Risk of maltreatment for siblings: Factors associated with similar and different childhood experiences in a dyadic sample of adult siblings

Risk of maltreatment for siblings: Factors associated with similar and different childhood experiences in a dyadic sample of adult siblings

Child Abuse & Neglect 76 (2018) 321–333 Contents lists available at ScienceDirect Child Abuse & Neglect journal homepage: www.elsevier.com/locate/ch...

272KB Sizes 0 Downloads 28 Views

Child Abuse & Neglect 76 (2018) 321–333

Contents lists available at ScienceDirect

Child Abuse & Neglect journal homepage: www.elsevier.com/locate/chiabuneg

Full Length Article

Risk of maltreatment for siblings: Factors associated with similar and different childhood experiences in a dyadic sample of adult siblings

T



Susanne Wittea,b, , Jörg M. Fegertb, Sabine Walpera a b

German Youth Institute, Germany Department of Child and Adolescent Psychiatry and Psychotherapy, Ulm University, Germany

AR TI CLE I NF O

AB S T R A CT

Keywords: Child maltreatment Sibling Risk factors

Siblings share the same environment and thus potentially a substantial number of risk factors for child maltreatment. Furthermore, the number of siblings and the sibling constellation itself might pose a risk for child maltreatment. Little is known about the likelihood that more than one child in a family is maltreated and which factors increase the risk. This study sought to investigate similarities and differences in maltreatment in siblings and risk factors associated with the maltreatment of more than one child from the same family. Data on maltreatment during childhood and adolescence, family background, and sibling constellation were collected from 870 pairs of siblings. In the dyadic analyses, siblings reported similar maltreatment experiences, especially when any type of maltreatment was considered. Parents’ mental health problems were significant predictors for maltreatment of at least one sibling. Father’s mental health problems were predictive of maltreatment of both or only the younger sibling, mother’s mental health problems of both or only the older sibling. Closeness in age and same gender of siblings did not emerge as a consistent predictor. The increasing number of siblings was a risk factor for any type of maltreatment of both siblings. The results highlight the need for preventive measures for families with a large number of children and with parents with mental health problems as well as a repeated risk assessment of all siblings in a family when one sibling was maltreated.

1. Introduction Siblings usually grow up in a similar environment (Bank & Kahn, 1994). Most siblings spent their childhood in the same household and difference in age is rather small. Even half and stepsiblings share similar experiences as they have at least one caretaker in common. Siblings with greater differences in age are still raised with similar values and child rearing behaviors as their siblings. Thus, they make similar but not the same experiences (Dunn & Kendrick, 1982; Zukow, 1989). Although, most risk factors for maltreatment are characteristics of the parents or the family (e.g. Stith et al., 2009) surprisingly little research has addressed the risk for sibling groups to be maltreated. Moreover, knowledge about risk factors for maltreatment of one or more siblings may be useful for making decisions about preventive measures in families in which maltreatment of one child was substantiated. In the following section a comprehensive literature review over findings on similarities and differences in maltreatment experiences of children in the same family for each type of maltreatment as well as risk factors contributing to the maltreatment of either one or more than one child in a family is provided. For the latter, Bronfenbrenner’s ecological model is used to structure the literature review.



Corresponding author at: German Youth Institute, Nockherstr. 2, 81541 Munich, Germany. E-mail address: [email protected] (S. Witte).

https://doi.org/10.1016/j.chiabu.2017.11.009 Received 30 May 2017; Received in revised form 9 October 2017; Accepted 14 November 2017 0145-2134/ © 2017 Published by Elsevier Ltd.

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

1.1. Similarities in maltreatment experiences The limited amount of research on similarities between siblings in terms of maltreatment focused mainly on sexual abuse. Rarely, more than one type of maltreatment was considered in the same study. To our knowledge, no study investigated differences between siblings in witnessing domestic violence. 1.1.1. Maltreatment Hamilton-Giachritsis and Browne (2005) found in their study of families referred to a child protection unit that in about 57% of the families with more than one child at least two siblings suffered maltreatment. Jean-Gilles and Crittenden (1990) reported similar findings. 1.1.2. Physical abuse The similarities of physical abuse between siblings vary widely based on the method used. Studies with administrative data reported that 47% to 83% of the siblings of an index child were also physically abused (Baldwin & Oliver, 1975; Hamilton-Giachritsis & Browne, 2005; Jean-Gilles & Crittenden,1990). Lindberg et al. (2012) reported lower rates of physical abuse of more than one sibling following the outcome of a medical examination. A dyadic study reported that over 50% of the index children had one physically abused sibling (MacMillan, Tanaka, Duku, Vaillancourt, & Boyle, 2013). In a college sample, the severity ratings of physical punishment of themselves and their siblings were very similar (Rausch & Knutson, 1991). Regardless of the method used for data collection, about 30% of severely abused children had at least one severely abused sibling (MacMillan et al., 2013; Smith, 1976; Smith & Hanson, 1974). 1.1.3. Sexual abuse Most studies of sexual abuse focused on girls and women. 23% to 48% of the sisters (biological or social) of sexually abused girls or women also reported sexual abuse (Herman, 1981; MacMillan et al., 2013; Muram, Speck, & Gold, 1991; Russell, 1986). Based on therapists’ reports, rates of sexual abuse of more than one sister in families undergoing treatment for sexual abuse varied between 22% to 82% (Meiselman, 1978; Phelan, 1986, 1995). The wide range reflects the differences in the definition of sexual abuse in those studies. A high variety in prevalence rates can also be found in research using perpetrators as the source of information (Proeve, 2009; Proeve, Day, Mohr, & Hawkins, 2006; Studer, Clelland, Aylwin, Reddon, & Monro, 2000). For intrafamilial sexual abuse, 22% to 34% of the perpetrators reported sexual abuse of another family member (Ballard et al., 1990; Studer et al., 2000). The only two studies with comparison groups found significant differences between the likelihood of sexual abuse of another family member between non-abused and abused participants (Boney-McCoy & Finkelhor, 1995; Harrison, Edwall, Hoffman, & Worthen, 1990). 1.1.4. Emotional abuse Only one study reported on emotional maltreatment using a large sample. In 64% of the families in which one child was emotionally maltreated another child also suffered from the same type of abuse (Hamilton-Giachritsis & Browne, 2005). 1.1.5. Neglect In 61% to 91% of the families in which one child was neglected, at least another child suffered from neglect as well (HamiltonGiachritsis & Browne, 2005; Jean-Gilles & Crittenden,1990). Sibling’s reports about their mother’s neglectful behavior correlated positively with each other (Hines, Kaufman Kantor, & Holt, 2006). In case studies, severe undernourishment of more than one sibling was reported (Barroso, Salvador, & Fagundes Neto, 2016) as well as selective neglect of one child within a family (Hollingsworth, Glass, & Heisler, 2008). 1.2. Risk factors for maltreatment of more than one child in a family Hamilton-Giachritsis and Browne (2005) assumed an increased risk to more than one or all siblings with an increasing number of risk factors on the level of parents and family. In contrast, individual characteristics of the child may lead to an increased risk for one child but not the other children within the family. 1.2.1. Child characteristics For sexual abuse, the gender of the identified victim as well as his or her siblings is considered a risk factor (Wilson, 2004). Sexual abuse of a boy is more likely to be associated with both more female and male victims within a family (Finkelhor, 1984; Proeve, 2009; Vander Mey & Neff, 1986). A twin study reported a greater similarity between same-sex twins than mixed-sex twins both for sexual abuse as well as for physical abuse (Nelson, Lynskey, Heath, Madden Pamela, & Martin, 2010). In contrast, gender constellation is also related to differences in parental behavior. In one study, boys with a sister reported more maternal emotional neglectful behavior than sibling pairs with any other gender constellation (Hines et al., 2006). Hamilton-Giachritsis and Browne (2005) did not find differences between the maltreatment of children based on children’s gender, age, or behavioral problems. 1.2.2. Characteristics of the parents and the parent-child relationship Given that parents play a key role in abuse and neglect – either as perpetrators or as protectors from external risk – it is no surprise that there is an increased risk of maltreatment of all children from one family based on parents’ characteristics. When parents abuse 322

