Are Children With a Number of Disabilities and Long-Term Illnesses at Increased Risk of Mental Violence,Disciplinary Violence,and Serious Violence?

Abstract

The focus of this study was on the question of whether or not children with a number of disabilities and long-term illnesses are at increased risk of child maltreatment (mental violence, disciplinary violence, and serious violence). This study was based on the Child Victim Survey of 2013 (FSD2943). The data consist of a nationally representative sample of pupils in Finland in sixth grade (12–13 years of age) and ninth grade (15–16 years of age). The sampling was undertaken as a stratified cluster by province, municipality type, and school size. The total number of respondents was 11,364. According to the logistic regression analyses, the children with at least three disabilities or long-term illnesses had an increased risk of violence compared with children with no disability: The risk of mental violence increased by 2.96 times, the risk of disciplinary violence by 4.30 times, and the risk of serious violence by 3.53 times. The effect of the category of at least three disabilities and illnesses remained statistically significant, although the analysis also accounted for several confounding factors. Thus, a child’s multiple morbidity (in the case of three or more disabilities and illnesses) can be categorized as one of the major risk factors for child maltreatment. The study complements the results of previous studies concerning the effect of children’s disabilities and long-term illnesses and their impact on child maltreatment. The results underline the importance of employees of health and social care having knowledge of multiple morbidity and its importance as a key factor regarding child maltreatment.

Keywords

physical abuse child abuse neglect family issues and mediators

Introduction

Child maltreatment is a complex issue, where no single factor is sufficient to explain why a child is being victimized. The ecological model of Bronfenbrenner (1977) can be used to understand both the potential risk factors and the promotive factors which affect a child’s life. For instance, parental binge drinking and the family financial situation are risk factors regarding child maltreatment (see, for example, Chan et al., 2016; Doidge et al., 2017; Svensson et al., 2011). On the contrary, a good quality of parent–child conversation reflects a positive interaction between a child and the parents, which reduces the risk of maltreatment (Giardino, Giardino, & Isaac, 2014, p. 175).

It is also well known that a child’s disability and long-term illness are significant risk factors for child maltreatment. In their literature review, Jones et al. (2012) showed that the child’s disability and illness increase the risk of physical violence and the risk of emotional violence. Furthermore, Chan et al. (2016) found that a child’s disability and illness increase the risk of serious physical violence. Furthermore, it is known that the number of disabilities and illnesses can increase the risk of violence, but less has been studied concerning the association between multiple morbidity and different forms of maltreatment (Doidge et al., 2017; Heinonen & Ellonen, 2013; Svensson et al., 2011).

The specific focus of the present study is on the association of the children with a number of disabilities and three forms of maltreatment (mental, disciplinary, and serious violence). It is hypothesized that the number of child disabilities and long-term illnesses poses a significant risk of mental, disciplinary, and serious violence. Hence, the study design is based on a model where other risk factors and promotive factors of maltreatment are also included. In this sense, the focus was on whether or not the association between child morbidity and the forms of violence remained statistically significant when several confounding factors are also controlled for. The present study complements previous research results in the literature that have reported the association between morbidity and maltreatment but with less emphasis on the significance of multiple morbidity and different forms of violence. This study provides knowledge on the role of multiple morbidity in child maltreatment, which health and social professionals can utilize in their usual daily practice when faced with their clients and patients.

Background

Risk Factors of Child Maltreatment

In the ecological development model, it is assumed that a child’s development will interact with his or her environment (Bronfenbrenner, 1977, 1986, 1992). The ecological model can be used to understand the complex process in which no single factor is sufficient to explain why a child is being victimized (Söderholm & Politi, 2012, pp. 83–84). The idea of promotive and risk factors is also presented in the resilience theory (Werner & Smith, 1982), which provides a conceptual framework for understanding why some people can control their ordinary life although they are exposed to risk factors (e.g., Rutter, 2006). According to the theory, promotive factors compensate for the consequences of risk factors.

Factors that influence a child can be both the potential risk factors and the protective factors (e.g., Van der Laan et al., 2010). The majority of children have risk factors in their lives. However, living in the sphere of one or two risk factors does not necessarily have a negative impact on a child’s long-term development. Thus, the existence of a risk factor does not necessarily imply that a child would be subjected to maltreatment, thereby jeopardizing his or her growth and development. But, the risk of maltreatment does increase if the risk factors accumulate and the protective factors are missing. Thus, the balance between risk factors and promotive factors largely influences whether the child is maltreated or not (Söderholm & Politi, 2012, pp. 84–85; Taylor & Lazenbatt, 2014, pp. 58–66). Risk factors can be divided into the sources related to the child, the parents, and the family, and also structural risk factors. Risk factors related to a child can include, for example, disabilities, illnesses, and behavioral problems. A child’s risk factors alone do not explain maltreatment, but a child’s disability or illness may be a risk factor which generates difficulties regarding constructive communication between the child and the parent (Söderholm & Politi, 2012, p. 85).

Mechanism Between a Child’s Disability and Maltreatment

A child’s disability or a child’s special health care needs present two of the many potential risk factors which threaten his or her safe growth and development. It may impair his or her ability to communicate, respond, and meet the expectations of parents or society, which increases their vulnerability. Because of a child’s disability or illness, he or she may have a weaker ability to avoid and resist maltreatment. Children with mental health problems and learning disabilities, in particular, may find it difficult to seek help and also to disclose cases of maltreatment (Giardino et al., 2014, pp. 169, 173; Heinonen & Ellonen, 2013, p. 275; Sobsey, 2002).

Child maltreatment can be explained by the accumulation of stress factors, where the child’s disability or illness is seen as one of the stress factors. The increased stress of the parents exposes the child to violence and other forms of maltreatment. Childhood illness increases the need for special care which increases parental stress. Physically, developmentally, or mentally challenging children have special needs that ultimately challenge parenthood. Furthermore, a child’s special needs may lead to lower levels of interaction between the child and his or her parent(s). Sometimes, the behavior of a child who is suffering from health problems can be challenging, which can also cause parental frustration. Parents may interpret a child’s behavioral and communication problems as a deliberate breach of the rules that the parents seek to impose and ultimately the parents may resort to using disciplinary violence against their children (Chan et al., 2016, p. 1029; Giardino et al., 2014, pp. 169–175).

Parents may be shocked by the fact that a child who is suffering from a disability or an illness may never achieve the same potential as other children having no such disability or illness. For instance, a child suffering from physical constraints may not meet his or her parents’ expectations of a physically healthy child. In turn, child behavioral disorders and cognitive disorders cause emotional constraints and can prevent a child’s normal development. Therefore, the child does not meet the parents’ expectations of a child’s development that can be considered as normal (Giardino et al., 2014, pp. 169, 173).

