Maternal Response in Cases of Suspected and Substantiated Child Sexual Abuse
 

^{Theresa Knott, PhD.*}

ABSTRACT: Secondary data analysis of the Canadian Incidence Study of Reported Child Abuse and Neglect, 1998 was conducted among 373 suspected and substantiated child sexual abuse investigations for which there was a female non-offending caregiver.  Bivariate analysis and hierarchical logistic regression was conducted to determine the association between caregiver, child, household, abuse, and investigation characteristics with negative maternal response.  The majority of female non-offending caregivers’ demonstrated belief of the sexual abuse disclosure, provided emotional support and protection from further abuse (87.1%).  The overall maternal response model was significant and accounted for 40.8% of the variance (Nagelkerke R2).  Predictors significantly associated with negative maternal response were caregiver mental illness, child age, sexualized behavior and developmental delay among children, duration of abuse, child’s relationship to the perpetrator, and co-occurring maltreatment.  Despite the number of risk factors examined, female non-offending caregivers investigated as part of the CIS-98 responded with belief of the disclosure, emotional support, and protection from further abuse.
 

Introduction

    Retrospective survey data indicates that 20-25% of women and 5-15% of men in the general population report sexual victimization as a child (Finkelhor, 1994).  Considerable variability exists in the observation of abuse related psychological sequelae among survivors of CSA.  While prediction of a developmental trajectory among survivors of child sexual abuse is imprecise, a constellation of risk factors have been documented.  Factors that influence outcome include the duration and frequency of sexual abuse (Caffaro-Rouget, Lang, & Van-Santen, 1989; Friedrich, Urquiza, & Beilke, 1986; Kendall-Tackett, Williams & Finkelhor, 1993), relationship of the child to the perpetrator (Browne & Finkelhor, 1996; Sedney & Brooks, 1984), severity of abuse (Fergusson, Horwood, & Lynskey, 1997; Fromuth, 1986; Russell, 1986), age at onset of the abuse (Cohen & Mannarino, 1998b; Kendall-Tackett, et al., 1993), the degree of force used during the abuse (Gomes-Schwartz, Horowitz, & Cardarelli, 1990; Kendall-Tackett et al., 1993; Ruggerio, McLeer, & Kixon, 2000; Spaccarelli & Kim, 1995), number of perpetrators (Wolfe, 1998; Wolfe & Birt, 1997), and threats to the child in the context of the abuse (Friedrich, Urquiza, & Beilke, 1986).

Maternal Response to CSA disclosure

    The negative impact of CSA varies considerably among survivors with evidence that an asymptomatic response is not uncommon (Kendall-Tackett, Williams, & Finkelhor, 1993).  The response of the non-perpetrating mother to the disclosure of child sexual abuse acts as a mediating factor in the short and long term psychological adjustment of the investigated child (Elliot & Carnes, 2001; Esparza, 1993; Everson, Hunter, Runyon, Edelsohn, & Coulter, 1989; Spaccarelli & Kim, 1995).

    Maternal response has been defined by CPS and reflected in the literature as: a) mother’s belief of the child’s disclosure, b) provision of emotional support, and c) protection from further abuse.  Driven primarily by mandated CPS investigation procedures the maternal response construct has not been subjected to rigorous theoretical analysis.  The absence of a fully developed theoretical formulation has inhibited the development of evidence-based best-practice guidelines for reacting to maternal response.  Intervention in cases of poor maternal response has focused on identifying a constellation of risk factors associated with poor maternal response and modifying child conduct to elicit maternal belief, emotional support and protection (e.g. CBT for children and non-offending parents).

    Research findings estimate that positive maternal response ranges from 27%-87% among non-offending caregivers of children investigated for sexual abuse (Elliot & Carnes, 2001; Everson et al., 1989; Heriot, 1996; Faller, 1988; Sirles & Franke, 1989; Trocmé, et al., 2001).  Use of various methodological procedures account for these wide variations in maternal response patterns.  Research evidence suggests that children exposed to positive maternal response experience fewer mental health and behavioural problems (Adams-Tucker, 1982; Barker- Collo, & Read, 2003; Bolen, & Lamb, 2004; Deblinger, Stauffer, & Landsberg, 1994; Deblinger, Steer, & Lipman, 1999; DeYoung, 1994; Elliott, & Carnes, 2001; Esparza, 1993; Everson, et al., 1989; Gomes-Schwartz, Horowitz, & Cardarelli, 1990; Heriot, 1996; Hiebert-Murphey, 1998; Leifer, Kilbane, & Grossman, 2001; Sirles & Franke, 1989; Spaccarelli, & Fuchs, 1997; Timmons-Mitchell, Chandler-Holtz, & Semple, 1996).  Comparatively, when maternal response is inadequate, research has demonstrated the association with child welfare services, including apprehension of CSA victims (Cross, Martell, McDonald, & Ahl, 1999; Hunter, Runyan, Coulter, Everson, 1990; Leifer, Shapiro, Martone, & Kassem, 1993; Pellegrin, & Wagner, 1990) and potential retraction of the sexual abuse disclosure (Berliner & Elliot, 1996; Lawson & Chaffin, 1992).

Study Objective

    This study examines the association between caregiver, child, household, abuse and investigation characteristics and maternal response among children for whom sexual abuse is suspected and substantiated in the CIS-98.

Research Questions

1.  What percentage of non-perpetrating mothers of children investigated for sexual abuse in the CIS-98 were identified by their social workers as having a negative maternal response?

2.  What characteristics are associated with maternal response?

2.1  In particular, what maternal characteristics (e.g. substance abuse, mental health, and domestic violence) are associated with maternal response?

2.2  What child characteristics (e.g. age, gender, sexualized behaviour, and developmental delay) are associated with maternal response?

2.3  What household characteristics are associated with maternal response e.g. (home tenure, household structure, and income)?