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

alcohol or drugs, suffer from mental health problems or intellectual disabilities, are criminal, report self-injuring behavior, conflicts, or violence in the romantic relationship all children are at risk of maltreatment (Chadik, 1997; Hamilton-Giachritsis & Browne, 2005). Moreover, caretakers’ childhood abuse and neglect pose a risk to all children (Chadik, 1997). The parent-child relationship is relatively stable between siblings (Dunn, Plomin, & Daniels, 1986; Ward, Vaugh, & Robb, 1988; van Ijzendoorn et al., 2000). Nevertheless, mothers support children less who were unwanted (Barber & East, 2009). Physically abused siblings are more likely to be unwanted children than their non abused brothers and sisters (Nakou, Adam, Stathacopoulou, & Agathonos, 1982). Hollingsworth et al. (2008) reported severe scapegoating of one child out of a group of siblings. The authors argued that the mother made the child the scapegoat who challenged her self-concept as a good mother. 1.2.3. Family characteristics Disorganization of the family structure and either rigid or chaotic family climate may contribute to maltreatment of all siblings (Hamilton-Giachritsis & Browne, 2005). However, family norms such as gender roles might lead to a different risk for specific children based on children's gender and age. For example, unstable living circumstances are associated with a risk for some but not all children in a family (Hamilton-Giachritsis & Browne, 2005). More evidence of family factors related to maltreatment of siblings is based on the composition of the sibling group. The number of siblings is a risk factor for abuse and neglect (Stith et al., 2009). A higher number of siblings is associated with other risk factors such as young age of parents, low socioeconomic status, and a high burden of care for parents (Bertram, 2008). The position within the sibling group was also considered in previous research. In clinical samples from intrafamilial sexual abuse, the oldest daughter was more likely to be abused than the younger ones (Browning & Boatman, 1977; de Young, 1982; DiPietro, 1987; Vander Mey & Neff, 1986). Some authors (Haugaard & Reppucci, 1988; Herman & Hirschman,1981) reported cases of intrafamilial sexual abuse in which the father started to abuse the oldest daughter. As she got older, he began to abuse her younger sisters. In cases of physical abuse, finding regarding position in the birth order contradict each other (Baldwin & Oliver, 1975; Nakou et al., 1982). Considering the shorter time in which physical abuse could be observed in younger siblings they seem to be more likely to experience physical abuse. During infancy and early childhood, twins are at greater risk of physical abuse (Groothuis et al., 1982; Lindberg et al., 2012; Nelson & Martin, 1985). Siblings of twins also have a higher risk for physical and sexual abuse when the twin siblings report maltreatment (Nelson et al., 2010). However, the studies did not control for effects of premature birth and parental burden of care. 1.2.4. Interaction effects Environmental factors may have different effects on siblings. Moreover, some factors might change with time, e.g. psychological problems of a parent, unstable living conditions. The maltreatment of one sibling might further cause the other children to develop strategies to avoid maltreatment (de Young, 1982; Monahan, 1997) or to rationalize the maltreatment of one sibling (Hollingsworth et al., 2008). As another interaction effect, the maltreated sibling might react to the maltreatment by victimizing his or her siblings (Crittenden, 1984; Green, 1984; Tucker, Finkelhor, Turner, & Shattuck, 2014). In some cases, the perpetrator used the relationship between the siblings to gain control or to ensure secrecy (Burgess & Clark, 1984; Hollingsworth et al., 2008; Monahan, 1997). 1.3. Study aims All studies reported high similarities between siblings in maltreatment experiences. However, as most of them are based on administrative data, protocols and guidelines may lead to an underestimation of the extent of maltreatment. For example, in the study by Lindberg et al. (2012) only siblings close in age underwent medical examination. Using more than one informant, led to an increase in the percentage of children from the same family identified as suffering from sexual abuse (Muram et al., 1991; Phelan, 1986, 1995), but was rarely done. In addition, many of the studies focus only on a short age span, e.g. considering only children under five years (e.g. Smith, 1976; Smith & Hanson, 1974), or – at the other extreme – no distinction is made between family members in general (Boney-McCoy & Finkelhor, 1995; Harrison et al., 1990). In sum, the previous literature on the risk of maltreatment of more than one child in a family points on the one hand to a high significance of the issue as similarities are high, on the other hand it has been rarely investigated using a population based sample and information from more than one informant from one family. Thus, the true extent of similarities between siblings is unclear and knowledge about risk factors associated with maltreatment of more than one child in a family might be biased by the use of clinical or administrative samples. The aim of this paper is to assess 1) the prevalence of maltreatment during childhood and adolescence of none, one or both siblings in a family and 2) risk factors associated with the maltreatment of one or both siblings in a family using a large adult sample with self-reported information from two siblings from the same family. We argue that differences on the child level might lead to differences in the maltreatment experience and that risk factors on the family or parental level are predictive of conjoint victimization of siblings in the family. 2. Methods 2.1. Participants The analyses are based on a subsample from a larger sample of 4568 participants (anchor persons) who answered an online survey on adverse events and sibling relationship in childhood. The participants were recruited via online advertisement, university mailing 323

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

lists, and the SoSciPanel. The latter is a noncommercial online panel, which may be accessed for successfully reviewed research projects allowing to send invitations for online surveys to a large group of participants in all German speaking countries. During completing the online survey, participants were asked if they would like to invite their sibling to take part in the study. If participants had more than one sibling, they were asked to select their oldest sibling. If they were the oldest sibling, they should choose the sibling closest in age. Participants could then enter the e-mail address of their sibling. The sibling then received an invitation to take part in the study. The data sets of both siblings were then linked through a computer-generated pseudonym. For 870 sibling pairs, a complete data set of both siblings was obtained. To take part in the study, all participants had to give informed consent, be over 18 years old and have at least one living sibling. The study has been voted positively by Ethics Committee of the University of Ulm. 2.2. Measures All participants completed the same questionnaire, which took them in average 27 min. 2.2.1. Demographic characteristics and sibling constellation Participants provided information on their age, gender, and socioeconomic status as well as the number of siblings, the participant's position in the birth order, and their relationship to their sibling (e.g. full biological siblings). 2.2.2. Family characteristics Participants were also asked about their family situation while growing up. This included the year of birth of the parents, separation or divorce of their parents, and family members’ mental health problems. Participants who mentioned that any of their family members had psychological problems including alcohol or drug problems during their childhood and adolescence were provided with a list of family members. For the dyadic analyses, the information of the siblings provided on their family members’ mental health problems was summarized in four categories: ‘both sibling reported’, ‘no sibling reported’, ‘only the older sibling reported’, and ‘only the younger sibling reported’. 2.2.3. History of maltreatment Different types of abuse were assessed using the Childhood Trauma Questionnaire (CTQ; Bernstein et al., 2003) in its German Version (Häuser, Schmutzer, Brähler, & Glaesmer, 2011). The 28-item measure is a well-established self-report questionnaire with good psychometric properties to quantify different types of child maltreatment (sexual abuse, physical abuse, emotional abuse, physical neglect and emotional neglect) during childhood and adolescence. The internal was moderate to excellent in the specific sample (physical abuse: α = 0.82; sexual abuse: α = 0.94: emotional abuse: α = 0.86; emotional neglect: α = 0.91, physical neglect: α = 0.47). Furthermore, two items were included about intimate partner violence between parents or caretakers. For further analysis, the cut-off scores of the CTQ (Häuser et al., 2011) were used to create dichotomous variables. The items regarding intimate partner violence were dichotomous as well, indicating either witnessing domestic violence against any caregiver or none. The number of different victimizations was calculated based on those dichotomous variables, leading to a range from 0 to 6. The participants also completed questionnaires on their past and present quality of sibling relationship, their present psychological well-being and their relationship with their parents. The results of these measures are beyond the scope of this article. 2.3. Data analysis In order to carry out dyadic analyses, data sets need to be prepared in a certain way to account for nonindependence of the data. The unit of analysis is no longer the individual, but the dyad. The data was organized in the structure ‘dyad’ according to Kenny, Kashy, & Cook (2006) with each row containing the answers of both siblings. In order to do so, a decision about the way members of the dyad are distinguished has to be made (Kenny et al., 2006). In other dyadic research this often done by the nature of the relationship of the dyad members, e.g. husband − wife or mother − child. For our sample, we choose the birth order to distinguish the members of the dyad, hence creating older sibling − younger sibling dyads. The birth order is a key feature of the sibling relationship, having proven influential in a great number of studies on various outcomes in sibling research (e.g. Bank & Kahn, 1994). All statistical analyses were performed using IBM SPSS Statistics 23. In order to predict maltreatment patterns within the sibling group, a multinomial logistic regression model was calculated for each criterion. It is an extension of the logistic regression and allows to predict more than one category in comparison to a reference category. Four categories were created based on the maltreatment experiences of both siblings: ‘no sibling maltreated’, ‘only older sibling maltreated’, ‘only younger sibling maltreated’, and ‘both siblings maltreated’. The ‘no sibling maltreated’ category was chosen as the reference category. As predictors demographic variables, characteristics of the sibling constellation, and characteristics of the parents were used. Due to the high correlation between the age of the older and the younger sibling, only the age of the older sibling was used as a predictor. The same applies to the age of the parents at the birth of the older and the younger sibling. For the mental health status of the parents three variables were created for each parent based on the agreement of the siblings about their mental health status. Thus, the three variables are named: ‘reported by both’, ‘reported by older only’, and ‘reported by younger only’. All assumptions, including singularities on the HesseMatrix, were tested before analyzing the data. 324