Previous Research on Childhood Disability as a Risk Factor for Maltreatment

Hershkowitz et al. (2007) studied the victimization of children with disabilities. They found that children with disabilities have experienced more serious physical violence and sexual abuse compared with children without disabilities. According to Reiter et al. (2007), children and adolescents with intellectual disabilities and other cognitive illnesses have more experiences related to maltreatment within their immediate environment, and such experiences are generally repeated over time. Furthermore, those authors note that the maltreatment of persons with disabilities seems to remain largely invisible.

In the results of their meta-analysis, Jones et al. (2012) find that children with disabilities are more likely to be victims of maltreatment compared with their peers who have no disabilities. Regarding disabilities, the odds ratio for pooled risk estimate is 3.56 for physical violence. Those authors also calculated the pooled risk estimates for different forms of maltreatment. If a child has any illness, he or she is 3.7 times more likely to be a victim of physical violence compared with children who do not have the illness. The child’s illness increases the risk of emotional violence by a factor of 4.4 and the risk of neglect by a factor of 4.6 (Jones et al., 2012).

According to the results of a study by Svensson et al. (2011), chronic health conditions in children increase the risk of physical maltreatment. The prevalence of physical maltreatment increases in relation to the number of illnesses. Regarding children having one long-term illness, 11.8% of those children have experienced physical maltreatment. For children who have experienced two long-term illnesses, 13.7% of those children have experienced physical maltreatment; for at least three diseases, it is 15.8%. A long-term illness increases the risk of physical maltreatment by a factor of 1.67 compared with children who experienced no illness. In addition, the results indicate that boys are more likely to experience physical maltreatment than girls by a factor of 1.49. Cuevas et al. (2009) studied the association between childhood maltreatment and psychiatric diagnoses (e.g., posttraumatic stress disorder [PTSD], depression). Their results show that children with a psychiatric diagnosis have significantly higher rates of victimization than children without a psychiatric diagnosis. The results highlight the need to consider psychiatric diagnoses as a risk marker for past and possible future victimization (Cuevas et al., 2009).

The results of a study by Sullivan and Knutson (2000) show a significant association between disability and maltreatment. Those authors found a 9% prevalence rate of maltreatment for nondisabled children and a 31% prevalence rate for disabled children. In addition, children with a behavioral disorder have a 7.3-fold risk of experiencing physical maltreatment and 7.0-fold risk of experiencing emotional maltreatment. Overall, children with disabilities are more likely to be maltreated than nondisabled peers by a factor of 3.4. According to Spencer et al. (2005), children experiencing disabling conditions appear to have an increased risk of being registered in terms of child abuse and neglect. A child’s psychological problems increase the risk of physical maltreatment by a factor of 3 and the risk of emotional maltreatment by a factor of 8. However, a child’s sensory disorders and autism spectrum disorders do not increase the risk of different forms of maltreatment (Spencer et al., 2005).

Chan et al. (2016) explored the maltreatment of children and young people throughout the lives of those children and also during the year prior to the study. Those authors found that children and young people without disabilities had experienced somewhat more mental violence and disciplinary violence during the year prior to that study compared with children and young people with disabilities. However, children and young people with disabilities had experienced significantly more serious violence. The results of their study show that disability increased the risk of serious violence by a factor of 1.57.

Doidge et al. (2017) explored child, parent, and family risk factors regarding maltreatment in an Australian population-based birth cohort study. They noted that adolescents’ psychical health problems increase the risk of physical maltreatment, and that cognitive and behavioral disorders increase the risk of emotional maltreatment. However, they found that if a child had at least two health problems by the age of 3 years, then there was no increase in the risk of physical and mental maltreatment (Doidge et al., 2017).

Also, Heinonen and Ellonen (2013) found that children with disabilities had experienced more disciplinary violence than children without disabilities. A child’s mental health problem increased the risk of disciplinary violence by a factor of 3.53, and a child’s learning or memory difficulties increased the risk by a factor of 1.45 compared with the children without disabilities. But, a child’s hearing disability, physical disability, and internal disease did not increase the risk of violence. Furthermore, the risk of disciplinary violence increased if the number of disabilities increased. If a child had one disability, it increased the risk of violence by a factor of 1.52. If a child had two disabilities, the risk increased by a factor of 2.37, and if the child had three disabilities, then the risk increased by a factor of 4.42.

Data, Research Question, and Method

Data and Hypothesis

Child maltreatment is a multifaceted problem. According to the World Health Organization (WHO), child maltreatment includes all types of physical and/or emotional ill-treatment, sexual abuse, neglect, negligence, and commercial or other exploitation, which results in actual or potential harm to the child’s health, survival, development, or dignity in the context of a relationship of responsibility, trust, or power (WHO, 2019). The present study has a focus on emotional and physical maltreatment (disciplinary violence and serious violence), which are also the subject of the Finnish version of the Conflict Tactics Scale (CTS; Straus, 1979, 1990) that is used in the questionnaire.

The present study was based on the Child Victim Survey of 2013 (FSD2943). Child Victim Surveys chart violent experiences of children in Finland. These experiences have been surveyed since 1988, and the actual Child Victim Survey has been conducted by the Police University College from 2008 onward. The latest data, which were collected in August and September 2013, consist of a nationally representative sample of pupils in sixth grade (12–13 years of age) and ninth grade (15–16 years of age). The sampling was undertaken as a stratified cluster by province, municipality type, and school size (for more details, consult Fagerlund et al., 2014). Schools were asked to provide responses from 1 to 3 classes, depending on school size. The original sample included 686 schools and 987 classes. Schools were asked to provide their permission beforehand for the study to be undertaken, and permission was granted by 78% of the schools. However, the exact response rate could not be calculated because the samples were based on the class information of the school year before data collection (Autumn 2012 to Spring 2013). The total number of respondents was 11,364. Respondents answered the Web-based questionnaire during school hours. The data for this study were retrieved from the Finnish Social Science Data Archive.

The study sample is described in Table 1. The number of girls is slightly higher than the number of boys. There are more sixth graders than ninth graders. The number of respondents born outside Finland was 3.4%, which corresponds well enough to the proportion of immigrants in the population (5.2%) at the time of the survey in 2012 (Statistics Finland, 2019). A significant proportion of the respondents (75%) live in a nuclear family. About 32% of respondents’ mothers and 36% of respondents’ fathers had reached the lowest level of education, and 72% of both of the respondents’ parents were working. The financial situation of the respondents’ families is mostly good or quite good (87%).

Table 1.

Description of the Sample.

