Abstract
The goal of this study is to better understand the characteristics of men who act as primary caregivers of maltreated children. We examined differences between male primary caregivers (fathers) for youth involved in the child welfare system and female primary caregivers (mothers). We conducted secondary data analyses of the National Survey of Child and Adolescent Wellbeing-II (NSCAW-II) baseline data. Overall, primary caregiving fathers and mothers were more similar than different, though a few differences were revealed. Compared to mothers, fathers tended to be older and were more likely to be employed, with a higher household income and older children. Fathers and mothers did not differ in terms of depression or parenting behavior, but there was evidence that mothers have more problems with drug use compared to fathers. Mothers also reported higher levels of internalizing and externalizing problems in their children compared to fathers. Children with male primary caregivers were more likely to have experienced physical abuse but less likely to have experienced emotional abuse or witnessed domestic violence than children with female primary caregivers. These findings may help to inform researchers, practitioners, and policymakers on how to address the needs of male caregivers and their children.
Keywords: caregivers, child welfare, child maltreatment, psychopathology, parenting
Introduction
In 2013, there were an estimated 3.5 million referrals—involving over six million children—to the child welfare system (CWS) in this country (U.S. Department of Health and Human Services, 2015). Children in the child welfare system are at high risk for the recurrence of maltreatment (Burns et al., 2004; Gewirtz & August, 2008) and caregivers strongly influence child outcomes (Lovejoy, Graczyk, O'Hare, & Neuman, 2000). Though families in which men are the primary caregivers represent a growing segment of the population (Laughlin, 2013), most research examining the role of parents in the lives of maltreated children has focused on mothers (Berger, 2004).
The role of fathers in maltreatment cases and childhood outcomes is complex. Fathers perpetrate almost half (46%) of child maltreatment cases overall (U.S. Department of Health and Human Services, 2005); yet, relative to their time spent in caregiving for young children, fathers are overrepresented as perpetrators of child maltreatment (Guterman & Lee, 2005). Fathers are more likely to engage in severe forms of maltreatment than mothers. For instance, fathers are much more likely than mothers to break or fracture their children’s bones (Starling, Sirotnak, Heisler, & Ames-Eley, 2007). Fathers are also more likely to be the identified perpetrator in fatal maltreatment cases (Schnitzer & Ewigman, 2005). However, despite the increased severity and physical injury associated with maltreatment perpetrated by fathers, one large study found that the involvement of fathers with children in the child welfare system (whether as noncustodial parents or primary caregivers) is not related to re-reports of child maltreatment (Bellamy, 2009). Moreover, a positive relationship between a child and his or her father can lead to improved health, both emotional and behavioral, for the child (Amato & Gilbreth, 1999). Children who experience positive relationships with their fathers demonstrate lower rates of aggressive behaviors, experience less psychological distress, and engage in fewer risky and delinquent behaviors (Black, Dubowitz, & Starr, 1999; Lundahl, Tollefson, Risser, & Lovejoy, 2008).
The lack of attention to fathers in the literature on child maltreatment may stem from misperceptions that male caregivers are “absent, dangerous, or unimportant” (Bellamy, 2009). Fathers in the child welfare system may also be understudied because of longstanding notions and practices regarding caregiving and mothers. Many social services-—including child welfare services—are either intentionally or unintentionally reserved for women (Johnson, 2001). Furthermore, and perhaps more importantly, men are often stigmatized when they access services (Johnson, 2001). Fathers may perceive the agency environment in which child welfare programs are typically provided as either unfriendly or unresponsive, and this may deter them from seeking services. Additionally, child welfare workers may not adequately engage with fathers due to discomfort or lack of knowledge about how to do so (Brown, Callahan, Strega, Walmsley, & Dominelli, 2009).
Parents of children in the child welfare system often experience substance abuse, mental health problems, and domestic violence, as well as other challenges related to low socioeconomic status (Chaffin, Kelleher, & Hollenberg, 1996; Dunn et al., 2002). All of these factors strongly relate to parenting and child mental health and maltreatment (Wilson & Durbin, 2010). For example, caregivers who suffer from substance use and mental health problems are more likely to use maladaptive parenting strategies, which further increase risk for child mental health problems and child maltreatment (Barnard & McKeganey, 2004; Berg-Nielsen, Vikan, & Dahl, 2002; Crouch & Behl, 2001; Prinz, Sanders, Shapiro, Whitaker, & Lutzker, 2009). Yet because of the lack of attention to fathers in the child welfare system, very little is known about their mental health, their parenting strategies, and other aspects of their lives (e.g., financial, employment) that likely have an impact on the wellbeing of the children they care for. A few studies have examined the association between fathers’ substance use and mental health problems and child mental health and child maltreatment, though none of these studies has focused specifically on fathers in primary caregiving roles. In one sample of high-risk families, fathers with substance use and mental health problems were more likely to use corporal punishment and aggressive discipline with their children (Lee, Kim, Taylor, & Perron, 2011; Lee, Perron, Taylor, & Guterman, 2011). Another study examined the mental health and substance use needs of the parents of over four thousand children placed into foster care in Illinois (Jarpe-Ratner, Bellamy, Yang, & Smithgall, 2015). The analyses in this study found that while most parents had low service needs in these two areas, a substantial proportion (about one third) of fathers had especially high substance use-related service needs. The authors noted that more work is necessary in order to better understand the mental health and substance use needs of fathers and how to address them.
Uncovering similarities and differences in female and male caregivers’ demographic characteristics (e.g., age, relationship to the child), as well as in their parenting behavior, mental health, substance use, and use of social services will help to determine how the child welfare system can more effectively engage fathers and address their most significant needs. Some research on mothers may be generalizable to fathers, but these two groups of caregivers differ in important ways. For instance, women are at greater risk for depression compared to men, whereas men are at greater risk for substance use disorders (Eaton et al., 2012). The few studies that have reported on service use among fathers in the general population suggest that fathers are less likely than mothers to seek medical care for themselves or for their children (Moore & Kotelchuck, 2004), and fathers’ attendance at pediatric visits can be inconsistent (Garfield & Isacco, 2006).
