CanJPsychiatry 2015;60(7):315–323
Original Research
Substantiated Reports of Child Maltreatment From the Canadian Incidence Study of Reported Child Abuse and Neglect 2008: Examining Child and Household Characteristics and Child Functional Impairment Tracie O Afifi, PhD1; Tamara Taillieu, MSc2; Kristene Cheung, BA3; Laurence Y Katz, MD, FRCPC4; Lil Tonmyr, PhD5; Jitender Sareen, MD, FRCPC6 1
Associate Professor, Departments of Community Health Sciences and Psychiatry, University of Manitoba, Winnipeg, Manitoba. Correspondence: S113 Medical Services Building, 750 Bannatyne Avenue, Winnipeg, MB R3E 0W3; [email protected].
2
Graduate Student, Department of Applied Health Sciences, University of Manitoba, Winnipeg, Manitoba. Graduate Student, Department of Psychology, University of Manitoba, Winnipeg, Manitoba.
3
Professor, Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba.
4
Senior Research Advisor, Health Surveillance and Epidemiology Division, Public Health Agency of Canada, Ottawa, Ontario.
5
Professor, Departments of Psychiatry, Psychology, and Community Health Sciences, University of Manitoba, Winnipeg, Manitoba.
6
Key Words: child abuse, child neglect, child welfare, suicidal ideation, self-injurious behaviour, depression, anxiety, alcohol abuse, drug abuse, single parent
Objective: Identifying child and household characteristics that are associated with specific child maltreatment types and child functional impairment are important for informing prevention and intervention efforts. Our objectives were to examine the distribution of several child and household characteristics among substantiated child maltreatment types in Canada; to determine if a specific child maltreatment type relative to all other types was associated with increased odds of child functional impairment; and to determine which child and household characteristics were associated with child functional impairment.
Received June 2014, revised, and accepted September 2014.
Method: Data were from the Canadian Incidence Study of Reported Child Abuse and Neglect (collection 2008) from 112 child welfare sites across Canada (n = 6163 children).
open access
Results: Physical abuse, sexual abuse, and emotional maltreatment were highly prevalent among children aged 10 to 15 years. For single types of child maltreatment, the highest prevalence of single-parent homes (50.6%), social assistance (43.0%), running out of money regularly (30.7%), and unsafe housing (30.9%) were reported for substantiated cases of neglect. Being male, older age, living in a single-parent home, household running out of money, moving 2 or more times in the past year, and household overcrowding were associated with increased odds of child functional impairment. Conclusions: More work is warranted to determine if providing particular resources for single-parent families, financial counselling, and facilitating adequate and stable housing for families with child maltreatment histories or at risk for child maltreatment could be effective for improving child functional outcomes. WWW
Rapports documentés de maltraitance d’enfants tirés de l’Étude canadienne sur l’incidence des signalements de cas de violence et de négligence envers les enfants de 2008 : examen des caractéristiques de l’enfant et du ménage, et de la déficience fonctionnelle de l’enfant Objectif : Identifier les caractéristiques de l’enfant et du ménage qui sont associées avec les types spécifiques de maltraitance d’enfants et la déficience fonctionnelle de l’enfant est important pour éclairer les efforts de prévention et d’intervention. Nos objectifs étaient d’examiner la répartition de plusieurs caractéristiques des enfants et des ménages parmi les types documentés de maltraitance d’enfants au Canada; déterminer si un type spécifique de maltraitance, relativement à tous les autres types, était associé à des probabilités accrues de déficience fonctionnelle de l’enfant; et déterminer les caractéristiques de l’enfant et du ménage qui étaient associées à la déficience fonctionnelle de l’enfant.
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 315
Original Research
Méthode : Les données de l’Étude canadienne sur l’incidence des signalements de cas de violence et de négligence envers les enfants (2008) obtenues de 112 centres de protection de l’enfance du Canada (n = 6163 enfants). Résultats : La violence physique, l’abus sexuel, et la maltraitance émotionnelle étaient hautement prévalents chez les enfants de 10 à 15 ans. Pour les types uniques de maltraitance d’enfants, la prévalence la plus élevée de foyers monoparentaux (50,6 %), d’aide sociale (43,0 %), de manque d’argent constant (30,7 %) et de logement non sécuritaire (30,9 %) a été signalée pour les cas de négligence documentés. Le sexe masculin, l’âge, un foyer monoparental, un ménage manquant d’argent, 2 déménagements ou plus l’année précédente, et un ménage surpeuplé étaient associés à des probabilités accrues de déficience fonctionnelle de l’enfant. Conclusions : Il faut plus de travail pour déterminer si l’offre de ressources particulières aux familles monoparentales, de conseils financiers, et d’aide au logement adéquat et stable aux familles qui ont des antécédents ou qui sont à risque de maltraitance d’enfants pourrait améliorer efficacement les résultats fonctionnels des enfants.
C
hild maltreatment includes abuse and neglect (that is, physical abuse, sexual abuse, neglect, emotional maltreatment, and exposure to IPV). Nationally representative Canadian data have indicated that 32% of the adult population in Canada has experienced child abuse, including physical abuse, sexual abuse, and exposure to IPV.1 Research has shown that childhood maltreatment is associated with mental disorders,2–10 physical health conditions,11–13 decreased quality of life,14 and suicidal behaviour15–17 that extends into adulthood. From a public health perspective, it is important to identify the child and household characteristics that are associated with specific types of child maltreatment and child functional impairment. This information can be used to develop and examine programs and policies aimed at protecting children and preventing maltreatment and associated impairment, such as mental illness from occurring.
The CIS is a rich data source that provides a unique opportunity to examine reported cases of child maltreatment within the Canadian context at a national level. Recently, a systematic review of the published studies examining CIS data indicated that child characteristics were associated with an increased likelihood of certain types of substantiated maltreatment.18 More specifically, older age of the child was associated with increased odds of substantiated physical abuse and sexual abuse, but not with neglect.19 Female sex was associated with increased odds of substantiated sexual abuse, decreased odds of physical abuse, but not with neglect.18 However, what is currently unknown is how a child’s age and sex is associated with all single types of substantiated child maltreatment cases and having multiple types of substantiated
Abbreviations AOR
adjusted odds ratio
CIS
Canadian Incidence Study of Reported Child Abuse and Neglect
IPV
intimate partner violence
316 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
Clinical Implications •
Knowledge about child and household characteristics can assist clinicians to identify people with greater odds of child functional impairment after experiencing child maltreatment.
•
Child characteristics associated with increased odds of functional impairment included being male and increasing age of the child.
•
Household characteristics associated with increased odds of child functional impairment included being in a single-parent home, being in a household that regularly runs out of money, household overcrowding, and moving 2 or more times in the past year.
Limitations •
Data are cross-sectional in nature.
•
Data are not representative of all child maltreatment in Canada.
•
Some types of functional impairment may have been underreported owing to lack of knowledge of the child welfare worker.
maltreatment. Additionally, the relations between household characteristics (for example, unsafe housing, number of children, or single parenting) and specific types of substantiated child maltreatment have not yet been examined. Child and household characteristics could also be useful beyond identifying increased likelihood of child maltreatment and may also provide insight into factors that may be associated with child functional impairment among children exposed to maltreatment. Previous research examining child maltreatment in Canada has provided some information on how child and household characteristics may be related to maltreatment and functional impairment. However, many gaps in our knowledge remain, which limits the ability to develop prevention and intervention strategies using evidencebased research. A more in-depth examination of child and household characteristics of families reported to child welfare is an approach to further knowledge of child www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
maltreatment within the Canadian context, which can have direct implications for intervention strategies within Canada and in other countries. Our study addresses existing gaps in our knowledge of reported child maltreatment to child welfare in Canada. The 3 main objectives of our study were to examine the distribution of several child and household characteristics among each specific single substantiated type of child maltreatment and among those with multiple types of substantiated maltreatment in Canada; to determine if a specific child maltreatment type relative to all other types of child maltreatment was associated with increased odds of child functional impairment; and to determine which child and household characteristics were associated with child functional impairment. It was hypothesized that the distribution of child and household characteristics would vary according to the specific child maltreatment type; all types of child maltreatment would be associated with increased odds of child functional impairment; and that all child and household characteristics would be associated with increased odds of child functional impairment.
