Brief Treatment and Crisis Intervention Advance Access originally published online on September 11, 2007
Brief Treatment and Crisis Intervention 2007 7(4):305-321; doi:10.1093/brief-treatment/mhm021
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Intimate Partner Violence and Child Maltreatment: Overlapping Risk
From the Department of Family Medicine, University of North Carolina at Chapel Hill School of Medicine (Zolotor), Department of Social Medicine, University of North Carolina at Chapel Hill School of Medicine (Theodore, Runyan), Department of Pediatrics, University of North Carolina at Chapel Hill School of Medicine (Theodore, Coyne-Beasley, Runyan), and Department of Internal Medicine, University of North Carolina at Chapel Hill School of Medicine (Coyne-Beasley)
Contact author: Adam Zolotor, Department of Family Medicine, University of North Carolina at Chapel Hill, CB# 7595, Chapel Hill, NC 27599-7595. E-mail: ajzolo{at}med.unc.edu.
Studies of intimate partner violence (IPV) and child maltreatment (CM) have examined the association between IPV and physical abuse. Children in homes with IPV may also experience other forms of CM. The objective is to determine the prevalence of CM in homes with and without IPV using cross-sectional analysis of survey data of mothers with partners (n = 1,232). The Conflict Tactics Scale and Parent Child Conflict Tactics Scale were used to determine IPV, physical, psychological, sexual abuse, and neglect. Mothers reporting IPV (either man to woman or woman to man) report 2.57 times the odds of physical abuse compared to those not reporting IPV (95% CI 1.11–5.97). Moms reporting IPV report 2.04 times the odds of neglect. Those reporting IPV report 9.58 times the odds of psychological abuse (95% CI 4.27–21.49). Mothers reporting IPV report 4.90 times the odds of sexual abuse (95% CI 0.43–55.67). IPV is associated with all forms of CM in this sample. Providers of IPV services for women with children should also assess for all forms of child maltreatment.
KEY WORDS: intimate partner violence, child maltreatment, abuse, domestic violence, family violence
Intimate partner violence (IPV) and child maltreatment (CM) are significant public health problems. Nearly 1.5 million women and 900,000 men in the United States are affected by physical IPV within any given year (Tjaden & Thoennes, 2000). Nearly 3 million children are reported for CM each year in the United States, 1 million of reported children have substantiated maltreatment reports (United States Department of Health and Human Services [U.S. DHHS], 2006). Child exposure to IPV is also very common. The National Crime Victimization Survey estimated that more than 50% of homes with calls to law enforcement for severe domestic violence contained children less than 12 years of age (United States Department Of Justice [U.S. DOJ], 2006). A more accurate picture of the intersecting landscape of family violence is elusive.
Early research in the area of family violence co-occurrence stems from exploration of specific populations, namely, information about the families of children reported to child protective services (CPS) for maltreatment and information from women seeking temporary refuge in battered women's shelters. A review by Edleson and colleagues of seven local and state CPS studies found that IPV occurred in 26–73% of the families reported to CPS. Edleson's review also reported on 25 studies of families where the woman was a victim of IPV, with a co-occurrence for CM rate for half of these studies in the range between 30% and 60%. In most of these studies, children suffered some type of physical abuse as reported by the mothers (1999). Similar results were found in a second review of studies. Appel and Holden summarized 17 studies of battered women with a median rate of co-occurrence of 40% and a range from 10% to 100% (1998). Due to the significant selection bias in these studies, it is difficult to know how these findings reflect the relationship of IPV and CM in community- and population-based settings. For example, a review of four representative community samples reported a co-occurrence rate between 5.6% and 11% (Edleson, Mbilinyi, Beeman, & Hagemeister, 2003).
Several recent studies have examined the co-occurrence of IPV and CM. The most recent report of co-occurrence for CPS samples is based on the National Survey of Child and Adolescent Well-Being, a national probability study of 5,501 children ages 0–14 who were randomly selected from among the families who entered the U.S. child welfare system between October 1999 and December 2000. IPV was assessed with the physical scale of the Conflict Tactic Scale (CTS) asked of the main caregiver (usually the mother) about her experiences as a victim. Hazen and colleagues reported a lifetime prevalence of physical IPV of 45% for mothers of children reported to CPS and a past year occurrence rate of 29% (Hazen, Connelly, Kelleher, Landsverk, & Barth, 2004). Similar results were obtained by English and colleagues in the state of Washington, where IPV was present in almost half (47%) of CPS cases accepted for investigation and assigned a moderate to high level of risk (2005). A reanalysis of the 1985 National Family Violence Surveys (NFVS) data by Tajima reported that 19.4% of the surveyed families reported some type of violence (2004). Of these families, 78% suffered IPV directed only toward the woman, 15% CM alone, and 7% had both IPV and CM. This study is limited by only studying physical abuse of children and not considering psychological abuse, sexual abuse, and neglect.