Background Variable	n	%	Missing Value
Gender			43 (0.4%)
Girl	5,746	50.8
Boy	5,575	49.2
Age (years)			285 (2.5%)
10–13	6,074	54.8
14–17	5,005	45.2
Country of birth			28 (0.2%)
Finland	10,956	96.6
Other country	380	3.4
Family structure			32 (0.3%)
Nuclear family	8,494	75.0
Other structure	2,836	25.0
Mother’s level of education			4,324 (38%)
Only comprehensive school	2,272	32.3
High school/vocational school or university or polytechnic	4,768	67.7
Father’s level of education			4,614 (40.6%)
Only comprehensive school	2,453	36.3
High school/vocational school or university or polytechnic	4,267	63.7
Employment of parents			131 (1.1%)
Both parents with work	8,138	72.4
One of the parents with work	2,693	24.0
Neither parent with work	402	3.6
Family’s financial situation			960 (8.5%)
Very good or quite good	9,083	87.3
Some difficulties	1,183	11.4
A lot of difficulties	138	1.3

The focus of the present study was on a child’s experience of violence undertaken by his or her parent(s). The aim of the study was to explore the association between the number of disabilities or illnesses with mental, disciplinary, and serious violence. Previous studies show the link between a child’s disability or illness and maltreatment (Jones et al., 2012), but to our knowledge there have been fewer studies on the association of the number of a child’s disabilities or long-term illnesses with respect to different forms of maltreatment. Thus, the hypothesis is expressed as follows:

Hypothesis 1 (H1). The number of a child’s disabilities and/or illnesses is associated with mental violence (H1a), disciplinary violence (H1b), and serious violence (H1c).

Dependent Variables

Parental violence was measured using a Finnish version of the CTS (Sariola & Uutela, 1992). The CTS was developed by Straus (1979, 1990; Straus et al., 1996) based on conflict theory, which assumes that all relationships, especially parent–child relationships, can contain conflicts. Conflict is an inevitable part of human association but violence as a tactic to deal with conflict is not (Creighton et al., 2003, pp. 30–31). The Finnish version of the CTS has been validated in previous studies (Ellonen et al., 2011; Lepistö & Paavilainen, 2012; Sariola & Uutela, 1992).

The CTS scale consists of 14 items, starting with parental aggressive verbalization toward the child and increasing to severe violent physical acts regarding the child. The CTS can be categorized into three forms of violence: mental violence, disciplinary violence, and serious violence. The child was asked separately about whether or not they were victimized by the mother and/or father during the last 12 months. The list of items is presented in Table 1. Respondents who had experienced at least one of those acts of parental violence, at least once during the 12 months prior to the study, were marked as 1 and respondents who had no such experiences were marked as 0. Three sum variables were formed—mental violence, disciplinary violence, and serious violence—by combining the violence of the mother and the father. The prevalence of different forms of violence is shown in Table 2.

Table 2.

The Dependent Variables by Items (Weighted Results).

Form of Violence	Violence Inflicted by Mother (n, %)	Violence Inflicted by Father (n, %)
Mental violence
Sulking or refusing to discuss the matter	808 (39.6)	495 (35.7)
Scolding, belittling, cursing, or insulting in other ways (no physical assault)	819 (40.2	586 (42.2)
Throwing, hitting, or kicking an object in anger (e.g., banging doors)	358 (17.6)	230 (16.6)
Threatening to use violence	54 (2.6)	77 (5.5)
Total	2,039 (100)	1,388 (100)
Disciplinary violence
Pushing, shoving, or shaking the you in anger	121 (29.1)	107 (30.2)
Pulling your hair	147 (35.3)	118 (33.3)
Slapping you	98 (23.6)	63 (17.8)
Whipping you	25 (6.0)	31 (8.8)
Giving you a severe spanking	25 (6.0)	35 (9.9)
Total	416 (100)	354 (100)
Serious violence
Punching you	22 (19.3)	31 (29.2)
Hitting you with an object	22 (19.3)	19 (17.9)
Kicking you once or repeatedly	37 (32.5)	25 (23.6)
Threatening to use a knife or a gun	16 (14.0)	15 (14.2)
Actually using a knife or a gun	17 (14.9)	16 (15.1)
Total	114 (100)	106 (100)

Missing data are a concern for the generalization of the results. The entire survey was extensive and consisted of a total of 902 variables and 11,364 cases. According to the Missing Value Analysis (MVA), the majority of elements of missing information is based on the fact that the respondent has not responded to the CTS. A total number of 4,710 respondents skipped all 14 questions regarding their mother’s violence (41.4% of respondents), and 4,880 respondents (42.9% of respondents) skipped questions regarding their father’s violence. However, the number of respondents can be considered sufficient. A total number of 6,906 respondents answered the questions on mental violence, 6,687 respondents answered the questions on disciplinary violence, and 6,628 answered the questions on serious violence.

The missing elements of information could not be coded into zero in the analysis because the respondents had an option to answer as “I have not experienced violence.” Missing elements of information may affect the generalization of the research results. According to the MVA, the share of missing information with respect to the boys (mental: 47.9%; disciplinary: 49.6%; serious: 50.1%) is higher than that of the girls (mental: 30.6; disciplinary: 32.7; serious: 33.3%). The proportion of age 10–13 years (mental: 42.1; disciplinary: 43.5; serious: 44.1%) is lower than that of age 14–17 years (mental: 35.0; disciplinary: 37.5; serious: 38.0%) in three forms of violence. However, in the number of disabilities, the missing information is evenly distributed in the categories, except in the category of three or more disabilities where the missing information is higher than in other categories. The missing data were handled by using the listwise deletion method.

Variables for Disabilities and Illnesses

The disabilities and illnesses were surveyed by asking the participant whether he or she had any of the following disabilities, illnesses, or special needs: visual impairment, hearing impairment, physical disability, reliance on a wheelchair, respiratory disease, internal disease, difficulty in learning and/or memory, attention deficit hyperactivity disorder (ADHD), mental health problem, or other disability or long-term illness. The respondent was able to choose more than one option and the answer was left blank if there was no disease. Thus, if the respondent had failed to respond to the entire set of questions, this was interpreted as having no illness.

In the questionnaire, the respondents were instructed to leave an item blank if they did not have a specific disability or illness. This is problematic because it cannot be known with any degree of certainty whether the participant did not have the specific disability or illness or whether they had merely skipped those items. Furthermore, it can be questioned whether all children have enough knowledge about their specific disabilities or illnesses so that they can report that knowledge clearly in the questionnaire. However, teachers were instructed to help children regarding their understanding the questions in the questionnaire (Fagerlund et al., 2014, p. 28).

A child’s disabilities and illnesses formed a sum variable with four categories that describes the total number of child disabilities and illnesses in the data. The categories of the variable describe the number of disabilities and illnesses: no disability or illnesses, one disability or illness, two disabilities or illnesses, and at least three disabilities or illnesses. The prevalence of disabilities is shown in Table 3.

Table 3.

Prevalence of Disabilities (Weighted Results).