Method Sample
The CIS is a child health surveillance activity of the Public Health Agency of Canada and is the product of collaboration among provincial and territorial governments, researchers, child welfare service providers, and policy makers from across the country.20 The CIS provides a nationwide sample of reports of child maltreatment to child welfare agencies across Canada. Data for our study came from the third and most recent cycle of the CIS, collected between October and December 2008 (CIS-2008). We used the CIS-2008 investigations of children aged 15 years and younger from 112 child welfare sites across Canada. Child welfare workers were provided a half-day training session and research assistants were available to respond to questions throughout the data collection period. At the end of the initial investigation, child welfare workers filled out a standardized questionnaire regarding investigation-related characteristics, including child and family or household characteristics, maltreatment variables, service dispositions, and child functional impairment. Child welfare workers determined the level of substantiation for each type of reported maltreatment: unfounded (evidence indicates this form has not occurred), suspected (not enough evidence to substantiate, but cannot be ruled out), and substantiated (evidence indicates this form has occurred). Up to 3 types of maltreatment could be substantiated per investigation. For our study, we were interested in examining single types of substantiated child maltreatment and multiple types of substantiated maltreatment (n = 6163 from 4095 different households, with 4974 single types of substantiated maltreatment cases and 1189 cases with more than 1 type of maltreatment). www.TheCJP.ca
Measures Child Maltreatment The CIS-2008 assessed 5 different types of child maltreatment: physical abuse, sexual abuse, neglect, emotional maltreatment, and exposure to IPV. Physical abuse included the following acts: shake, push, grab, throw; hit with hand; punch, kick, bite; hit with object; choke, poison, stab; and other physical abuse. Sexual abuse included the following acts: penetration, attempted penetration, oral sex, fondling, sex talk or images, voyeurism, exhibitionism, and other sexual abuse. Neglect included the following: failure to supervise, leading to physical harm; failure to supervise, leading to sexual abuse; permit criminal behaviour; physical neglect; medical or dental neglect; failure to provide psychological treatment; abandonment; and educational neglect. Emotional maltreatment included the following: terrorize, threat of violence; verbal abuse, belittling; isolation, confinement; inadequate nurturing; exploit, corrupting behaviour; and exposure to nonpartner violence. Exposure to IPV included the following: direct witness to violence; indirect exposure to violence; and exposure to emotional violence. Child and Household Characteristics Child demographic characteristics included sex (male or female) and age (in years). Household characteristics included single-parent home (yes or no), total number of children younger than 20 years old living in the home (1, 2, 3, or 4 or more), social assistance (yes or no), household regularly runs out of money for basic necessities (yes or no), household overcrowding (yes or no), number of moves in the past year (0, 1, 2, or 3 or more), and unsafe housing (yes or no). The social assistance variable was based on the household’s main income source. In our study, social assistance or no income was compared with employed full-time, part-time, multiple jobs, seasonal, employment insurance benefits, or other benefits. Unsafe housing included guns or weapons in the home; needles, legal, or illegal drugs in home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial solvents or chemicals in home; poisons, fire implements, or electrical hazards in the home; or quality of the home presents a health risk to the child. A positive response to any of the aforementioned items was coded as living in an unsafe home. Child Functional Impairment Child welfare workers assessed numerous different child functional impairments at the end of the initial investigation. Functional impairments included the following: depression, anxiety, or withdrawal; suicidal thoughts; self-harming behaviour; attention-deficit disorder, attention-deficit hyperactivity disorder; attachment issues; aggression; running away (multiple incidents); inappropriate sexual behaviour; Youth Criminal Justice Act involvement; intellectual or developmental disability; failure to meet developmental milestones; academic difficulties; fetal alcohol syndrome or fetal alcohol effects; positive The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 317
Original Research
toxicology at birth; physical disability; alcohol abuse; drug or solvent abuse; and any other functioning conditions. A full description of the variables can be found elsewhere.20 The assessment of variables differed; for instance, depression, anxiety, or withdrawal were assessed by feelings of anxiety and (or) depression that persist for most of every day for 2 weeks or longer, and interfere with the child’s ability to manage at home and at school. Suicidal thoughts were assessed as follows: the child has expressed thoughts of suicide, ranging from fleeting thoughts to a detailed plan. Owing to the nature of certain functional impairments (that is, attention-deficit disorder or attention-deficit hyperactivity disorder; intellectual or developmental disability; failure to meet developmental milestones; fetal alcohol syndrome or fetal alcohol effects; and positive toxicology at birth), no referent period was specified in the questionnaire. For all other functional impairments, the referent period was the past 6 months. Child welfare workers documented the presence or absence of each type of child functional impairment using the following categories: no (the child welfare worker believes there is no problem), suspected (condition may be present, but has not been confirmed), confirmed (disclosed by caregiver, diagnosed, or observed), or unknown (unsure or have not attempted to determine its presence). In our study, children were coded as having a functional impairment if the child welfare worker suspected or confirmed that one or more specific functional impairment was present. Investigated children were coded as no functional impairment if the child welfare worker documented no evidence of any assessed child functional impairment. Children with an unknown status of functional impairment were not included in the analysis, as it was not possible to determine with certainty to which group they would belong. Most of the child functioning issues have excellent test–retest reliability, with kappas showing excellent-to-moderate agreement (0.81–0.46) during a 3-week period.21
Statistical Analyses
First, cross-tabulations and chi-square tests of significance were computed to determine the distribution of child and household characteristics by substantiated child maltreatment type. Owing to the lack of independence of the observations and the positive correlation expected for children from the same household, a generalized mixed model with a binomial response were applied to the data. The generalized mixed model comprised fixed effects (that is, independent variable effects) as well as a random effect (that is, household) to account for its additional variation (that is, extra binomial variation). The first set of generalized mixed models were computed to examine the relation between child maltreatment type and child functional impairment. The models were first computed unadjusted and then adjusting for child age and sex. The second set of generalized mixed models were computed to determine if specific child and household characteristics were associated with child functional impairment. To determine if the relation between 318 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
each specific child and household characteristic was related to functional impairment, independent of other child and household characteristics, we ran the first models, adjusted, for all other child and household characteristics (AOR-1). These models were then further adjusted for types of child maltreatment (AOR-2).