Most empirical studies to date suggest that the presence of IPV increases the risk of CM, with some variance based on the perpetrator, type of violence, and type of maltreatment. Researchers re-examining data from the 1985 NFVS concluded that the presence of IPV in a home increased the odds that CM was present in the home as well (Berger, 2005; Ross, 1996; Tajima, 2000). Rumm and colleagues analyzed families on active duty in the U.S. army using a centralized database of CM and IPV reports. They found that reported spouse abuse increased the rate ratio of child abuse twofold after adjusting for the age and rank (proxy for economic status) of the military parent. Interestingly, they found that this association was true for physical abuse (rate ratio [RR] = 2.36) and sexual abuse (RR = 1.46), but not for neglect (2000). In another study, researchers used a large high-risk cohort participating in a CM prevention program (n = 2,544) to assess the risk of IPV in the first 6 months of a baby's life and the risk of child physical abuse, psychological abuse, and neglect up to age 5. This study found that IPV increased the odds of child physical abuse (OR 3.38), psychological abuse (OR 2.20), and neglect (OR 2.18) (McGuigan & Pratt, 2001). Another recent study by Lee and colleagues evaluated IPV as a risk factor for reported CM among a cohort of children considered at risk of CM. IPV and control variables were measured at child age 6 years. In the presence of IPV, this analysis showed a threefold increase in the odds of physical abuse but not in the risk of neglect or total maltreatment as determined by a research-based coding system applied to CPS reports (2004).
The principle objective of this paper is to describe the overlap of IPV and CM in a two-state population-based sample and estimate the risk of CM in the presence of IPV. We hypothesize presence of IPV will increase the risk for all types of CM (physical abuse, psychological abuse, neglect, and sexual abuse).
 |
Methods |
|---|
 TOP Methods ResultsDiscussionPractice ImplicationsFundingReferences |
|---|
Study Design and Sampling
An anonymous telephone survey on child rearing was administered to a probability sample of North and South Carolina mothers. This survey has previously been described in a report on the epidemiology of harsh physical punishment (Theodore et al., 2005). Slightly over half of the sample came from a sampling frame of working telephone numbers of households with children under the age of 18 purchased from GENESYS Sampling Systems, Fort Wayne, Pennsylvania, a vendor supplying telephone numbers for research. The use of this purchased list was intended to restrain the costs of the study. However, midcourse, it became clear that poor families were underrepresented in the purchased sampling frame. A second sampling frame using random-digit dial to ensure that a broad range of families with differing incomes was included. Sample weights were assigned to each subject in the study. The sample weights took into consideration the disproportionality in the sample arising from the sampling process, and the sample was weighted to the state census characteristics for race and income (Theodore et al., 2005).
The sample was derived from 8,262 phone numbers. Of which, 4,611 numbers were ineligible (i.e., numbers not in service, business) and 1,335 were of unknown eligibility (maximum call attempts resulted in no answer); 881 were eligible but did not complete an interview (refused, unavailable, etc.). A combined response rate of 52% (n = 1,435) was achieved for the two sampling frames following the computational guidelines of the American Association for Public Opinion Research for response rate three (American Association for Public Opinion Research, 2005). Using this method for calculating the response rate, the ineligible and unknown eligibility numbers are subtracted from the total numbers (8,262 – 4,611 – 1,335 = 2,316). Of the numbers with unknown eligibility, this method assumes that the same proportion of numbers is eligible as the proportion for numbers with known eligibility (0.33 x 1,335 = 441). This is considered a conservative estimate of eligibility. The total adjusted response rate is the number of respondents divided by the total number of estimated and known eligible numbers [1,435/(1,435 + 881 + 441) = 52%]. For this study, a subset of the full data set, including only those mothers reporting the presence of a husband or partner in the home was used (n = 1,232). Sampling weights were applied to the entire sample prior to creating a subset of the data.
Data Collection
The survey was conducted from August to December of 2002 using Blaise, a computer-assisted telephone interview software package by Statistics Netherlands. A minimum of 12 callback attempts were placed. Eligible households were located in the North and South Carolina, with a child under 18 in the household, and a mother or female guardian present. At the start of the interview, after eligibility was determined, phone numbers were purged from the interviewer's computer to provide anonymity to the respondent. Where more than one eligible child resided in the home, a referent child was selected by a computerized algorithm to achieve a distribution of children in terms of birth order and age. If the parent or guardian spoke only Spanish, the interview was administered in that language. In preparation for Spanish-speaking families, the survey was translated into Spanish and independently back translated. All respondents were given phone numbers for parenting resources as part of a routine "debriefing." The survey was anonymous, and after connection with a potential respondent the phone number and identity of the respondent was purged from the computer system. The respondent was therefore not traceable, and this allowed for a truly anonymous survey without requirements for reporting suspected abuse as the identity of the respondent could not be known by the interviewer. This study was approved by the University of North Carolina School of Medicine Institutional Review Board.