The majority of children in the child welfare system (87%) remain in their homes following an investigation of child abuse or neglect (Dolan, Smith, Casanueva, & Ringeisen, 2011) and 8% of these youth have male primary caregivers (Bellamy, 2009). Evidence-based approaches that engage parents to prevent adverse child mental health and maltreatment outcomes may need to be enhanced or implemented differently in order to reach fathers who are acting as primary caregivers, but such adaptations must take into account the needs of these fathers and their children, and how they are similar to or differ from the needs of other family structures. For example, if families with male primary caregivers tend to have children of a different age and gender, or if they tend to live under different socioeconomic conditions, compared to families with female primary caregivers, these families may require different interventions and engagement strategies. At a time when policymakers are calling on men to play more active parenting roles (Connell & Goodman, 2002; Thompson, 2010) and when an increasing number of men are primary caregivers for their families (Laughlin, 2013), it is critical to understand the unique mental health, parenting, and other (e.g., financial, employment) needs of families in which fathers are the primary caregivers, particularly within the highly vulnerable child welfare system population. The goal of this exploratory and descriptive study is to examine characteristics of this understudied family structure in order to address this gap in the literature.
Method
To describe the demographic and parenting characteristics, as well as the mental health and substance use needs, of men identified as primary in-home caregivers for children in the child welfare system, we conducted secondary data analyses of the National Survey of Child and Adolescent Wellbeing-II (NSCAW-II) baseline data. The NSCAW-II is a longitudinal, national probability study of children and families investigated for child maltreatment (see Dolan et al., 2011 for a detailed description of the study design).
Sample
The NSCAW-II baseline sample includes 5,873 children, aged birth to 17.5 years old, who were the subject of completed maltreatment investigations from 2008 to 2009, regardless of the outcome of the investigation. This includes children whose cases were substantiated and unsubstantiated (Kohl, Jonson-Reid, & Drake, 2009). Of the children in the NSCAW-II baseline, 3,636 were living with their primary caregivers (e.g., not in foster care), 334 (9.2%) of whom were identified as male. Our sample includes the 322 fathers and 3,297 mothers who had non-missing data and were identified as in-home primary caregivers, regardless of their relationship to the child (e.g., biological, step, adoptive) and whether other caregivers were present in the home.
Measures
We used demographic information obtained from caseworker and caregiver reports to compare fathers and mothers on child age, race/ethnicity, gender, and type of child maltreatment and number of occurrences; caregiver age, race/ethnicity, education, employment, relationship to the child, and whether they had been identified as a perpetrator of child maltreatment; family income, presence of another caregiver in the home, and number of children in the household.
Primary caregiver depression
was measured by caregiver self-report on the Composite International Diagnostic Interview-Short Form (CIDI-SF) (Kessler, Andrews, Mroczek, Ustun, & Wittchen, 1998). The CIDI-SF is a diagnostic interview that screens for psychiatric disorders as per the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association, 2000). The CIDI-SF is valid and reliable, and classifies respondents as having major depressive disorder with 93% accuracy (Kessler et al., 1998).
Primary caregiver substance dependence
was measured by caregiver report on the Alcohol Use Disorders Identification Test (AUDIT) (Babor, Higgins-Biddle, Saunders, & Monteiro, 2001) (total score 8 or higher indicates alcohol dependence) and Drug Abuse Screening Test (fathers’ Cronbach’s α=0.71; mothers’ α=0.88; DAST-20; total score 6 or higher indicates drug dependence) (Skinner, 1982). These self-report measures have good psychometric properties (Yudko, Lozhkina, & Fouts, 2007). In the NSCAW-II, respondents screened out of the AUDIT after the first three questions if they reported having two or fewer drinks on a typical day and if they reported that they never have six or more drinks on one occasion.
Primary caregiver parenting
was measured by the self-reported Parent-Child Conflict Tactics Scale (CTSPC) (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998). The Nonviolent Discipline (fathers’ Cronbach’s α=0.82; mothers’ α=0.83), Psychological Aggression (fathers’ α=0.63; mothers’ α=0.72), Physical Assault (fathers’ α=0.57; mothers’ α=0.66), and Neglectful Parenting (fathers’ α=0.44; mothers’ α=0.33) subscale scores were used. The CTSPC generally has sound psychometric properties, and low reliability estimates for the Neglectful Parenting subscale have been attributed to the low prevalence of these behaviors (Straus et al., 1998). In the current dataset, however, the CTSPC subscale scores displayed high skewness that common transformations did not adequately address. Therefore, we computed quartile scores for each dimension. We compared caregiver groups on proportion with scores above the 75th percentile for each subscale.
Primary caregiver service use
Caregivers and caseworkers reported whether the caregiver received caseworker-referred services for drug/alcohol or psychiatric problems, or for parenting skills.
Child mental health
The parent-report Child Behavior Checklist (CBCL) and Youth Self Report (YSR) (T.M. Achenbach & Rescorla, 2001) internalizing (fathers’ αs=.99 [CBCL and YSR]; mothers’ αs=.96 [CBCL] and .99 [YSR]) and externalizing problems (fathers’ αs=.99 [CBCL and YSR]; mothers’ αs=.97 [CBCL] and .99 [YSR]) raw scores were used to measure child mental health. The CBCL preschool version was used for youth ages 1.5 to 5 and the school-age version was used for ages 6 to18. The YSR was used for youth over age 11. The psychometric properties of the CBCL and YSR are strong and well established (T. M. Achenbach & Rescorla, 2000).
Data Analysis
SUDAAN, SAS/STAT Survey Sampling Procedures (SAS 9.3) and R accounted for clustering, stratification, and unequal selection probabilities in order to calculate correct variance estimates and preserve the national-representativeness of NSCAW-II. Cases with missing data were deleted listwise. For demographic variables, one-way ANOVAs and chi-square tests were conducted to compare groups. Post-hoc pairwise comparisons enabled interpretation of significant findings for analyses including categorical variables. To assess differences in caregiver mental health, substance use, parenting, and service use, logistic regressions were conducted. Linear, Poisson, and logistic regressions tested differences in child mental health and maltreatment experiences. We report quasi-Poisson standard errors to account for over-dispersion in the count models (Venables & Ripley, 2013). For analyses with potential bias due to small cell sizes, we performed sensitivity analyses using Firth-type penalized logistic regressions (Heinze & Schemper, 2002) in analyses that ignore the complex survey design.