Results
Table 1 presents the descriptive statistics for child and household characteristics by substantiated child maltreatment type. The findings indicate that, with the exception of a higher prevalence of sexual abuse among females, the prevalence of all other types of substantiated child maltreatment were about equal for males and females. Physical abuse, sexual abuse, and emotional maltreatment were more prevalent among older children, aged 10 to 15 years. The highest prevalence of single-parent homes were reported for substantiated cases of neglect (50.6%). The highest prevalence of unsafe housing and regularly running out of money regularly were reported for substantiated cases of neglect (30.9% and 30.7%, respectively) and more than 1 type of maltreatment (27.2% and 34.7%, respectively). In the sample, child welfare workers documented no evidence of functional impairments in 45.6% (n = 2429) of the cases of substantiated maltreatment and documented the presence of at least 1 suspected and (or) confirmed functional impairment in 54.4% (n = 2894) of the cases. In the sample, the mean number of functional impairments was 1.16 (SD 1.86). Table 2 presents the descriptive statistics and odds ratios from the generalized mixed models for the relations between types of child maltreatment and child functional impairment. The results indicated that physical abuse, sexual abuse, neglect, and more than one type of child maltreatment, compared with all other single types of child maltreatment, were associated with increased odds of child functional impairment. These findings remained significant after adjusting for child age and sex. Emotional maltreatment relative to all other single types of child maltreatment was not associated with increased odds of child functional impairment in the unadjusted and adjusted models. Exposure to IPV, compared with all other single types of child maltreatment, was associated with decreased odds of child functional impairment in the unadjusted and adjusted models. Table 3 presents the descriptive statistics and the odds ratios from the generalized mixed model for the relations between child and household characteristics and child functional impairment. The findings indicated that males relative to females, older children (ages 5 to 15 years) relative to younger children (ages 0 to 4 years), single-parent homes, household overcrowding, 2 or more moves in the past year, and regularly running out of money were all associated with increased odds of child functional impairment after adjusting for all child and household characteristics and for type of child maltreatment. Unsafe housing was associated with increased odds of child functional impairment in models adjusting for other child and household characteristics, but was no longer significant after further www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
Table 1 Child and household characteristics by substantiated maltreatment type, n (%) Substantiated maltreatment type Physical abuse
Characteristic
Sexual abuse
Neglect
Emotional maltreatment
IPV exposure
More than 1 type
Total
Child sex
53.648, 5
Female
413 (47.1)
125 (75.8)
833 (47.0)
200 (52.1)
885 (49.8)
572 (48.1)
3028 (49.1)
Male
463 (52.9)
40 (24.2)
940 (53.0)
184 (47.9)
891 (50.2)
617 (51.9)
3135 (50.9)
Child age, years
379.447, 10
0 to 4
104 (11.9)
19 (11.5)
627 (35.4)
97 (25.3)
763 (43.0)
336 (28.3)
1946 (31.6)
5 to 9
310 (35.4)
63 (38.2)
528 (29.8)
101 (26.3)
561 (31.6)
381 (32.0)
1944 (31.5)
10 to 15
462 (52.7)
83 (50.3)
618 (34.9)
186 (48.4)
452 (25.5)
472 (39.7)
2273 (36.9)
Single-parent home
206.540, 5
No
620 (70.8)
97 (58.8)
875 (49.4)
231 (60.2)
1255 (70.7)
728 (61.2)
3806 (61.8)
Yes
256 (29.2)
68 (41.2)
898 (50.6)
153 (39.8)
521 (29.3)
461 (38.8)
2357 (38.2)
40 (24.2)
464 (26.2)
94 (24.5)
392 (22.1)
250 (21.0)
1450 (23.5)
Total children in home 1
χ2, df All P ≤ 0.001
210 (24.0)
58.940, 15
2
366 (41.8)
68 (41.2)
554 (31.2)
142 (37.0)
676 (38.1)
414 (34.8)
2220 (36.0)
3
172 (19.6)
35 (21.2)
379 (21.4)
84 (21.9)
401 (22.6)
285 (24.0)
1356 (22.0)
≥4
128 (14.6)
22 (13.3)
376 (21.2)
64 (16.7)
307 (17.3)
240 (20.2)
1137 (18.4)
Social assistancea
239.094, 5
No
704 (85.5)
115 (74.2)
944 (57.0)
272 (73.9)
1251 (73.3)
773 (67.8)
4059 (69.4)
Yes
119 (14.5)
40 (25.8)
711 (43.0)
96 (26.1)
455 (26.7)
367 (32.2)
1788 (30.6)
Household regularly runs out of money
391.848, 5
No
710 (94.0)
131 (92.9)
955 (69.3)
305 (89.7)
1316 (86.8)
639 (65.3)
4056 (79.4)
Yes
45 (6.0)
10 (7.1)
423 (30.7)
35 (10.3)
201 (13.2)
340 (34.7)
1054 (20.6)
Household overcrowding
85.116, 5
No
811 (94.4)
147 (94.8)
1467 (86.7)
335 (87.9)
1621 (93.6)
994 (86.7)
5375 (90.1)
Yes
48 (5.6)
8 (5.2)
226 (13.3)
46 (12.1)
111 (6.4)
153 (13.3)
592 (9.9)
Number of moves in past year
133.789, 15
0
499 (73.0)
98 (73.1)
734 (54.9)
222 (66.1)
884 (61.2)
570 (58.9)
3007 (61.3)
1
154 (22.5)
21 (15.7)
355 (26.6)
81 (24.1)
393 (27.2)
227 (23.5)
1231 (25.1)
2
17 (2.5)
9 (6.7)
130 (9.7)
23 (6.8)
107 (7.4)
99 (10.2)
385 (7.9)
≥3
14 (2.0)
6 (4.5)
118 (8.8)
10 (3.0)
61 (4.2)
72 (7.4)
281 (5.7)
Unsafe housingb
465.321, 5
No
697 (95.9)
110 (90.9)
954 (69.1)
288 (89.4)
1390 (93.3)
696 (72.8)
4135 (82.8)
Yes
30 (4.1)
11 (9.1)
426 (30.9)
34 (10.6)
100 (6.7)
260 (27.2)
861 (17.2)
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. Owing to rounding, the percentages may not total to 100%. a
Social assistance or no income, compared with employed full- or part-time, multiple jobs, seasonal, employment insurance, or other benefits.
b
Unsafe housing includes the following: guns and (or) weapons in the home; needles, legal and (or) illegal drugs in the home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial chemicals or solvents in the home; poisons, fire implements, or electrical hazards in the home; or quality of the home poses health risk to a child.
IPV = intimate partner violence
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 319
Original Research
Table 2 Substantiated child maltreatment type and child functional impairment accounting for the positive correlation among family members as a random effect in a generalized mixed model Child functional impairment No impairment n (%)
Impairment n (%)
OR (95% CI)
Physical abuse, compared with all other single types
297 (37.1)
504 (62.9)
Sexual abuse, compared with all other single types
49 (33.6)
Neglect, compared with all other single types
Type of maltreatment
AOR (95% CI)
1.85 (1.54–2.22)
a
1.29 (1.06–1.57)b
97 (66.4)
1.86 (1.26–2.76)
b
1.63 (1.08–2.45)c
563 (37.8)
928 (62.2)
2.01 (1.73–2.34)a
2.27 (1.93–2.68)a
Emotional maltreatment, compared with all other single types
150 (43.2)
197 (56.8)
1.27 (0.98–1.66)
d
1.15 (0.87–1.52)e
Exposure to IPV, compared with all other single types
1015 (68.9)
458 (31.1)
0.26 (0.22–0.30)a
0.31 (0.26–0.37)a
More than 1 type, compared with all other single types
355 (33.3)
710 (66.7)
1.91 (1.61–2.26)
1.91 (1.59–2.28)a
a
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. For cases where a single type of maltreatment was substantiated, odds ratios compare the odds of functional impairment for each individual type of maltreatment relative to the other 4 types combined (for example, physical abuse relative to sexual, neglect, emotional, and exposure combined). For cases where more than 1 type of maltreatment was substantiated, the odds ratios compare the odds of functional impairment for more than 1 type relative to the other 5 types combined. a
P ≤ 0.001; b P ≤ 0.01; c P ≤ 0.05; d P = 0.70; e P = 0.33
AOR = adjusted odds ratio; IPV = intimate partner violence
adjustment for type of maltreatment. Increasing number of children in the household decreased the odds of child functional impairment. Social assistance was not associated with child functional impairment in the adjusted models. The estimated variance (0.74) and standard error (0.11) of the household intercept (that is, random effects) in the fully adjusted model was significant at the P < 0.001 level, confirming that observations from children from the same household were positively correlated.