Measures
The survey was designed to assess parenting behaviors, disciplinary practices, and family and community characteristics. The parenting behaviors were assessed using questions selected from the Parent Child Conflict Tactics Scales (PCCTS) (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998) with additional questions suggested by international focus groups. CarolinaSAFE was part of a consortium of studies from six countries with common measures and similar procedures. Each country held a minimum of one focus group of women with children to learn more about the local practices regarding parent–child and intimate partner conflict resolution, translation of those practices, and social acceptability of those practices. The focus groups were led by trained facilitators using a common focus group guide. Focus groups were transcribed and reviewed by the researchers. The focus group results were used to add six new items to the PCCTS and to separate one item to account for a range of conflict resolution tactics used in other countries (Sadowski, Hunter, Bangdiwala, & Munoz, 2004). The PCCTS has been widely used to assess parenting practices (Hassan, Refaat, El-Sayed, & El-Defrawi, 1999; Theodore et al., 2005; Zolotor & Runyan, 2006). The PCCTS has moderate reliability with physical assault scale alpha = 0.55, psychological aggression scale alpha = 0.60, nonviolent scale alpha = 0.70, and the poorest reliability for the neglect scale = 0.22 and severe physical assault = –0.2 (Straus et al., 1998). Several studies have shown good construct validity for the PCCTS (Caliso & Milner, 1992; Jouriles & Norwood, 1995; Straus et al., 1998). One study has shown fair discriminant validity for the PCCTS by demonstrating that mothers with a history of maltreating a child had higher scores on the scales for neglect and psychological aggression. However, in this same study, mothers with a history of maltreating a child had higher scores for nonviolent discipline (Bennett, Sullivan, & Lewis, 2006). This undermines the discriminant validity of nonviolent discipline in this instrument, but those questions are not used in the current analysis.
The PCCTS asks parents about a variety of positive and negative discipline techniques, including positive and negative reinforcement, corporal punishment, and potentially abusive behaviors. In addition, the supplemental questions for neglect and sexual maltreatment were asked from the PCCTS. The core questions from the PCCTS were asked of both the responding mother's behaviors toward the index child and the behavior of her partner toward the index child (two separate questions). The PCCTS does not assign blame to the neglect questions, but asks them from the perspective of basic childhood needs. Similarly, the sexual maltreatment questions ask about various types of sexual abuse by adults, without assuming caregiver relationships. The use of anonymous surveys to assess potentially abusive caregiver behaviors utilizes the work of Straus, which shows that caregivers are willing to report harsh and socially disapproved forms of discipline (Straus & Gelles, 1990).
IPV was assessed using the CTS. As with the PCCTS, the CTS normalizes interpersonal conflict and then queries the respondent on a variety of techniques used in conflict resolution (Straus, Hamby, Boney-McCoy, & Sugarman, 1995). The CTS has shown better reliability than the PCCTS with alpha for husband-to-wife violence = 0.87 and alpha for wife-to-husband violence = 0.88 (Straus, 1979). Concurrent validity has been shown to be moderate for verbal aggression and violence subscales of the CTS by surveying college students and their parents regarding the parent's behavior over the last year. For verbal aggression, the correlation was 0.43–0.51 and for violence it was 0.33–0.64 (Straus, Gelles, & Steinmetz, 1980). The respondent is asked how many times she has used each technique in the past month (or not in the past month but in the past year). Techniques for conflict resolution include discussing, arguing, belittling, hitting, threatening, and in the case of man to woman forced sex. Each question is asked of the responding mother pertaining to her behavior toward her partner and her partner's behavior toward her (separate questions).
Analysis
Analyses were performed using STATA 8.2 (College Station, TX). Descriptive and analytic statistics were weighted based on socioeconomic status and race/ethnicity in order to create a sample that represented North and South Carolina. Analysis included demographic variables and inventories of IPV, harsh physical punishment, harsh psychological punishment, neglect, and sexual maltreatment.
Types of potential maltreatment were grouped into the following summary types: use of harsh physical punishment, neglect, harsh psychological punishment, and sexual maltreatment. Harsh physical punishment includes parents who report beating, burning, kicking, hitting with an object elsewhere than the buttocks, or shaking a child less than two more than once in the last month. Harsh physical punishment also includes hitting a child frequently on the buttocks with an object (>5 times in the last year). Harsh psychological punishment includes threatening to leave or abandon, threatening to kick out of the house, locking out of the house, or calling stupid, ugly, or useless. Neglect includes lack of needed medical care in the last year, lack of food in the last month, leaving a child less then 10 years old home alone without an adult for greater than 1 hr in the last month, or inadequate supervision such that a child got hurt more than once in the last month. Sexual maltreatment includes touching a child or a child who is made to touch an adult in a sexual way or forcing a child to have sex with an adult in the past year. For each inventory, respondents were dichotomized to those who endorsed one or more of the behaviors listed and those who endorsed none. The parenting inventories include all parents reporting one or more of the aforementioned commissions or omissions of care. Some families may endorse more then one behavior on an inventory.
IPV was measured using questions regarding harsher types of violence from the CTS. An affirmative answer to a question covering "insulting, belittling, and demeaning" was so common in this sample that this behavior was included only if it was reported to occur more then 10 times in the last year and this was categorized as frequent belittling IPV. Verbal threatening and physical violence (slapping, kicking, or hitting with a fist) were categorized as severe IPV. In addition, forcing sex (man to woman only) was included as severe IPV. The variables for IPV were measured as partner to mother, mother to partner, either, or both and categorized as a dichotomous variable for analysis.