Results
Caregiver Demographic Characteristics
Fathers and mothers differed significantly in age; on average, fathers were approximately five years older than mothers (see Table 1 for caregiver demographics). Fathers and mothers also differed significantly in household income and employment status. Pairwise comparisons showed that the difference in employment was due to a greater proportion of fathers reporting full-time work (P<.001) and a greater proportion of mothers reporting that they did not work (P<.001). About half of fathers were employed full time, while only 30% of mothers were employed full time. Nearly one third of mothers did not work, while only 14.7% of fathers did not work. A similar pattern was revealed with household income. Mothers were more likely than fathers to report their household income at 50% of the poverty line or lower (28% of mothers and 12.4% of fathers; P<.01), and mothers were less likely than fathers to report a household income at 200% above the poverty line or higher (13.6% of mothers and 24.6% of fathers; P=.001). To examine differences in the relationships between caregivers and children, due to small cell sizes, we collapsed step- and adoptive parents into a “non-biological parent” group and we combined the remaining non-biological types of relationships into an “other” category. The vast majority of both mothers and fathers (96% of fathers and 97.7% of mothers) belonged in a third category: biological parent relationships. While the proportion of fathers and mothers who were biological parents did not differ, a significantly greater proportion of fathers were step- or adoptive parents compared to mothers (3.9% of fathers and 1.3% of mothers; P<.01), and fathers were less likely to be “other” non-parental types of caregivers (e.g., sibling, grandparent) compared to mothers (0% of fathers and 0.2% of mothers; P<.001). Fathers and mothers did not differ significantly in race/ethnicity, education level, whether they were identified as a perpetrator of child maltreatment, or whether there was another caregiver present in the home. About half of both groups were White, approximately one quarter were Hispanic, and around one-fifth were Black. The majority of mothers and fathers had a high school education or less, and were not the perpetrator of the alleged child maltreatment. About 35% of each primary caregiver type reported that another caregiver was present in their home.
Table 1.
Caregiver Demographics
| Fathers (N=322) | Mothers (N=3,297) | |
|---|---|---|
| Mean (SD) | Mean (SD) | |
| Age *** | 37.1 (9.7), Min=18,Max=73 | 31.8 (7.8), Min=18,Max=69 |
| Total N | 321 | 3,283 |
| Raw N (Weighted %) | Raw N (Weighted %) | |
| Race/Ethnicity | ||
| Black | 68 (17.6%) | 860 (20.1%) |
| White | 155 (46.8%) | 1,436 (48.7%) |
| Hispanic | 68 (25.3%) | 788 (24.9%) |
| Other | 29 (10.3%) | 207 (6.3%) |
| Total N | 320 | 3,291 |
| Education | ||
| Unknown | 0 (0.0%) | 2 (0.0%) |
| Less than High School | 76 (29.0%) | 1,017 (28.1%) |
| High School | 170 (49.6%) | 1,456 (44.4%) |
| More than High School | 76 (21.4%) | 819 (27.4%) |
| Total N | 322 | 3,294 |
| Employment *** | ||
| Full Time | 147 (49.2%) | 786 (30.3%) |
| Part Time | 52 (17.8%) | 505 (15.3%) |
| Unemployed | 53 (15.6%) | 809 (19.8%) |
| Does not Work | 54 (14.7%) | 1,096 (32.5%) |
| Other | 15 (2.8%) | 99 (2.2%) |
| Total N | 321 | 3,295 |
| Household Income ** | ||
| <50% poverty line | 50 (12.4%) | 940 (28.0%) |
| 50%-<100% poverty line | 81 (32.3%) | 1,019 (34.0%) |
| 100%-200% poverty line | 85 (30.7%) | 720 (24.4%) |
| >200% poverty line | 84 (24.6%) | 353 (13.6%) |
| Total N | 300 | 3,032 |
| Relationship to Child *** | ||
| Biological parent | 304 (96.1%) | 3,172 (97.7%) |
| Step-parent | 9 (3.9%) | 27 (1.3%) |
| Adoptive parent | 0 (0.0%) | 0 (0.0%) |
| Foster parent | 0 (0.0%) | 0 (0.0%) |
| Aunt or Uncle | 0 (0.0%) | 9 (0.2%) |
| Grandparent | 0 (0.0%) | 0 (0.0%) |
| Sibling | 0 (0.0%) | 2 (0.0%) |
| Other Relative | 2 (0.0%) | 28 (0.8%) |
| Total N | 315 | 3,238 |
| Presence of Other Caregiver in Home | ||
| Yes | 85 (34.1%) | 789 (35.9%) |
| No | 164 (65.9%) | 1410 (64.1%) |
| Total N | 249 | 2199 |
| Identified as Perpetrator of Maltreatment | ||
| Not perpetrator | 235 (70.8%) | 2,228 (64.8%) |
| Perpetrator | 78 (29.2%) | 981 (35.2%) |
| Total N | 313 | 3,209 |
Note: P<.05,
P<.01,
P<.001.
Means, SDs, and percentages are all weighted. CG=Caregiver; SD=Standard Deviation.
Child Demographic Characteristics
Children of fathers were older than children of mothers (mean child age=8.7 years vs. 7.4 years) (see Table 2 for child demographic characteristics). However, children of fathers and mothers did not differ significantly in gender, race/ethnicity, or in the number of children present in the household.
Table 2.