Discussion
Novel findings from this research indicate that the distribution of child and household characteristics vary according to the specific child maltreatment type. For example, the prevalence of physical abuse, sexual abuse, and emotional maltreatment increased with older age of the child. However, the prevalence of neglect remained fairly consistent across all 3 age categories. Notably, high prevalence of single-parent homes, social assistance, regularly running out of money, and unsafe housing were found for substantiated cases of neglect. High prevalence of household overcrowding was noted among cases of neglect and emotional maltreatment. In addition, the prevalence of regularly running out of money, household overcrowding, and unsafe housing was the highest in families where children experienced more than one type of child maltreatment, compared with some of the other single child maltreatment types. Descriptively, these data provide insight into child and household vulnerabilities that may be more likely among families at risk for child maltreatment and that certain household characteristics, as noted above, may be more common among certain child maltreatment types. Other main findings from this research indicate that in adjusted models, physical abuse, sexual abuse, neglect, and more than one type of abuse relative to all other types of substantiated maltreatment combined were associated with 320 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
increased odds of child functional impairment. Exposure to IPV relative to all other single types of child maltreatment combined was associated with decreased odds of child functional impairment. Child characteristics associated with increased odds of functional impairment included being a male and being older. Household characteristics associated with increased odds of child functional impairment included being in a single-parent home, household overcrowding, moving 2 or more times in the past year, and being in a household that regularly runs out of money. Greater number of children in the home was associated with significantly decreased odds of child functional impairment. The finding that exposure to IPV is related to decreased odds of child functional impairment relative to all other types of child maltreatment needs to be considered in the context of the study design. When interpreting this finding, it is important to note that the comparison is exposure to IPV, compared with other single types of substantiated child maltreatment, and not that exposure to IPV is protective against child functional impairment. It may be that exposure to IPV may not have the same association with child functional impairment relative to other types of child maltreatment that are directed specifically at the child. Regardless, it would be incorrect to interpret this finding as exposure to IPV is protective against child functional impairment as the comparison group was other child maltreatment and not a no child maltreatment group. Male sex was associated with increased likelihood of functional impairment. When interpreting this finding, it seems that understanding the functional impairments assessed in the study are important. The types of functional impairments that are often more common among males are more externalizing in nature and likely more readily assessed by the child welfare worker. It may not be that maltreated males, compared with maltreated females, inherently have www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
Table 3 Child and household characteristics by child functional impairment among substantiated reports of child maltreatment accounting for the positive correlation among family members as a random effect in a generalized mixed model Child functional impairment
Impairment
Impairment n (%)
No impairment n (%)
AOR-1 (95% CI)
AOR-2 (95% CI)
Female
1268 (49.1)
1317 (50.9)
1.00
1.00
Male
1626 (59.4)
1112 (40.6)
1.76 (1.50–2.07)
0 to 4
492 (31.9)
5 to 9
941 (55.1)
10 to 15
Child or household characteristic Child sex
a
1.83 (1.55–2.16)a
1052 (68.1)
1.00
1.00
767 (44.9)
3.35 (2.69–4.16)a
3.11 (2.49–3.88)a
1461 (70.5)
610 (29.5)
7.67 (6.15–9.56)a
6.51 (5.19–8.17)a
No
1673 (50.5)
1637 (49.5)
1.00
1.00
Yes
1221 (60.7)
792 (39.3)
1.34 (1.11–1.62)b
1.22 (1.01–1.49)c
1
729 (57.3)
543 (42.7)
1.00
1.00
2
1065 (54.8)
880 (45.2)
0.80 (0.64–1.00)d
0.84 (0.67–1.06)e
3
582 (49.4)
595 (50.6)
0.59 (0.46–0.76)a
0.59 (0.45–0.76)a
518 (55.8)
411 (44.2)
0.73 (0.55–0.98)c
0.74 (0.55–1.00)f
No
1915 (53.0)
1701 (47.0)
1.00
1.00
Yes
860 (57.9)
625 (42.1)
1.13 (0.90–1.41)
No
1832 (49.7)
Yes
611 (66.3)
No
2487 (52.9)
2215 (47.1)
1.00
1.00
Yes
320 (65.4)
169 (34.6)
1.47 (1.05–2.05)c
1.43 (1.03–2.01)c
0
1482 (54.3)
1249 (45.7)
1.00
1.00
1
559 (51.7)
2 ≥3
Child age, years
Single-parent home
Total children in home
≥4 Social assistance
d
g
1.09 (0.87–1.37)h
1853 (50.3)
1.00
1.00
310 (33.7)
1.78 (1.38–2.29)a
1.67 (1.29–2.16)a
Household regularly runs out of money
Household overcrowding
Number of moves in past year 523 (48.3)
0.92 (0.75–1.13)
i
0.95 (0.77–1.17)j
222 (66.3)
113 (33.7)
1.47 (1.02–2.12)c
1.47 (1.02–2.13)c
161 (67.6)
77 (32.4)
1.62 (1.04–2.52)
c
1.63 (1.04–2.56)c
No
1844 (48.9)
1926 (51.1)
1.00
1.00
Yes
476 (64.9)
258 (35.1)
1.57 (1.21–2.04)a
1.23 (0.94–1.60)k
Unsafe housing
e
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. No functional impairment is the reference category with an odds ratio of 1.00. a P ≤ 0.001; b P ≤ 0.01; c P ≤ 0.05; d P = 0.053; e P = 0.13; f P = 0.052; g P = 0.29; h P = 0.44; i P = 0.41; j P = 0.62; k P = 0.14 d
Social assistance or no income, compared with employed full-time, part-time, multiple jobs, seasonal, employment insurance, or other benefits.
e
Unsafe housing includes the following: guns and (or) weapons in the home; needles, legal and (or) illegal drugs in the home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial chemicals or solvents in the home; poisons, fire implements, or electrical hazards in the home; or quality of the home poses a health risk to a child.