2.4  What abuse characteristics are associated with maternal response (duration of abuse, child’s relationship to the perpetrator)?

2.5  What investigation characteristic (co-occurring maltreatment) is associated with maternal response?

Methodology

    The Canadian Incidence Study of Reported Child Abuse and Neglect – 1998 (CIS-98) found that primary sexual abuse accounted for 10% of child maltreatment investigations (Trocmé, et al., 2001).  More recently, the CIS-2003 documented a decline in substantiation of child sexual abuse with CSA accounting for 3% of all substantiated primary maltreatment (Fallon, Lajoie, Trocmé, Chaze, MacLaurin, & Black, 2005).  A substantial decline in the CSA victimization rate has similarly been documented in the United States (Jones & Finkelhor, 2003; Trocmé, Fallon, MacLaurin, & Copp, 2002).

    This study is based on secondary data analysis of the Canadian Incidence Study of Reported Child Abuse and Neglect - 1998.  The CIS-1998 is a nationwide Canadian study of the incidence of child abuse and neglect.  A three page data collection instrument was completed by child welfare workers from 51 randomly selected child welfare service areas.  A stratified cluster sampling design was used to select maltreatment investigations and a four-stage sampling process was initiated.  A total of 7,672 child maltreatment investigations were conducted between October and December, 1998 from among the 51 randomly selected child welfare source areas (Trocmé, et al., 2001, p. 12).

    The CIS-98 included only reports investigated for maltreatment by child welfare workers and determined to be unsubstantiated, suspected, or substantiated.  The CIS-98 did not include cases that were undetected, unreported, or screened out by child welfare service.  For each maltreatment investigation, the CIS-98 collected a wide range of information including information on child, caregiver, household, and maltreatment characteristics.  Training was provided on-site to inform child welfare workers of the study and to familiarize them with the survey instrument.  A study guide, describing all the items in the survey was provided.  The on-site presence of the project team was useful for ensuring participation in the CIS-98 and to ensure the accuracy of the data collected.  Additional verification for inconsistencies was done at the central offices of the CIS-98.  The study participation rate varied from 75-100%, with an average participation rate of 90%.  Weighting procedures were conducted to establish national estimates of the annual incidence of child maltreatment and were derived from regionalization and annualization calculations.  Consequently, it is estimated that 135,573 child welfare investigations were conducted in Canada during 1998, an annual incidence rate of 21.52 children investigated for maltreatment per 1000 children in Canada (Trocmé, et al., 2001, p. xiv, 21).

    Results of the CIS-98 indicated that primary neglect accounted for 40% of all maltreatment investigations, of which 43% were substantiated.  Primary physical abuse accounted for 31% of investigations, of which 34% were substantiated.  Primary emotional abuse accounted for 19% and primary sexual abuse 10% of investigations, of which 54% and 38% were substantiated.  These data were limited to child maltreatment investigations excluding cases that either did not come to the attention of child welfare, or were screened out prior to investigation or handled only by the family courts where child welfare investigation was absent (Trocmé et al, 2001, p xv).

Study Sample

    Though there were 883 cases of investigated child sexual abuse in the CIS-1998, this study’s data set was restricted to suspected and substantiated investigations only (n=524).  The unit of analysis in this study was the child maltreatment investigation.  Only  investigations for which there was a female non-offending caregiver (i.e. biological mother, grandmother, stepmother, foster mother, and adoptive mother) were identified and retained in the study sample (n=373).  As most (93.6%) of these women were biological mothers to the investigated child, the term “maternal non-offending caregiver” was used to describe these persons.  The rationale for retaining only these investigations in the study sample was to allow for relevant comparison with existing research that examines the maternal response of non-offending caregivers of children who are investigated for child sexual abuse (Elliot & Carnes, 2001; Esparza, 1993; Everson, Hunter, Runyon, Edelsohn, & Coulter, 1989).  Among the 373 cases included in the study sample, 321 (86.3%) investigations were for primary child sexual abuse.  Among the remaining 52 (13.7%) investigations, child sexual abuse and co-occurring maltreatment was suspected and substantiated.

Operationalization of the Maternal Response Variable

    As conceptualized in the literature, the maternal response construct reflects a) belief of the child’s disclosure, b) provision of emotional support, and c) protection from further abuse.  In the current study, the operationalization of positive maternal response required all of a), b) and c) to be evident.  For example, positive maternal response was evident when the investigating child welfare worker documented belief of the disclosure, provision of emotional support, and protection from further abuse.  Negative maternal response was evident when the investigating child welfare worker documented a “no” to at least one of belief, emotional support, or protection.  Negative maternal response was similarly evident when the child welfare worker documented a “no,” and a non-response to belief, emotional support, or protection.  While we recognize the presence of the nonparallel treatment of missing data for the maternal response variable, the proportion of missing data was small and the model retained sufficient power.

    The derived outcome variable, negative maternal response was a dichotomous variable: 0=positive maternal response, 1=negative maternal response.

Operationalization of the Non-Offending Caregiver Variable

    This study examined the association between a series of predictor variables with the criterion variable, negative maternal response.  Using the CIS-98 data collection instrument, the investigating child welfare worker was asked to complete information about the response of the caregivers subsequent to the disclosure of child sexual abuse.  If Caregiver A was the perpetrator involved in the CSA maltreatment investigation, then the response of Caregiver B was included if Caregiver B was female and non-offending.  If the child welfare worker determined that Caregiver B was the perpetrator then the response of Caregiver A was included if Caregiver A was female and non-offending.  If both Caregivers A and B were identified as perpetrators, this investigation was not included in the sub-sample utilized in this study.

Rationale for Inclusion of Predictive Variables

    Many of this study’s predictive variables are theoretically relevant factors associated with maternal response to CSA (See Knott, 2008 for a theoretical analysis).  The influence of other predictors such as demographic factors and investigation characteristics has never been examined, to our knowledge.