Type of Disability or Illness	n	%
Visual impairment	537	15.2
Hearing impairment	159	4.5
Physical disability	300	8.5
Reliance on a wheelchair	43	1.2
Respiratory disease	829	23.5
Internal disease	112	3.2
Difficulties in learning and/or memory	415	11.7
Attention deficit hyperactivity disorder (ADHD)	200	5.7
Mental health problem	166	4.7
Other disability or long-term illness	772	21.9
Total	3,533	100
Number of Disabilities or Illnesses	n	%
No disability or illness	8,650	76.1
One disability or illness	2,241	19.7
Two disabilities or illnesses	349	3.1
At least three disabilities or illnesses	123	1.1
Total	11,364	100

Confounding Variables

In the present study, confounding variables were also used that, based on the body of literature, have also been shown to be risk factors for child maltreatment. Child gender and age, parental binge drinking, family financial situation, and witnessing intimate partner violence have been found to be associated with child maltreatment (see, for example, Black, Heyman, & Slep, 2001; Black, Slep, & Heyman, 2001; Chan et al., 2016; Doidge et al., 2017; Drake & Jonson-Reid, 2014; Svensson et al., 2011). Also, lack of parent–child conversation reflects a weakened interaction between the child and the parent, which increases the risk of violence (Giardino et al., 2014, p. 175). If the family rarely dines together or the child spends most of his or her free time outside the family, there is less interaction between the child and the parent(s), which may contribute to the occurrence of maltreatment toward the child.

The variable of age was transformed into two categories: children aged 10–13 years received a value of 0 and children aged 14–17 years received a value of 1. Parental binge drinking during the 12 months prior to the study had six response options in the material, which were transformed into four categories. The family financial situation was assessed by asking the question, “What do you think about the family’s financial situation?” The variable contained five response options, which were coded into three categories. The question on family dining was presented as follows: “How often do you usually eat in the evening with one or both parents (or with the adults you live with?).” The question included five options that were transformed into three categories. A question concerning leisure time was focused on with whom a child spends his or her leisure time. The original variable of six categories was transformed into three categories. The child’s conversation with parents was studied by asking, “Can you discuss with your parents about your own affairs?” The variable contained four response options, which were converted into two response options. Witnessing intimate partner violence was studied by asking whether the child had seen or heard that the mother, father, sister, or brother had been subjected to various acts of violence at home in the 12 months prior to the study. The sum variable for witnessing intimate partner violence was formed by combining the various acts of violence against the mother and the father. The variable was encoded into two categories, with the received response options of “no, yes.”

Method

Cross-tabulation analysis was used to study the association between the child’s morbidity and the various forms of violence. Pearson’s chi-squared test and Fisher’s exact test were used to analyze the group differences in the categorical variables. Logistic regression analysis was used to explore the child’s morbidity as a risk factor for the forms of violence. Various regression models were built into the analysis, within which the variables were added on a step-by-step basis. First, only the background variables of gender and age were added. Second, the background variables and the parental binge drinking and the family financial situation were controlled for. Third, the child’s leisure time, the family dining, and the conversational connection were also added into the model. Fourth, witnessing intimate partner violence was added into the model. Fifth, the morbidity of the child was added into the model. The weights suggested by Fagerlund et al. (2014) were used in the analyses for correcting the possible bias caused by background factors.

Ethical Considerations

Child maltreatment is a sensitive research topic, where the consideration of ethical issues is particularly important (Alderson & Morrow, 2004). It is noteworthy that taking part in research should not be a new source of violence for children who have already been subjected to violence. On the contrary, it can also be seen as maltreatment if their experiences of violence could be silenced and ignored by researchers (Mudaly & Goddard, 2006).

The data were collected in accordance with the ethical guidelines of Policy University College (Fagerlund et al., 2014, pp. 35–36) and the data for this study were retrieved from the Finnish Social Science Archive (FSD2943). The survey was carried out in schools as a web-based survey which was led by the teacher during the lesson (Fagerlund et al., 2014, p. 28). Participation in the study was voluntary and a pupil could stop attending the survey at any time. The parents of the children were informed of the study at the end of the lesson by a bulletin distributed to the students. With the exception of the City of Helsinki, the parents were not asked for prior consent (Fagerlund et al., 2014, p. 28). The procedure was supported by a sensitive research topic in which, for example, the cases of domestic violence may create a conflict between the interests of the parents and the child. The Finnish Advisory Board on Research Integrity (TENK) has stated that the questionnaires conducted as part of normal school work that do not collect personalized identifiable information can be carried out under Finnish law without the consent of the guardian (Fagerlund et al., 2014, p. 35; Finnish Advisory Board on Research Integrity, 2019).

The school environment is problematic for ethical issues. The teacher’s role is based on authority. Also, group pressure, which is based on the views and behavior of other pupils, plays a key role in the life of a pupil. In this sense, it can be questioned to what extent the participation of children in the survey has been voluntary. At the same time, however, the school’s network of adult and peer relationships can provide the best possible environment for dealing with a child’s feelings that are likely raised by the study. This is especially important when it comes to domestic violence, where the role of other adults is emphasized along with that of their own parents (Morris et al., 2012). Various help desks and service points were presented to children through the survey website, from which they can ask for advice or seek help on issues related to violence. Also, teachers and parents were encouraged to answer any questions that might arise (Fagerlund et al., 2014, p. 28). At the end of the survey, respondents were asked how they felt about answering the questions. The responsible institution of the survey has not reported concerns about the respondent experiences related to sensitive issues (see Fagerlund et al., 2014).

Results

Descriptive Analysis

Table 4 shows the children’s experiences in different forms of violence according to explanatory variables. The more a child has disabilities or long-term illnesses, the greater the proportion of those who have experienced mental violence, disciplinary violence, or serious violence. Thus, in these data, the increase in the number of disabilities and illnesses increases the prevalence of various forms of violence. The highest percentage is found in mental violence, as 56.1% of children with at least three illnesses have experienced mental violence. The corresponding proportion in the case of disciplinary violence is 28.6% and for serious violence is 9.8%. The more serious the form of violence, the more the increase in the number of disabilities and illnesses affects the prevalence of violence. In turn, children without disabilities and illnesses have experienced less regarding all forms of violence. The number of illnesses has a statistically significant connection with mental violence, χ²(3, N = 6,878) = 116.42, p < .001, disciplinary violence, χ²(3, N = 6,655) = 84.42, p < .001, and serious violence, χ²(3, N = 6,602) = 39.02, p < .001 (Fisher’s exact test).

Table 4.

The Relationship Between the Explanatory Variables and the Forms of Violence (Weighted Results).