AOR-1 = adjusted odds ratio (adjusted for all other child and household characteristics in model); AOR-2 = adjusted odds ratio (adjusted for child and household characteristics and types of maltreatment)
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 321
Original Research
more functional impairment, but rather the many types of functional impairment assessed in our study may be both more common among males and easier for a child welfare worker to assess. Further research is needed on the specific functional impairment among males and females. Increasing age of the child was associated with more functional impairment in our study, which is contradictory to many studies that indicate that maltreatment occurring in the younger age categories is more damaging.22,23 However, it may not be that this finding is inconsistent, but rather assessing a different phenomenon. It may be that the older age of the child is an indicator of longer exposure of maltreatment, as maltreatment is likely to be ongoing in these reported substantiated cases. As well, older children are easier to interview, which may make the recognition of impairment easier. It may also be that it takes time for the impairment to develop. Further, many of the functional impairments assessed are somewhat age-specific and biased toward the older age categories (that is, academic difficulties, running away from home, suicidal thoughts, and alcohol or drug abuse). Therefore, it may be that functional impairment is more likely in the older age categories (5 to 9 years and 10 to 15 years) relative to the youngest age category (0 to 4 years) simply owing to the nature of the types of child functional impairment assessed. Single-parent homes, household overcrowding, and regularly running out of money were 3 household characteristics that were associated with increased child functional impairment. These characteristics may be indicators of limited resources that can have an impact on the health and well-being of a child. Interestingly, social assistance was not associated with child functional impairment. It may be that families involved with social service agencies receive some additional funds and may qualify for more services and opportunities to help families. The findings from our study should be considered in light of important limitations. First, the data from the study were cross-sectional in nature, which precludes inferences regarding causation. We are unable to conclude that the child maltreatment caused the child functional impairment. Second, the data collected were based on the knowledge of the child welfare worker at the time of the investigation. It is possible that some types of functional impairment may have been underreported owing to lack of knowledge of the child welfare worker. Also, some types of functional impairment (externalizing in nature) may have been easier to evaluate and report, compared with other types of functional impairment (internalizing in nature). Third, these data consisted of only reported cases of child maltreatment without a comparison group of children without child maltreatment histories. This should be considered in interpreting the findings. Finally, the data were reported cases of child maltreatment in Canada. Many incidents of child maltreatment go unreported and it is often the more severe and chronic cases that are reported to authorities. Therefore, these data are not representative of all child maltreatment in Canada. 322 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
Our study provides insight into child and household factors that may be associated with child maltreatment reported to child welfare agencies and child functional impairment. Importantly, when thinking of our study finding, in relation to better child functioning, being female, younger child age, being in a 2-parent household, not regularly running out of money, living in an uncrowded home, and not moving multiple times in 1 year were associated with increased odds of better child functioning. These findings should be considered when developing prevention policies and programs aimed at reducing child maltreatment in Canada. There especially seems to be an opportunity to intervene with young children before they are exposed to maltreatment and develop functional impairment, including mental illness across the lifespan. Prevention activities, such as the Nurse Family Partnership, adapted for Canada, is presently being evaluated in a randomized controlled trial in British Columbia.24 More work is warranted to determine if providing particular resources for single-parent families, financial counselling, and facilitating adequate and stable housing for families with child maltreatment histories or at risk for child maltreatment could be effective for improving child functional outcomes.
Acknowledgements
Preparation of this article was supported by a Canadian Institutes of Health Research New Investigator Award (Dr Afifi), a Research Manitoba (formerly Manitoba Health Research Council) Operating and Establishment Award (Dr Afifi), and a Research Manitoba Chair Award (Dr Sareen). Data used in this publication are from the Canadian Incidence Study of Reported Child Abuse and Neglect, and are used with the permission of the Public Health Agency of Canada. The study was funded by the federal, provincial, and territorial governments of Canada; the Social Sciences and Humanities Research Council of Canada; and the Canadian Foundation for Innovation. The analyses and interpretations presented in this work do not necessarily reflect the opinions of the above-mentioned sponsors. The authors thank Sarah Turner for her assistance with the preparation of this manuscript.
References 1. Afifi TO, MacMillan HL, Boyle M, et al. Child abuse and mental disorders in Canada. CMAJ. 2014;186:E324–E332. 2. Afifi TO, Brownridge DA, Cox BJ, et al. Physical punishment, childhood abuse, and psychiatric disorders. Child Abuse Negl. 2006;30(10):1093–1103. 3. Afifi TO, Boman J, Fleisher W, et al. The relationship between child abuse, parental divorce, and lifetime mental disorders and suicidality in a nationally representative adult sample. Child Abuse Negl. 2009;33:139–147. 4. Afifi TO, Brownridge DA, MacMillan H, et al. The relationship of gambling to intimate partner violence and child maltreatment in a nationally representative sample. J Psychiatr Res. 2010;44:331–337. 5. Afifi TO, Mather A, Boman J, et al. Childhood adversity and personality disorders: results from a nationally representative population-based study. J Psychiatr Res. 2011;45:814–822. 6. Afifi TO, Mota NP, Dasiewicz P, et al. Physical punishment and mental disorders: results from a nationally representative US sample. Pediatrics. 2012;130:184–192. www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008 7. Kessler RC, Davis CG, Kendler KS. Childhood adversity and adult psychiatric disorders in the US National Comorbidity Survey. Psychol Med. 1997;27(5):1101–1119.
16. Rhodes AE, Boyle MH, Bethell J, et al. Child maltreatment and onset of emergency department presentations for suicide-related behaviors. Child Abuse Negl. 2012;36:542–551.
8. MacMillan HL, Boyle MH, Wong MYY, et al. Slapping and spanking in childhood and its association with lifetime prevalence of psychiatric disorders in a general population sample. CMAJ. 1999;161(7):805–809.
17. Katz LY, Au W, Singal D, et al. Suicide and suicide attempts in children and adolescents in the child welfare system. CMAJ. 2011;183:1977–1981.
9. MacMillan HL, Fleming JE, Streiner DL, et al. Childhood abuse and lifetime psychopathology in a community sample. Am J Psychiatry. 2001;158(11):1878–1883. 10. Scott KM, Smith DR, Ellis PM. Prospectively ascertained child maltreatment and its association with DSM-IV mental disorders in young adults. Arch Gen Psychiatry. 2010;67:712–719.
18. Tonmyr L, Ouimet C, Ugnat A-M. A review of findings from the Canadian Incidence Study of Reported Child Abuse and Neglect (CIS). Can J Public Health. 2012;103:103–112. 19. Tonmyr L, De Marco R, Hovdestad WE, et al. Policy makers’ perspectives on the utility of a national study of child maltreatment. Child Maltreat. 2004;9:304–308.
11. Afifi TO, Mota NP, MacMillan HL, et al. Harsh physical punishment in childhood and adult physical health. Pediatrics. 2013;132(2):e333–e340.
20. Public Health Agency of Canada. Canadian Incidence Study of Reported Child Abuse and Neglect—2008 (CIS-2008): user’s manual and codebook. Ottawa (ON): Injury and Child Maltreatment Section, Centre for Chronic Disease Prevention and Control; 2008.
12. Chartier MJ, Walker JR, Naimark B. Childhood abuse, adult health, and health care utilization: results from a representative community sample. Am J Epidemiol. 2007;165:1031–1038.
21. Knoke D, Trocmé N, MacLaurin B, et al. Reliability of the Canadian Incidence Study data collection instrument. Can J Program Eval. 2009;23:87–112.
13. Tonmyr L, Jamieson E, Mery L, et al. The relationship between childhood adverse experiences and disability due to physical health problems in a community sample of women. Women Health. 2005;41:23–35.
22. English DJ, Graham JC, Litrownik AJ, et al. Defining maltreatment chronicity: are there differences in child outcomes? Child Abuse Negl. 2005;29:575–595.
14. Afifi TO, Enns MW, Cox BJ, et al. Child abuse and health-related quality of life. J Nerv Ment Dis. 2007;195:797–804.
23. Kaplow JB, Widom CS. Age of onset of child maltreatment predicts long-term mental health outcomes. J Abnorm Psychol. 2007;116:176–187.
15. Enns MW, Cox BJ, Afifi TO, et al. Childhood adversities and risk for suicidal ideation and attempts: a longitudinal population-based study. Psychol Med. 2006;32:1769–1778.
24. Jack SM, Busser D, Sheehan D, et al. Adaption and implementation of the nurse–family partnership in Canada. Can J Public Health. 2012;103:542–548.
CPA
Membership Matters ●
Leading-edge professional development to support improved quality of care
APC
Nos membres, notre force ●
Perfectionnement professionnel de pointe pour soutenir une meilleure qualité de soins
A three-day annual conference
Full-day educational programs through the CPA CPD Institute
Un congrès annuel d’une durée de trois jours
Des programmes de formation d’une journée organisés par l’Institut APC PPC
●
National voice that speaks for psychiatry and mental health
●
Celebrates member excellence with awards, grants and fellowship designations
●
Une voix nationale porte-parole de la psychiatrie et de la santé mentale ● Célèbre l’excellence des membres par des prix, des bourses et des titres de Fellow.