Child Demographics
| Fathers (N=322) | Mothers (N=3,297) | |
|---|---|---|
| Mean (SD) | Mean (SD) | |
| Child Age *** | 8.7 (4.4); Min=0, Max=17 | 7.4 (4.8); Min=0, Max=18 |
| Total N | 322 | 3,297 |
| Raw N (Weighted %) | Raw N (Weighted %) | |
| Child Gender | ||
| Male | 170 (50.9%) | 1,685 (50.7%) |
| Female | 152 (49.1%) | 1,612 (49.3%) |
| Total N | 322 | 3,297 |
| Child Race/Ethnicity | ||
| Black | 69 (18.9%) | 932 (22.2%) |
| White | 143 (43.7%) | 1,198 (42.3%) |
| Hispanic | 78 (26.3%) | 927 (28.1%) |
| Other | 30 (11.1%) | 235 (7.3%) |
| Total N | 320 | 3,292 |
| # Children in Household | ||
| 1 Child | 142 (37.1%) | 1,077 (29.1%) |
| 2 Children | 77 (28.6%) | 797 (24.5%) |
| 3 Children | 45 (16.9%) | 682 (23.3%) |
| 4 Children | 24 (10.4%) | 404 (12.2%) |
| 5+ Children | 34 (7.0%) | 337 (11.0%) |
| Total N | 322 | 3,297 |
Note: P<.001. Means, SDs, and percentages are all weighted. SD=Standard Deviation.
Caregiver Depression, Substance Dependence, Parenting Behavior, and Service Use
Table 3 presents details on caregiver depression, substance dependence, parenting, and service use among mothers and fathers. Table 4 presents the results of regressions testing whether mothers and fathers differed in any of these domains.
Table 3.
Caregiver depression, substance dependence, parenting behavior, and service use
| Fathers (N=322) | Mothers (N=3,297) | |
|---|---|---|
| Raw N (Weighted %) | Raw N (Weighted %) | |
| Depression | ||
| No | 235 (82.1%) | 2,093 (73.0%) |
| Yes | 48 (17.9%) | 891 (27.0%) |
| Total N | 283 | 2984 |
| Alcohol Dependence | ||
| No Dependence | 288 (90.5%) | 3,089 (96.5%) |
| Dependence | 21 (9.5%) | 132 (3.5%) |
| Total N | 309 | 3,221 |
| Drug Dependence | ||
| No Dependence | 290 (99.5%) | 2,826 (97.1%) |
| Dependence | 4 (0.5%) | 189 (2.9%) |
| Total N | 294 | 3,015 |
| Non-Violent Discipline | ||
| Below 75th percentile | 255 (82.8%) | 2,588 (75.1%) |
| Above 75th percentile | 56 (17.2%) | 621 (24.9%) |
| Total N | 311 | 3,209 |
| Psychological Aggression | ||
| Below 75th percentile | 251 (77.9%) | 2,611 (74.9%) |
| Above 75th percentile | 60 (22.1%) | 613 (25.1%) |
| Total N | 311 | 3,224 |
| Physical Assault | ||
| Below 75th percentile | 254 (76.0%) | 2,727 (78.0%) |
| Above 75th percentile | 60 (24.0%) | 503 (22.0%) |
| Total N | 314 | 3,230 |
| Child Neglect | ||
| Below 75th percentile | 264 (79.8%) | 2,696 (81.2%) |
| Above 75th percentile | 50 (20.2%) | 528 (18.8%) |
| Total N | 314 | 3,224 |
| Drug/Alcohol Services | ||
| No | 9 (37.6%) | 79 (30.9%) |
| Yes | 16 (62.4%) | 350 (69.1%) |
| Total N | 25 | 429 |
| Psychiatric Services | ||
| No | 4 (20.0%) | 73 (20.9%) |
| Yes | 16 (80.0%) | 318 (79.1%) |
| Total N | 20 | 391 |
| Parent Training Services | ||
| No | 137 (77.2%) | 1,429 (77.9%) |
| Yes | 76 (22.8%) | 888 (22.1%) |
| Total N | 213 | 2,317 |
Note: Means, SDs, and percentages are all weighted. SD=Standard Deviation.
Table 4.
Weighted adjusted logistic regression results comparing mothers and fathers on depression, substance dependence, parenting behavior, and service use
| Depressio n |
Alcohol Dep. |
Drug Dep. |
NV Discipline |
Psych Agg |
Phys Assault |
Neglect | Drug/ Alcohol Svcs |
Psych Svcs |
Parenting Svcs |
|
|---|---|---|---|---|---|---|---|---|---|---|
| OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
|
| Fathers (ref= mothers) |
.64 (.35- 1.17) |
2.93 (.99- 8.66)+ |
.25 (.07- .90)* |
.62 (.33- 1.14) |
.82 (.46- 1.46) |
1.32 (.77- 2.27) |
.94 (.56- 1.56) |
.50 (.20- 1.31) |
.78 (.15- 4.00) |
1.05 (.58- 1.89) |
P=.05,
P<.05
Note: All models adjusted for caregiver age, employment, and household income. OR=odds ratio; CI=confidence interval; Dep.=Dependence; NV=Nonviolent; Agg=Aggression; Phys=Physical; Svcs=Services; Psych=Psychiatric.
Depression (17.9% of fathers and 27% of mothers), alcohol dependence (9.5% of fathers and 3.5% of mothers) and drug dependence (0.5% of fathers and 2.9% of mothers) were reported by a minority of mothers and fathers (see Table 3). The majority of caregivers who were referred for drug/alcohol services received those services (62.5% of referred fathers and 69.1% of referred mothers). This was also true for psychiatric services (80% of referred fathers and 79% of referred mothers). However, less than one quarter of caregivers received recommended parenting services.
To examine differences in mental health, substance dependence, parenting behavior, and service use among fathers and mothers, weighted adjusted logistic regressions (see Table 4) controlled for caregiver age, employment, and income because mothers and fathers differed significantly on these demographic variables. After controlling for caregiver age, employment, and income, fathers were approximately one quarter as likely as mothers to report drug use at a level that met criteria for dependence (see Table 3). The difference between fathers and mothers in terms of alcohol dependence was borderline significant (p=.05), suggesting that fathers may be more likely than mothers to meet criteria for alcohol dependence. We found no other significant differences between fathers and mothers in terms of mental health, substance dependence, parenting behavior, or service use. The models of alcohol and drug dependence and use of psychiatric services have relatively modest cell sizes when cross-tabulated with caregiver type. Compared to the analogous maximum likelihood estimates, the Firth point estimates yielded from sensitivity analyses were slightly smaller in relative magnitude (less than 10%). The penalization also yielded reductions in the standard errors that resulted in smaller p-values for the hypotheses of interest—even when combined with the smaller point estimates—suggesting that the maximum likelihood logistic regression biases due to small cell sizes were modest in our analyses.
