Child abuse is described as deliberate violence to a child and/or the intentional withholding of care by parents or adults responsible for the child (Kempe and Hefler, 1972). There has been a significant increase in reported cases of various types of child abuse including physical, sexual and the newer categories of mental injuries (emotional abuse/neglect) in most parts of the world. More than 3.1 million reports of child abuse were recorded in the USA in 1994, an increase of more than 50 percent over the 1985 figure (Alexander, 1995).
In the Caribbean island of Dominica, the incidence of reported cases doubled between the years 1990 and 1993 from 127 to 252 (Taylor, 1997). This figure reached a high of 416 in 1994 and dropped to 260 in 1995. The reported incidence continued to rise in subsequent years reaching 267 in 1997 and 303 in 1998, a trend that is probably due to issues of sensitization and service access.
Sexual abuse was the most prevalent form of child abuse in Dominica in 1998, comprising 41.6% of cases registered in the Child Abuse Register (1999). It is defined as "sexual exploitation of a child through violent and nonviolent molestation. This includes a spectrum of behaviors from violent rape to inappropriate touching or seduction" (Eldemire, 1992).
Neglect represents the second highest number of cases reported in Dominica for the periods 1995 to 1998. It is defined as gross or repeated failure to provide for child’s physical or emotional needs such that there is harm or threatened harm to the child’s health or welfare (Sharpe, 1992). Physical abuse is the next most common form of child abuse. It is defined as any act by "parent" which results in non-accidental physical injury to the child. The word "parent" in the broad sense means both parent and caretaker or guardian (Eldemire, 1992).
In all cases of abuse, the psychological scars last a lifetime. This dictates that the event must not be allowed to occur. The significant predictors of child sexual abuse perpetrated by a family member were physical abuse, having no one to confide in, having no caring female adult, and having an alcoholic father (Fleming, 1997).
In order to manage the problem of child abuse in Dominica more effectively, a comprehensive health plan must be established which would attempt to identify the children at risk. It is believed that biological, psychological and social indicators of child welfare are strongly associated with the occurrence of childhood abuse. The loss of significant relations early in childhood, their living arrangements coupled with the use and/or abuse of alcohol and other drugs by parents predisposes children to abuse.
The aim therefore was to determine whether living arrangements and personal experiences of childhood influence the occurrence of abuse of children in Dominica. The objectives of the study were to: 1) determine the living arrangements of the abused children; 2) determine the psycho-social status of abused children; 3) compare abused children, with their age-specific population.
This is a case-control study of children who were reported to the Ministry of Community Development as having suffered abuse during the period January 1st to December 31st 1998. The cases were sampled using a random stratified method with individually matched control pairs. It is a non-intervention study, conducted in December 1999 and January 2000.
The stratification of the population was based on the gender, types of abuse and geographic distribution of the population. This ensured that the sample was representative of the population.
The control group was selected by individually matching the cases with children from the same community, with similar age and socioeconomic background. Excluded from the study were the children below the age of four years, as they were considered too young to appreciate the complex issues under investigation, and older than sixteen years at the time of registration. Children who were too scared, too traumatized or had too low an IQ were also included but a sibling was asked to assist in completing the questionnaire.
The data was collected using investigator-assisted questionnaire. This instrument was pre-tested and modified. Permission was obtained from the relevant bodies; the Ministry of Community Development, the teacher and the parents. The issue of confidentiality was ascertained.
The researcher administered all the interviews. The interviews with the abused children and their controls carefully avoided any aspect of the abuse experienced by the subject. Most of these interviews were conducted while the children were at school. In order to protect the identity of the abused children, the controls were always identified at the same time as the cases.
The data from the completed questionnaires were entered into a computer and analyzed using EPI-INFO software. Summary statistics such as central tendency, proportion, frequency and percentages were used to describe while Pearson’chi square were used to determine relationships, associations and differences between responses in order to establish levels of statistical significance in the data which were collected. Significance were taken as p < 0.05. Odds ratio analysis were done to determine the strength of the association and the attributable risk were used to measure how much the risk of abuse were attributable to the factors studied.
The majority of children in this study attended a primary or a secondary school. However 12.5% of the abused children were not involved in the school system. Most were first-born (46.9%) while the control group was evenly distributed into first, middle or last child (31.3%, 31.3%, 37.5%).
There were significant differences in the living arrangements between the children who were registered as abused and those who had no reports of abuse. A significantly lesser number of abused children lived with both parents (1.0:2.1) p: 0.02. The odds ratio (OR) is 0.29 and a confidence interval (CI) 0.10, 0.85 (Table 3.1). The attributable risk (AR) is 70.6%.