    We established derived variables for child age, child gender, child functioning concerns, caregiver age, caregiver education, caregiver race, caregiver employment, caregiver functioning concerns, caregiver abused as a child, caregiver report of domestic violence, home tenure, household structure, abuse type, perpetrator of abuse, duration of maltreatment, previous child welfare opening, police investigation, and co-occurring maltreatment.  Please see Knott (2008) for variable definitions and a detailed description of coding procedures.  Only the predictive variables statistically significant at the p=.05 level in bivariate analysis were retained in the multivariate model (See Table 1).

  Table 1
  Independent variables used to predict maternal response in logistic regression analysis.

    Note. Subsequent to bivariate analysis, household characteristics were not retained in the multivariate model.

Analytic Strategy

    Bivariate analysis (chi-square) and multivariate analysis (logistic regression) were used to determine the factors associated with social workers perceptions of negative maternal response among the non-perpetrating caregivers of children for whom sexual abuse was suspected and substantiated.  Pearson chi-square analysis was conducted to test for group differences among caregiver, child, household, abuse, and investigation characteristics in the proportion of investigations where negative maternal response was identified.  Only those predictive variables that were statistically significant at the bivariate level (p<.05) were included in the multivariate model.  Effect size calculations revealed that there was sufficient power to detect both a small and medium effect for most of the main predictive variables.

Logistic Regression Model

    A hierarchical regression model was developed to predict the likelihood that a female nonperpetrating caregiver demonstrated a negative maternal response to the disclosure of child sexual abuse.  This model utilized four blocks including caregiver, child, abuse, and investigation characteristics representing each consecutive block.  Blocks were compared to determine their independent contribution to negative maternal response.

    Adjusted odds ratios were reported and the pseudo R² (Nagelkerke adjustment) statistic was cited.  Multicolinearity tests were conducted to obtain an estimate of correlation between the independent variables.  The presence of multicolinearity was identified between the variables physical injury and domestic violence.  As a result, physical injury was removed from the multivariate model.

Findings

    Univariate analysis determined the prevalence of negative maternal response.  Bivariate analysis (chi-square) was conducted to determine the correlation between caregiver, child, abuse and investigation factors and negative maternal response while a logit model yielded probability estimates between each independent variable and block of variables with negative maternal response.  Of the 373 investigations included in the study sample, 325 (87.1%) of non-offending female caregivers responded positively to the disclosure of child sexual abuse as reported by the investigating child welfare worker while 48 (12.9%) of caregivers responded negatively.  For a detailed description of results from the univariate analysis see Knott, (2008).

    As indicated in Table 2, the caregiver variables statistically significant at the bivariate level were caregiver education, caregiver mental health, caregiver physical health, caregiver abused as a child, domestic violence, caregiver involved in criminal activity, caregiver with few social supports while caregiver substance abuse approached statistical significance (p<.03, p<.001, p<.01, p=.04, p<.01, p<.001, p<.03, p<.06, respectively).

Table 2: Percentage with negative maternal response by caregiver characteristics among those
with suspected or substantiated sexual abuse
Source: CIS-1998

 
    Table 3 reveals that the statistically significant child characteristics were child age, child with a developmental delay, child with depression/anxiety, inappropriate sexual behavior, and self-harming behavior (p<.001, p<.001, p<.01, p<.01, p<.01, respectively).

Table 3: Percentage with negative maternal response by child characteristics among those with suspected or substantiated sexual abuse
Source: CIS-1998

    Among the three perpetrator characteristics examined, suspected and substantiated abuse by a father figure, and suspected and substantiated abuse by a stranger/unknown/other acquaintance/or multiple perpetrators were found to be significantly associated with negative maternal response (p<.001, p<.03, respectively) as shown in Table 4.

Table 4
Percentage with negative maternal response by alleged perpetrator characteristics among those with suspected or substantiated sexual abuse
Source: CIS-1998

    As demonstrated in Table 5, the abuse characteristics that reached the level of statistical significance were duration of abuse, and maltreatment co-occurrence (p<.05, p<.001, respectively).

Table 5
Percentage with negative maternal response by investigated abuse characteristics among those with suspected or substantiated sexual abuse
Source: CIS-1998

    A hierarchical logistic regression model was established with four distinct categories (caregiver, child, abuse, and investigation characteristics) to predict the likelihood that a female non-offending caregiver would demonstrate negative maternal response to a child for whom sexual abuse was suspected and substantiated¹.  As a result of the small number of investigations involving negative maternal response (n=48), the regression model had insufficient power to rely on theoretical relevance to determine inclusion of variables.  Inclusion in multivariate analysis was determined by the level of statistical significance at the bivariate stage (p<.05). Investigations with missing data were not included in multivariate analysis.

    Table 6 illustrates the inclusion of caregiver, child, abuse, and investigation characteristics, which individually represented each of the four steps in the regression model.  As a whole, the block containing the caregiver characteristics (caregiver mental health and caregiver domestic violence) was found to be statistically significant (Block 1 X² 23.21, p<.001). Individually, caregiver mental health was a significant predictor of negative maternal response at each of the four steps in the regression equation.  In the fourth and final step of the regression model, caregivers with a mental health problem were two times more likely to demonstrate a negative response to CSA disclosure when compared with caregivers without a mental health problem.  The inclusion of caregiver domestic violence resulted in significantly higher odds of negative maternal response in the first, and second phase of the regression equation.  Domestic violence ceased to be a significant predictor of negative maternal response (p=.07²) subsequent to the inclusion of abuse characteristics in the third step of the regression model.