Child mental health and maltreatment history
Table 5 displays the mean internalizing and externalizing scores for youth as reported by their parents (CBCL) and the self-reported scores for children ages 11 years and older (YSR). The average number of times that alleged physical or sexual abuse occurred was less than one for children of primary caregiving mothers and fathers, but there was a particularly wide range on these variables for children of mothers due to a handful of outliers (see Table 5). Physical abuse and neglect (physical neglect and failure to provide) accounted for the majority of maltreatment reports among children regardless of the gender of their primary caregiver.
Table 5.
Child mental health and maltreatment history
| Fathers (N=322) | Mothers (N=3,297) | |
|---|---|---|
| Child Mental Health | Mean (SD) | Mean (SD) |
| CBCL Internalizing | 6.7 (5.3), Min=0, Max=33 | 8.6 (7.7), Min=0, Max=50 |
| Total N | 248 | 2,094 |
| YSR Internalizing (age 11+) | 13.9 (11.2), Min=0, Max=41 | 11.7 (8.7), Min=0, Max=54 |
| Total N | 79 | 616 |
| CBCL Externalizing | 10.2 (9.0), Min=0, Max=45 | 12.3 (10.2), Min=0, Max=59 |
| Total N | 250 | 2,094 |
| YSR Externalizing (age 11+) | 14.5 (8.7), Min=0, Max=44 | 14.0 (8.8), Min=0, Max=49 |
| Total N | 79 | 616 |
| Child Maltreatment | ||
| # Times Physical Abuse Occurred | 0.3 (1.7), Min=0,Max=20 | 0.4 (5.9), Min=0,Max=300 |
| Total N | 265 | 2,790 |
| # Times Sexual Abuse Occurred | 0.2 (1.2), Min=0,Max=12 | 0.3 (7.2), Min=0,Max=300 |
| Total N | 274 | 2,857 |
| Type of Maltreatment | Raw N (Weighted %) | Raw N (Weighted %) |
| Physical | 66 (29.7%) | 517 (21.8%) |
| Sexual | 20 (6.8%) | 182 (7.3%) |
| Emotional | 6 (1.0%) | 101 (5.9%) |
| Physical Neglect (Failure to Provide) |
22 (8.8%) | 239 (9.0%) |
| Neglect (Lack of Supervision) | 44 (19.3%) | 607 (24.3%) |
| Witness Domestic Violence | 20 (3.1%) | 348 (8.8%) |
Note: Means, SDs, and percentages are all weighted. CG=Caregiver; SD=Standard Deviation, Min=minimum, Max=Maximum; CBCL=Child Behavior Checklist; YSR=Youth Self Report.
Table 6 presents result from linear, Poisson, and logistic regressions testing differences in mental health and maltreatment history among children of male and female primary caregivers. In these regression models, we controlled for child age and gender and caregiver employment and income due to previous findings on significant group differences and because of established gender differences in child mental health (Card, Stucky, Sawalani, & Little, 2008; Nolen-Hoeksema & Girgus, 1994). Based on parent reports on the CBCL, children of male primary caregivers had lower levels of internalizing and externalizing problems compared to the children of female primary caregivers. However, primary caregiver gender was not significantly linked to internalizing and externalizing symptoms as reported by children older than 11. There were no significant caregiver differences in the number of times a child was reported to have experienced sexual or physical abuse. The reported number of physical and sexual abuse instances each contain several large, outlying values. While the point estimates in Table 6 are sensitive to these outliers, the lack of statistical significance was corroborated by ordered logistic models that recoded the outcomes to take the values: no instances of physical/sexual abuse; 1 instance; 2 or more instances.
Table 6.
Regression results comparing mothers and fathers on mental health and maltreatment history of their children
| CHILD MENTAL HEALTH (Linear regression) |
FREQUENCY OF ALLEGED CHILD MALTREATMENT (Poisson regression) |
TYPE OF ALLEGED CHILD MALTREATMENT (Logistic regression) |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CBCL Internaliz ing |
YSR Internaliz ing (age 11+) |
CBCL Externali zing |
YSR Externalizi ng (age 11+) |
# Times Physical Abuse Occurred |
# Times Sexual Abuse Occurred |
Physical | Sexual | Emot ional |
Physical Neglect (Failure to Provide) |
Neglect (Lack of Supervisi on) |
Witness Domestic Violence |
|
| B (SE) | B (SE) | B (SE) | B (SE) | B (SE) | B (SE) | OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
|
| Fathers (ref= mothers) |
B=−1.85 (SE=.47) *** |
B=2.42 (SE=1.66) |
B=−2.12 (SE=1.00) * |
B=.44 (SE=1.28) |
B=−.48 (SE=.71) |
B=−1.24 (SE=1.50) |
1.70 (1.03- 2.83)* |
.95 (.5- 1.66) |
.19 (.06- .62)** |
1.36 (.69- 2.65) |
.94 (.48- 1.82) |
.38 (.17- .86)* |
P<.05,
P<.01,
P<.001.
Note: All models adjusted for child age and gender, caregiver employment, and household income. CBCL=Child Behavior Checklist; YSR=Youth Self Report; SE=Standard Error; OR=Odds Ratio; CI=Confidence Interval.
We did find differences in the type of maltreatment experienced by children of female primary caregivers and male primary caregivers. The overall multinomial logistic regression showed caregiver type was significantly related to maltreatment type (p<.001), and post hoc logistic regressions for each specific maltreatment type revealed that this was primarily due to differences in physical abuse (p<.05), emotional abuse (p<.01), and witnessing domestic violence (p<.05). Children of male primary caregivers were 1.7 times more likely than children of female primary caregivers to have experienced physical abuse compared to other types of maltreatment. On the other hand, male primary caregivers’ children were about 20 percent as likely as female primary caregivers’ children to have experienced emotional maltreatment, and 40 percent as likely to have witnessed domestic violence compared to other types of maltreatment.