Survey-weighted logistic regression was used to assess the simple relationships between each category of potential CM and each category of IPV. Survey-weighted multivariate logistic regression was used with the CM categories as dependent variables, considering the category and direction of IPV as the independent variable of interest. Control variables in the multivariate analysis included child age, maternal education, and receipt of public assistance as a proxy for poverty and race. These control variables were chosen as known or suspected risk factors for CM and IPV that might confound the relationship as measured (Belsky, 1993; Straus & Gelles, 1990). Results of bivariate and multivariate analyses are reported as odds ratios (ORs) and adjusted ORs with 95% confidence intervals (CIs). It is important to note that the CM outcomes are not mutually exclusive, and in many cases, they do overlap.
|
Results |
|---|
|
TOPMethodsResultsDiscussionPractice ImplicationsFundingReferences |
|---|
Child and household characteristics are summarized in Table 1. Total sample size was 1,232. The child age was evenly distributed. The children were just over half boys. Most respondents (95.2%) were biological mothers of the child subject of the survey. All respondents in this subset reported husbands or partners in the home. Nearly 96% of these partners were husbands, 2.7% boyfriends, and 1.4% other (mostly other female partners). The sample was relatively well-educated; roughly 4% reported less then a 12th grade education and over 40% reported college degrees or beyond. Almost 85% of the respondents identified their race/ethnicity as white, 9.6% as black. Just over 80% of respondents answering the income questions reported a household income of more than $40,000 in the prior year and 160 respondents were unable or unwilling to answer the questions related to income. The receipt of public assistance was used as a proxy for poverty in further analysis as the data were more complete and indexed to relative percents of the federal poverty level. In total, 14.5% of respondents reported that they or one child had received Women Infant and Children's nutrition program, Temporary Assistance to Needy Families, or Medicaid in the prior year. The sample underrepresented poor North and South Carolinians; however, the data were weighted to reflect the income distribution of the Carolinas.
|
Table 2 shows the weighted and crude percents for IPV and CM categories. Parents reporting "insulting, belittling, or demeaning" is exceedingly common in this sample; over 40% of mothers reported belittling their partners in the last year and 37% reported being belittled by their partner. Given that this was normative behavior and relatively less related to CM in our preliminary analysis, we focused on homes with more frequent verbal abuse. In total, 5.5% of mothers reported doing this more then 10 times the last year, and 7.0% reported that their partner did this to them 10 or more times. This is categorized as frequent belittling IPV. In total, 5.7% of mothers reported perpetrating severe IPV (hitting or threatening) against their partner and 7.4% reported being the victims of severe IPV (hitting, threatening, or forced sex). Frequent belittling and severe IPV in either direction (as a cumulative total), and in both directions (bidirectional violence) was assessed. Five percent of mothers reported frequent belittling IPV in both directions and 3.3% reported severe IPV in both directions, perhaps representing the most violent homes in the sample. See Table 2 for details.
|
Summary rates of CM are also reported in Table 2. Nearly 11% of mothers reported that the child has received one or more forms of harsh physical punishment, 10.9% reported that the child has received one or more forms of harsh psychological punishment, 0.9% reported that the child is a victim of one or more forms of sexual abuse, and 11.0% reported one or more major omissions in care consistent with neglectful parenting. Nearly 30% of children were victims of one or more types of potential CM. It should be noted from Table 2 that many categories have small numbers of affirmative responses leading to potentially unstable estimates. This is true for child sexual abuse, and some subcategories of severe IPV as indicated on Table 2.
Tables 3 and 4 show crude and adjusted ORs for each category of IPV and the odds of each type of CM comparing mothers reporting that type of IPV with those who do not. The ORs in Table 4 are calculated using survey-weighted logistic regression while controlling for child age, child gender, maternal education, and poverty. These variables are chosen because they may confound the relationship between IPV and CM. The most clear and consistent result from this study is the relationship between all levels and types of IPV and harsh psychological punishment. Mothers reporting frequent belittling IPV in either direction have 5.01 times the odds of reporting harsh psychological parenting compared to mothers not reporting frequent belittling IPV (95% CI 2.47–10.16). With severe IPV in both directions, the OR is 9.81 (95% CI 2.99–32.25). The results are nearly equivalent in the adjusted analysis (OR 8.44, 95% CI 3.60–19.82).
|
|
Likewise, the relationship between neglectful parenting and IPV is also strong and consistent for nearly all measures of IPV. For example, with any partner to mother IPV, mothers are at 2.71 times the odds of reporting neglect (95% CI 1.24–5.92). In the case of mother to partner severe IPV, the odds of reporting neglect is 3.16 times higher (95% CI 1.21–8.21). Again, results are similar in the adjusted analysis. In several categories of IPV, the relationship is similar in magnitude but lacks precision for statistical significance in predicting neglectful parenting.