Table 6: Logistic regression: Factors associated with negative maternal response among
those with suspected or substantiated child sexual abuse (N=373)

    The variables age of the child, presence of sexualized behavior, and developmental delay were included in the regression procedure at step two.  The block containing the child characteristics was found to be statistically significant (Block 2 X2=32.48, p<.001).  Table 6 indicates that children 8-11 & 12-15 had significantly higher odds of exposure to negative maternal response when compared with children in the reference category (age 0-7).  In the final step of the regression procedure, the adolescent group (12-15 years of age) was eight times more likely to receive a negative maternal response to the disclosure of child sexual abuse when compared with younger children 0-7 years of age (O.R. 8.53).  Children for whom inappropriate sexual behavior was documented had higher odds of negative maternal response when compared with children who did not manifest this behavior (O.R. 3.23).  Inappropriate sexualized behavior remained a statistically significant predictor of negative maternal response throughout the entire regression procedure (p=.02).  In the fourth and final step of the regression model, the child characteristic developmental delay was found to be a statistically significant predictor of negative maternal response (p=.02).

    Abuse characteristics such as duration of suspected and substantiated child sexual abuse and relationship to the perpetrator were examined in the third step of the regression model.  The block containing these abuse characteristics was statistically significant (Block 3 X² 13.29, p=.001).  In the final step of the regression procedure, sexual abuse occurring for longer than a single incident had higher odds of negative maternal response when compared with single incident abuse.  Table 6 demonstrates that children whose relative was the perpetrator of suspected and substantiated CSA were over two times more likely to be exposed to negative maternal response (O.R. 2.60).

    Co-occurring maltreatment was examined in the fourth and final block of the regression equation with the block containing this investigation characteristic yielding a statistically significant result (Block 4 X2=22.79, p<.001).  Children for whom co-occurring maltreatment was suspected and substantiated were over eight times more likely to be exposed to negative maternal response when compared with children investigated for sexual abuse only.

    The caregiver and child characteristics explained the largest proportion of variance in negative maternal response (11.3% and 14.7% respectively).  Abuse characteristics added 5.6% to the variance in the third step (Pseudo R2=31.6%) while co-occurring maltreatment contributed an additional 9.2%.  The overall model explained a cumulative total of 40.8% of the variance and was significant at p<.01.

Discussion

    Consistent with the extant CSA literature (Bolen, 2002; Elliot & Carnes, 2001; Leifer, Kilbane & Grossman, 2001; Sirles & Franke, 1989), the current study demonstrated that the vast majority of non-offending caregivers of children for whom CSA is suspected and substantiated (87.1%) responded positively.  Despite the substantial number of risk factors examined in this study non-offending mothers overwhelmingly responded to the disclosure of child sexual abuse with belief, emotional support, and protection from further abuse.  The regression model was statistically significant and accounted for 40.8% of the variance in predicting negative maternal response (Nagelkerke R2).  Among the factors investigated, child and caregiver characteristics contributed most substantially to the variance in outcome (14.7% and 11.3% respectively).  Regression analysis of the variables child age (12-15 years of age), inappropriate sexualized behavior, children’s developmental delay, and caregiver mental health resulted in a significant association with negative maternal response (p<.001, p=.02, p=.02, and p=.04 respectively).  Despite its significant association with negative maternal response in steps 1 and 2 of the regression procedure, domestic violence failed to reach the level of significance in the 3rd and 4th step of the regression equation (p=.07, p=.94).  Co-occurring maltreatment was strongly associated with negative maternal response (p<.001).  Children for whom co-occurring maltreatment was suspected and substantiated were eight times more likely to be exposed to negative maternal response.

    Findings of this study demonstrated the vulnerability of adolescents given their heightened risk of exposure to negative maternal response.  Not surprisingly, age of the investigated child was one of the most significant predictors of negative maternal response (p<.001).  Adolescents 12-15 years of age were over eight times more likely to receive a negative response from their non-offending maternal caregiver when compared with children 0-7 years of age.  As indicated by Summit (1983), older children often exhibit secretive behavior in response to perpetrator threats.  Perhaps, maternal response is compromised as a result of this avoidant coping response often exhibited by adolescents, and interpreted by the non-offending caregiver as culpability.  Adolescents’ externalizing behaviors and running away has been noted as a potential trigger of poor maternal response by non-offending caregivers (Johnson & Kenkel, 1991; Shapiro & Levendosky, 1999; Spaccarelli, 1994).  Conversely, preschool and school age children’s disclosures are more frequently believed by their non-offending caregiver.  Lovett (2004) refers to the continuum of disclosure with young children more likely to accidentally disclose CSA, and school-age children typically yielding a purposeful disclosure.  In the absence of maternal belief, support and protection the risk for recantation is heightened (Reiser, 1991).  Social service professionals should be aware of the continuum of disclosure, in particular, the varied age-related disclosure patterns, and the potential meaning associated with secretive and externalizing behaviors.  With increased awareness of the benefits of positive maternal response in contributing to child wellbeing and preventing recantation, professional resources should be designed to strengthen the relationship between non-offending mothers and victimized children.

    This study’s findings replicated evidence on the influences of children’s sexualized behaviour on maternal response.  Findings of this study indicated that children who manifested sexualized behaviour were significantly more likely to be exposed to negative maternal response when compared with children who did not display this behaviour (32.4% versus 12.3%).  Sexually inappropriate behaviour was associated with higher odds of negative maternal response throughout each stage of the regression procedure (O.R.2.3).  The presence of inappropriate sexualized behaviour as potential sequelae of child sexual abuse has been documented in CSA literature (Einbender & Friedrich, 1989; Goldston, Turnquist, & Knutson, 1989).  The effect of sexualized behaviour can influence the non-offending mother’s perception of the culpability of her child and consequently contribute to poor maternal response (Cohen & Mannarino, 1993; Friedrich, 2007; Wolfe & Birt, 1997).  Despite the effect of inappropriate sexualized behaviour on maternal response, evidence exists regarding the efficacy of Cognitive Behavioural Therapy in mitigating this conduct, and contributing to improved parenting (Cohen & Mannarino, 1996a, 1996b; Deblinger, Stauffer & Steer, 2001; Deblinger, Steer & Lippman, 1999).