Discussion
The goal of this study is twofold: to describe fathers who are the primary caregivers of children in the child welfare system; and to understand how these male caregivers are similar to or different from their female counterparts, who, for the most part, have provided the basis for research on families in the child welfare system. We first explored the demographic characteristics of male caregivers and female caregivers and their children. Our findings suggested that these two groups of caregivers were demographically similar in many ways. We found no statistically significant differences between fathers and mothers in race/ethnicity, education level, likelihood of being a biological parent, number of children in the home, presence of another caregiver in the home, likelihood of being identified as a perpetrator of child maltreatment, or child gender. However, we found that the “average” father is older, more likely to be employed, more likely to have a higher household income, more likely to be a step- or adoptive parent, less likely to be a non-parental caregiver (e.g., sibling or grandparent), and more likely to have an older child, as compared to the “average” mother in this sample. There are a number of potential explanations for this finding. For one, when deciding whether a child should remain in his or her home after a maltreatment investigation, child welfare workers may have different standards when the primary caregiver is male rather than female. Research on divorce has shown, for example, that fathers are more likely to receive physical custody of their children when the child is older and when the father is employed (Fox & Kelly, 1995). Additionally, most family structures have a woman as the primary caregiving (as noted above, only 8% of primary caregivers in the NSCAW-II sample are men), so the fathers who feel willing and able to be the primary caregiver may be somewhat unique. Older age and greater financial and employment security (compared to mothers) could give fathers the confidence they need to take on a primary caregiving role. However, given the scarcity of research on this topic, future studies should investigate these and other potential explanations for the observed demographic differences in male and female primary caregivers.
Overall, fathers and mothers were more similar than expected (Eaton et al., 2012) in terms of mental health, parenting behavior, and service use. Drug dependence was more common among mothers compared to fathers, but no other significant differences were revealed. This appears to differ from to findings from Jarpe-Ratner and colleagues (2015), who identified substance use as an area of particular need among both fathers and mothers in the child welfare system: in that study 38% of fathers reported substance use needs versus approximately 10% of fathers in the current study. However, it is worth noting that the sample examined by Jarpe-Ratner and colleagues (2015) consisted of parents whose children had been placed in foster care, whereas the current study excluded such cases, focusing specifically on in-home primary caregivers. Thus, it may not be surprising that our study found caregiver substance use to be less prevalent.
Our finding that mothers and fathers tend to be more similar than different with respect to mental health, parenting behavior, and service use suggests that, in general, the key services and supports needed by and available to mothers are likely needed by and should be made available to fathers. For instance, we found that substantial portions of both mothers and fathers have clinically significant depressive symptoms, and that there may be an unmet need for parenting support in both groups. We did not examine caregivers’ level of engagement in and satisfaction with services, however, so a promising direction for future research would be to determine whether male and female primary caregivers differ in this regard, and then to identify treatment engagement strategies that might work for both genders.
Mothers in our study reported greater mental health needs in their children compared to fathers. However, there were not differences in the extent of mental health needs between youth with female primary caregivers compared to those with male primary caregivers when self-reported by youth (11 and older). The explanation for mothers’ higher ratings of their children’s mental health problems (compared to fathers’ ratings of their children’s mental health problems) may be similar to the potential explanations for fathers’ higher incomes, employment, and age. For instance, this difference may reflect systematic tendencies in decisions by child welfare professionals about whether a child should stay in his or her home following a maltreatment investigation: child welfare workers may feel more confident deciding that a child with mental health needs should stay in his or her home when the primary caregiver is female rather than male. And this finding may also reflect fathers’ hesitancy to serve as primary caregivers in the first place. Additionally, it is possible that this finding stems from a response bias- that is, differences in men and women’s perceptions and ratings of their children’s mental health. For example, some studies have found that, depending on the survey methodology used, men tend to underreport their own mental health symptoms compared to women (Hunt, Auriemma, & Cashaw, 2003; Sigmon et al., 2005). However, we found no research specifically examining whether there is a response bias in how men and women rate their children’s mental health. More research is needed in this area.
Finally, our findings on maltreatment type are important. Children with female primary caregivers were more likely to have experienced emotional abuse and domestic violence, but less likely to have experienced physical abuse, compared to children with male primary caregivers. The greater likelihood of exposure to domestic violence among children with female primary caregivers could be in part related to the fact that most perpetrators of domestic violence are male (Kimmel, 2002); following an investigation, the (usually male) perpetrator may be deemed unfit to serve as the primary caregiver for the child. However, this explanation does not hold for physical and emotional abuse; men are more likely than women to perpetrate physical abuse, and there are no established gender differences in terms of perpetrators of emotional abuse (U.S. Department of Health and Human Services, 2005). Interestingly, there were no differences among male and female caregivers in terms of childrens’ exposure to sexual abuse, where perpetrators are most often male, or neglect, where perpetrators are most often female (U.S. Department of Health and Human Services, 2005). It is possible that some of the differences in child maltreatment—and thus mental health—can be attributed to greater levels of financial stress experienced by families with female primary caregivers, as a result of mothers’ higher unemployment and lower income compared to fathers (Costello, Compton, Keeler, & Angold, 2003; Yoshikawa, Aber, & Beardslee, 2012). While we controlled for employment and income in our analyses, we were not able to examine the potential long-term, cumulative impact that economic stress and unemployment can have on youth (McLeod & Shanahan, 1993). This is another potentially promising line of research.
Limitations
In addition to limitations that have already been noted, such as the potential for our findings to be confounded by other variables and the role of reporter bias or systematic bias in placements, it is important to note limitations related to the exploratory, cross-sectional study design. The cross-gender findings related to behaviors and child outcomes that we have reported should be interpreted with caution, as we were not able to draw conclusions about causality. Longitudinal studies are needed to better understand whether any such gender-based differences predict differential child mental health and maltreatment outcomes. In addition, this study is based on a child welfare sample, which is not generalizable to the general population. The sampling weights applied to the analyses allow us only to generalize to families in the child welfare system in this country.