Table 3.1: Distribution of children by Living arrangement
|
Living Arrangements |
Cases |
% |
Control |
% |
Total |
|
Both Parents |
8 |
25.0 |
17 |
53.1 |
25 |
|
Not Both Parents |
24 |
75.0 |
15 |
46.9 |
39 |
|
Total |
32 |
32 |
64 |
X2 : 5.32 p: 0.02
A greater number of the cases of abused children lived in other homes apart from their present home (1.3:1.0). The control group spent a longer period than the abused children living in the other homes (1.0yrs:1.5yrs). The AR is 46.8%. The children from the control group reported greater interest in living with their parents (1.0:1.2) p: 0.07. The OR is 0.24 and the AR is 76.2%.
The majority of the children reported to have lost one or both of their parents by death, migration or family separation (67.1%). However, the two groups, differed significantly in terms of their experience of loss during their short lives. A significantly higher proportion (1.5:1.0) of the abused children (Table 3.2) lost their parent(s) (p: 0.01). Family separation was a development more associated with abused children than their control partners (1.4:1.0).
Table 3.2 : Distribution of children by experience of loss
|
Loss of Parent(s) |
Cases |
% |
Control |
% |
Total |
|
Yes |
26 |
81.3 |
17 |
53.1 |
43 |
|
No |
6 |
18.7 |
15 |
46.9 |
21 |
|
Total |
32 |
32 |
64 |
X2 : 5.74, p : 0.01
The majority of the children reported to have lost their fathers (61%) while twenty-eight percent lost their mother. A higher proportion of abused children lost their fathers (1.4:1.0) p: 0.07. A similar pattern was observed in respect to the mothers (1.7:1.0). More of the abused children lost both parents (1.7:1.0) Approximately twenty-two percent of the reported abuses occurred around the time of loss of the parent(s). These children had fewer friends, as the proportion with three or more friends is 1.0:1.7.
The majority of the children reported sharing various types of activities with their parents including personal conversation (69%), schoolwork at home (72%) and social activities (78%). They however differed in terms of those activities. Those in the control group (Table 3.3) had significantly deeper relationship with their parents in terms of sharing social activities such as playing games and having fun time (1.0:1.4) p: 0.01. The OR is 0.20 and the AR is 80.3%.
Table 3.3: Distribution of children by activities shared with parents
|
Social Activities |
Cases |
% |
Control |
% |
Total |
X2 |
P |
|
Yes |
21 |
65.7 |
29 |
90.63 |
50 |
||
|
No |
11 |
34.3 |
3 |
9.4 |
14 |
||
|
Total |
32 |
32 |
64 |
5.85 |
0.01 |
||
|
|
|||||||
|
Homework |
|||||||
|
Yes |
20 |
62.5 |
26 |
81.25 |
46 |
||
|
No |
12 |
37.5 |
6 |
28.8 |
18 |
||
|
Total |
32 |
32 |
64 |
2.78 |
0.09 |
||
|
Personal conversation |
|||||||
|
Yes |
20 |
62.5 |
24 |
75.0 |
44 |
||
|
No |
12 |
37.5 |
8 |
25.0 |
20 |
||
|
Total |
32 |
32 |
64 |
1.16 |
0.28 |
There was also a higher number of parents of children of the control group who participated in their children’s schoolwork at home (1.0:1.3) and who also shared personal conversation (1.0:1.2) with their children (p : 0.09), (p : 0.28).
More of the children who suffered abuse claimed to attend group meetings at church or otherwise (4.1:1.0) immediately after school while more of the control group admitted to going straight home (1.0:1.2). The non-abused children were more likely to attend church services (1.0:2.4) 3-4 times monthly (p: 0.0001). The OR is 0.13 and the preventive fraction risk is 87%.
The use of substances of abuse by parents represented a significant problem. Alcohol use (Table 3.4) among the parents of abused children was significantly higher (1.5:1.0) than their non-abused partners (p: 0.04). These children also identified alcohol use at a significantly higher proportion among their mothers (7.0:1.0) p: 0.02. The OR is 8.68 and the AR is 88.5%.