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 1 Patterns of maltreatment within the sibling dyad only older sibling. maltreatment type

any type of maltreatment physical abuse sexual abuse emotional abuse emotional neglect physical neglect witnessing domestic violence

none

only older sibling

only younger sibling

both siblings

95%-CI

N

%

%

%

%

OR

LL

UL

870 870 869 870 870 870 870

20.69 85.98 80.55 49.08 42.76 54.37 82.76

18.74 4.71 8.75 17.59 16.09 16.67 7.59

14.14 5.40 8.75 15.17 14.02 15.63 4.02

46.44 3.91 1.96 18.16 27.13 13.33 5.63

3.63 13.20 2.06 3.34 5.14 2.78 15.27

[2.71, [7.68, [1.16, [2.49, [3.84, [2.04, [9.25,

4.86] 22.69] 3.67] 4.49] 6.89] 3.79] 25.22]

Notes. OR = odds ratio; L = lower limit; U = upper limit; CI = confidence interval.

3. Results 3.1. Demographic characteristics The sample consisted of 447 sister pairs (51.38%), 163 siblings with an older brother and a younger sister (18.74%), 162 siblings with an older sister and a younger brother (18.62%), and 93 brother pairs (10.69%). In five sibling pairs, one sibling was transgender. The average age of the older siblings was 33.44 years (SD = 11.63; n = 870) and 29.74 years for the younger siblings (SD = 11.06; n = 870). The mean age difference between the siblings was 3.70 years (SD = 2.73; n = 870). 96.09% were full biological siblings

Table 2 Factors associated with any maltreatment in the sibling dyad. older sibling only

younger sibling only

both siblings

predictor

OR

95% CI

OR

95% CI

OR

95% CI

age of older sibling

1.03*

[1.00,

1.05]

1.00

[0.98,

1.03]

1.04***

[1.02,

1.06]

gender older: female younger: female

0.79 0.47*

[0.44, [0.26,

1.43] 0.83]

0.96 0.70

[0.54, [0.40,

1.70] 1.22]

0.90 0.52**

[0.55, [0.32,

1.49] 0.85]

high educational degree older younger

1.10 0.94

[0.45, [0.41,

2.70] 2.15]

1.44 0.69

[0.59, [0.33,

3.49] 1.45]

0.59 0.72

[0.30, [0.38,

1.15] 1.38]

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

0.93 0.96 0.96 1.77 1.26*

[0.52, [0.85, [0.88, [0.16, [1.00,

1.65] 1.08] 1.04] 19.24] 1.59]

0.83 1.04 1.02 4.66 1.32*

[0.48, [0.92, [0.95, [0.49, [1.07,

1.46] 1.16] 1.11] 44.32] 1.64]

1.00 0.99 0.99 4.91 1.24*

[0.61, [0.89, [0.93, [0.60, [1.02,

1.63] 1.09] 1.06] 40.25] 1.51]

age at birth of older sibling mother father

1.04 0.96

[0.96, [0.90,

1.13] 1.02]

1.03 1.00

[0.96, [0.96,

1.10] 1.05]

0.99 1.01

[0.93, [0.96,

1.05] 1.05]

mental health reported by reported by reported by

problems mother both older only younger only

4.05** 2.73 0.82

[1.46, [0.85, [0.26,

11.24] 8.76] 2.64]

1.33 0.82 1.59

[0.44, [0.20, [0.64,

4.07] 3.29] 3.93]

5.50*** 3.67* 2.37*

[2.24, [1.34, [1.07,

13.49] 10.10] 5.26]

mental health reported by reported by reported by

problems father both older only younger only

1.29 1.07 0.55

[0.27, [0.34, [0.13,

6.10] 3.38] 2.35]

3.95* 0.28 2.49

[1.05, [0.05, [0.89,

14.94] 1.39] 7.00]

5.08** 1.91 1.58

[1.49, [0.77, [0.60,

17.31] 4.75] 4.20]

5.64** 3.25

[1.95, [0.92,

16.35] 11.55]

3.07* 0.71

[1.06, [0.15,

8.88] 3.38]

4.79** 2.15

[1.81, [0.68,

12.65] 6.85]

a

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.24; Log-Likelihood: 1975.56; Omnibus-Test: Χ2(60) = 211.76; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

325

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 3 Factors associated with physical abuse in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI **

[1.02,

1.07]

1.03

0.64 0.86

[0.32, [0.43,

1.31] 1.71]

high educational degree older younger

0.55 0.67

[0.26, [0.29,

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

1.25 0.94 1.47 0.93 0.92

age at birth of older sibling mother father mental health reported by reported by reported by mental health reported by reported by reported by

95% CI *

95% CI ***

[1.01,

1.07]

1.07

[1.03,

1.10]

0.99 0.77

[0.47, [0.37,

2.12] 1.62]

0.72 1.42

[0.24, [0.51,

2.11] 3.95]

1.20] 1.52]

1.54 0.90

[0.50, [0.32,

4.71] 2.50]

0.39 0.77

[0.14, [0.27,

1.08] 2.23]

[0.62, [0.82, [0.40, [0.84, [0.70,

2.49] 1.07] 5.48] 1.02] 1.21]

1.36 0.86 0.45 0.94 1.07

[0.65, [0.73, [0.05, [0.85, [0.82,

2.83] 1.00] 3.91] 1.05] 1.39]

1.53 0.85 2.13 0.99 1.38**

[0.55, [0.70, [0.42, [0.86, [1.11,

4.22] 1.03] 10.71] 1.14] 1.73]

0.98 0.97

[0.88, [0.89,

1.09] 1.06]

0.99 0.96

[0.89, [0.88,

1.11] 1.05]

0.82** 1.06

[0.71, [0.97,

0.95] 1.16]

problems mother both older only younger only

2.96* 1.29 2.16

[1.25, [0.34, [0.83,

7.03] 4.93] 5.59]

1.44 1.01 3.61**

[0.51, [0.27, [1.48,

4.08] 3.77] 8.78]

6.84** 7.90** 2.07

[2.31, [2.46, [0.46,

20.23] 25.36] 9.31]

problems father both older only younger only

0.30 1.82 1.46

[0.04, [0.55, [0.46,

2.34] 5.96] 4.62]

2.09 1.79 1.83

[0.72, [0.48, [0.60,

6.03] 6.64] 5.52]

3.96* 8.45*** 6.93**

[1.12, [2.71, [1.84,

14.02] 26.35] 26.10]

0.98 2.89

[0.35, [0.88,

2.74] 9.43]

2.12 1.58

[0.89, [0.33,

5.05] 7.45]