The relationship with harsh physical punishment is less consistent. For many categories of IPV, those with IPV had higher odds of reporting harsh physical punishment. Most comparisons lack statistical significance or meaningful differences among mothers who reported IPV compared to those who did not report IPV. However, those mothers who reported mild or severe IPV in both directions are at 2.63 times the odds of reporting physical abuse (95% CI 1.00–7.07). Adjusted analyses again yield similar results (OR 2.57; 95% CI 1.11–5.97).
Sexual maltreatment is rarely reported by mothers in this sample. There are therefore no statistically significant comparisons in the homes of children with and without IPV. It is, however, worth noting that the directionality and magnitude of comparisons is relatively consistent. Mothers who reported more IPV reported higher rates of sexual maltreatment though none of these differences rise to the level of statistical significance.
Figure 1 demonstrates the prevalence of CM in homes with IPV. Over 55% of mothers who reported IPV reported one or more forms of CM. Figure 2 demonstrates the overlap between IPV and CM. In our study, CM as defined is much more common, but roughly half of the homes with IPV reported CM. Nearly one third of homes with CM reported IPV. Over one third of mothers reported either IPV or CM, demonstrating that family violence is exceedingly common. The occurrence of family violence in this sample is quite high when compared to federal statistics for child abuse and neglect (U.S. DHHS, 2006). The reason for this is twofold. First, federal data only capture those children brought to the attention of authorities, and our study methodology allows us to identify children who may never present to social services. Also, the use of the PCCTS allows us to include children whose victimization may not rise to the level of social service concern. For example, the way we defined child supervision may be considered neglectful in some social service jurisdictions and not in other jurisdictions.
|
|
|
Discussion |
|---|
|
TOPMethodsResultsDiscussionPractice ImplicationsFundingReferences |
|---|
IPV between parents of North and South Carolina children is common. Nearly 40% of mothers report that they have been belittled by their partners and a similar percentage has belittled their partners. Fortunately, frequent belittling (>10 times in the last year) and more severe forms of interpersonal conflict are much less common; in just 3.7% of households has the mother reported hitting her partner and in just 3.4% of households has the mother reported being hit by her partner in the past 12 months. These rates are somewhat higher than those reported by the 1995 Violence Against Woman Study where 1.3% of adult women reported experiencing hitting as a part of IPV in the last month (Tjaden & Thoennes, 2000). This difference in part is due to the fact that for inclusion in the current study, woman had to report a husband or partner in the home. This eliminated 14% of the original sample. Also, it may reflect differences in IPV rates in younger women and/or women with children. Lastly, it may reflect regional differences in rates of IPV.
It is clear that IPV and CM are closely related phenomena with great importance to the lives of children and families. This study demonstrates, in a population-based survey, that families reporting IPV report much greater rates of harsh psychological parenting and neglectful parenting. For example, 37.9% of mothers reporting severe IPV in either direction report harsh psychological parenting and 26.0% report neglectful parenting. This compares to 7.9% and 9.4%, respectively, in families that do not report severe IPV. The relationship between IPV and harsh psychological punishment is the strongest and most consistent. The relationship between IPV and neglectful parenting is also strong. In contradiction to several previous studies, the relationship between IPV and harsh physical punishment is less clear in the current study (Berger, 2005; Lee, Kotch, & Cox, 2004; McGuigan & Pratt, 2001; Ross, 1996; Rumm, Cummings, Krauss, Bell, & Rivara, 2000; Tajima, 2000). Our study fails to shed much light on the risk of sexual abuse in families experiencing IPV because of the very small number of mothers endorsing sexual abuse in this population-based sample. The rates of mother reporting sexual abuse of children is quite low (<1%) and is likely a conservative estimate due to the stigma associated with this type of abuse.
In our sample, over one third (35.4%) of families endorsed one or more types of family violence. Over half (55%) of families reporting IPV also report one or more types of CM, indicating a critical need for addressing child victimization in the setting of IPV. Studies have shown the deleterious effects on children of witnessing IPV (Casanueva, Foshee, & Barth, 2005; Dube et al., 2001; Dubowitz et al., 2001; English, Marshall, & Stewart, 2003; Johnson et al., 2002; Jouriles et al., 2001; Levendosky, Huth-Bocks, Shapiro, & Semel, 2003; Thompson et al., 2005) and in most homes with IPV children are exposed to the IPV whether or not their parents are aware of this (Johnson et al., 2002). In addition to the trauma of living in a home with IPV, over half of such children are also victims of CM.