Explanatory Variables	Mental Violence					Disciplinary Violence					Serious Violence
	No		Yes	χ²		No		Yes	χ²		No		Yes		χ²
	n	(%)	n	(%)		n	(%)	n	(%)		n	(%)	n	(%)	χ²
Gender					<.001					<.001					.364
Girl	2,836	(71.3)	1,142	(28.7)		3,599	(93.3)	260	(6.7)		3,781	(98.8)	47	(1.2)
Boy	2,385	(82.8)	495	(17.2)		2,648	(95.4)	129	(4.6)		2,714	(98.5)	41	(1.5)
Age (years)					<.001					<.001					<.001
10–13	2,992	(85.4)	512	(14.6)		3,258	(95.4)	158	(4.6)		3,359	(99.2)	26	(0.8)
14–17	2,138	(66.0)	1,102	(34.0)		2,886	(92.7)	226	(7.3)		3,028	(98.0)	62	(2.0)
Parents’ binge drinking					<.001					<.001					<.001
Never	1,701	(77.7)	487	(22.3)		2,005	(94.4)	120	(5.6)		2,078	(99.0)	21	(1.0)
Once or few times a year	959	(65.8)	499	(34.2)		1,287	(92.5)	105	(7.5)		1,363	(98.8)	16	(1.2)
At least once a month	170	(51.1)	163	(48.9)		263	(85.1)	46	(14.9)		287	(93.8)	19	(6.2)
Family’s financial situation					<.001					<.001					<.001
Very good or pretty good	4,301	(78.1)	1,205	(21.9)		5,080	(95.1)	261	(4.9)		5,257	(99.1)	49	(0.9)
Some difficulties	449	(60.1)	298	(39.9)		624	(88.6)	80	(11.4)		671	(96.4)	25	(3.6)
A lot of difficulties	45	(46.4)	52	(53.6)		78	(83.0)	16	(17.0)		83	(92.1)	7	(7.9)
Child’s leisure time					<.001					<.001					<.001
With family	1,923	(85.2)	335	(14.8)		2,148	(96.8)	70	(3.2)		2,199	(99.6)	9	(0.4)
With companion or friends	2,822	(72.6)	1,047	(27.4)		3,498	(93.5)	242	(6.5)		3,654	(98.4)	60	(1.6)
Alone	461	(66.9)	228	(33.1)		584	(88.5)	76	(11.5)		625	(96.9)	20	(3.1)
Family dining					<.001					<.001					<.001
Daily or several times a week	4,594	(78.5)	1,258	(21.5)		5,403	(95.1)	276	(4.9)		5,580	(99.0)	58	(1.0)
Once a week	407	(64.2)	227	(35.8)		544	(89.8)	62	(10.2)		585	(97.2)	17	(2.8)
Once a month or less	220	(59.3)	157	(40.7)		300	(85.5)	51	(14.5)		330	(96.2)	13	(3.8)
Conversation with parents					<.001					<.001					<.001^a
Yes or to some extent	5,066	(77.7)	1,454	(22.3)		5,995	(94.9)	321	(5.1)		6,202	(98.9)	70	(1.1)
Not or not at all	156	(46.3)	181	(53.7)		252	(79.2)	66	(20.8)		291	(94.2)	18	(5.8)
Witnessing intimate partner violence					<.001					<.001					<.001^a
No	5,116	(78.9)	1,366	(21.1)		5,971	(95.1)	308	(4.9)		6,169	(98.9)	68	(1.1)
Yes	121	(30.6)	274	(69.4)		294	(78.2)	82	(21.8)		344	(94.2)	21	(5.8)
Number of disabilities or illnesses					<.001					<.001					<.001^a
Not at all	4,124	(78.9)	1,105	(21.1)		4,817	(95.1)	249	(4.9)		4,985	(98.9)	57	(1.1)
One	951	(69.7)	414	(30.3)		1,218	(92.4)	100	(7.6)		1,273	(98.5)	20	(1.5)
Two	134	(61.5)	84	(38.5)		185	(88.9)	23	(11.1)		200	(97.1)	6	(2.9)
At least three	29	(43.9)	37	(56.1)		45	(71.4)	18	(28.6)		55	(90.2)	6	(9.8)

Fisher’s exact test.

There is a statistically significant connection between all confounding variables and different forms of violence, except for gender and serious violence. Girls have reported significantly more mental violence compared with boys. Girls have also experienced more disciplinary violence compared with boys. Respondents aged 14–17 years have reported that they have experienced more of all forms of violence than those respondents aged 10–13 years. Children have experienced more different forms of violence when parents consumed more alcohol in the past 12 months, the family has financial difficulties, or a child has intimately witnessed partner violence. On the contrary, social interaction between a child and his or her parents seems to be a significant promotive factor. Children who are able to discuss their own affairs with parents have experienced all forms of violence to a lesser degree than children who are unable to discuss issues with their parents. Also, children who spend most of their leisure time with their family and who eat together daily or several times a week with their family have reported less different forms of violence (Table 4).

Logistic Regression Analyses

The logistic regression analysis was done separately with respect to all response variables. First, the analysis was focused on mental violence. According to Table 5, a child’s disability or illness increases the risk of mental violence. If a child has one disability or illness, he or she has a 1.26-fold risk of mental violence compared with children who do not have a disability or illness. If a child has at least three disabilities or illnesses, he or she has a 2.96-fold risk of mental violence compared with children without a disability or illness. However, the category of two disabilities or illnesses was not statistically significant. The confounding variables such as child gender and age, parental binge drinking, family financial situation, the child leisure time, family dining, conversation between the parents and the child, and witnessing intimate partner violence are also statistically significant in the final model, where all confounding variables are controlled for.

Table 5.

Final Models of Logistic Regression Analyses: Odds Ratios (OR) and Confidence Intervals (CI/95%) for Mental Violence, Disciplinary Violence, and Serious Violence (Weighted Results).

Variables	Mental Violence	Disciplinary Violence	Serious Violence
Gender
Girl	1	1	1
Boy	0.58 [0.49, 0.68]	0.84 [0.62, 1.13]	3.97 [2.07, 7.63]
Age (years)
10–13	1	1	1
14–17	1.80 [1.52, 2.13]	0.97 [0.71, 1.32]	1.59 [0.76, 3.34]
Parental binge drinking
Never	1	1	1
Once or few times a year	1.37 [1.16, 1.62]	0.98 [0.72, 1.32]	0.79 [0.39, 1.60]
At least once a month	1.85 [1.39, 2.44]	1.24 [0.80, 1.94]	2.24 [1.03, 4.84]
Family’s financial situation
Very good or pretty good	1	1	1
Some difficulties	1.22 [0.98, 1.51]	1.40 [0.99, 1.99]	1.94 [0.97, 3.88]
A lot of difficulties	1.67 [0.91, 3.07]	1.59 [0.77, 3.32]	1.96 [0.54, 7.18]
Child’s leisure time
With family	1	1	1
With companion or friends	1.51 [1.24, 1.83]	1.15 [0.81, 1.65]	1.60 [0.65, 3.96]
Alone	1.56 [1.18, 2.07]	1.57 [0.98, 2.51]	1.43 [0.46, 4.41]
Family dining
Daily or several times a week	1	1	1
Once a week	1.27 [1.00, 1.62]	1.61 [1.08, 2.40]	2.12 [1.01, 4.45]
Once a month or less	1.19 [0.87, 1.63]	2.31 [1.50, 3.55]	1.85 [0.76, 4.51]
Conversation with parents
Yes or to some extent	1	1	1
Not or not at all	1.80 [1.31, 2.48]	2.50 [1.65, 3.79]	2.75 [1.22, 6.18]
Witnessing intimate partner violence
No	1	1	1
Yes	4.88 [3.66, 6.50]	3.03 [2.10, 4.37]	3.62 [1.75, 7.48]
Number of disabilities or illnesses
Not at all	1	1	1
One	1.26 [1.05, 1.51]	1.40 [1.02, 1.92]	0.66 [0.29, 1.47]
Two	1.44 [0.98, 2.11]	1.62 [0.92, 2.85]	1.77 [0.62, 5.05]
At least three	2.96 [1.48, 5.93]	4.30 [2.00, 9.22]	3.53 [1.02, 12.2]
Pseudo-R², Nagelkerke	.192	.119	.154