0.70 5.60*

[0.20, [1.43,

2.46] 22.03]

age of older sibling

1.04

gendera older: female younger: female

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.26; Log-Likelihood: 801.55; Omnibus-Test: Χ2(60) = 166.43; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

(n = 836), 2.64% maternal half-siblings (n = 23) and 0.92% paternal half-siblings (n = 8). Two sibling pairs were stepsiblings (0.23%) and one were adoptive siblings (0.11%). In 12.41% of the sibling pairs (n = 108) the parents got divorced while both were under 18 years old, in 4.02% of the dyads (n = 35) only the younger sibling was under 18 years old. 3.2. History of child maltreatment Patterns of maltreatment experiences in the sibling dyad are presented in Table 1. The number of different victimizations of the older siblings with the younger correlated significantly (r = 0.51; p < 0.001; n = 870). In general, all types of maltreatment of one sibling were associated with an increased likelihood that the other sibling would also report this type of maltreatment. Using the severity scales of the CTQ, moderate to high correlations between the scores of both siblings were found (physical abuse: r = 0.48; p < 0.001; sexual abuse: r = 0.13; p < 0.001; emotional abuse: r = 0.39; p < 0.001; emotional neglect: r = 0.51; p < 0.001; physical neglect: r = 0.35; p < 0.001). 3.3. Predicting patterns of maltreatment between siblings The results of the multinomial regression analyses for any type of maltreatment and the specific types are depicted in Tables 2–8 . 3.3.1. Sibling individual characteristics Age of the older sibling was a significant predictor for all criteria for the category both siblings, but also for some types of maltreatment of the older and the younger sibling only. Older sisters were more likely to report sexual abuse and emotional abuse solely than any other category. Younger brothers, more than younger sisters, were associated with a greater likelihood of any type of maltreatment of either only the older or only the younger sibling. In full biological siblings, either solely the older sibling or both siblings were more likely to report witnessing domestic violence than in any other sibling constellation. 326

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 4 Factors associated with sexual abuse in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI

age of older sibling

1.06

***

[1.03,

1.08]

1.03

gendera older: female younger: female

4.18*** 0.82

[1.97, [0.40,

8.84] 1.71]

high educational degree older younger

0.97 2.15

[0.46, [0.90,

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

0.69 0.87* 2.45 0.96 1.04

age at birth of older sibling mother father mental health reported by reported by reported by mental health reported by reported by reported by

95% CI **

95% CI **

[1.01,

1.05]

1.07

[1.02,

1.12]

1.01 1.82

[0.54, [0.98,

1.91] 3.36]

1.43 1.60

[0.31, [0.36,

6.56] 7.09]

2.06] 5.13]

0.81 1.12

[0.39, [0.51,

1.67] 2.47]

1.85 1.44

[0.34, [0.32,

10.07] 6.40]

[0.33, [0.77, [0.79, [0.89, [0.85,

1.43] 0.98] 7.61] 1.05] 1.28]

0.41** 1.01 1.61 1.03 1.20*

[0.22, [0.91, [0.50, [0.95, [1.01,

0.75] 1.13] 5.12] 1.10] 1.42]

0.92 0.85 1.99 0.86 0.57

[0.21, [0.68, [0.27, [0.73, [0.30,

4.04] 1.08] 14.71] 1.02] 1.11]

1.00 0.91*

[0.91, [0.84,

1.10] 0.99]

1.08 0.92*

[0.99, [0.86,

1.17] 0.99]

1.12 0.85

[0.92, [0.71,

1.37] 1.01]

problems mother both older only younger only

1.87 1.97 2.11

[0.85, [0.82, [0.89,

4.15] 4.71] 4.98]

1.31 1.64 2.31*

[0.56, [0.63, [1.11,

3.06] 4.23] 4.81]

9.99*** 4.53 1.78

[2.74, [0.93, [0.19,

36.45] 22.09] 16.90]

problems father both older only younger only

0.89 1.67 2.40

[0.31, [0.67, [0.93,

2.56] 4.20] 6.18]

1.56 2.00 3.15**

[0.60, [0.78, [1.39,

4.06] 5.12] 7.12]

3.28 4.42 1.64

[0.72, [0.98, [0.18,

14.92] 19.81] 14.84]

1.44 1.85

[0.66, [0.61,

3.16] 5.57]

1.38 1.32

[0.65, [0.42,

2.92] 4.16]

1.83 0.98

[0.44, [0.10,

7.63] 10.09]

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.23; Log-Likelihood: 1000.69; Omnibus-Test: Χ2(60) = 162.90; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

3.3.2. Sibling constellation characteristics Age difference had divergent effects for different kinds of abuse. A larger age difference between siblings was predictive domestic violence witnessed by the older sibling only and reduced the likelihood of a report of sexual abuse of the older sibling only. Mixedgender sibling composition was a risk factor for sexual abuse of the younger sibling only. The number of siblings predicted any maltreatment reported by either one or both siblings. For the specific types of maltreatment, the number of siblings was a significant predictor of sexual abuse, emotional abuse and emotional neglect of the younger sibling only. 3.3.3. Characteristics of the parents The age of the father at the birth of the older sibling was a significant predictor of sexual abuse of only the older and only the younger sibling. The mother’s young age at birth of the older child was associated with physical abuse of both siblings. Mother’s mental health problems during the childhood of both siblings predicted an increased risk for each and any type of maltreatment of both siblings and the older sibling only. Similarly, father’s mental health problems were predictive for each and any type of maltreatment of both siblings (except sexual abuse). Different from mother’s mental health problems, father’s mental health problems increased the risk for each and any type of maltreatment of solely the younger sibling. The different perceptions and experiences of the siblings with their parents' mental health problems predicted some of the differences between the siblings. However, the report of the older sibling was predictive of any maltreatment of both siblings. The parents’ divorce or separation was a predictor for any type of maltreatment for either only one or both siblings. A divorce during the childhood of the younger sibling was also a significant predictor for emotional and physical neglect of the older one only. 4. Discussion This study retrospectively identified the risk of child maltreatment for siblings who grew up together in one family. In line with previous findings, the risk for maltreatment for their sibling is increased. 327

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 5 Factors associated with emotional abuse in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI

95% CI

95% CI *

age of older sibling

1.01

[0.99,

1.03]

0.99

[0.97,

1.01]

1.02

[1.00,

1.04]

gendera older: female younger: female

2.19** 0.69

[1.30, [0.42,

3.70] 1.15]

0.95 1.23

[0.59, [0.77,

1.53] 1.97]

1.50 1.46

[0.89, [0.88,

2.52] 2.42]

high educational degree older younger

0.77 1.55

[0.40, [0.77,

1.47] 3.11]

1.03 0.92

[0.54, [0.49,

1.96] 1.74]

0.90 0.93

[0.48, [0.49,

1.69] 1.75]

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

0.92 0.99 0.50 0.94 1.09

[0.55, [0.90, [0.13, [0.88, [0.92,

1.54] 1.09] 1.91] 1.01] 1.30]

1.62* 1.01 1.73 0.95 1.24**

[1.02, [0.92, [0.60, [0.89, [1.06,

2.59] 1.11] 5.00] 1.02] 1.44]

0.70 0.98 1.33 0.94 1.09

[0.42, [0.89, [0.47, [0.88, [0.91,

1.16] 1.07] 3.80] 1.01] 1.29]

age at birth of older sibling mother father

0.98 1.00

[0.91, [0.95,

1.04] 1.06]

0.98 1.02

[0.92, [0.98,

1.04] 1.06]

0.96 1.00

[0.89, [0.95,

1.03] 1.05]

mental health reported by reported by reported by

problems mother both older only younger only

3.62*** 3.50** 1.38

[1.77, [1.65, [0.55,

7.42] 7.45] 3.47]

1.24 0.58 2.00

[0.56, [0.19, [0.98,

2.74] 1.73] 4.08]

8.62*** 4.56*** 5.24***

[4.54, [2.13, [2.66,

16.37] 9.75] 10.29]

mental health reported by reported by reported by

problems father both older only younger only

1.31 3.08** 0.67

[0.48, [1.38, [0.18,

3.60] 6.87] 2.50]