Why does our study fail to show a clear and consistent relationship between IPV and harsh physical punishment when several previous studies have demonstrated this relationship? There are several possible explanations. Rates of IPV have declined by roughly 50% between 1993 and 2001 (Rennison, 2003). Likewise, rates of protective service substantiated physical abuse have declined slightly (U.S. DHHS, 2006). These declines may explain the lack of precision in the current study when compared with analysis from the 1985 NFVS because both types of violence are somewhat less common (Berger, 2005; Ross, 1996; Tajima, 2000). Also, most of the previous studies demonstrating this relationship use either CPS populations or domestic violence shelter populations (Appel & Holden, 1998; Edleson, 1999). With increased rates for both IPV and CM in these select populations (and possibly more severe forms of both), these studies have more precision to identify these relationships. However, these studies may also be less broadly generalizeable then the current study in terms of reflecting violence in the homes of the population at large. It should be noted that our study does not demonstrate (nor seek to demonstrate) an absence of such a relationship, but merely in this case fails to support the hypothesis given the lack of a consistent and significant relationship between IPV and harsh physical punishment. There is a suggestive relationship between IPV and harsh physical punishment, notably with mother to partner frequent belittling IPV and mother to partner any IPV (adjusted OR 2.31, 95% CI 1.00–5.54 and 2.13, 95% CI 1.04–4.35, respectively). Also, mothers who report frequent belittling or severe IPV in both directions are more likely to report harsh physical punishment (adjusted OR 2.57, 95% CI 1.11–5.97). These relationships are similar in magnitude to the findings from the 1985 NFVS (Berger, 2005; Ross, 1996; Tajima, 2000). In the case of all other categories of IPV, more mothers reporting IPV also report harsh physical abuse, but the difference is often small.
This study is the first to our knowledge to evaluate the association between multiple forms of CM and IPV in population-based survey data. Several previous studies have evaluated this association between physical abuse and IPV with consistent results demonstrating an association (Berger, 2005; Kerker, Horwitz, Leventhal, Plichta, & Leaf, 2000; Ross, 1996; Tajima, 2000). Several longitudinal high-risk cohort studies have evaluated the link between IPV and other forms of CM, with mixed results (Lee et al., 2004; McGuigan & Pratt, 2001; Rumm et al., 2000). The study allows for a broader generalization of the findings and strengthens the previously demonstrated association between IPV and psychological abuse and neglect.
This study has several important limitations. Investigators using the CTS and PCCTS have chosen a variety of cut points for omissions of care and commissions of violence to measure IPV and CM. Our cut points are based on our clinical and research experience and may not equal protective service or legal definitions. However, these cut points are similar to our prior work with these instruments (Zolotor & Runyan, 2006) and the work of others studying IPV and CM (Tajima, 2000; Theodore et al., 2005). The CTS and PCCTS do not capture all potential forms of violence and neglect but represent a summary way of inquiring about some of the most common forms of family violence and neglect. Our measurement of both IPV and CM was limited by collecting data from only one informant, the child's mother. Likewise, data were collected about only parenting toward one index child. Parenting may differ within a household based on age, gender, temperament, and vulnerabilities.
Additionally, our sample is limited to the Carolinas and may not be representative of the United States. Also, our study sample was relatively wealthy, white, and educated compared to the Carolinas as a whole. We attempted to account for this by using survey weights that adjusted for socioeconomic status for all analyses. It should be underscored that this study uses cross-sectional survey data, and all results demonstrate associations but do not infer causality. Also, our study is based totally on anonymous self-report. We depend on the prior work of Straus et al. (1998) demonstrating that people are remarkably willing to admit to even socially reprehensible behaviors when guaranteed the anonymity of a survey. This raises the possibility of underreporting; however, if true rates of IPV and CM were higher then reported, and it would likely serve to strengthen the associations reported. The generally wide CIs of many associations should be noted. This lack of precision is due to the sample size and low prevalence of many findings. Lastly, the psychometric properties of the PCCTS are an important limitation on the findings of this study. Low alpha values for subscales of the PCCTS indicate that items in a subscale may be substantially unrelated to a latent construct. Developers of the PCCTS have argued that such items as beating a child and burning a child may be unrelated acts and therefore not part of a latent construct (Strauss, 1998). However, this leaves us to assess the construct of each subscale with an inventory of relatively unrelated single-item queries and may yield less reliable results.
|
Practice Implications |
|---|
|
TOPMethodsResultsDiscussionPractice ImplicationsFundingReferences |
|---|
Upon evaluating an adult victim of IPV, a direct service worker (social worker, nurse, physician, etc.) should consider the presence of a child in the home to represent a potential crisis for both the adult and the child. Not only is the child a victim of the potentially deleterious effects of IPV, but in over half of cases, they may be the victims of CM as well. Ten percent of the time, they are victims of more than one type of CM (see Figure 2). In addition, they may be uprooted from their home, their school, their friends, and a parent. Clearly this is a time of family crisis. In the 1990s, many domestic violence shelters began to expand services to children. However, knowledge of program effectiveness is minimal (Graham-Bermann, 2001; Graham-Bermann & Hughes, 2003). IPV services may be recommended to families at the time of entry into a child protective service evaluation. A recent study using data from National Survey of Child and Adolescent Well-Being showed that only 12% of caseworkers identified IPV and 31% of mothers self-reported IPV during study assessment. Only half of women with caseworker-identified IPV received a referral for IPV-related services (Kohl, Edleson, English, & Barth, 2005). Although it is not clear what services are most effective in treating the dual crisis of IPV and CM, our study clearly demonstrates the importance of identifying CM in families with IPV.
