarily further the best interests of a child. From a social science perspective, child-specific variables, such as age, prior standard of living, and the presence of unique physical needs, handicaps, or disabilities that necessitate financial expenditures, would be more valid indicators of such interests [Ref. 38, p 31].
In addition, other literature suggests that the effect of income on children's adjustment following divorce is mediated by other correlated variables, such as the possibility that parents under economic stress may be less supportive and less available.7
A review of the literature on income and visitation suggested that nonresident fathers' payment of child support has consistently shown that such support is linked to children's well-being, educational progress, and good health.41 In the current study, fathers' visitation and access rights were predicted only by reported income, with rights significantly more likely to be recommended if the mother reported a lower income and significantly less likely to be recommended if the father reported a lower income (Table 3). The literature suggests that increased child support from the nonresident father is favorably related to a child's well-being following divorce.
Education
In this study, the fathers were less likely to have custody before clinic evaluation if the mother reported a lower level of education. After the clinic's evaluation, maternal sole custody was less likely to be recommended if the father reported a lower level of education. These results do not support the hypothesis that level of education correlates positively with the likelihood of receiving custody. They also stand in contrast to the finding by Fox and Kelly18 that paternal sole physical custody is less likely when the mother has a college degree. The authors of this article were unable to find previous research on the impact of education, as an isolated factor, on parental fitness.
Joint Custody and Postdivorce Parental Conflict
A meta-analysis of 33 studies comparing joint physical and sole maternal custody indicated that children in joint physical custody arrangements were better adjusted than were those in sole maternal custody arrangements; in addition, joint-custody parents also reported less past and current conflict than did sole-custody parents.42 Two other studies found joint physical custody to be more beneficial to children and adolescents than sole maternal custody when conflict was low, but these benefits were suppressed by high levels of conflict.43,44
With respect to postdivorce conflict, findings in studies of the impact of postdivorce parental conflict on children's adjustment have been mixed.9 Booth and Amato19 reported no association between postdivorce conflict and later adjustment in young adults while Hetherington20 found that postdivorce conflict had more adverse effects on children than did conflict in married families. Ultimately, the literature suggests that postdivorce conflict is more likely to be destructive when parents involve their children in the conflict.8 However, if parents are able to encapsulate their conflict and not put their children in the middle, the children do not differ from children whose parents have low or no conflict.20
In this study, joint physical custody (both before and after clinic evaluation) was significantly more likely when at least one parent reported communication on some aspect of the child's functioning. Ostensibly, communication may be appreciated as an appropriate predictor of a joint arrangement, which requires such. However, the literature generally suggests that it is the nature of the communication (hostile versus cooperative and the degree to which children are exposed to such hostility), rather than the presence of communication, that is the relevant predictor of child adjustment following divorce or separation,45 and this association was not found in this study. Given the potential benefit of joint arrangements for children's adjustment, more studies are needed to examine whether postdivorce parental cooperation (including cooperation to encapsulate interparental conflict) predicts the likelihood of a joint award of custody.
The presence of communication, as discussed in the preceding paragraph, was also found to predict a decreased likelihood of maternal custody before clinic evaluation. Because the presence of communication did not predict a greater likelihood of paternal custody, this result presumably reflects a shift toward joint custody, as we have discussed.
Visitation and Contact with the Nonresident Parent
According to Gould and Martindale, “There is increasingly robust research literature about the importance of nonresidential parent involvement in the lives of divorced children” (Ref. 46, p 184) and Bauserman42 found that limited contact with their father affects various aspects of children's lives, such as self-esteem, scholastic achievement, emotional stability, and psychological well-being. In the context of low conflict, it appears that frequent visits between nonresident fathers and children are associated with better child adjustment; but where interparental conflict is intense, more frequent visits can be linked to poorer adjustment, hypothesized by Hetherington and Kelly3 to be the result of increasing opportunities for more direct exposure of the children to parental aggression and pressures. More specifically, in low-conflict situations, Wallerstein and Kelly47 found that with boys and younger children in particular, frequent and regular contact with nonresident fathers is associated with more positive adjustment.
Conversely, when intense conflict persists between parents, frequent contact with the nonresident parent has been found to be associated with poorer adjustment, presumably because of more opportunities for parental hostility to be expressed in front of children during exchanges.48 Likewise, children do not benefit from frequent contact with nonresident parents who are mentally ill or abusive or those whose parenting is compromised.49
As noted previously, in the current study, fathers' visitation and access rights were predicted only by reported income, with rights significantly more likely to be recommended if the mother reported a lower income and significantly less likely to be recommended if the father reported a lower income (Table 3). Although the literature suggests that increased child support from the nonresident father is favorably related to child well-being following divorce, it also suggests that postdivorce parental cooperation is highly related to children's adjustment after divorce; contrary to the literature, the two interparental dynamic predictor variables (cooperation and hostility) were not found to predict the recommendation of paternal visitation rights.