Table 3.4: Distribution of alcohol use by parents
|
Alcohol Use |
Cases |
% |
Control |
% |
Total |
X2 |
p |
|
Parents |
|||||||
|
Yes |
23 |
71.8 |
15 |
46.8 |
38 |
||
|
No |
9 |
28.2 |
17 |
53.2 |
26 |
||
|
Total |
32 |
32 |
64 |
4.15 |
0.04 |
||
|
Mother specific |
|||||||
|
Yes |
7 |
21.8 |
1 |
3.1 |
8 |
||
|
No |
25 |
78.2 |
31 |
96.9 |
56 |
||
|
Total |
32 |
32 |
64 |
5.14 |
0.02 |
||
|
Father specific |
|||||||
|
Yes |
18 |
56.2 |
13 |
40.6 |
31 |
||
|
No |
14 |
43.6 |
19 |
59.3 |
33 |
||
|
Total |
32 |
32 |
64 |
1.56 |
0.21 |
||
The fathers were more likely to be alcohol users than mothers in both groups, but the prevalence was higher in the abused children (1.3:1.0). In addition, there were significant differences between the two groups as more abused children remembered seeing their parents in a state of drunkenness (2.2:1.0) p: 0.001. The AR is 82.5%. Furthermore a significant number of abused children acknowledged that their parents (Table 3.5) used illicit drugs (4.3:1.0) p: 0.004. The OR is 6.39 and the AR is 84.3%.
Table 3.5: Distribution of illicit drug use by parents
|
Illicit drug use |
Cases |
% |
Control |
% |
Total |
|
Yes |
13 |
40.6 |
3 |
9.6 |
16 |
|
No |
19 |
59.4 |
29 |
90.4 |
48 |
|
Total |
32 |
32 |
64 |
X2 : 7.96, p : 0.004
A much higher number of abused children remembered seeing their parents involved in family violence (2.2:1.0) p: 0.08. The attributable risk is 64.6%.
This study has demonstrated that the living arrangements of children, the absence of one or both parents, the shared activities between parents and children and the use of drugs by parents are predictive of childhood abuse in Dominica.
Only one-quarter of abused children lived with both parents when compared with more than one-half of the control group. This finding was significant and illustrates the importance of family setting as a protective factor in the wholesome development of children. The nuclear family type is probably not the most common form of family organization in the Caribbean (McKenzie,1990) and social problems such as economic deprivation, child abuse among others can probably be traced to the other forms of relationships that prevail in the society.
As expected the children from the control group expressed greater interest in their family relationship. Those who lived in other homes apart from parents’ actually lived for longer period in those homes than their abused counterpart who themselves were more likely to live in other homes. This symbolized the significance of stability to these children and its' protective impact on their lives.
The majority of the children interviewed lost one or both parents (67.1%). This experience had a significantly worst outcome for the abused children (81.3%). Although family separation was the most frequent form of loss reported, migration was a major factor contributing to the lost of one or both parent.
This finding substantiated an earlier report on the Situation Analysis of Children and their Families in Dominica (UNICEF, 1996). It reported that out-migration impacted significantly on family life in Dominica as women were more likely to migrate than males. Dominica's children therefore are likely to lose their main caregivers at crucial stages in their development exposing them to all kinds of social and economic risk.
In this study the majority of abused children lost their mothers and even more lost their fathers (72%). Almost forty percent of the abused children were growing up with neither their biological mother nor father. It is interesting to note that more than one fifth of the reports of abuse occurred around the time of parental loss. In an Argentine study (Morales et al, 1997) social support was noted as a crucial protective factor against child abuse.
It is quite normal for children to share activities with their parents and in this study the majority of the children interviewed were actively involved with their parents. However the control group of children reported deeper relationship with their parents. This difference was most significant in terms of social activities shared such as playing games and having fun-time.
The activities assessed in this study are all parameters to measuring the functional state of the family. The noted trend of parents assisting their children with schoolwork at home or sharing conversations and social activities favored the non-abused children. This is similar to findings by Paradise et al. (1994) where he illustrated those preexisting poor family interactions may contribute importantly to persistent problematic behavior among abused children.
The abused children were more likely to be involved in extracurricular activities on schooldays rather than go straight home. On the other hand the control group attended church more frequently and this may have served as a protective factor to child abuse. It is clear therefore that social activities by the children can impact positively or negatively on their personality shaping. The church can be seen as a medium of controlled interaction while the extra-curricular activities after school may not be so well supervised and posed more risk for unpleasant experiences.
The findings of this study supports previous work done on the predictive value of substances of abuse on the occurrence of child abuse (Chaffin, 1996; Fleming, 1997; Miller, 1999; Kotch, 1999; Ammerman, 1999). The abused children were more likely to report alcohol use in their parents. It is interesting to note however that the mothers of the abused children were consumers of alcohol at a significantly higher proportion than their female counterparts of the control group. Miller et al (1999) reported that mothers with that problem were more punitive towards their children.