The second analysis was focused on disciplinary violence. Also in this case, the category of at least three disabilities or illnesses increased the risk of violence (4.30-fold) compared with children without a disability or illness. The category of one disability and illness was statistically associated with the risk, but the category of two disabilities and illnesses was not. The confounding variables such as family financial situation, child leisure time, family dining, conversation between the child and the parents, and witnessing intimate partner violence are statistically significant in the final model.

The third analysis was focused on serious violence (Table 4). In this case, only the category of at least three disabilities and illnesses increased the risk of serious violence, by 3.54-fold. The confounding variables such as gender and witnessing domestic violence were related to serious violence. Boys have a 3.97-fold risk of serious violence compared with girls. Witnessing intimate partner violence increased the risk of serious violence by 3.62-fold.

According to the analyses, child multiple morbidity (disability and long-term illness) increases the risk of child maltreatment. If a child has at least three disabilities and illnesses, he or she has 2.96-fold risk of mental violence, 4.30-fold risk of disciplinary violence, and 3.53-fold risk of serious violence compared with children without any disability. However, it is noteworthy that the category of two disabilities was not a statistically significant risk factor for violence. Hence, the results only partially supported the hypotheses that the number of a child’s disabilities and illnesses is associated with mental violence (H1a), disciplinary violence (H1b), and serious violence (H1c). All three logistic regression analyses produced a clear result that a child who has at least three disabilities and illnesses has a significant risk of child maltreatment; however, the fit of the regression models (Nagelkerke R²) remains quite modest (mental violence 19.2%, disciplinary violence 11.9%, and serious violence 15.4%).

Discussion

Child disability and child long-term illness are associated with child maltreatment (e.g., Chan et al., 2016; Jones et al., 2012). Also, according to the results of the present study, one disability or illness increases the risk of mental and disciplinary violence, compared with the children without a disability or long-term illness, but instead it was not found that one disability or long-term illness would have increased the risk of serious violence.

However, the specific focus of the present study was on the association of the number of disabilities and long-term illnesses with the following forms of child maltreatment: mental violence, disciplinary violence, and serious violence. It was found that if a child has at least three disabilities or long-term illnesses, it increases the risk of all forms of violence multiple times compared with children without any disability. The risk of mental violence increased by 2.96-fold, the risk of disciplinary violence increased by 4.30-fold, and the risk of serious violence increased by 3.53-fold. But, interestingly, having two disabilities was not statistically significantly associated with any forms of violence.

The results of the present study complement previous studies. Heinonen and Ellonen (2013) found that the number of disabilities increases the prevalence of disciplinary violence. In their study, children with one or two disabilities were even more likely to have been subjected to disciplinary violence than in the present study. However, their study was focused only on disciplinary violence. The differences in the risk of disciplinary violence between their study and the present study may be explained by the fact that Heinonen and Ellonen (2013) used the previous version of Child Victim Survey (2008) and further they had significantly less control variables compared with the present study which was based on the data of 2013.

Svensson et al. (2011) found that the prevalence of physical violence is associated with the number of child disabilities and illnesses, which is consistent with the results of the present study. Also, they found that long-term illness increases the risk of child maltreatment by 1.67-fold which is consistent with the results of the present study in the cases of mental and disciplinary violence. However, their response variable is different, and their study also differs regarding how the association between the number of illnesses and the risk of physical violence is explored. Furthermore, Sullivan and Knutson (2000) found that child disabilities and illnesses increase the risk of physical violence 3.79-fold and the risk of emotional violence 3.88-fold. Their odds ratios are in line with the present study. However, Sullivan and Knutson did not divide physical violence into categories of disciplinary violence and serious violence and they did not define the number of child disabilities and illnesses.

Doidge et al. (2017) in the results of their study find that if a child has at least two identified health problems by the age of 3 years, then it did not increase the risk of physical and emotional violence. The result is consistent with the present study, in which the number of two disabilities and illnesses was not associated with any forms of violence. However, the categories for the numbers of disabilities are different between their study and the present study. In the present study, the number of disabilities or illnesses being 2 is not statistically significant would be statistically significant for any form of violence. In the case of serious violence, only three disabilities or illnesses would be statistically significant, which might be interpreted as, one or two risk factors would not necessarily have a negative impact on a child’s life (cf. Van der Laan et al., 2010). But in the cases of mental and disciplinary violence, the nonsignificance of two disabilities or illnesses is a challenging result. It is possible that there are differences in the significance of the combinations of disabilities and illnesses (Doidge et al., 2017). For instance, there might be a situation where some combinations of disabilities are significant and some combinations are nonsignificant which, as a sum, leads to a nonsignificant entire result. On the other hand, it could be possible that there are qualitative differences within the types of diseases and illnesses (minor, moderate, severe) which the questionnaire did not reach. In this sense, further research is needed to clarify this issue.

Despite the differences in the implementation with the present and previous studies, it can be argued that the number of three or more disabilities and long-term illnesses poses a significant risk of violence. In this sense, the results can be seen to support the mechanism between disability and violence, where behavioral and communicational problems are associated with child disability with respect to causing parental concern and frustration, which in turn increases the risk of disciplinary violence and other physical violence. Child behavior and child communication problems can make it difficult for a child to settle for parental rules that parents may experience as intentional noncompliance (cf. Giardino et al., 2014, pp. 174–175).

However, there are also some limitations regarding this study. For instance, the present study focused only on the number of disabilities and illnesses but not on the specific combinations of disabilities and illnesses. However, the previous studies have shown that different disabilities and illnesses increase the risk of maltreatment in different ways (Doidge et al., 2017; Jaudes & Mackay-Bilaver, 2008; Spencer et al., 2005; Sullivan & Knutson, 2000). Hershkowitz et al. (2007) find that the severity of the disability affects the prevalence of maltreatment. Hence, further research is needed for exploring which combinations of disabilities and illnesses increase the risk in different forms of maltreatment.