INFORMATION: [email protected] www.cpa-apc.org
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 323
Original Research
Substantiated Reports of Child Maltreatment From the Canadian Incidence Study of Reported Child Abuse and Neglect 2008: Examining Child and Household Characteristics and Child Functional Impairment Tracie O Afifi, PhD1; Tamara Taillieu, MSc2; Kristene Cheung, BA3; Laurence Y Katz, MD, FRCPC4; Lil Tonmyr, PhD5; Jitender Sareen, MD, FRCPC6 1
Associate Professor, Departments of Community Health Sciences and Psychiatry, University of Manitoba, Winnipeg, Manitoba. Correspondence: S113 Medical Services Building, 750 Bannatyne Avenue, Winnipeg, MB R3E 0W3; [email protected].
2
Graduate Student, Department of Applied Health Sciences, University of Manitoba, Winnipeg, Manitoba. Graduate Student, Department of Psychology, University of Manitoba, Winnipeg, Manitoba.
3
Professor, Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba.
4
Senior Research Advisor, Health Surveillance and Epidemiology Division, Public Health Agency of Canada, Ottawa, Ontario.
5
Professor, Departments of Psychiatry, Psychology, and Community Health Sciences, University of Manitoba, Winnipeg, Manitoba.
6
Key Words: child abuse, child neglect, child welfare, suicidal ideation, self-injurious behaviour, depression, anxiety, alcohol abuse, drug abuse, single parent
Objective: Identifying child and household characteristics that are associated with specific child maltreatment types and child functional impairment are important for informing prevention and intervention efforts. Our objectives were to examine the distribution of several child and household characteristics among substantiated child maltreatment types in Canada; to determine if a specific child maltreatment type relative to all other types was associated with increased odds of child functional impairment; and to determine which child and household characteristics were associated with child functional impairment.
Received June 2014, revised, and accepted September 2014.
Method: Data were from the Canadian Incidence Study of Reported Child Abuse and Neglect (collection 2008) from 112 child welfare sites across Canada (n = 6163 children).
open access
Results: Physical abuse, sexual abuse, and emotional maltreatment were highly prevalent among children aged 10 to 15 years. For single types of child maltreatment, the highest prevalence of single-parent homes (50.6%), social assistance (43.0%), running out of money regularly (30.7%), and unsafe housing (30.9%) were reported for substantiated cases of neglect. Being male, older age, living in a single-parent home, household running out of money, moving 2 or more times in the past year, and household overcrowding were associated with increased odds of child functional impairment. Conclusions: More work is warranted to determine if providing particular resources for single-parent families, financial counselling, and facilitating adequate and stable housing for families with child maltreatment histories or at risk for child maltreatment could be effective for improving child functional outcomes. WWW
Rapports documentés de maltraitance d’enfants tirés de l’Étude canadienne sur l’incidence des signalements de cas de violence et de négligence envers les enfants de 2008 : examen des caractéristiques de l’enfant et du ménage, et de la déficience fonctionnelle de l’enfant Objectif : Identifier les caractéristiques de l’enfant et du ménage qui sont associées avec les types spécifiques de maltraitance d’enfants et la déficience fonctionnelle de l’enfant est important pour éclairer les efforts de prévention et d’intervention. Nos objectifs étaient d’examiner la répartition de plusieurs caractéristiques des enfants et des ménages parmi les types documentés de maltraitance d’enfants au Canada; déterminer si un type spécifique de maltraitance, relativement à tous les autres types, était associé à des probabilités accrues de déficience fonctionnelle de l’enfant; et déterminer les caractéristiques de l’enfant et du ménage qui étaient associées à la déficience fonctionnelle de l’enfant.
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 315
Original Research
Méthode : Les données de l’Étude canadienne sur l’incidence des signalements de cas de violence et de négligence envers les enfants (2008) obtenues de 112 centres de protection de l’enfance du Canada (n = 6163 enfants). Résultats : La violence physique, l’abus sexuel, et la maltraitance émotionnelle étaient hautement prévalents chez les enfants de 10 à 15 ans. Pour les types uniques de maltraitance d’enfants, la prévalence la plus élevée de foyers monoparentaux (50,6 %), d’aide sociale (43,0 %), de manque d’argent constant (30,7 %) et de logement non sécuritaire (30,9 %) a été signalée pour les cas de négligence documentés. Le sexe masculin, l’âge, un foyer monoparental, un ménage manquant d’argent, 2 déménagements ou plus l’année précédente, et un ménage surpeuplé étaient associés à des probabilités accrues de déficience fonctionnelle de l’enfant. Conclusions : Il faut plus de travail pour déterminer si l’offre de ressources particulières aux familles monoparentales, de conseils financiers, et d’aide au logement adéquat et stable aux familles qui ont des antécédents ou qui sont à risque de maltraitance d’enfants pourrait améliorer efficacement les résultats fonctionnels des enfants.
C
hild maltreatment includes abuse and neglect (that is, physical abuse, sexual abuse, neglect, emotional maltreatment, and exposure to IPV). Nationally representative Canadian data have indicated that 32% of the adult population in Canada has experienced child abuse, including physical abuse, sexual abuse, and exposure to IPV.1 Research has shown that childhood maltreatment is associated with mental disorders,2–10 physical health conditions,11–13 decreased quality of life,14 and suicidal behaviour15–17 that extends into adulthood. From a public health perspective, it is important to identify the child and household characteristics that are associated with specific types of child maltreatment and child functional impairment. This information can be used to develop and examine programs and policies aimed at protecting children and preventing maltreatment and associated impairment, such as mental illness from occurring.
The CIS is a rich data source that provides a unique opportunity to examine reported cases of child maltreatment within the Canadian context at a national level. Recently, a systematic review of the published studies examining CIS data indicated that child characteristics were associated with an increased likelihood of certain types of substantiated maltreatment.18 More specifically, older age of the child was associated with increased odds of substantiated physical abuse and sexual abuse, but not with neglect.19 Female sex was associated with increased odds of substantiated sexual abuse, decreased odds of physical abuse, but not with neglect.18 However, what is currently unknown is how a child’s age and sex is associated with all single types of substantiated child maltreatment cases and having multiple types of substantiated
Abbreviations AOR
adjusted odds ratio
CIS
Canadian Incidence Study of Reported Child Abuse and Neglect
IPV
intimate partner violence
316 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
Clinical Implications •
Knowledge about child and household characteristics can assist clinicians to identify people with greater odds of child functional impairment after experiencing child maltreatment.
•
Child characteristics associated with increased odds of functional impairment included being male and increasing age of the child.
•
Household characteristics associated with increased odds of child functional impairment included being in a single-parent home, being in a household that regularly runs out of money, household overcrowding, and moving 2 or more times in the past year.
Limitations •
Data are cross-sectional in nature.
•
Data are not representative of all child maltreatment in Canada.
•
Some types of functional impairment may have been underreported owing to lack of knowledge of the child welfare worker.
maltreatment. Additionally, the relations between household characteristics (for example, unsafe housing, number of children, or single parenting) and specific types of substantiated child maltreatment have not yet been examined. Child and household characteristics could also be useful beyond identifying increased likelihood of child maltreatment and may also provide insight into factors that may be associated with child functional impairment among children exposed to maltreatment. Previous research examining child maltreatment in Canada has provided some information on how child and household characteristics may be related to maltreatment and functional impairment. However, many gaps in our knowledge remain, which limits the ability to develop prevention and intervention strategies using evidencebased research. A more in-depth examination of child and household characteristics of families reported to child welfare is an approach to further knowledge of child www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
maltreatment within the Canadian context, which can have direct implications for intervention strategies within Canada and in other countries. Our study addresses existing gaps in our knowledge of reported child maltreatment to child welfare in Canada. The 3 main objectives of our study were to examine the distribution of several child and household characteristics among each specific single substantiated type of child maltreatment and among those with multiple types of substantiated maltreatment in Canada; to determine if a specific child maltreatment type relative to all other types of child maltreatment was associated with increased odds of child functional impairment; and to determine which child and household characteristics were associated with child functional impairment. It was hypothesized that the distribution of child and household characteristics would vary according to the specific child maltreatment type; all types of child maltreatment would be associated with increased odds of child functional impairment; and that all child and household characteristics would be associated with increased odds of child functional impairment.
