Predictors of recantation after child sexual abuse disclosure among a Brazilian sample

https://doi.org/10.1016/j.chiabu.2021.105006Get rights and content

Abstract

Background

Recantation after child sexual abuse (CSA) disclosure refers to the complete denial of prior reports by a victim. In substantiated cases of CSA, recantation has negative effects on criminal proceedings and the protection of current and future victims. Past studies have identified the socio-family factors and CSA characteristics that are associated with recantation. However, the systematization of these risk factors remains limited.

Objective

This study aimed to identify the predictors of recantation using a sample of criminal cases of CSA.

Participants and settings

The sample consisted of 108 criminal cases of CSA with (n = 46) and without recantation (n = 62). All the cases had been judged in a Brazilian court of law between 2010 and 2016.

Results

The final logistic regression model consisted of the following five significant predictors of recantation: a lack of support from the caregiver (odds ratio [OR] = 15.90, p < .001), disclosure to a single individual (OR = 13.75, p = .001), parental offending (OR = 12.44, p = .001), a long time interval (i.e., >one week) between disclosure and the reporting of the abuse to the authorities (OR = 8.64, p = .003), and a higher number of formal interviews (OR = 1.50, p = .052).

Conclusions

The present findings underscore not only the effects of socio-family factors and the number of interviews on recantation but also the importance of examining the contexts within which previous disclosures of CSA have been made to the assessment of the risk of recantation.

Introduction

At the core of child sexual abuse (CSA) is secrecy (Furniss, 1991; Sgroi, Blick, & Porter, 1982; Summit, 1983), which results from a combination of external (e.g., a lack of physical evidence, pressures that deter a child from disclosing abuse), relational (e.g., grooming strategies), and internal (e.g., a child may take responsibility for the sexual abuse or be afraid of the possible consequences of disclosure; Furniss, 1991) factors. Disclosure by child victims play a fundamental role to identify the CSA and enable the protection and justice system to implement child protection/intervention procedures (Alaggia, Collin-Vézina, & Lateef, 2017; Faller, 2020; Leach, Powell, Sharman, & Anglim, 2016; London, Bruck, Wright, & Ceci, 2008; Magnusson, Ernberg, & Landstrom, 2017; McGuire & London, 2020)

CSA disclosure is a dialogical and an interactional process that involves the child victim and other parties in court proceedings (Brattfjell & Flam, 2019; Grandgenett, Pittenger, Dworkin, & Hansen, 2019; Jensen, Gulbrandsen, Mossige, Reichelt, & Tjersland, 2005; Reitsema & Grietens, 2016; Staller & Nelson-Gardell, 2005). Past studies have found that, after reporting events that can be regarded as sexual abuse, some children begin to deny their previous reports when they are questioned again in an informal environment (e.g., home, school) or during formal interviews (Bradley & Wood, 1996; Elliott & Briere, 1994; Hershkowitz, Lanes, & Lamb, 2007; Katz, 2014; Malloy, Lyon, & Quas, 2007; Malloy, Mugno, Rivard, Lyon, & Quas, 2016; McGuire & London, 2020; Sgroi et al., 1982; Summit, 1983). This process of recantation also occurs when CSA has been substantiated (Elliott & Briere, 1994; Malloy et al., 2007) and previous reports have been found to be credible (Hershkowitz et al., 2007; Katz, 2014).

The consequences of recantation after CSA disclosure can be dramatic, because it undermines the credibility of the child’s testimony. As a result, the actual sex offender remains a free man or woman, and current and future victims remain unprotected (Celik et al., 2018; Molinaro & Malloy, 2016). Thus, the forensic investigation of suspected CSA should include an assessment of the presence of conditions that increase the likelihood of recantation (Celik et al., 2018; Katz, 2014; Molinaro & Malloy, 2016).

Based on observations from the clinical context, recantation is characterized as a child's response to family, community and aggressor pressure (Sgroi et al., 1982; Summit, 1983). These pressures are intended to lead the child to change the initial report, and may be accompanied by verbalizations of discredit and blaming the child (Sgroi et al., 1982; Summit, 1983). These pressures can also emerge in situations of extra-family CSA, but are more intense in cases of intra-family CSA, since the disclosure can result in family conflicts, divorces, changes in the child's life project (Celik et al., 2018; Katz, 2014; Sgroi et al., 1982). As the child feels the consequences of the disclosure, the greater the likelihood that he will deny his previous reports (Faller, 2007; Summit, 1983). Empirical studies have identified some factors related to recantation, based on CSA samples from different contexts (Bradley & Wood, 1996; Celik et al., 2018; Elliott & Briere, 1994; Gonzalez, Waterman, Kelly, & Oliveri, 1993; Gordon & Jaudes, 1996; Gries, Goh, & Cavanaugh, 1996; Gries, Goh, Andrews, Gilbert, Praver, & Stelzer, 2000; Hershkowitz et al., 2007; Katz, 2014; Malloy et al., 2007, 2016; Sorensen & Snow, 1991).According to these studies, recantation is most common among children who are aware of the negative consequences of CSA disclosure (Hershkowitz et al., 2007), experience shame in response to the reactions of their caregivers (Hershkowitz et al., 2007), exhibit behaviors that foster the maintenance of secrecy about SA, and/or have experienced polyvictimization (Celik et al., 2018). With regard to age, past studies have reported different averages (Malloy et al., 2007: 4.2 years; Gordon & Jaudes, 1996: 4.8 years; Gries et al., 1996: 6.2 years).