Furthermore, the cross-sectional study has difficulties in determining the causal relationships between phenomena (Darlington & Hayes, 2017). On the one hand, several studies have shown that childhood disabilities and illnesses are a risk factor for maltreatment (Jones et al., 2012). On the other hand, many studies have also shown that maltreatment increases the risk of physical and mental illness at different stages of life (Afifi et al., 2006; Gilbert et al., 2009; Lanier et al., 2010; Springer et al., 2003). Hence, the connection between a child disability and child maltreatment is a two-way scenario. The present study was focused on violence which a child had experienced over the past 12 months instead of violence experienced during the whole life of the child. Thus, it was justified to assume that the morbidity would act as a risk factor for recent maltreatment. However, the fit of the regression models (Nagelkerke R²) also leaves room for other explanatory factors. Therefore, the risk effect of multiple morbidity on the specific form of violence should be interpreted with caution.

Furthermore, the response variables do not form a completely homogeneous group of respondents. For instance, the group of respondents who have experienced mental violence consists of the respondents who have experienced only mental violence, the respondents who have experienced also disciplinary violence, and the respondents who have experienced all forms of violence. The more serious the form of violence, the more the respondents have also experienced the other forms of violence. On the other hand, the respondent experiences of violence illustrate a very real situation, as polyvictimization is common (Finkelhor, Ormrod, & Turner, 2007, p. 8).

The result of the regression analysis largely depends on the variables used in the model. Therefore, the interpretation of the present study can only be linked to the combination of variables used in this study. The combination of confounding variables was based on the notion that maltreatment is a sum of many different factors, which can be seen as protective and risk factors. However, all potential confounding factors were not taken into consideration in the analyses. For instance, the Child Victim Survey has not mapped parental mental health problems, which have been identified as a significant risk factor for maltreatment according to previous studies in the literature (e.g., Doidge et al., 2017). On the other hand, parental psychiatric morbidity is hereditary in contributing to child morbidity. Therefore, it is justified that parental mental health problems are not included in the analysis, where they may in fact be contributing to child morbidity and child maltreatment.

The data were collected as a survey, which also sets some limitations. The sensitivity of the topic affects the reliability of the study, as it may cause the respondents to refrain completely from responding and could cause under-reporting and over-reporting. When children and adolescents are asked about experiences of violence, the threshold for reporting such sensitive issues in the survey may be high for some respondents (McNeeley, 2012). Furthermore, the proportion of missing information is large in the present study. The length of the survey, the sensitivity of the topic, and the extent of the questions set are likely to explain this dilemma. On the other hand, the large size of the dataset ensures that randomly biased responses do not distort the results of the overall data. In the analyses of the study, the weight of the analysis was used to compensate for possible distortions caused by background variables. The survey questions used in the study can be considered valid as they relate to a clear concrete violence situation and they do not contain concepts that could be interpreted differently by both children and adolescents. The validity of the questionnaire is also supported by the fact that it was conducted mainly in the same way as the 2008 Child Victim Survey, which was tested before the actual data collection (Fagerlund et al., 2014, p. 33).

From a diversity perspective, it is noteworthy that the prevalence of mental violence and disciplinary violence is almost twice more common among girls compared with boys. On the contrary, in the case of serious violence, there are no differences between the groups, but instead boys have a 3.97-fold risk of serious violence compared with girls in the case where the factors of the present study are taken into the analysis. The distinction between these results is interesting and demands further research.

Conclusion

Child maltreatment is a complex issue because various factors influence concomitantly with respect to the existence of maltreatment (Söderholm & Politi, 2012). According to the ecological development model of Bronfenbrenner (1992), child maltreatment can be seen as a result of an accumulation of risk factors and a lack of promotive factors. In the present study, all factors were related to promotive or risk factors in child-based microsystems. The main argument of the present study is based on the result that child multiple morbidity (i.e., in the case of three or more disabilities and illnesses) is a significant risk for all forms of violence although several confounding factors were also controlled for. In this sense, child multiple morbidity can be categorized as one of the major risk factors for maltreatment. The result is consistent with the classic study of Rutter (1981) on child psychiatric disorders, in which he argues that there were no differences between the risk factors, and the main explanatory issue was the number of risk factors.

The connection between multiple morbidity and maltreatment reflects the links between health and social problems in families, highlighting the importance of cooperation between social care and health care. Therefore, it is important that the practitioners of health and social care have some knowledge regarding multiple morbidity as a key factor for child maltreatment (cf. Paavilainen & Flinck, 2013). The research result can be utilized by those working with children and adolescents, especially in child protection, which investigates and assesses the occurrence and severity of maltreatment, but it is crucial also in child and youth psychiatry, which examines whether or not a child has been subjected to violence.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Timo Toikko

Author Biographies

Piia Seppälä earned her Master’s degree in social work from the University of Eastern Finland in 2019. She is interested in the issues related to child maltreatment and social work with children.

Riitta Vornanen is a professor of social work at the University of Eastern Finland (Kuopio Finland). Her interest areas include social work with children and young people, insecurity and risks and cooperation among child welfare organizations.

Timo Toikko is a professor of social work at the University of Eastern Finland (Kuopio, Finland). His interest areas are in the methodology of research-based development and the current trends of health and social services.

References

Afifi

Brownridge

D. A.

Cox

Sareen

(2006). Physical punishment, childhood abuse, and psychiatric disorders. Child Abuse & Neglect, 30(10), 1093–1103.

Alderson

Morrow

(2004). Ethics, social research and consulting with children and young people. Barnardo’s.

Black

Heyman

Slep

A. M. S.

(2001). Risk factors for child physical abuse. Aggression and Violent Behavior, 6(2-3), 121–188.

Black

Slep

A. M. S.

Heyman

(2001). Risk factors for child psychological abuse. Aggression and Violent Behavior, 6(2-3), 189–201.

Bronfenbrenner

(1977). Toward an experimental ecology of human development. American Psychologist, 32(7), 513–522.

Bronfenbrenner

(1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 22(6), 723–742.

Bronfenbrenner

(1992). Ecological systems theory. Jessica Kingsley.

Chan

K. L.

Emery

C. R.

(2016). Children with disability are more at risk of violence victimization: Evidence from a study of school-aged Chinese children. Journal of Interpersonal Violence, 31(6), 1026–1046.

Creighton

Ghate

Hazel

Field

Finch

(2003). Putting the Conflict Tactics Scale in context in violence from parent to child. In Raymond

M. L.

Stanko

A. E.

(Eds.), Researching violence: Essays on methodology and measurement (pp. 30–48). Routledge.

10.

Cuevas

C. A.

Finkelhor

Ormrod

Turner

(2009). Psychiatric diagnosis as a risk marker for victimization in a national sample of children. Journal of Interpersonal Violence, 24(4), 636–652.

11.

Darlington

R. B.

Hayes

A. F.

(2017). Regression analysis and linear models: Concepts, applications, and implementation. Guilford Press.