Method Sample
The CIS is a child health surveillance activity of the Public Health Agency of Canada and is the product of collaboration among provincial and territorial governments, researchers, child welfare service providers, and policy makers from across the country.20 The CIS provides a nationwide sample of reports of child maltreatment to child welfare agencies across Canada. Data for our study came from the third and most recent cycle of the CIS, collected between October and December 2008 (CIS-2008). We used the CIS-2008 investigations of children aged 15 years and younger from 112 child welfare sites across Canada. Child welfare workers were provided a half-day training session and research assistants were available to respond to questions throughout the data collection period. At the end of the initial investigation, child welfare workers filled out a standardized questionnaire regarding investigation-related characteristics, including child and family or household characteristics, maltreatment variables, service dispositions, and child functional impairment. Child welfare workers determined the level of substantiation for each type of reported maltreatment: unfounded (evidence indicates this form has not occurred), suspected (not enough evidence to substantiate, but cannot be ruled out), and substantiated (evidence indicates this form has occurred). Up to 3 types of maltreatment could be substantiated per investigation. For our study, we were interested in examining single types of substantiated child maltreatment and multiple types of substantiated maltreatment (n = 6163 from 4095 different households, with 4974 single types of substantiated maltreatment cases and 1189 cases with more than 1 type of maltreatment). www.TheCJP.ca
Measures Child Maltreatment The CIS-2008 assessed 5 different types of child maltreatment: physical abuse, sexual abuse, neglect, emotional maltreatment, and exposure to IPV. Physical abuse included the following acts: shake, push, grab, throw; hit with hand; punch, kick, bite; hit with object; choke, poison, stab; and other physical abuse. Sexual abuse included the following acts: penetration, attempted penetration, oral sex, fondling, sex talk or images, voyeurism, exhibitionism, and other sexual abuse. Neglect included the following: failure to supervise, leading to physical harm; failure to supervise, leading to sexual abuse; permit criminal behaviour; physical neglect; medical or dental neglect; failure to provide psychological treatment; abandonment; and educational neglect. Emotional maltreatment included the following: terrorize, threat of violence; verbal abuse, belittling; isolation, confinement; inadequate nurturing; exploit, corrupting behaviour; and exposure to nonpartner violence. Exposure to IPV included the following: direct witness to violence; indirect exposure to violence; and exposure to emotional violence. Child and Household Characteristics Child demographic characteristics included sex (male or female) and age (in years). Household characteristics included single-parent home (yes or no), total number of children younger than 20 years old living in the home (1, 2, 3, or 4 or more), social assistance (yes or no), household regularly runs out of money for basic necessities (yes or no), household overcrowding (yes or no), number of moves in the past year (0, 1, 2, or 3 or more), and unsafe housing (yes or no). The social assistance variable was based on the household’s main income source. In our study, social assistance or no income was compared with employed full-time, part-time, multiple jobs, seasonal, employment insurance benefits, or other benefits. Unsafe housing included guns or weapons in the home; needles, legal, or illegal drugs in home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial solvents or chemicals in home; poisons, fire implements, or electrical hazards in the home; or quality of the home presents a health risk to the child. A positive response to any of the aforementioned items was coded as living in an unsafe home. Child Functional Impairment Child welfare workers assessed numerous different child functional impairments at the end of the initial investigation. Functional impairments included the following: depression, anxiety, or withdrawal; suicidal thoughts; self-harming behaviour; attention-deficit disorder, attention-deficit hyperactivity disorder; attachment issues; aggression; running away (multiple incidents); inappropriate sexual behaviour; Youth Criminal Justice Act involvement; intellectual or developmental disability; failure to meet developmental milestones; academic difficulties; fetal alcohol syndrome or fetal alcohol effects; positive The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 317
Original Research
toxicology at birth; physical disability; alcohol abuse; drug or solvent abuse; and any other functioning conditions. A full description of the variables can be found elsewhere.20 The assessment of variables differed; for instance, depression, anxiety, or withdrawal were assessed by feelings of anxiety and (or) depression that persist for most of every day for 2 weeks or longer, and interfere with the child’s ability to manage at home and at school. Suicidal thoughts were assessed as follows: the child has expressed thoughts of suicide, ranging from fleeting thoughts to a detailed plan. Owing to the nature of certain functional impairments (that is, attention-deficit disorder or attention-deficit hyperactivity disorder; intellectual or developmental disability; failure to meet developmental milestones; fetal alcohol syndrome or fetal alcohol effects; and positive toxicology at birth), no referent period was specified in the questionnaire. For all other functional impairments, the referent period was the past 6 months. Child welfare workers documented the presence or absence of each type of child functional impairment using the following categories: no (the child welfare worker believes there is no problem), suspected (condition may be present, but has not been confirmed), confirmed (disclosed by caregiver, diagnosed, or observed), or unknown (unsure or have not attempted to determine its presence). In our study, children were coded as having a functional impairment if the child welfare worker suspected or confirmed that one or more specific functional impairment was present. Investigated children were coded as no functional impairment if the child welfare worker documented no evidence of any assessed child functional impairment. Children with an unknown status of functional impairment were not included in the analysis, as it was not possible to determine with certainty to which group they would belong. Most of the child functioning issues have excellent test–retest reliability, with kappas showing excellent-to-moderate agreement (0.81–0.46) during a 3-week period.21
Statistical Analyses
First, cross-tabulations and chi-square tests of significance were computed to determine the distribution of child and household characteristics by substantiated child maltreatment type. Owing to the lack of independence of the observations and the positive correlation expected for children from the same household, a generalized mixed model with a binomial response were applied to the data. The generalized mixed model comprised fixed effects (that is, independent variable effects) as well as a random effect (that is, household) to account for its additional variation (that is, extra binomial variation). The first set of generalized mixed models were computed to examine the relation between child maltreatment type and child functional impairment. The models were first computed unadjusted and then adjusting for child age and sex. The second set of generalized mixed models were computed to determine if specific child and household characteristics were associated with child functional impairment. To determine if the relation between 318 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
each specific child and household characteristic was related to functional impairment, independent of other child and household characteristics, we ran the first models, adjusted, for all other child and household characteristics (AOR-1). These models were then further adjusted for types of child maltreatment (AOR-2).