The Greenbook Demonstration Initiative published by the National Council of Juvenile and Family Court Judges promotes system integration and focuses on the entire family in cases where IPV and CM occur simultaneously (Schechter & Edleson, 1999). In 2000, the U.S. DHHS selected for funding six demonstration programs for implementation of the Greenbook guidelines. Results of these programs are not yet available, but implementation includes cross-training of caseworkers, revision of screening and assessment practices, and cross-agency information sharing (Greenbook National Evaluation Team, 2004). Increasingly, state and local agencies that serve IPV victims and CM victims are screening at the case level for other forms of family violence (National Clearinghouse on Child Abuse and Neglect Information, 2004). The type of coordination, information sharing, and family focus that is the subject of the Greenbook Initiative remains more difficult to achieve; however, comprehensive assessment and coordinated treatment should remain the goal of all agencies working with families in the turbulence of family violence (Schechter & Edleson, 1999). Understanding the common co-occurrence of all types of CM with IPV will help practitioners who work with IPV victim assess and respond to the co-victimization of children.
|
Funding |
|---|
|
TOPMethodsResultsDiscussionPractice ImplicationsFundingReferences |
|---|
The Duke Endowment, Inc., Charlotte, NC; Sunshine Lady Foundation to A.J.Z.
|
Acknowledgments |
|---|
The conclusions and opinions expressed in the manuscript are those of the authors and do not necessarily represent the views of The Duke Endowment, its officers, or members of the Board of Directors. We are grateful to Karolyn Forbes for her editorial assistanceConflict of Interest: None declared.
|
References |
|---|
|
TOPMethodsResultsDiscussionPractice ImplicationsFundingReferences |
|---|
-
American Association of Public Opinion Research. Standard definitions: Final dispositions of case codes and outcome rates for surveys (2005) Lenexa, KS: American Association for Public Opinion Research.
Appel AE, Holden GW. The co-occurrence of spouse and physical child abuse: A review and appraisal. Journal of Family Psychology (1998) 12:578–599.[CrossRef][ISI][find it @ Stanford]
Belsky J. Etiology of child maltreatment: A developmental-ecological analysis. Psychological Bulletin (1993) 114:413–434.[CrossRef][ISI][Medline][find it @ Stanford]
Bennett DS, Sullivan MW, Lewis M. Relations of parental report and observation of parenting to maltreatment history. Child Maltreatment (2006) 11:63–75.
Berger LM. Income, family characteristics, and physical violence toward children. Child Abuse and Neglect (2005) 29:107–133.[CrossRef][ISI][Medline][find it @ Stanford]
Caliso JA, Milner JS. Childhood history of abuse and child abuse screening. Child Abuse and Neglect (1992) 16:647–659.[CrossRef][ISI][Medline][find it @ Stanford]
Casanueva C, Foshee VA, Barth RP. Intimate partner violence as a risk factor for children's use of the emergency room and injuries. Children and Youth Services Review (2005) 27:1223–1242.[CrossRef][ISI][find it @ Stanford]
Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: Findings from the adverse childhood experiences study. Journal of the American Medical Association (2001) 286:3089–3096.
Dubowitz H, Black MM, Kerr MA, Hussey JM, Morrel TM, Everson MD, et al. Type and timing of mothers' victimization: Effects on mothers and children. Pediatrics (2001) 107:728–735.
Edleson JL. The overlap between child maltreatment and woman battering. Violence Against Women (1999) 5:134–154.
Edleson JL, Mbilinyi JF, Beeman SK, Hagemeister AK. How children are involved in adult domestic violence—Results from a four-city telephone survey. Journal of Interpersonal Violence (2003) 18:18–32.[Abstract]
English DJ, Edleson JL, Herrick ME. Domestic violence in one state's child protective caseload: A study of differential case dispositions and outcomes. Children and Youth Services Review (2005) 27:1183–1201.[CrossRef][ISI][find it @ Stanford]
English DJ, Marshall DB, Stewart AJ. Effects of family violence on child behavior and health during early childhood. Journal of Family Violence (2003) 18:43–57.[CrossRef][ISI][find it @ Stanford]
Graham-Bermann SA. Designing intervention evaluations for children exposed to domestic violence: Applications of research and theory. In: Domestic violence in the lives of children: The future of research, intervention, and social policy—Graham-Bermann SA, Edleson JL, eds. (2001) Washington, DC: American Psychological Association. 237–267.
Graham-Bermann SA, Hughes HM. Intervention for children exposed to interparental violence: Assessment of needs and research priorities. Clinical Child and Family Psychology Review (2003) 6:189–204.[CrossRef][ISI][Medline][find it @ Stanford]
Greenbook National Evaluation Team. The greenbook demonstration initiative: Interim evaluation report (2004) Washington, DC: Caliber Associates.
Hassan F, Refaat A, El-Sayed H, El-Defrawi M. Disciplinary practices and child maltreatment among Egyptian families in an urban area in Ismaila. Egyptian Journal of Psychiatry (1999) 22:177–193.[find it @ Stanford]
Hazen AL, Connelly CD, Kelleher K, Landsverk J, Barth R. Intimate partner violence among female caregivers of children reported for child maltreatment. Child Abuse and Neglect (2004) 28:301–319.[CrossRef][ISI][Medline][find it @ Stanford]
Johnson RM, Kotch JB, Catellier DJ, Winsor JR, Dufort V, Hunter W, et al. Adverse behavioral and emotional outcomes from child abuse and witnessed violence. Child Maltreatment (2002) 7:179–186.