5.29*** 1.48 3.99**

[2.37, [0.53, [1.82,

11.81] 4.13] 8.76]

4.38*** 3.00* 2.55*

[1.93, [1.29, [1.11,

9.93] 6.93] 5.88]

2.06* 1.13

[1.06, [0.37,

4.02] 3.43]

1.47 0.50

[0.77, [0.13,

2.82] 1.88]

2.31* 2.32

[1.22, [0.94,

4.37] 5.69]

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.28; Log-Likelihood: 1894.32; Omnibus-Test: Χ2(60) = 257.80; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

4.1. Similarities between siblings In about 59% of the sibling pairs in which one of the siblings reported any type of maltreatment the other sibling reported maltreatment as well in the presented study. This result is similar to the findings of Hamilton-Giachritsis and Browne (2005) as well as Jean-Gilles and Crittenden (1990). For physical abuse, the rate for both siblings suffering from the same abuse in this study is consistent with those found in other studies (MacMillan et al., 2013; Smith & Hanson, 1974). Results based on analyses of administrative data from child protection units found higher similarities between siblings (Baldwin & Oliver, 1975; Hamilton-Giachritsis & Browne,2005) but results from medical examinations showed fewer similarities (Lindberg et al., 2012). In about 10% of the sibling pairs in which one reported sexual abuse the other sibling also reported sexual abuse. This rate is slightly lower than those rates reported by Boney-McCoy and Finkelhor (1995) and MacMillan et al. (2013). The difference in the rates compared to the presented sample may be caused by the nature of the sampling strategy that could have contributed to biased drop out in cases in which both siblings were maltreated. In clinical samples (Herman & Hirschman, 1981; Phelan, 1986), in analyses of administrative data (Finkelhor, 1984; Hamilton-Giachritsis & Browne, 2005), and in studies conducted with perpetrators (Ballard et al., 1990; Proeve et al., 2006; Studer et al., 2000) the rate was considerably higher. The similarities between siblings for emotional abuse, emotional neglect, and physical abuse in the presented sample are lower than the rates reported by Hamilton-Giachritsis and Browne (2005). This may be due to different methods used in sampling and the fact that in this study only two siblings of the sibling group were analyzed. 4.2. Risk factors for maltreatment In line with changing norms in child rearing, including a decline in the acceptance of corporal punishment (Bussmann, 2005), older age was associated with higher rates of maltreatment, especially for both siblings in the multinomial regression analyses. A 328

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 6 Factors associated with emotional neglect in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI ***

[1.02,

1.07]

1.02

0.57* 0.71

[0.34, [0.43,

0.96] 1.16]

high educational degree older younger

0.58 1.49

[0.29, [0.70,

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

1.09 0.92 1.55 0.89** 1.13

age at birth of older sibling mother father mental health reported by reported by reported by mental health reported by reported by reported by

95% CI *

95% CI ***

[1.00,

1.05]

1.04

[1.02,

1.06]

1.31 0.68

[0.80, [0.42,

2.14] 1.10]

1.04 0.67

[0.66, [0.43,

1.62] 1.02]

1.13] 3.19]

0.94 1.64

[0.45, [0.75,

1.97] 3.59]

0.62 0.62

[0.35, [0.36,

1.11] 1.07]

[0.66, [0.83, [0.52, [0.83, [0.94,

1.79] 1.02] 4.60] 0.96] 1.35]

0.78 1.00 0.62 0.95 1.20*

[0.48, [0.91, [0.14, [0.89, [1.01,

1.26] 1.11] 2.69] 1.02] 1.42]

1.40 0.99 1.13 1.00 1.10

[0.91, [0.91, [0.40, [0.94, [0.95,

2.16] 1.08] 3.21] 1.06] 1.29]

1.00 0.96

[0.93, [0.91,

1.08] 1.02]

1.07 0.96

[1.00, [0.92,

1.14] 1.02]

0.99 0.99

[0.93, [0.94,

1.05] 1.03]

problems mother both older only younger only

3.30** 3.76** 0.79

[1.51, [1.60, [0.29,

7.19] 8.85] 2.10]

2.46* 0.62 1.93

[1.14, [0.17, [0.94,

5.30] 2.30] 3.95]

5.00*** 3.99*** 2.54**

[2.64, [1.90, [1.32,

9.47] 8.35] 4.88]

problems father both older only younger only

0.67 2.27 0.42

[0.17, [0.96, [0.11,

2.59] 5.40] 1.54]

2.94* 0.75 1.79

[1.19, [0.24, [0.79,

7.30] 2.38] 4.05]

4.92*** 1.96 1.55

[2.24, [0.90, [0.71,

10.78] 4.25] 3.39]

1.84 4.57**

[0.91, [1.54,

3.68] 13.54]

1.10 1.34

[0.54, [0.37,

2.27] 4.79]

1.83* 3.27*

[1.02, [1.20,

3.30] 8.90]

age of older sibling

1.04

gendera older: female younger: female

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.28; Log-Likelihood: 1955.56; Omnibus-Test: Χ2(60) = 253.33; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

higher prevalence of physical neglect has also been found among older cohorts in other samples in Germany (Häuser et al., 2011). 4.2.1. Sibling individual characteristics Gender differences in the prevalence of sexual abuse have been widely reported (Häuser et al., 2011; Stoltenborgh, BakermansKranenburg, Alink, & van IJzendoorn, 2012). In the presented sample, older sisters were more likely to exclusively report sexual and emotional abuse without younger siblings being affected. In mixed-gender dyads, it was more likely that only the younger siblings were sexually abused. This is probably due to sexual preferences of the perpetrator. Lower level of education of the older sibling was associated a higher likelihood for physical neglect for both siblings. This effect was also reported in other studies (Häuser et al., 2011; Stith et al., 2009). The data was collected retrospectively, so the deteriorating effect of physical neglect on educational achievement has to be considered as a possible bidirectional effect (Baker et al., 2016; Corrales et al., 2016). 4.2.2. Sibling constellation characteristics Full biological siblings reported more often that they witnessed domestic violence. Parents with more than one child live together longer so there might be a longer period in which intimate partner violence can occur. Moreover, having a full biological sibling and a large age difference between siblings was also associated with reports about witnessing domestic violence by the older sibling only. Domestic violence might have occurred between caretakers before the younger sibling was born or could remember the incidents. Other researchers found that older siblings tend to experience domestic violence as more extreme (Skopp, McDonald, Manke, & Jouriles, 2005). In contrast to other studies (Groothuis et al., 1982; Lindberg et al., 2012; Nelson & Martin, 1985), difference in age did not predict physical abuse. It is possible that the risk for twins is extremely high in the first years after their birth, but decreases over time and thus lead to no difference in patterns on maltreatment between siblings when the whole childhood and adolescence are considered. A small difference in age was predictive of sexual abuse of the older sibling only, which is in line with the findings of Haugaard and Reppucci (1988) and Herman and Hirschman (1981). The authors found that in cases of intrafamilial abuse that the father started to abuse the older daughter. When she moved out of the household, he started to abuse the younger sister. Hence, if 329

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 7 Factors associated with physical neglect in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI **

95% CI

[1.01,

1.05]

1.02

[1.00,

1.04]

1.03

0.75 0.62*

[0.47, [0.39,

1.20] 0.98]

0.81 0.62*

[0.52, [0.40,

1.27] 0.97]

high educational degree older younger

0.68 0.97

[0.35, [0.53,

1.29] 1.80]

0.73 1.11

[0.38, [0.59,

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

1.01 0.94 1.90 0.95 1.06

[0.64, [0.86, [0.67, [0.89, [0.90,

1.59] 1.03] 5.43] 1.01] 1.24]

0.84 0.96 1.98 0.97 1.06

age at birth of older sibling mother father

1.01 0.99

[0.95, [0.94,

1.08] 1.04]

mental health reported by reported by reported by

problems mother both older only younger only

2.62** 2.22* 0.68

[1.37, [1.05, [0.28,

mental health reported by reported by reported by

problems father both older only younger only

1.85 1.92 0.84 1.66 3.17*

age of older sibling

1.03

gendera older: female younger: female

separation/divorce of parents childhood of both childhood of younger

95% CI **

[1.01,

1.05]