Implications for Policy and Practice
To our knowledge, this is the first study to explore and report on the characteristics of fathers who are the primary caregivers of children in the child welfare system and to examined similarities and differences between these fathers and their female counterparts. Our findings suggest that there are more similarities than differences between these groups, yet female primary caregivers may require greater assistance with drug use, financial issues, and child mental health problems compared to male primary caregivers. Overall, our findings also indicate that the families with male primary caregivers in the child welfare system appear to be doing relatively well, counter to some incorrect beliefs that fathers are absent, dangerous, or unimportant (Bellamy, 2009). Children and families are therefore likely to be strengthened—not harmed—by enhanced clinical and policy efforts to engage fathers in caregiving.
Acknowledgments
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Institutes of Health (NIH), R03MH101542.
References
- Achenbach TM, Rescorla LA. Manual for ASEBA preschool forms & profiles. University of Vermont Center for Children, Youth, & Families; Burlington, VT: 2000. [Google Scholar]
- Achenbach TM, Rescorla LA. Manual for the ASEBA School-Age Forms & Profiles. University of Vermont, Research Center for Children, Youth, & Families; Burlington, VT: 2001. [Google Scholar]
- Amato PR, Gilbreth JG. Nonresident fathers and children's well-being: A meta-analysis. Journal of Marriage and the Family. 1999;61(3):557–573. [Google Scholar]
- Association, American Psychiatric . Diagnostic and Statistical Manual of Mental Disorders IV, Text Revision (DSM-IV-TR) American Psychiatric Association; Washington, DC: 2000. [Google Scholar]
- Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The Alcohol Use Disorders Identification Test. World Health Organization; Geneve, Switzerland: 2001. [Google Scholar]
- Barnard M, McKeganey N. The impact of parental problem drug use on children: what is the problem and what can be done to help? Addiction. 2004;99(5):552–559. doi: 10.1111/j.1360-0443.2004.00664.x. [DOI] [PubMed] [Google Scholar]
- Bellamy JL. A national study of male involvement among families in contact with the child welfare system. Child maltreatment. 2009;14(3):255–262. doi: 10.1177/1077559508326288. [DOI] [PubMed] [Google Scholar]
- Berg-Nielsen TS, Vikan A, Dahl AA. Parenting related to child and parental psychopathology: A descriptive review of the literature. Clinical Child Psychology and Psychiatry. 2002;7:529–552. [Google Scholar]
- Berger Lawrence M. Income, family structure, and child maltreatment risk. Children and Youth Services Review. 2004;26(8):725–748. doi: 10.1016/j.childyouth.2004.02.017. [DOI] [Google Scholar]
- Black MM, Dubowitz H, Starr RH. African American fathers in low income, urban families: Development, behavior, and home environment of their three-year-old children. Child Development. 1999;70(4):967–978. doi: 10.1111/1467-8624.00070. [DOI] [PubMed] [Google Scholar]
- Brown L, Callahan M, Strega S, Walmsley C, Dominelli L. Manufacturing ghost fathers: the paradox of father presence and absence in child welfare. Child & Family Social Work. 2009;14(1):25–34. doi: 10.1111/j.1365-2206.2008.00578.x. [DOI] [Google Scholar]
- Burns BJ, Phillips SD, Wagner HR, Barth RP, Kolko DJ, Campbell Y, Landsverk J. Mental health need and access to mental health services by youths involved with child welfare: a national survey. Journal of the American Academy of Child and Adolescent Psychiatry. 2004;43(8):960–970. doi: 10.1097/01.chi.0000127590.95585.65. [DOI] [PubMed] [Google Scholar]
- Card NA, Stucky BD, Sawalani GM, Little TD. Direct and indirect aggression during childhood and adolescence: a meta-analytic review of gender differences, intercorrelations, and relations to maladjustment. Child Dev. 2008;79(5):1185–1229. doi: 10.1111/j.1467-8624.2008.01184.x. [DOI] [PubMed] [Google Scholar]
- Carlson MJ. Family structure, father involvement, and adolescent behavioral outcomes. Journal of Marriage and Family. 2006;68(137-154) [Google Scholar]
- Connell AM, Goodman SH. The association between psychopathology in fathers versus mothers and children's internalizing and externalizing behavior problems: a meta-analysis. Psychological bulletin. 2002;128(5):746–773. doi: 10.1037/0033-2909.128.5.746. [DOI] [PubMed] [Google Scholar]
- Costello EJ, Compton SN, Keeler G, Angold A. Relationships between poverty and psychopathology: a natural experiment. JAMA : the journal of the American Medical Association. 2003;290(15):2023–2029. doi: 10.1001/jama.290.15.2023. [DOI] [PubMed] [Google Scholar]
- Crouch JL, Behl LE. Relationships among parental beliefs in corporal punishment, reported stress, and physical child abuse potential. Child abuse & neglect. 2001;25(3):413–419. doi: 10.1016/S0145-2134(00)00256-8. [DOI] [PubMed] [Google Scholar]
- Dolan M, Smith K, Casanueva C, Ringeisen H. NSCAW II Baseline Report: Introduction to NSCAW II. Washington, DC: 2011. p. 19. [Google Scholar]
- Eaton NR, Keyes KM, Krueger RF, Balsis S, Skodol AE, Markon KE, Hasin DS. An invariant dimensional liability model of gender differences in mental disorder prevalence: evidence from a national sample. J Abnorm Psychol. 2012;121(1):282–288. doi: 10.1037/a0024780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fox GL, Kelly RF. Determinants of Child-Custody Arrangements at Divorce. Journal of Marriage and the Family. 1995;57(3):693–708. doi: 10.2307/353924. [DOI] [Google Scholar]
- Garfield CF, Isacco A. Fathers and the well-child visit. Pediatrics. 2006;117(4):e637–645. doi: 10.1542/peds.2005-1612. [DOI] [PubMed] [Google Scholar]