Furthermore, the abused children reported that a significantly higher proportion of their parents were seen in a drunken state (p= 0.001) and that they used illicit drugs (p= 0.004). Ammerman et al (1999) confirmed that emotional dysregulation resulting from substance use disorders predicted the occurrence of child abuse. In another study done in the United State of America (Chaffin et al, 1996) substance abuse disorders were strongly associated with the onset of both abuse and neglect. Sexual abuse was the most common form of abuse suffered by the children in this study and Fleming et al (1997) reported in an Australian study that having an alcoholic father and/or alcoholic mother were significant predictors of childhood sexual abuse.
The exposure of children to physical confrontation between their parents appears to be associated with child abuse. There was a tendency, in this study, for abused children to report that form of family violence. In a New Zealand study, Rodriguez and Green (1997) reported that anger expression was positively correlated with child abuse potential. Miller et al (1999) also supported this finding when he reported that history of partners' violence and parental violence among mothers predicted a higher level of mother-child punitiveness.
In this study, the psycho-social factors that predisposed children to abuse were loss of one or both parents, not living with both parents, not sharing social activities with their parents or irregular attendance at church services. Also children who identified their parents as users of alcohol, who saw their parents in a drunken state or reported their parents as illicit drug users were at an increase risk of abuse. Children who were not interesting in living with their parents trended towards having an abusive experience.
Alexander R.C. (1995) Current and Emerging concepts in child abuse. Compr. Ther.; 21 (12): 726-30.
Ammerman R.T., Kolko D.J., Kirisci L., and Blackson T.C. (1999) Child abuse potential in parents with histories of substance use disorder. Child Abuse Negl; 23 (12): 1225-38.
Chaffin M., Kelleher K., and Hollenberg J. (1996) Onset of physical abuse and neglect: psychiatric, substance abuse, and social risk factors from prospective community data. Child Abuse Negl; 20 (3): 191-203.
Eldemire, A. (1992) Overview of Physical Abuse - Definition and Indicators. In Desai, P. and Ramlal, AM.(Eds). Child Abuse: A Regional Problem. UNICEF, Barbados, 1995, pp 4-5.
Fleming J., Muller P., and Bammer G. (1997) A study of potential risk factors for sexual abuse in childhood. Child Abuse Negl; 21 (1): 49-58.
Kempe C.H. and Hefler R. (1972) Helping the battered child and his family. Lippincott Co., Philadelphia, pp. 86-104 and 198-215.
Kotch J.B., Browne D.C., Ringwatt C.L., and Stewart P.W. (1995) Risk of child abuse or neglect in a cohort of low income children. Child Abuse Negl; 19(9): 1115-30.
McKenzie H. and McKenzie H.(1990) Sociology and the Caribbean Family. In The Caribbean Family Workshop. Bureau of Health Education, Ministry of Health, Jamaica.
Miller B.A., Smyth N.J., and Mudar P.J. (1999) Mothers' alcohol and other drug problems and their punitiveness toward their children. J Stud Alcohol; 60 (5): 632-42e.
Ministry of Community Development and Gender Affairs, Commonwealth of Dominica (1999) Child Abuse Register, 1999.
Morales J.M., Zunzunegui M.V., and Martinez V. (1997) Child abuse: socioeconomic factors and health status. An Esp Pediatr; 47 (1): 33-41.
Paradise J.E., Rodriguez C.M., and Green A.J. (1997) Parenting stress and anger expression as predictors of child abuse potential. Child Abuse Negl; 21 (4): 367-77.
Rodriguez C.M., and Green A.J. (1997) Parenting stress and anger expression as predictors of child abuse potential. Child Abuse Negl; 21 (4): 367-77.
Sharpe, J. (1992) Overview of Emotional Abuse and Neglect. In: Desai, P. and Ramlal, A.M. (Eds), Child Abuse: A Regional Problem, pp 6-7 UNICEF, Barbados, 1995.
Taylor, D.C. (1997) Evaluation Report on Child Abuse Prevention Programme, pp33. Division of Welfare, Ministry of Community Development and Women’s Affairs, Commonwealth of Dominica.
UNICEF (1996) Situational Analysis of Children and their Families in Dominica. Commonwealth of Dominica, 1991-96
URL http://www.uwichill.edu.bb/bnccde/dominica/conference/papers/Benjamin.html
© Griffin Benjmain, 2001. HTML prepared using 1st Page 2000, revised February 5th, 2001.