The literature also underscores the significant relationship that exists between recantation and the following factors: a) a victim’s familiarity with the abuser (Celik et al., 2018; Malloy et al., 2007); b) continued sexual abuse (Celik et al., 2018; Hershkowitz et al., 2007); c) nonspontaneous disclosure and disclosure to individuals who are not relatives; d) a long time interval between abuse and disclosure (Hershkowitz et al., 2007); e) a lack of support from one’s parental caregiver or another family member after the disclosure (Celik et al., 2018; Elliott & Briere, 1994; Malloy et al., 2007); and f) disbelief on the part of a relative other than the primary caregiver (Malloy et al., 2016). Within the context of the protection and justice system, the following predictors of recantation have been identified: (a) the recommendation to maintain the child and the offender in contact made at the time of the first court hearing (Malloy et al., 2016) and (b) the number of interviews conducted with the child throughout the investigation of the criminal case (Malloy et al., 2007).

The literature review shows that there are still few investigations that focus on exploring the recantation process and systematizing the factors associated with it, although this issue has been debated since the 1980s (Sgroi et al., 1982; Summit, 1983). Following this limitation, Malloy et al. (2007) carried out a research in which they found a high probability of recantation in cases involving younger children, in which the aggressor had parental functions and the nonabusive caregiver it did not guarantee child support. Despite the relevance of these results, this study did not explore the influence of other variables on recantation, namely the pattern of disclosure (spontaneous vs. non-spontaneous), the person to whom the child made the disclosure in the informal context, and the time interval between the occurrence of the CSA and disclosure (Hershkowitz et al., 2007). In addition, because disclosure constitutes a crisis for the family system (Furniss, 1991; Summit, 1983), some families may experience greater difficulties in reporting cases to the authorities (Marra and Costa, 2016). Thus, it is crucial to understand whether the time interval between the child's disclosure and communication to the authorities may constitute a risk of recantation.

The present study aimed to identify predictive factors for recantation after CSA disclosure, by analyzing criminal cases of CSA. This study provides an important contribution to the literature by integrating the process of the CSA disclosure in the informal context, the temporal aspects of disclosure, and other factors of the family context. An examination of interrelationship between risk factors for recantation in different ecological contexts must be prioritized (Katz, 2014), considering that children are agents that adjust their response according to the challenges they experience before, during and after the disclosure (Jensen et al., 2005; McElvaney, Greene, & Hogan, 2012; Staller & Nelson-Gardell, 2005). Most of the empirical studies investigated the emergence of recantation in specific settings (Celik et al., 2018; Elliott & Briere, 1994; Gordon & Jaudes, 1996; Gries et al., 1996; Hershkowitz et al., 2007; Sorensen & Snow, 1991), which can limit the analysis of recantation, since this phenomenon develops over time (Malloy et al., 2007). Thus, this study seeks to fill this gape by analyzing a sample of criminal cases of CSA, which integrates information related to different settings of formal interviews performed by the child from the first communication of the cases to the authorities until the final judgment of the sentence.

Based on previous studies, we explored the relationship between recantation and factors related to the characteristics of the victims (age, gender), sexual abuse (type of offender, severity), dynamics of disclosure in the informal context, family context and aspects of the protection and justice system. We consider, as hypotheses, that the following factors will predict recantation in criminal cases of CSA: younger victims, parental offender, lack of support from the primary caregiver, disbelief on the part of a family member, continued sexual abuse, polyvictimization, a longer time interval between abuse and disclosure, a higher number of interviews with the child, and no recommendation to keep the child away from the offender (Celik et al., 2018; Elliott & Briere, 1994; Gordon & Jaudes, 1996; Gries et al., 1996; Hershkowitz et al., 2007; Malloy et al., 2007, 2016). In addition, this study adopts as a hypothesis that recantation will be associated with financial dependence and a longer time between the CSA disclosure and the first communication to the authorities (Ajdukovic, Rajter, & Rzo, 2018; Furniss, 1991; Katz, 2014; Magnusson et al., 2017; McGuire & London, 2020; Sgroi et al., 1982; Summit, 1983)?