    It is well established that developmental and cognitive disability contributes to the level of children’s vulnerability to sexual abuse (Goldman, 1994; Sullivan & Knutson, 2000; Sullivan, & Knutson, 1998; Westat, 1993; Westcott & Jones, 1999; Zirpoli, 1990) and to skepticism about the veracity of CSA disclosure (Bottoms, Carris, Harris, & Tyda, 2003; Tharinger, Horton, & Millea, 1990; Verdugo & Bermejo, 1997).  In the current study, developmental disability included behavior disorders, learning disabilities, mental retardation, speech/language difficulties, mental illness, orthopedic problems, hearing and visual disabilities, and autism. Findings of our study revealed that children with a developmental disability were much more likely to receive a negative maternal response to CSA disclosure when compared to children without a developmental disability (38.5% versus 11%).  Regression analysis indicated that children for whom a developmental disability was identified were three times more likely to be exposed to a negative maternal response according to the report of the investigating child welfare worker.  Parental disbelief in the capacity of such children to accurately recount events and a myth that children with a developmental delay possess an asexual quality are potential explanations for poor maternal response (Tharinger, Horton, & Milea, 1990).  Given the enhanced risk of maltreatment assumed by those with a developmental delay, increased support and education for the identified child and non-offending caregiver is required.  Further research is needed to inform evidence based practice for this vulnerable sub-group.

    A minority of the non-offending female caregivers in the current study were identified by the investigating child welfare worker as having a mental illness (12.1%).  Subsequent regression analysis indicated that children of mothers with a mental illness were two times more likely to be exposed to a negative maternal response.  This research confirms that despite the small minority of mothers with mental health problems in this sample, children continue to be at risk for poor maternal response.  Findings of this study are consistent with research evidence documenting the relationship between mental illness and negative maternal response (Davies, 1995; Manion et al, 1996; Zuravin & Fontella, 1999).  The parenting challenges faced by caregivers with a mental illness have been identified while the developmental trajectories among children of parents with a serious mental illness examined (Beardslee, Versage, & Gladstone, 1998; Berg-Nielsen, Vikan, & Dahl, 2002; Gladstone, Boydell, & McKeever, 2006; Hall, 2004; Rutter, 1990; Rutter & Quinton, 1984).  Despite the balance of evidence documenting parenting challenges among caregivers with a mental illness, an emerging body of research has challenged empirical data linking mental illness with compromised parenting (Aldridge, 2006; Gladstone, et al., 2006).  Perhaps disparate study methodologies and the varied operationalization of variables have contributed to these inconsistent findings.  Given that CSA disclosures are most frequently made to the non-offending mother (Sauzier, 1989), appropriate and timely social support for mothers who struggle with a serious mental illness is critical.

    While domestic violence was found to be significantly associated with negative maternal response in steps one and two of the regression equation, it failed to yield a statistically significant result in the final step of multivariate analysis – although a trend towards statistical significance was demonstrated.  This study’s findings are contrary to empirical data documenting poor maternal response among non-offending caregivers of children investigated for CSA who are themselves victims of domestic violence (Alaggia & Turton, 2005; de Young, 1994; Hiebert-Murphy, 2001; Kellogg & Menard, 2003).  Given the small sample size and number of measures utilized in this study there may have been inadequate power to retain a significant association between domestic violence and negative maternal response when the model adjusted for abuse characteristics.  Negative maternal response associated with child sexual abuse when domestic violence is present is an emerging area of research.  Social service professionals may consider utilizing screening procedures for domestic violence once a CSA disclosure is made.  Avoidance of the re-victimization of the non-offending mother is critical as potential exists for professionals to attribute blame towards the non-offending caregiver for remaining in the abusive situation.  More precise research is needed to help expand practice measures for this population of non-offending mothers who experience domestic violence and whose children disclose CSA.

    As hypothesized, children investigated for suspected and substantiated CSA and another form of maltreatment were significantly more likely to be exposed to negative maternal response when compared with children who were investigated for sexual abuse only (O.R. 8.43).  In cases where child sexual abuse exists concurrent with domestic violence empirical data suggests that maternal response is compromised (Alaggia & Turton, 2005).  Clinical practice strategies should aim to address the cumulative effect of exposure to multiple forms of maltreatment, and prolonged duration.  Treatment professionals should not discount the potential for financial or other dependence on the perpetrator of co-occurring maltreatment.

Study Limitations

    Missing/unknown data occurred most frequently among the variables, caregiver education, caregiver income, caregiver employment and caregiver ethno-racial heritage.  Given the proportion of missing data among these variables, chi-square analysis containing the missing/unknown responses was conducted to ensure no systematic bias in reporting the missing/unknown demographic data relative to negative maternal response.  For example, chi-square analysis conducted with negative maternal response and the missing/unknown caregiver education data was not statistically significant (p=.36).  As a result, these investigations were removed.  The same procedure was followed among the remaining variables with substantial missing data, and analysis revealed no significant associations (p=.53; p=.21; p=.10, respectively).  Multiple imputation for missing data was not used in this study.

    This study’s sample included a minority of investigations involving suspected and substantiated extrafamilial child sexual abuse.  Extrafamilial child sexual abuse referred to those investigations for which a non-kin related perpetrator was identified.  When compared with the outcomes associated with incest, research literature documents that the potential for psychological sequelae is diminished among survivors of extrafamilial child sexual abuse (Finkelhor, et al., 1990).  Lower levels of negative maternal response may be reported in this study than reported elsewhere because of the diminished psychological sequelae associated with extrafamilial abuse.