- Gewirtz AH, August GJ. Incorporating multifaceted mental health prevention services in community sectors-of-care. Clinical child and family psychology review. 2008;11(1-2):1–11. doi: 10.1007/s10567-007-0029-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Guterman NB, Lee Y. The role of fathers in risk for physical child abuse and neglect: Possible pathways and unanswered questions. Child maltreatment. 2005;10(2):136–149. doi: 10.1177/1077559505274623. [DOI] [PubMed] [Google Scholar]
- Heinze G, Schemper M. A solution to the problem of separation in logistic regression. Statistics in Medicine. 2002;21(16):2409–2419. doi: 10.1002/sim.1047. [DOI] [PubMed] [Google Scholar]
- Jarpe-Ratner E, Bellamy JL, Yang DH, Smithgall C. Using child welfare assessments and latent class analysis to identify prevalence and comorbidity of parent service needs. Children and Youth Services Review. 2015;57:75–82. doi: 10.1016/j.childyouth.2015.07.019. [DOI] [Google Scholar]
- Johnson W. Engaging nonresident fathers in social work with fragile families. In: Briggs H, Cocoran K, editors. Social Work Practice: Treating Common Client Problems. Portland State University; 2001. pp. 324–338. [Google Scholar]
- Kessler RC, Andrews G, Mroczek D, Ustun B, Wittchen H. The World Health Organization Composite International Diagnostic Interview short-form (CIDI-SF) International Journal of Methods in Psychiatric Research. 1998;7(4):171–185. [Google Scholar]
- Kimmel MS. "Gender symmetry" in domestic violence - A substantive and methodological research review. Violence against Women. 2002;8(11):1332–1363. doi: 10.1177/107780102237407. [DOI] [Google Scholar]
- Kohl PL, Jonson-Reid M, Drake B. Time to leave substantiation behind: findings from a national probability study. Child maltreatment. 2009;14(1):17–26. doi: 10.1177/1077559508326030. [DOI] [PubMed] [Google Scholar]
- Laughlin Lynda. Who's Minding the Kids? Child Care Arrangements: Spring 2011 Current Population Reports. Washington, D.C.: 2013. pp. 70–135. [Google Scholar]
- Lee SJ, Kim J, Taylor CA, Perron BE. Profiles of disciplinary behaviors among biological fathers. Child maltreatment. 2011;16(1):51–62. doi: 10.1177/1077559510385841. [DOI] [PubMed] [Google Scholar]
- Lee SJ, Perron BE, Taylor CA, Guterman NB. Paternal psychosocial characteristics and corporal punishment of their 3-year-old children. Journal of interpersonal violence. 2011;26(1):71–87. doi: 10.1177/0886260510362888. [DOI] [PubMed] [Google Scholar]
- Lovejoy M. Christine, Graczyk Patricia A., O'Hare Elizabeth, Neuman George. Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review. 2000;20(5):561–592. doi: 10.1016/S0272-7358(98)00100-7. [DOI] [PubMed] [Google Scholar]
- Lundahl BW, Tollefson D, Risser H, Lovejoy MC. A meta-analysis of father involvement in parent training. Research on Social Work Practice. 2008;18(2):97–106. doi: 10.1177/1049731507309828. [DOI] [Google Scholar]
- McLeod JD, Shanahan MJ. Poverty, parenting, and children's mental health. American sociological review. 1993:351–366. [Google Scholar]
- Moore T, Kotelchuck M. Predictors of Urban fathers' involvement in their child's health care. Pediatrics. 2004;113(3 Pt 1):574–580. doi: 10.1542/peds.113.3.574. [DOI] [PubMed] [Google Scholar]
- Nolen-Hoeksema S, Girgus JS. The emergence of gender differences in depression during adolescence. Psychol Bull. 1994;115(3):424–443. doi: 10.1037/0033-2909.115.3.424. [DOI] [PubMed] [Google Scholar]
- Prinz RJ, Sanders MR, Shapiro CJ, Whitaker DJ, Lutzker JR. Population-based prevention of child maltreatment: the U. S. Triple p system population trial. Prevention science : the official journal of the Society for Prevention Research. 2009;10(1):1–12. doi: 10.1007/s11121-009-0123-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schnitzer PG, Ewigman BG. Child deaths resulting from inflicted injuries: Household risk factors and perpetrator characteristics. Pediatrics. 2005;116(5):E687–E693. doi: 10.1542/ped.205-0296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Skinner HA. The drug abuse screening test. Addictive behaviors. 1982;7(4):363–371. doi: 10.1016/0306-4603(82)90005-3. [DOI] [PubMed] [Google Scholar]
- Starling SP, Sirotnak AP, Heisler KW, Ames-Eley MLB. Inflicted skeletal trauma: The relationship of perpetrators to their victims. Child abuse & neglect. 2007;31(9):993–999. doi: 10.1016/j.chiabu.2007.02.010. [DOI] [PubMed] [Google Scholar]
- Straus MA, Hamby SL, Finkelhor D, Moore DW, Runyan D. Identification of child maltreatment with the Parent-Child Conflict Tactics Scales: development and psychometric data for a national sample of American parents. Child abuse & neglect. 1998;22(4):249–270. doi: 10.1016/s0145-2134(97)00174-9. [DOI] [PubMed] [Google Scholar]
- Thompson K. Obama steps up fatherhood advocacy with new mentoring initiative. The Washington Post. 2010 [Google Scholar]
- U.S. Department of Health and Human Services Male Perpetrators of Child Maltreatment: Findings from NCANDS. 2005.
- U.S. Department of Health and Human Services. Administration for Children and Families. Administration on Children, Youth and Families. Children’s Bureau Child maltreatment 2013. 2015.
- Venables WN, Ripley BD. Modern applied statistics with S-PLUS. Springer Science & Business Media; 2013. [Google Scholar]
- Wilson S, Durbin CE. Effects of paternal depression on fathers' parenting behaviors: a meta-analytic review. Clin Psychol Rev. 2010;30(2):167–180. doi: 10.1016/j.cpr.2009.10.007. [DOI] [PubMed] [Google Scholar]
- Yoshikawa H, Aber JL, Beardslee WR. The effects of poverty on the mental, emotional, and behavioral health of children and youth: implications for prevention. Am Psychol. 2012;67(4):272–284. doi: 10.1037/a0028015. [DOI] [PubMed] [Google Scholar]
- Yudko E, Lozhkina O, Fouts A. A comprehensive review of the psychometric properties of the Drug Abuse Screening Test. Journal of substance abuse treatment. 2007;32(2):189–198. doi: 10.1016/j.jsat.2006.08.002. [DOI] [PubMed] [Google Scholar]