Conclusions
An analysis of 202 custody arrangements from a family court clinic in the Northeast region of the United States found that mental health professionals and family court judges considered parental mental health, income, and antisocial behavior to be important in determining custody awards. A growing body of literature suggests that these factors are also predictive of a child's adjustment following divorce. The analysis also found that the presence of a child with mental health needs and the presence of postdivorce parenting communication (and not cooperation) were important in determining custody awards; it is not clear that awarding custody based on these factors is predictive of a child's postdivorce adjustment. Finally, visitation rights for the father, a factor that has been shown to be predictive of the child's adjustment, were found to be a function of income.
The findings in this study support the following statements: mental health, as gauged by a report of psychiatric hospitalizations and substance use, predict a shift away from maternal custody, but the same is not true of paternal custody; antisocial behavior, as gauged by a report of arrests, is deemed relevant for a father's capacity to be the custodial parent but not for the mother; a father is more likely to be recommended visitation rights if the mother reports a lower income, and he is less likely to be recommended visitation rights if he reports a lower income; a mother is less likely to have custody if she reports a low income; the presence of at least one child with mental health needs is associated with a shift in custody away from the father; and, finally, joint custody is predicted only by the reported presence of communication between parents and is not a function of the nature of that communication (i.e., whether the parents' describe their separation or divorce as hostile).
Limitations
These findings should be read in the context of several limitations. Generalizability is limited in this study, given the particular characteristics of the study sample (less educated, less affluent, a higher rate of involvement with family law services, substance use, mental illness, and arrests, compared with the general population); however, these results may be generalizable to the segment of the population that utilizes a family court clinic in custody proceedings. As a result of this limitation, the findings in the study may be applicable only to a specific subset of the general population (for example, the segment that turns to a family court clinic to facilitate custody proceedings).
All independent predictor variables in this study are based on parent self-report, and parents are likely to have an interest in minimizing certain aspects of their history for the purpose of portraying greater parental fitness (e.g., underreporting past psychiatric or legal history). Mitigating this limitation is the fact that the clinic routinely corroborates self-report using collateral sources of information (e.g., reviewing medical and legal records). This practice may encourage parents to be more forthcoming and accurate in their self-reporting practices.
The study presented here is a chart review and therefore has all of the limitations that this research method entails: incomplete documentation, missing charts, information that is unrecoverable or unrecorded, difficulty with interpretation of information found in the documents (e.g., jargon, acronyms, illegibility), problematic verification of information (e.g., self-report in this case), and variance in the quality of information recorded.50
Future Directions
Regarding parental mental health, future studies should examine the effects of specific psychiatric diagnoses and symptoms, along with their severity, on parenting capacity. Competencies (of which parenting is one of many) are not simply dependent on a person's abilities, but on the match or mismatch between such abilities and the demands of the situation.51 The effect of specific psychiatric signs and symptoms should therefore be investigated with respect to their relationship to parenting capacity, as should substance use (both type and severity). Additional studies might also investigate the effects of parental compliance with psychiatric care and available support systems (e.g., family, friends, and community) as mitigating factors when parental mental illness and substance use are present.
Future studies might also examine whether mental illness and substance use are more relevant factors for mothers than for fathers regarding parental capacity, as this study suggests, or whether this differential association is more an artifact of increased supervision for fathers, as discussed previously. Finally, with respect to the relevance of mental health in custody disputes, future studies should examine whether maternal custody in the case of a child with mental health needs serves the child's best interests or merely represents a persistence of the now-outdated tender years doctrine.
Regarding criminal behavior and moral character, this study found a report of such to be a relevant factor for fathers but not for mothers. This finding stands in contrast to the literature that suggests that criminal behavior should be regarded as a relevant negative factor for both parents. Future studies should examine the influence of parental criminal behavior on child outcomes and whether there is a differential effect on children if the criminal behavior is present in the mother or the father.
Concerning financial resources, reported lower income predicted a shift away from maternal custody in this study. Although it might be posited that low income in mothers serves as a surrogate marker for other areas of dysfunction, even when controlling for other potentially confounding variables, income was specifically associated with a loss of custody for mothers. Future studies may investigate more systematically the specific effects of lower income (both in the mother and the father) on child outcomes and, further, examine the finding that visitation rights for the father were predicted only by income and not by other variables that may be seen as influential in deciding whether fathers should have access to their children.
Acknowledgments
Gratitude is extended to the Family Court Clinic for allowing access to their records after IRB approval was received. Gratitude is also expressed to Dr. Cynthia Telingator and Dr. J. Richard Ciccone for their support and guidance and to Dr. Stephen Herman for his guidance.
Footnotes
Disclosures of financial or other potential conflicts of interest: None.
- © 2013 American Academy of Psychiatry and the Law