    When completing the CIS-98 data collection instrument social workers conducted an assessment of characteristics associated with the functioning of children and caregivers (e.g. developmental & cognitive delay among children, mental health problems among caregivers).  To make this designation child welfare workers relied on independent observation, disclosure from families or professionals, or case histories.  It was not expected that the investigating child welfare worker received professional training to confirm a diagnosis of a mental illness or to assess cognitive capacity.  When the investigating child welfare worker had received an independent confirmation of a diagnosis, the worker indicated “confirmed” on the CIS-98 data collection instrument.  In the absence of this knowledge, “suspected,” or “no” was indicated.

    Due to the structure of the CIS-98 data collection instrument, it was not possible to determine who the investigating child welfare worker was referring to when responding to the caregiver functioning items.  As a result, the worker may have been referring to a non-offending caregiver, or the perpetrator of suspected and substantiated CSA.  It is not possible to determine if this person was male or female.  Caution is warranted in the interpretation of the findings associated with caregiver mental health, and caregiver substance abuse.

    The small sample size and the number of risk factors included in the multivariate model, and the absence of a rigorous assessment tool to measure maternal response, may influence the possibility of non-findings.

References

Adams-Tucker, C. (1982). Proximate effects of sexual abuse in childhood: A report on 28 children. American Journal of Psychiatry, 139(10), 1252-1256.

Alaggia, R. & Turton, J. (2005). Against the odds: the impact of woman abuse on maternal response to disclosure of child sexual abuse. Journal of Child Sexual Abuse, 14(4), 95-113.

Aldridge, J. (2006). The experiences of children living with and caring for parents with mental illness. Child Abuse Review, 15, 79-88.

Barker-Collo, S., & Read, J. (2003). Models of response to child sexual abuse: Their implications for treatment. Trauma, Violence and Abuse, 4(2), 95-111.

Beardslee, W. R., Versage, E.M., & Gladstone, T.R.G. (1998). Children of affectively ill parents: A review of the past 10 years. Journal of the American Academy of Child and Adolescent Psychiatry, 37(11), 1134-1141.

Berg-Nielsen, T.S., Vikan, A., & Dahl, A.A. (2002). Parenting related to child and parental psychopathology: A descriptive review of the literature. Clinical Child Psychology & Psychiatry, 7(4), 529-552.

Berliner, L., & Elliott, D.M. (2002). Sexual abuse of children. In J. Myers, L. Berliner, J. Briere, C. Hendrix, C. Jenny, & T. Reid (Eds.), The APSAC Handbook on Child Maltreatment ()()(), 2nd Edition. (pp. 55-78). Thousand Oaks, CA: Sage

Berliner, L., & Elliott, D.M. (1996). The sexual abuse of children. In J. Briere, L. Berliner, J.A. Bulkley, C. Jenny, & T. Reid (Eds.), The APSAC Handbook on Child Maltreatment ()()() (pp. 51-71). Thousand Oaks, CA: Sage.

Bolen, R. (2002). Guardian support of sexually abused children: A definition in search of a construct. Trauma, Violence and Abuse, 3(1), 40-67.

Bolen, R., & Lamb, J.L. (2004). Ambivalence of nonoffending guardians after child sexual abuse disclosure. Journal of Interpersonal Violence, 19(2), 185-211.

Bottoms, B.L., Nysse-Carris, K.L., Harris, T., & Tyda, K. (2003). Jurors’ perceptions of adolescent sexual assault victims who have intellectual disabilities. Law and Human Behavior, 27, 2, 205-226.

Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99, 66-77.

Caffaro-Rouget, A., Lang, R. A. & Van-Santen, V. (1989). The impact of child sexual abuse on victims' adjustment. Annals of Sex Research, 2, 29-47.

Cohen, J. A. & Mannarino, A. P. (1998a) Interventions for sexually abused children: initial treatment outcome findings. Child Maltreatment, 3, 17 -26.

Cohen, J. A. & Mannarino, A. P. (1998b) Factors that mediate treatment outcome of sexually abused preschool children; six- and 12-month follow-up. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 44 -51.

Cohen, J. A. & Mannarino, A. P. (1996a) A treatment outcome study for sexually abused preschool children: initial findings. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 42 -50.

Cohen, J.A., & Mannarino, A. P. (1996b). Factors that mediate treatment outcome of sexually abused preschool children. Journal of the American Academy of Child and Adolescent Psychiatry, 34(10), 1402-1410.

Cohen, J. A. & Mannarino, A. P. (1993). A treatment model for sexually abused preschoolers. Journal of Interpersonal Violence, 8, 115-131.

Cross, T.P., Martell, D., McDonald, E., & Ahl, M. (1999). The criminal justice system and child placement in child sexual abuse cases. Child Maltreatment, 4(1), 32-44.

Davies, M. (1995). Parental distress and ability to cope following disclosure of extra-familial sexual abuse. Child Abuse & Neglect, 19(4), 399-408.

Deblinger, E., Stauffer, L., & Landsberg, C. (1994). The impact of a history of child sexual abuse on maternal response to allegations of sexual abuse concerning her child. Journal of Child Sexual Abuse, 3, 67-75.

Deblinger, E., Stauffer, L. B., & Steer, R.A. (2001). Comparative efficacies of supportive and cognitive behavioural group therapies for young children who have been sexually abused and their non-offending mothers. Child Maltreatment, 6(4), 332-343.

Deblinger, E., Steer, R. A. & Lippmann, J. (1999) Two-year follow-up study of cognitive behavioural therapy for sexually abused children suffering post-traumatic stress symptoms. Child Abuse & Neglect, 23, 1371 -1378.

DeMarco, R., Tonmyr, L., Fallon, B., &, Trocmé, N. (2007). The effect of maltreatment co-occurrence on emotional harm among sexually abused children. Victims and Offenders, 2(1), 45-62.

de Young, M. (1994a). Women as mothers and wives in paternally incestuous families: Coping with role conflict. Child Abuse & Neglect, 18, 73-83.

de Young, M. (1994b). Immediate maternal reactions to the disclosure or discovery of incest. Journal of Family Violence, 9, 21-31.