0.81 0.62

[0.47, [0.37,

1.39] 1.04]

1.38] 2.10]

0.33** 1.29

[0.18, [0.63,

0.62] 2.65]

[0.54, [0.88, [0.67, [0.91, [0.90,

1.31] 1.06] 5.83] 1.03] 1.24]

1.61 0.91 1.87 0.99 1.05

[0.95, [0.82, [0.61, [0.92, [0.88,

2.73] 1.01] 5.70] 1.06] 1.24]

0.96 1.04

[0.90, [1.00,

1.02] 1.08]

0.98 1.00

[0.91, [0.94,

1.06] 1.05]

5.00] 4.69] 1.63]

1.77 1.20 1.12

[0.89, [0.51, [0.56,

3.53] 2.84] 2.26]

4.53*** 2.96** 2.41*

[2.36, [1.37, [1.17,

8.71] 6.41] 4.93]

[0.83, [0.82, [0.29,

4.12] 4.50] 2.38]

1.96 2.13 2.22*

[0.88, [0.91, [1.03,

4.39] 5.00] 4.80]

3.79** 4.20** 2.27

[1.75, [1.84, [0.99,

8.20] 9.58] 5.24]

[0.88, [1.27,

3.15] 7.88]

1.17 1.75

[0.61, [0.62,

2.25] 4.94]

2.43** 2.83

[1.29, [1.00,

4.56] 8.04]

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.20; Log-Likelihood: 1870.36; Omnibus-Test: Χ2(60) = 168.39; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

both siblings are very close in age it is more likely that both moved out at the same time and only the older was sexually abused. Emotional neglect of the older sibling only was associated with a shorter duration of living together. This is because the older sibling moved out of the household at an earlier age. However, due to the nature of an online survey it was not possible to collect information on the nature of the relationship to the perpetrator of the sexual abuse. As in other studies (Stith et al., 2009), the number of siblings was a risk factor for maltreatment. It was a risk factor for all types of patterns between siblings if any type of maltreatment was considered. For the specific types, it predicted substantially emotional neglect, sexual abuse, and emotional abuse of the younger sibling only. It might be that younger siblings in large families are at a higher risk for those types of maltreatment. However, a large number of siblings was predictive of reports of physical abuse of both siblings. 4.2.3. Characteristics of parents Mother's young age at birth of the older siblings predicted physical abuse of both siblings. In other studies on risk factors for physical abuse, the caretakers' young age was also considered a risk factor (Stith et al., 2009). Father's young age at birth of the older sibling was a risk factor for sexual abuse of the older or the younger sibling only. Mental health problems of both parents were risk factors for any type of maltreatment of both siblings and all specific types of maltreatment except for the father's mental health problems for sexual abuse. Overall, mother's mental health problems were more predictive for maltreatment of solely the older sibling than solely the younger sibling. The father's mental health problems were slightly more predictive of maltreatment of the younger sibling only. To our knowledge, there has been no study on the differential influence of mental health problems of parents on siblings. For future research interactive effects of birth order of a child and parents' mental health problems leading to specific risks for the younger or the older sibling have to be further investigated. Divorce or separation of the parents was a risk factor for any type of maltreatment for all patterns between siblings. This finding is in line with other research (Stith et al., 2009). The influence of the divorce of the parents during the childhood of the younger sibling on the older sibling is very likely linked to the fact that the effects of the divorce also originate in conflicts before the divorce (Amato 330

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Table 8 Factors associated with witnessing domestic violence in the sibling dyad.

predictor

older sibling only

younger sibling only

both siblings

OR

OR

OR

95% CI

95% CI

95% CI *

age of older sibling

1.01

[0.98,

1.04]

1.02

[0.99,

1.06]

1.03

[1.00,

1.06]

gendera older: female younger: female

1.39 0.53

[0.71, [0.28,

2.69] 1.02]

1.50 0.92

[0.63, [0.39,

3.57] 2.14]

1.14 1.43

[0.48, [0.61,

2.73] 3.36]

high educational degree older younger

1.02 0.69

[0.45, [0.31,

2.30] 1.50]

0.96 0.57

[0.34, [0.22,

2.76] 1.44]

0.75 1.96

[0.30, [0.66,

1.89] 5.88]

sibling constellation same-gender age difference full biological siblings duration living together number of siblings

1.21 1.16* 4.50** 1.01 0.96

[0.63, [1.03, [1.56, [0.93, [0.78,

2.31] 1.30] 12.93] 1.09] 1.20]

0.71 1.02 2.35 1.02 0.96

[0.30, [0.87, [0.43, [0.92, [0.72,

1.66] 1.20] 12.74] 1.14] 1.29]

0.99 0.89 7.08** 0.92 0.90

[0.42, [0.77, [2.17, [0.83, [0.69,

2.33] 1.04] 23.10] 1.02] 1.17]

age at birth of older sibling mother father

0.95 1.03

[0.86, [0.97,

1.04] 1.10]

0.98 1.02

[0.87, [0.94,

1.10] 1.10]

0.92 1.06

[0.83, [0.99,

1.02] 1.13]

mental health reported by reported by reported by

problems mother both older only younger only

2.32* 2.04 1.03

[1.05, [0.86, [0.37,

5.13] 4.85] 2.86]

3.96** 3.22* 2.29

[1.55, [1.03, [0.75,

10.13] 10.09] 6.97]

6.03*** 4.02** 1.64

[2.68, [1.55, [0.51,

13.58] 10.46] 5.33]

mental health reported by reported by reported by

problems father both older only younger only

3.08* 5.17*** 2.16

[1.31, [2.20, [0.71,

7.23] 12.20] 6.56]

2.37 1.13 2.10

[0.78, [0.24, [0.63,

7.13] 5.47] 7.02]

3.22* 5.44** 2.66

[1.20, [2.03, [0.88,

8.66] 14.60] 8.08]

3.62*** 2.31

[1.80, [0.69,

7.26] 7.75]

2.11 1.65

[0.80, [0.34,

5.55] 7.93]

3.06** 5.50**

[1.34, [1.82,

6.98] 16.64]

separation/divorce of parents childhood of both childhood of younger

Notes. reference category ‘none of the siblings’; n = 859 (98.74%); OR: Odds Ratio; 95%-CI: 95%-confidence interval; R2 Nagelkerke: 0.27; Log-Likelihood: 925.98; Omnibus-Test: Χ2(60) = 189.58; p < 0.001. a male and other as reference category. * p < 0.050. ** p < 0.010. *** p < 0.001.

& Keith, 1991). In sum, very few risk factors differentiate clearly between sibling dyads with maltreatment of one or both siblings in respect to all types of maltreatment. Strong risk factors for maltreatment of both siblings are characteristics of the parents and the family situation. Differences between siblings can be explained only partially by individual characteristics of each child. A tendency exists for younger siblings being more at risk in families with a large number of children and with fathers with mental health problems. Mother’s mental health problems are more strongly associated with maltreatment of the older sibling only or both siblings. 4.3. Strengths and limitations The sample is not representative for the German population, with more female participants and participants with higher educational background. The comparison of the prevalence rates showed no deviation from population based samples in Germany for the physical abuse, sexual abuse, physical neglect, and emotional neglect, when age and gender of the participants were considered (Häuser et al., 2011; Witt, Brown, Plener, Brähler & Fegert, 2017). The prevalence of emotional abuse was higher than in other samples, yet the difference was less pronounced in comparison to more recent samples (Witte, 2017 in press). The higher rate may be even more enforced by the method of data collection and the use of a convenience sample, which has been found to be a confounding factor on emotional abuse in previous studies (Stoltenborgh et al., 2012). However, as information on the number of siblings was not collected in any other study, the influence of the effects of having siblings and other confounding factors remains unclear. The current study demonstrates the feasibility of recruiting adult sibling pairs in an easy and cost effective way. To our knowledge, this was the first study to recruit sibling pairs using a study design with an online survey in which the second sibling was invited by the first. This led to a low threshold for participation in the study with siblings being able to take part in the study even if they lived abroad. As the study has been conducted online, there might be a bias in sampling due to internet usage. However, over 89.7% of the German population are frequently using the internet (International Telecommunication Union, 2017). The specificity analysis (Witte, in press) showed that siblings with a better relationship were more likely to take part. However, maltreatment status was not a significant 331