12.

Doidge

J. C.

Higgings

D. J.

Delfabbro

Segal

(2017). Risk factors for child maltreatment in an Australian population-based birth cohort. Child Abuse & Neglect, 64, 47–60.

13.

Drake

Jonson-Reid

(2014). Poverty and child maltreatment. In Korbin

J. E.

Krugman

R. D.

Miller-Fellows

(Eds.), Handbook of child maltreatment (pp. 152–168). Springer.

14.

Ellonen

Kääriäinen

Sariola

Helweg-Larsen

Bøving Larsen

(2011). Adolescents’ experiences of parental violence in Danish and Finnish families: A comparative perspective. Journal of Scandinavian Studies in Criminology and Crime Prevention, 12(2), 173–197.

15.

Fagerlund

Peltola

Kääriäinen

Ellonen

Sariola

(2014). Lasten ja nuorten väkivaltakokemukset: Lapsiuhritutkimuksen tuloksia 2013 [The report of Finnish Child Victim Survey 2013] (Poliisiammattikorkeakoulun raportteja 110). Poliisiammattikorkeakoulu.

16.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Poly-victimization: A neglected component in child victimization. Child abuse & neglect, 31(1), 7-26.

17.

Finnish Advisory Board on Research Integrity. (2019). Ethical review in human sciences. https://www.tenk.fi/sites/tenk.fi/files/eettisetperiaatteet.pdf

18.

Giardino

A. P.

Giardino

E. R.

Isaac

(2014). Child Maltreatment and Disabilities: Increased Risk. In Korbin

Krugman

(Eds.), Handbook of child maltreatment (pp. 169-185). Dordrecht: Springer.

19.

Gilbert

Widom

C. S.

Browne

Fergusson

Webb

Janson

(2009). Burden and consequences of child maltreatment in high-income countries. The Lancet, 373(9657), 68–81.

20.

Heinonen

Ellonen

(2013). Are children with disabilities and long-term illnesses at increased risk of disciplinary violence? Journal of Scandinavian Studies in Criminology and Crime Prevention, 14(2), 172–187.

21.

Hershkowitz

Lamb

M. E.

Horowitz

(2007). Victimization of children with disabilities. American Journal of Orthopsychiatry, 77(4), 629–635.

22.

Jaudes

P. K.

Mackey-Bilaver

(2008). Do chronic conditions increase young children’s risk of being maltreated? Child Abuse & Neglect, 32(7), 671–681.

23.

Jones

Bellis

M. A.

Wood

Hughes

McCoy

Eckley

Bates

Mikton

Shakespeare

Officer

(2012). Prevalence and risk of violence against children with disabilities: A systematic review and meta-analysis of observational studies. The Lancet, 380(9845), 899–907.

24.

Lanier

Jonson-Reid

Stahlschmidt

M. J.

Drake

Constantino

(2010). Child maltreatment and pediatric health outcomes: A longitudinal study of low-income children. Journal of Pediatric Psychology, 35(5), 511–522.

25.

Lepistö

Paavilainen

(2012). The association between child maltreatment and coping strategies among Finnish 9th graders. Child Welfare, 91(6), 103–123.

26.

McNeeley

(2012). Sensitive issues in surveys: Reducing refusals while increasing reliability and quality of responses to sensitive survey items. In Gideon

(Ed.), Handbook of survey methodology for the social sciences (pp. 377–396). Springer.

27.

Morris

Hegarty

Humphreys

(2012). Ethical and safe: Research with children about domestic violence. Research Ethics, 8(2), 125–139.

28.

Mudaly

Goddard

(2006). The truth in longer than a lie: Children’s experiences of abuse and professional interventions. Jessica Kingsley.

29.

Paavilainen

Flinck

(2013). National clinical nursing guideline for identifying and intervening in child maltreatment within the family in Finland. Child Abuse Review, 22(3), 209–220.

30.

Reiter

Bryen

D. N.

Shachar

(2007). Adolescents with intellectual disabilities as victims of abuse. Journal of Intellectual Disabilities, 11(4), 371–387.

31.

Rutter

(1981). The city and the child. American Journal of Orthopsychiatry, 51, 610–625.

32.

Rutter

(2006). Implications of resilience concepts for scientific understanding. Annals of the New York Academy of Sciences, 1094(1), 1–12.

33.

Sariola

Uutela

(1992). The prevalence and context of family violence against children in Finland. Child Abuse & Neglect, 16(6), 823–832.

34.

Sobsey

(2002). Exceptionality, education, and maltreatment. Exceptionality, 10, 29–46.

35.

Söderholm

Politi

(2012). Lapsen laiminlyönti [Child Abuse]. In Söderholm

Kivitie-Kallio

(Eds.), Lapsen kaltoinkohtelu [Child Maltreatment] (pp. 76-98). Helsinki: Duodecim.

36.

Spencer

Devereux

Wallace

Sundrum

Shenoy

Bacchus

Logan

(2005). Disabling conditions and registration for child abuse and neglect: A population-based study. Pediatrics, 116(3), 609–613.

37.

Springer

K. W.

Sheridan

Kuo

Carnes

(2003). The long-term health outcomes of childhood abuse. Journal of General Internal Medicine, 18(10), 864–870.

38.

Statistics Finland. (2019). Numbers and shares of persons with immigrant background by area 1990-2017. https://www.stat.fi/tup/maahanmuutto/maahanmuuttajat-vaestossa_en.html

39.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics Scale. Journal of Marriage and Family, 41, 75–88.

40.

Straus

M. A.

(1990). The Conflict Tactics Scale and its critics: An evaluation and new data on validity and reliability. In Straus

M. A.

Gilles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8145 families (pp. 49–73). Transaction.

41.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The revised Conflict Tactics Scales (CTS2) development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316.

42.

Sullivan

P. M.

Knutson

J. F.

(2000). Maltreatment and disabilities: A population-based epidemiological study. Child Abuse & Neglect, 24(10), 1257–1273.

43.

Svensson

Bornehag

C. G.

Janson

(2011). Chronic conditions in children increase the risk for physical abuse–but vary with socio-economic circumstances. Acta Paediatrica, 100(3), 407–412.

44.

Taylor

Lazenbatt

(2014). Child maltreatment and high risk families. Dunedin Academic Press.

45.

Van der Laan

A. M.

Veenstra

Bogaerts

Verhulst

F. C.

Ormel

. (2010). Serious, minor, and non-delinquents in early adolescence: The impact of cumulative risk and promotive factors. The TRAILS study. Journal of Abnormal Child Psychology, 38(3), 339–351.

46.

Werner

Smith

R. S.

(1982). Vulnerable but invincible: A longitudinal study of children and youth. Adams Bannister Cox Pubs.

47.

World Health Organization. (2019). Child maltreatment. https://www.who.int/news-room/fact-sheets/detail/child-maltreatment