Section snippets

Sample

The initial sample consisted of 523 criminal cases of CSA, which had been judged in a Court of Justice in a city in Brazil between 2010 and 2016. Data were collected between January and June 2017. For the composition of the final samples, two inclusion criteria were applied: (1) only criminal files comprising documents that provided detailed information on the variables explored, and (2) in accordance with the methodology that has been used in past studies, only cases with corroborative

Results

The results are presented across two sections. First, the results of the statistical analyses that were conducted to examine group differences (i.e., criminal cases with and without recantation) in the aforementioned categorical and continuous variables (Table 1) are presented. The second section presents the results of logistic regression analysis and delineates the emergent predictors of recantation (Table 2).

Discussion

This study aimed to identify the predictors of recantation in criminal cases of CSA. The final regression model consisted of five predictors of recantation, three of which have previously been identified (Celik et al., 2018; Elliott & Briere, 1994; Malloy et al., 2007, 2016), namely, parental offending, a lack of support from the primary caregiver of the child, and the number of interviews conducted with the child. In addition, other predictors of recantation (e.g., the number of individuals to

Conclusions

This study aimed to identify the predictors of recantation after CSA disclosure. Information about the possible risk factors for recantation after sexual abuse disclosure can sensitize the protection and justice system to the need to assess and prevent child revictimization and recantation, even in cases of abuse. Using logistic regression analysis, we analyzed data extracted from court case reports related to CSA and found that the following variables were predictors of recantation: a lack of

Acknowledgement

This research was carried out through a License for Studies granted to the first author by the Court of Justice of the State of Pará (Brazil).

Pedro Augusto Dias Baía, PhD student, University of Coimbra, Faculty of Psychology and Educational Sciences, Portugal.

References (38)

  • M. Magnusson et al.

    Preschooler’s disclosures of child sexual abuse: Examining corroborated cases from Swedish courts

    Child Abuse & Neglect

    (2017)
  • L.C. Malloy et al.

    Filial dependency and recantation of child sexual abuse allegations

    Journal of the American Academy of Child and Adolescent Psychiatry

    (2007)
  • K.M. Staller et al.

    “A burden in your heart”: Lessons of disclosure from female preadolescent and adolescent survivors of sexual abuse

    Child Abuse & Neglect

    (2005)
  • R.C. Summit

    The child sexual abuse accommodation syndrome

    Child Abuse & Neglect

    (1983)
  • R. Alaggia et al.

    Facilitators and barriers to child sexual abuse (CSA) disclosures: A research update (2000–2016)

    Trauma, Violence & Abuse

    (2017)
  • G. Celik et al.

    Recantation of sexual abuse disclosure among child victims: Accommodation syndrome

    Journal of Child Sexual Abuse

    (2018)
  • M. Cyr et al.

    Psychological and physical health of nonoffending parents after disclosure of sexual abuse of their child

    Journal of Child Sexual Abuse

    (2016)
  • D.M. Elliott et al.

    Forensic sexual abuse evaluations of older children: Disclosures and symptomatology

    Behavioural Sciences and the Law

    (1994)
  • Children who do not want to disclose

  • Cited by (4)

    • Prevalence of sexual abuse and its association with health-risk behaviors among Brazilian adolescents: A populational study

      2021, Child Abuse and Neglect
      Citation Excerpt :

      Child sexual abuse has short- and long-term effects with physical and mental health consequences, including anxiety, post-traumatic stress, depression, eating disorders, sleep disorders, and suicidal and self-harm ideation (World Health Organization, 2017a). Although previous studies have investigated the prevalence of sexual abuse in adolescents and its associated factors, few studies have been conducted in a school environment in Latin America (Baía et al., 2021; Delziovo et al., 2017; Justino et al., 2015; Oliveira, 2016; Trindade et al., 2014) and without representative populational designs. A previous study that analyzed the evolution of the cases of sexual abuse in Brazil from 2009 to 2013, with a general population, identified an increase of more than five times of sexual abuse and revealed different prevalences between regions (Gaspar & Pereira, 2018).

    • Child sexual abuse in mexican women: Type of experience, age, perpetrator, and disclosure

      2021, International Journal of Environmental Research and Public Health

    Pedro Augusto Dias Baía, PhD student, University of Coimbra, Faculty of Psychology and Educational Sciences, Portugal.

    Isabel Maria Marques Alberto, PhD, University of Coimbra, Faculty of Psychology and Educational Sciences, Portugal.

    Débora Dalbosco Dell’Aglio, PhD, Graduate Program in Psychology, Federal University of Rio Grande do Sul, Brazil.

    View full text