Eidenbender, A. J., & Friedrich, W.N. (1989). Psychological functioning and behavior of sexually abused girls. Journal of Consulting and Clinical Psychology, 57(1), 155-157.

Elliott, A., N., & Carnes, C. N. (2001). Reactions of non-offending parents to the sexual abuse of their child: A review of the literature. Child Maltreatment, 6(4), 314-331.

Esparza, D. (1993). Maternal support and stress response in sexually abused girls ages 6–12. Issues in Mental Health Nursing, 14, 85-107.

Everson, M. D., Hunter, W. M., Runyon, D. K., Edelsohn, G. A., & Coulter, M. L. (1989). Maternal support following disclosure of incest. American Journal of Orthopsychiatry, 59, 197- 207.

Faller, K.C. (1988). The myth of the “collusive mother”: Variability in the functioning of mothers of victims of intrafamilial sexual abuse. Journal of Interpersonal Violence, 3, 190-196.

Fallon, B., Lajoie, J., Trocmé, N., Chaze, F., MacLaurin, B., & Black, T. (2005). Sexual abuse of children in Canada. CECW Information Sheet 25E.

Fergusson, D., Horwood, J., & Lynskey, M. (1997). Childhood sexual abuse, adolescent sexual behaviors and sexual revictimization. Child Abuse & Neglect, 21(8), 789-803.

Finkelhor, D. (1994). Current information on the scope and nature of child sexual abuse. The Future of Children, 4(2):31, 46-48.

Friedrich, W.N. (2007). Sexual behavior in sexually abused children. New Directions for Mental Health Services, 1991(51), 15-27.

Friedrich, W. N., Urquiza, A. J., & Beilke, R. (1986). Behavior problems in sexually abused young children. Journal of Pediatric Psychology, 11, 47-57.

Fromuth, M. (1986). The relationship of childhood sexual abuse with later psychological and sexual adjustment in a sample of college women. Child Abuse & Neglect, 10(1), 5-15.

Gladstone-McConnell, B., Boydell, K.M., & McKeever, P. (2006). Recasting research into children’s experiences of parental mental illness: Beyond risk and resilience. Social Science & Medicine, 62, 2540-2550.

Goldman, R. (1994). Children and youth with intellectual disabilities: Targets for sexual abuse. International Journal of Disability, Development and Education, 41(2), 89-102.

Goldston, D.B., Turnquist, D.C., & Knutson, J.F. (1989). Presenting problems of sexually abused girls receiving psychiatric services. Journal of Abnormal Psychology, 98(3), 314-317.

Gomes-Schwartz, B., Horowitz, J.M., & Cardarelli, A.P. (1990). Child Sexual Abuse: The Initial Effects ()(). Newbury Park, CA: Sage.

Hall, A. (2004). Parental psychiatric disorder and the developing child. In M. Göpfert, J. Webster, & M. Seeman (Eds.), Parental Psychiatric Disorder: Distressed Parents and their Families () () (pp. 22-49). Cambridge University Press.

Heriot, J. K. (1996). Maternal protectiveness following the disclosure of intrafamilial child sexual abuse. Journal of Interpersonal Violence, 11, 181-194.

Hiebert-Murphy, D. (2001). Partner abuse among women whose children have been sexually abused: An exploratory study. Journal of Child Sexual Abuse, 10(1), 109-118.

Hiebert-Murphey, D. (1998). Emotional distress among mothers whose children have been sexually abused: The role of a history of child sexual abuse, social support and coping. Child Abuse & Neglect, 22(5), 423-435.

Hunter, W.M., Coulter, M. L., Runyan, D.K., Everson, M.D. (1990). Determinants of placement for sexually abused children. Child Abuse & Neglect, 14, 407-417.

Johnson, B. K., & Kenkel, M. B. (1991). Stress, coping, and adjustment in female adolescent incest victims. Child Abuse & Neglect, 15, 293-305.

Jones, L. M., & Finkelhor, D. (2003). Putting together evidence on declining trends in sexual abuse: A complex puzzle. Child Abuse & Neglect, 27, 133-135.

Kellogg, N.D., & Menard, S.W. (2003). Violence among family members of children and adolescents evaluated for sexual abuse. Child Abuse & Neglect, 27, 1367-1376.

Kendall-Tackett, K. A., Williams, L. M., & Finkelhor, D. (1993). Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psychological Bulletin, 113, 164- 180.

Knott, T. (2008). Testing the maternal response hypothesis in cases of suspected and substantiated child sexual abuse: Secondary data analysis of the Canadian Incidence Study of Reported Child Abuse and Neglect, 1998.  https://tspace.library.utoronto.ca/bitstream/1807/17325/1/Knott_Theresa_F_200811_PhD_thesis.
pdf

Lawson, L., & Chaffin, M. (1992). False negatives in sexual abuse disclosure interviews: Incidence and influence of caretaker’s belief in abuse in cases of accidental abuse discovery by diagnosis of STD. Journal of Interpersonal Violence, 7, 532-542.

Leifer, M., Kilbane, T., & Grossman, G. (2001). A three-generational study comparing the families of supportive and unsupportive mothers of sexually abused children. Child Maltreatment, 6(4), 353-364.

Leifer, M., Shapiro, J.P., & Kassem, L. (1993). The impact of maternal history and behaviour upon foster placement and adjustment in sexually abused girls. Child Abuse & Neglect, 17(6), 755-766.

Lovett, B. B. (2004). Child sexual abuse disclosure: Maternal response and other variables impacting the victim. Child & Adolescent Social Work Journal, 21(4).

Manion, I.G., McIntyre, J., Firestone, P., Ligezinska, M., Ensom, R., & Wells, G. (1996). Secondary traumatization in parents following the disclosure of extrafamilial child sexual abuse: Initial effects. Child Abuse & Neglect, 20 (11), 1095-1109.