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

predictor. Another limitation is that more sisters took part in the study. This is a common issue for sibling research especially with adult sisters seeming to be more interested in questions of the sibling relationship. Most measures in the study were retrospective as well as self-report. This might influence the results, as the answers may be prone to memory bias, bias due to social desirability, or other response patterns. However, a careful plausibility analysis was carried out. Moreover, the CTQ has been shown to be a reliable instrument to assess childhood trauma (Häuser et al., 2011). The reports of the participants were not compared to official reports using official administrative data. This was not possible due to a lack of an overall person-centered data collection in Germany. Furthermore, this data would be also prone to bias as not all types of maltreatment might not have been detected by official authorities. In this study design, only one sibling of a person was included. This may have led to an underestimation of the extent of maltreatment of all children within a family. However, many participants only had one sibling. In addition, information from other family members would have been useful to confirm the mental health status of the parents during childhood, which was only assessed using information from the siblings. Asking more than two family members was not feasible, as it might have led to an increase of the attrition. 4.4. Implications for practice and research Social workers and other professionals working with children and adolescents should consider a risk of maltreatment of all children in a family as soon as one child is reported. As the understanding of risk factors associated with maltreatment of one or more siblings is limited, a risk assessment should be carried out for every child regardless of age, gender or age difference between siblings. Offering support and advice to families with a large number of children may be crucial to prevent child maltreatment. Special consideration should be given to the later born children. The results as previous studies highlight the need for preventive measures for families with parents with mental health issues and young parents. An important aspect is to consider the possible role of the father for maltreatment of the younger sibling. To fully understand the process of maltreatment of one, more than one, and all children in a family more research has to be carried out on risk factors and dynamics within the family. Future research should also consider the timing and the perpetrators of maltreatment in more detail. References Amato, P. R., & Keith, B. (1991). Parental divorce and the well-being of children: A meta-analysis. Psychological Bulletin, 110(1), 26–46. http://dx.doi.org/10.1037/0033-2909. 110.1.26. Baker, M. R., Frazier, P. A., Greer, C., Paulsen, J. A., Howard, K., Meredith, L. N., et al. (2016). Sexual victimization history predicts academic performance in college women. Journal of Counseling Psychology, 1–8. http://dx.doi.org/10.1037/cou0000146. Baldwin, J. A., & Oliver, J. E. (1975). Epidemiology and family characteristics of severely-abused children. Journal of Epidemiology & Community Health, 29(4), 205–221. http:// dx.doi.org/10.1136/jech.29.4.205. Ballard, D. T., Blair, G. D., Devereaux, S., Valentine, L. K., Horton, A. L., & Johnson, B. L. (1990). A comparative profile of the incest perpetrator: Background, characteristics, abuse history, and use of social skills. In A. L. Horton (Ed.). The incest perpetrator. A family member no one wants to treat (pp. 43–64). (1st ed.). Newbury Park: Sage. Bank, S. P., & Kahn, M. D. (1994). Geschwister-Bindung (Ungekürzte Ausg). Dtv: 35072: Dialog und Praxis. München: Dt. Taschenbuch-Verl. Barber, J. S., & East, P. L. (2009). Home and parenting resources available to siblings depending on their birth intention status. Child Development, 80(3), 921–939. http://dx.doi. org/10.1111/j.1467-8624.2009.01306.x. Barroso, M. M. B., Salvador, L. M., & Fagundes Neto, U. (2016). Severe protein-calorie malnutrition in two brothers due to abuse by starvation [Desnutricao proteico-calorica grave em dois irmaos devido ao abuso por privacao alimentar]. Revista paulista de pediatria: Orgao oficial da Sociedade de Pediatria de Sao Paulohttp://dx.doi.org/10.1016/j.rpped. 2016.05.001. Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., et al. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse & Neglect, 27(2), 169–190. http://dx.doi.org/10.1016/S0145-2134(02)00541-0. Bertram, H. (2008). Die Mehrkinderfamilie in Deutschland. Zur demographischen Bedeutung der Familie mit drei und mehr Kindern und zu ihrer ökonomischen Situation. Retrieved from https://www.bmfsfj.de/blob/jump/76420/mehrkinderfamilie-bertram-data.pdf. Boney-McCoy, S., & Finkelhor, D. (1995). Prior victimization: A risk factor for child sexual abuse and for PTSD-related symptomatology among sexually abused youth. Child Abuse & Neglect, 19(12), 1401–1421. http://dx.doi.org/10.1016/0145-2134(95)00104-9. Browning, D. H., & Boatman, B. (1977). Incest: Children at risk. The American Journal of Psychiatry, 134(1), 69–72. http://dx.doi.org/10.1176/ajp.134.1.69. Burgess, A. W., & Clark, M. L. (1984). Child pornography and sex rings. Lexington, Mass: LexingtonBooks. Bussmann, K.-D. (2005). Report über die Auswirkungen des Gesetzes zur Ächtung der Gewalt in der Erziehung. Retrieved from www.uni-bielefeld.de/ikg/zick/Bussmann-Report.pdf. Chadik, C. A. (1997). Child abuse risk factors for siblings of identified victims. Dissertation Abstracts International Section A: Humanities and Social Sciences, Vol. 57Dissertation Abstracts International (9-A), Mar, 1997. Corrales, T., Waterford, M., Goodwin-Smith, I., Wood, L., Yourell, T., & Ho, C. (2016). Childhood adversity, sense of belonging and psychosocial outcomes in emerging adulthood: A test of mediated pathways. Children and Youth Services Review, 63, 110–119. http://dx.doi.org/10.1016/j.childyouth.2016.02.021. Crittenden, P. M. (1984). Sibling interaction: Evidence of a generational effect in maltreating infants. Child Abuse & Neglect, 8(4), 433–438. http://dx.doi.org/10.1016/01452134(84)90024-3. de Young, M. (1982). The sexual victimization of children. Jefferson, N.C: McFarland. DiPietro, S. B. (1987). The effects of intrafamilial child sexual abuse on the adjustment and attitudes of adolescents. Violence and Victims, 2(1), 59–78. Dunn, J., & Kendrick, C. (1982). Siblings: Love, envy, & understanding. Cambridge, Mass: Harvard University Press. Dunn, J. F., Plomin, R., & Daniels, D. (1986). Consistency and change in mothers' behavior toward young siblings. Child Development, 57(2), 348. Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press. Green, A. H. (1984). Child abuse by siblings. Child Abuse & Neglect, 8(3), 311–317. http://dx.doi.org/10.1016/0145-2134(84)90072-3. Groothuis, J. R., Altemeier, W. A., Robarge, J. P., O'Connor, S., Sandler, H., Vietze, P., et al. (1982). Increased child abuse in families with twins. Pediatrics, 70(5), 769–773. http://dx.doi.org/10.1002/nur.4770050406. Häuser, W., Schmutzer, G., Brähler, E., & Glaesmer, H. (2011). Maltreatment in childhood and adolescence: Results from a survey of a representative sample of the german population. Deutsches Ärzteblatt, 108(17), 287–294. http://dx.doi.org/10.3238/arztebl.2011.0287. Hamilton-Giachritsis, C. E., & Browne, K. D. (2005). A retrospective study of risk to siblings in abusing families. Journal of Family Psychology, 19(4), 619–624. http://dx.doi.org/ 10.1037/0893-3200.19.4.619. Harrison, P. A., Edwall, G. E., Hoffman, N. G., & Worthen, M. D. (1990). Correlates of sexual abuse among boys in treatment for chemical dependency. Journal of Adolescent Chemical Dependency, 1(1), 53–67. http://dx.doi.org/10.1300/J272v01n01_04.