Jouriles EN, McDonald R, Spiller L, Norwood WD, Swank PR, Stephens N, et al. Reducing conduct problems among children of battered women. Journal of Consulting and Clinical Psychology (2001) 69:774–785.[CrossRef][ISI][Medline][find it @ Stanford]
Jouriles EN, Norwood WD. Physical aggression toward boys and girls in families characterized by the battering of women. Journal of Family Psychology (1995) 9:69–78.[CrossRef][ISI][find it @ Stanford]
Kerker BD, Horwitz SM, Leventhal JM, Plichta S, Leaf PJ. Identification of violence in the home: Pediatric and parental reports. Archives of Pediatric and Adolescent Medicine (2000) 154:457–462.
Kohl PL, Edleson JL, English DJ, Barth RP. Domestic violence and pathways into child welfare services: Findings from the National Survey of Child and Adolescent Well-Being. Children and Youth Services Review (2005) 27:1167–1182.[CrossRef][ISI][find it @ Stanford]
Lee LC, Kotch JB, Cox CE. Child maltreatment in families experiencing domestic violence. Violence and Victims (2004) 19:573–591.[Medline][find it @ Stanford]
Levendosky AA, Huth-Bocks AC, Shapiro DL, Semel MA. The impact of domestic violence on the maternal-child relationship and preschool-age children's functioning. Journal of Family Psychology (2003) 17:275–287.[CrossRef][ISI][Medline][find it @ Stanford]
McGuigan WM, Pratt CC. The predictive impact of domestic violence on three types of child maltreatment. Child Abuse and Neglect (2001) 25:869–883.[CrossRef][ISI][Medline][find it @ Stanford]
National Clearinghouse on Child Abuse and Neglect Information. Children and domestic violence (2004) Washington, DC: U.S. Department of Health and Human Services.
Rennison C. Intimate partner violence, 1993 –2001 (2003) Washington (DC): Bureau of Justice Statistics, Department of Justice (US). Publication No. NCJ197838.
Ross SM. Risk of physical abuse to children of spouse abusing parents. Child Abuse and Neglect (1996) 20:589–598.[CrossRef][ISI][Medline][find it @ Stanford]
Rumm PD, Cummings P, Krauss MR, Bell MA, Rivara FP. Identified spouse abuse as a risk factor for child abuse. Child Abuse and Neglect (2000) 24:1375–1381.[CrossRef][ISI][Medline][find it @ Stanford]
Sadowski LS, Hunter WM, Bangdiwala SI, Munoz SR. The world studies of abuse in the family environment: A model study of a multi-national study of family violence. Injury Control and Safety Prevention (2004) 11:81–90.[CrossRef][find it @ Stanford]
Schechter S, Edleson JL. Effective intervention in domestic violence and child maltreatment cases: Guidelines for policy and practice (1999) Reno, NV: National Council of Juvenile and Family Court Judges.
Straus MA. Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. Journal of Marriage and the Family (1979) 41:75–88.[CrossRef][ISI][find it @ Stanford]
Straus MA, Gelles RJ. Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (1990) New Brunswick, NJ: Transaction.
Straus MA, Gelles RJ, Steinmetz SK. Behind closed doors: Violence in American families (1980) Garden City, NY: Anchor Press/Doubleday.
Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The Revised Conflict Tactics Scale (CTS2) (1995) Durham, NH: Family Research Laboratory, University of New Hampshire.
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 and Neglect (1998) 22:249–270.[CrossRef][ISI][Medline][find it @ Stanford]
Tajima EA. The relative importance of wife abuse as a risk factor for violence against children. Child Abuse and Neglect (2000) 24:1383–1398.[CrossRef][ISI][Medline][find it @ Stanford]
Tajima EA. Correlates of the co-occurrence of wife abuse and child abuse among a representative sample. Journal of Family Violence (2004) 19:399–410.[ISI][find it @ Stanford]
Theodore AD, Chang JJ, Runyan DK, Hunter WM, Bangdiwala SI, Agans R. Epidemiologic features of the physical and sexual maltreatment of children in the Carolinas. Pediatrics (2005) 115:e331–e337.
Thompson R, Briggs E, English DJ, Dubowitz H, Lee LC, Brody K, et al. Suicidal ideation among 8-year-olds who are maltreated and at risk: Findings from the LONGSCAN studies. Child Maltreatment (2005) 10:26–36.
Tjaden P, Thoennes N. Extent, nature, and consequences of intimate partner violence: Findings from the National Violence Against Women Survey (2000) Washington, DC: U.S. Department of Justice.
United States Department of Health and Human Services. Child maltreatment 2004 (2006) Washington, DC: U.S. Department of Health and Human Services.
United States Department of Justice. Criminal victimization in the United States (2006) Washington, DC: Bureau of Justice Statistics, Department of Justice.
Zolotor AJ, Runyan DK. Social capital, family violence, and neglect. Pediatrics (2006) 117:e1124–e1131.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||