Pellegrin, A., & Wagner, W.G. (1990). Child sexual abuse: Factors affecting victims’ removal from home. Child Abuse & Neglect, 14,(1), 53-60.

Reiser, M. (1991). Recantation in child sexual abuse cases. Child Welfare, 70(6).

Ruggiero, K. J., McLeer, S.V., & Dixon, J.F. (2000). Sexual abuse characteristics associated with survivor psychopathology. Child Abuse & Neglect, 24, 951-964.

Russell, D. (1986). The Secret Trauma: Incest in the Lives of Girls and Women ()(). New York: Basic Books.

Rutter, M. (1990). Psychosocial resilience and protective mechanisms. In J. Rolf, A.S. Masten, D. Cicchetti, K.H. Nuechterlein, & S. Weintraub (Eds.), Risk and Protective Factors in the Development of Psychopathology ()() (pp. 181-214). New York: Cambridge University Press.

Rutter, M., & Quinton, D. (1984). Parental psychiatric disorder: Effects on children. Psychological Medicine, 14, 853-880.

Sauzier, M. (1989). Disclosure of child sexual abuse: For better or worse. Psychiatric Clinics of North America, 12, 455-469.

Sedney, M., & Brooks, B. (1984). Factors associated with a history of childhood sexual experience in a non-clinical female population. Journal of the American Academy of Child Psychiatry, 23(2), 215-218.

Shapiro, D. L., & Levendosky, A. A. (1999). Adolescent survivors of childhood sexual abuse: The mediating role of attachment style and coping in psychological and interpersonal functioning. Child Abuse & Neglect, 23, 1175-1191.

Sirles, E. A., & Franke, P. J. (1989). Factors influencing mothers’ reaction to intrafamilial child sexual abuse. Child Abuse & Neglect, 13, 131-189.

Spaccarelli, S. (1994). Stress, appraisal, and coping in child sexual abuse: A theoretical and empirical review. Psychological Bulletin, 116, 340-362.

Spaccarelli, S., & Fuchs, C. (1997). Variability in symptom expression among sexually abused girls: Developing multivariate models. Journal of Clinical Child Psychology, 26(1), 24-35.

Spaccarelli, S., & Kim, S. (1995). Resilience criteria and factors associated with resilience in sexually abused girls. Child Abuse & Neglect, 19, 1171-1182.

Sullivan, P.M., & Knutson, J.F. (1998). The association between child maltreatment and disabilities in a hospital-based epidemiological study. Child Abuse & Neglect, 22, 271-288.

Sullivan, P.M., & Knutson, J.F. (2000). Maltreatment and disabilities: A population-based epidemiological study. Child Abuse & Neglect, 24(10), 1257-1273.

Summit, R.C. (1983). The child sexual abuse accommodation syndrome. Child Abuse & Neglect, 7(2), 177-193

Tharinger, D., Horton, C., & Millea, S. (1990). Sexual abuse and exploitation of children and adults with mental retardation and other handicaps. Child Abuse & Neglect, 14, 301-312.

Timmons-Mitchell, J., Chandler-Holtz, D., & Semple, W.E. (1996). Post-traumatic stress symptoms in mothers following children’s reports of sexual abuse: An exploratory study. American Journal of Orthopsychiatry, 66, 463-467.

Trocmé, N., Fallon, B., MacLaurin, B., & Copp, B. (2002). The changing face of child welfare investigations in Ontario: Ontario Incidence Studies of Reported Child Abuse and Neglect (OIS 1993/1998). Toronto, ON: Centre of Excellence for Child Welfare, Faculty of Social Work, University of Toronto, 23 pages.

Trocmé, N., MacLaurin, B., Fallon, B., Daciuk, J., Billingsley, D., Tourigny, M., Mayer, M., Wright, J., Barter, K., Burford, G., Hornick, J., Sullivan, R., & McKenzie, B. (2001) Canadian incidence study of reported child abuse and neglect: Final report. Ottawa, Ontario: Health Canada, 133 pages.

Verdugo, M.A., & Bermegjo, B.G. (1997). The mentally retarded person as a victim of maltreatment. Aggression and Violent Behavior, 2(2), 143-165.

Westat, Inc. (1993). A report on the maltreatment of children with disabilities. Washington, DC: National Center on Child Abuse and Neglect.

Westcott, H. L., & Jones, D. P. H. (1999). Annotation: The abuse of disabled children. Journal of Child Psychology and Psychiatry and Allied Disciplines, 40, 497–506.

Wolfe, V.V. (1998). Child sexual abuse. In E.J. Mash & R.A. Barkley (Eds.), Treatment of Childhood Disorders ()() (2nd ed., pp. 545-597). New York: The Guilford Press.

Wolfe, V., & Birt, J. (1997). Child sexual abuse. In Eric J. Mash & Leif G. Terdal (Eds.). Assessment of Childhood Disorders ()()()()(), (3rd Ed). pp. 569-623. New York, NY: Guilford Press.

Zirpoli, T. J. (1990). Physical abuse: Are children with disabilities at greater risk? Interventions in School and Clinic, 26, 6-11.

Zuravin, S., & Fontanella, C. (1999). Parenting behaviors and perceived parenting
competence of child sexual abuse survivors. Child Abuse & Neglect, 23(7), 623-632.

Footnotes

¹ As a result of bivariate analyses, household characteristics were excluded from logistic regression procedures.  Given the presence of multicolinearity among the variables physical injury and domestic violence, physical injury was excluded from the multivariate model.  [Back]

² A trend towards statistical significance was noted.  [Back]

³ Reference category for age was 0-7 years.  [Back]

⁴ Reference category for duration was single incident abuse.  [Back]

⁵ The reference category for relative perpetrator is non-relative, or non-kin perpetrator.  [Back]

⁶ Reference category for sexual abuse and other co-occurring maltreatment was sexual abuse.  [Back]