332

Child Abuse & Neglect 76 (2018) 321–333

S. Witte et al.

Haugaard, J. J., & Reppucci, N. D. (1988). The sexual abuse of children: A comprehensive guide to current knowledge and intervention strategiesThe Jossey-Bass social and behavioral science series (1st ed.). San Francisco: Jossey-Bass Publishers. Herman, J. L., & Hirschman, L. (1981). Families at risk for father-daughter incest. American Journal of Psychiatry, 138(7), 967–970. http://dx.doi.org/10.1007/978-1-4684-47545_17. Herman, J. L. (1981). Father-daughter incest. Cambridge, Mass: Harvard Univ. Pr. Hines, D. A., Kaufman Kantor, G., & Holt, M. K. (2006). Similarities in siblings’ experiences of neglectful parenting behaviors. Child Abuse & Neglect, 30(6), 619–637. http://dx.doi. org/10.1016/j.chiabu.2005.11.008. Hollingsworth, J., Glass, J., & Heisler, K. W. (2008). Empathy deficits in siblings of severely scapegoated children. Journal of Emotional Abuse, 7(4), 69–88. http://dx.doi.org/10. 1300/J135v07n04_04. International Telecommunication Union (2016). Individuals using the internet from 2000 to 2016. Key ICT indicators for developed and developing countries and the world (totals and percentage rates)http://www.itu.int/en/ITU-D/Statistics/Pages/stat/default.aspx. Jean-Gilles, M., & Crittenden, P. M. (1990). Maltreating families: A look at siblings. Family Relations: An Interdisciplinary Journal of Applied Family Studies, 39(3), 323–329. http:// dx.doi.org/10.2307/584879. Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. Methodology in the social sciences. New York: Guilford Press. Lindberg, D. M., Shapiro, R. A., Laskey, A. L., Pallin, D. J., Blood, E. A., & Berger, R. P. (2012). Prevalence of abusive injuries in siblings and household contacts of physically abused children. Pediatrics, 130(2), 193–201. http://dx.doi.org/10.1542/peds.2012-0085. MacMillan, H. L., Tanaka, M., Duku, E., Vaillancourt, T., & Boyle, M. H. (2013). Child physical and sexual abuse in a community sample of young adults: Results from the Ontario Child Health Study. Child Abuse & Neglect, 37(1), 14–21. http://dx.doi.org/10.1037/t14441-000. Meiselman, K. C. (1978). Incest: A psychological study of causes and effects with treatment recommendationsJossey-Bass social and behavioral science series (1st ed.). San Francisco: Jossey-Bass Publishers. Monahan, K. (1997). Crocodile talk: Attributions of incestuously abused and nonabused sisters. Child Abuse & Neglect, 21(1), 19–34. http://dx.doi.org/10.1016/S0145-2134(96) 00132-9. Muram, D., Speck, P. M., & Gold, S. S. (1991). Genital abnormalities in female siblings and friends of child victims of sexual abuse. Child Abuse & Neglect, 15(1/2), 105–110. http://dx.doi.org/10.1016/0145-2134(91)90095-U. Nakou, S., Adam, H., Stathacopoulou, N., & Agathonos, H. (1982). Health status of abused and neglected children and their siblings. Child Abuse & Neglect, 6(3), 279–284. http:// dx.doi.org/10.1016/0145-2134(82)90030-8. Nelson, H. B., & Martin, C. A. (1985). Increased child abuse in twins. Child Abuse & Neglect, 9(4), 501–505. Nelson, E. C., Lynskey, M. T., Heath, A. C., Madden Pamela, A. F., & Martin, N. G. (2010). A family study of adult twins with and without a history of childhood abuse: Stability of retrospective reports of maltreatment and associated family measures. Twin Research and Human Genetics, 13(2), 121–130. http://dx.doi.org/10.1037/t06510-000. Phelan, P. (1986). The process of incest: Biologic father and stepfather families. Child Abuse & Neglect, 10(4), 531–539. http://dx.doi.org/10.1016/0145-2134(86)90058-X. Phelan, P. (1995). Incest and its meaning: The perspectives of fathers and daughters. Child Abuse & Neglect, 19(1), 7–24. http://dx.doi.org/10.1016/0145-2134(94)00096-D. Proeve, M., Day, A., Mohr, P., & Hawkins, K. (2006). Specific risk assessment based on victim type in child sexual offenders. Psychiatry, Psychology and Law, 13(1), 28–40. http:// dx.doi.org/10.1375/pplt.13. Proeve, M. (2009). A preliminary examination of specific risk assessment for sexual offenders against children. Journal of Child Sexual Abuse, 18(6), 583–593. http://dx.doi.org/ 10.1080/10926770903307898. Rausch, K., & Knutson, J. F. (1991). The self-report of personal punitive childhood experiences and those of siblings. Child Abuse & Neglect, 15(1-2), 29–36. http://dx.doi.org/10. 1016/0145-2134(91)90087-T. Russell, D. E. (1986). The secret trauma: Incest in the lives of girls and women. New York: Basic Books. Skopp, N. A., McDonald, R., Manke, B., & Jouriles, E. N. (2005). Siblings in domestically violent families: Experiences of interparent conflict and adjustment problems. Journal of Family Psychology, 19(2), 324–333. http://dx.doi.org/10.1037/0893-3200.19.2.324. Smith, S. M., & Hanson, R. (1974). 134 battered children: A medical and psychological study. British Medical Journal, 3(5932), 666–670. Smith, S. M. (1976). The battered child syndrome-some research aspects. The Bulletin of the American Academy of Psychiatry and the Law, 4(3), 235–243. Stith, S. M., Liu, T., Davies, L. C., Boykin, E. L., Alder, M. C., Harris, J. M., et al. (2009). Risk factors in child maltreatment: A meta-analytic review of the literature. Aggression and Violent Behavior, 14(1), 13–29. http://dx.doi.org/10.1016/j.avb.2006.03.006. Stoltenborgh, M., Bakermans-Kranenburg, M. J., Alink, L. R. A., & van IJzendoorn, M. H. (2012). The universality of childhood emotional abuse: A meta-analysis of worldwide prevalence. Journal of Aggression, Maltreatment & Trauma, 21(8), 870–890. http://dx.doi.org/10.1080/10926771.2012.708014. Studer, L. H., Clelland, S. R., Aylwin, A., Reddon, J. R., & Monro, A. (2000). Rethinking risk assessment for incest offenders. International Journal of Law and Psychiatry, 23(1), 15–22. http://dx.doi.org/10.1016/S0160-2527(99)00002-3. Tucker, C. J., Finkelhor, D., Turner, H., & Shattuck, A. M. (2014). Family dynamics and young children's sibling victimization. Journal of Family Psychology, 28(5), 625–633. http://dx.doi.org/10.1037/fam0000016. van Ijzendoorn, M. H., Moran, G., Belsky, J., Pederson, D., Bakermans-Kranenburg, M. J., & Kneppers, K. (2000). The similarity of siblings' attachments to their mother. Child Development, 71(4), 1086–1098. Vander Mey, B. J., & Neff, R. L. (1986). Incest as child abuse: Research and applications. New York: Praeger. Ward, M. J., Vaugh, B. E., & Robb, M. D. (1988). Social-emotional adaptation and infant-mother attachment in siblings: Role of the mother in cross-sibling consistency. Child Development, 59(3), 643–651. Wilson, R. F. (2004). Recognizing the threat posed by an incestuous parent to the victim's siblings: Part I: Appraising the risk. Journal of Child and Family Studies, 13(2), 143–162. http://dx.doi.org/10.1023/B:JCFS.0000015704.40389.56. Witte, S. (2017). Geschwister im Kontext von Misshandlung, Missbrauch und Vernachlässigung. Risikokonstellationen, Qualität der Geschwisterbeziehungen und psychische Belastung. Weinheim: Beltz [in press]. Witt, A., Brown, R. C., Plener, P. L., Brähler, E., & Fegert, J. M. (2017). Child maltreatment in Germany. Prevalencerates in the general population. Child and Adolescent Psychiatry and Mental Health, 11(1), 47–55. http://dx.doi.org/10.1186/s13034-017-0185-0. Zukow, P. G. (1989). Sibling interaction across cultures: Theoretical and methodological issues. New York, NY: Springer.

333