Results
Table 1 presents the descriptive statistics for child and household characteristics by substantiated child maltreatment type. The findings indicate that, with the exception of a higher prevalence of sexual abuse among females, the prevalence of all other types of substantiated child maltreatment were about equal for males and females. Physical abuse, sexual abuse, and emotional maltreatment were more prevalent among older children, aged 10 to 15 years. The highest prevalence of single-parent homes were reported for substantiated cases of neglect (50.6%). The highest prevalence of unsafe housing and regularly running out of money regularly were reported for substantiated cases of neglect (30.9% and 30.7%, respectively) and more than 1 type of maltreatment (27.2% and 34.7%, respectively). In the sample, child welfare workers documented no evidence of functional impairments in 45.6% (n = 2429) of the cases of substantiated maltreatment and documented the presence of at least 1 suspected and (or) confirmed functional impairment in 54.4% (n = 2894) of the cases. In the sample, the mean number of functional impairments was 1.16 (SD 1.86). Table 2 presents the descriptive statistics and odds ratios from the generalized mixed models for the relations between types of child maltreatment and child functional impairment. The results indicated that physical abuse, sexual abuse, neglect, and more than one type of child maltreatment, compared with all other single types of child maltreatment, were associated with increased odds of child functional impairment. These findings remained significant after adjusting for child age and sex. Emotional maltreatment relative to all other single types of child maltreatment was not associated with increased odds of child functional impairment in the unadjusted and adjusted models. Exposure to IPV, compared with all other single types of child maltreatment, was associated with decreased odds of child functional impairment in the unadjusted and adjusted models. Table 3 presents the descriptive statistics and the odds ratios from the generalized mixed model for the relations between child and household characteristics and child functional impairment. The findings indicated that males relative to females, older children (ages 5 to 15 years) relative to younger children (ages 0 to 4 years), single-parent homes, household overcrowding, 2 or more moves in the past year, and regularly running out of money were all associated with increased odds of child functional impairment after adjusting for all child and household characteristics and for type of child maltreatment. Unsafe housing was associated with increased odds of child functional impairment in models adjusting for other child and household characteristics, but was no longer significant after further www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
Table 1 Child and household characteristics by substantiated maltreatment type, n (%) Substantiated maltreatment type Physical abuse
Characteristic
Sexual abuse
Neglect
Emotional maltreatment
IPV exposure
More than 1 type
Total
Child sex
53.648, 5
Female
413 (47.1)
125 (75.8)
833 (47.0)
200 (52.1)
885 (49.8)
572 (48.1)
3028 (49.1)
Male
463 (52.9)
40 (24.2)
940 (53.0)
184 (47.9)
891 (50.2)
617 (51.9)
3135 (50.9)
Child age, years
379.447, 10
0 to 4
104 (11.9)
19 (11.5)
627 (35.4)
97 (25.3)
763 (43.0)
336 (28.3)
1946 (31.6)
5 to 9
310 (35.4)
63 (38.2)
528 (29.8)
101 (26.3)
561 (31.6)
381 (32.0)
1944 (31.5)
10 to 15
462 (52.7)
83 (50.3)
618 (34.9)
186 (48.4)
452 (25.5)
472 (39.7)
2273 (36.9)
Single-parent home
206.540, 5
No
620 (70.8)
97 (58.8)
875 (49.4)
231 (60.2)
1255 (70.7)
728 (61.2)
3806 (61.8)
Yes
256 (29.2)
68 (41.2)
898 (50.6)
153 (39.8)
521 (29.3)
461 (38.8)
2357 (38.2)
40 (24.2)
464 (26.2)
94 (24.5)
392 (22.1)
250 (21.0)
1450 (23.5)
Total children in home 1
χ2, df All P ≤ 0.001
210 (24.0)
58.940, 15
2
366 (41.8)
68 (41.2)
554 (31.2)
142 (37.0)
676 (38.1)
414 (34.8)
2220 (36.0)
3
172 (19.6)
35 (21.2)
379 (21.4)
84 (21.9)
401 (22.6)
285 (24.0)
1356 (22.0)
≥4
128 (14.6)
22 (13.3)
376 (21.2)
64 (16.7)
307 (17.3)
240 (20.2)
1137 (18.4)
Social assistancea
239.094, 5
No
704 (85.5)
115 (74.2)
944 (57.0)
272 (73.9)
1251 (73.3)
773 (67.8)
4059 (69.4)
Yes
119 (14.5)
40 (25.8)
711 (43.0)
96 (26.1)
455 (26.7)
367 (32.2)
1788 (30.6)
Household regularly runs out of money
391.848, 5
No
710 (94.0)
131 (92.9)
955 (69.3)
305 (89.7)
1316 (86.8)
639 (65.3)
4056 (79.4)
Yes
45 (6.0)
10 (7.1)
423 (30.7)
35 (10.3)
201 (13.2)
340 (34.7)
1054 (20.6)
Household overcrowding
85.116, 5
No
811 (94.4)
147 (94.8)
1467 (86.7)
335 (87.9)
1621 (93.6)
994 (86.7)
5375 (90.1)
Yes
48 (5.6)
8 (5.2)
226 (13.3)
46 (12.1)
111 (6.4)
153 (13.3)
592 (9.9)
Number of moves in past year
133.789, 15
0
499 (73.0)
98 (73.1)
734 (54.9)
222 (66.1)
884 (61.2)
570 (58.9)
3007 (61.3)
1
154 (22.5)
21 (15.7)
355 (26.6)
81 (24.1)
393 (27.2)
227 (23.5)
1231 (25.1)
2
17 (2.5)
9 (6.7)
130 (9.7)
23 (6.8)
107 (7.4)
99 (10.2)
385 (7.9)
≥3
14 (2.0)
6 (4.5)
118 (8.8)
10 (3.0)
61 (4.2)
72 (7.4)
281 (5.7)
Unsafe housingb
465.321, 5
No
697 (95.9)
110 (90.9)
954 (69.1)
288 (89.4)
1390 (93.3)
696 (72.8)
4135 (82.8)
Yes
30 (4.1)
11 (9.1)
426 (30.9)
34 (10.6)
100 (6.7)
260 (27.2)
861 (17.2)
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. Owing to rounding, the percentages may not total to 100%. a
Social assistance or no income, compared with employed full- or part-time, multiple jobs, seasonal, employment insurance, or other benefits.
b
Unsafe housing includes the following: guns and (or) weapons in the home; needles, legal and (or) illegal drugs in the home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial chemicals or solvents in the home; poisons, fire implements, or electrical hazards in the home; or quality of the home poses health risk to a child.
IPV = intimate partner violence
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 319
Original Research
Table 2 Substantiated child maltreatment type and child functional impairment accounting for the positive correlation among family members as a random effect in a generalized mixed model Child functional impairment No impairment n (%)
Impairment n (%)
OR (95% CI)
Physical abuse, compared with all other single types
297 (37.1)
504 (62.9)
Sexual abuse, compared with all other single types
49 (33.6)
Neglect, compared with all other single types
Type of maltreatment
AOR (95% CI)
1.85 (1.54–2.22)
a
1.29 (1.06–1.57)b
97 (66.4)
1.86 (1.26–2.76)
b
1.63 (1.08–2.45)c
563 (37.8)
928 (62.2)
2.01 (1.73–2.34)a
2.27 (1.93–2.68)a
Emotional maltreatment, compared with all other single types
150 (43.2)
197 (56.8)
1.27 (0.98–1.66)
d
1.15 (0.87–1.52)e
Exposure to IPV, compared with all other single types
1015 (68.9)
458 (31.1)
0.26 (0.22–0.30)a
0.31 (0.26–0.37)a
More than 1 type, compared with all other single types
355 (33.3)
710 (66.7)
1.91 (1.61–2.26)
1.91 (1.59–2.28)a
a
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. For cases where a single type of maltreatment was substantiated, odds ratios compare the odds of functional impairment for each individual type of maltreatment relative to the other 4 types combined (for example, physical abuse relative to sexual, neglect, emotional, and exposure combined). For cases where more than 1 type of maltreatment was substantiated, the odds ratios compare the odds of functional impairment for more than 1 type relative to the other 5 types combined. a
P ≤ 0.001; b P ≤ 0.01; c P ≤ 0.05; d P = 0.70; e P = 0.33
AOR = adjusted odds ratio; IPV = intimate partner violence
adjustment for type of maltreatment. Increasing number of children in the household decreased the odds of child functional impairment. Social assistance was not associated with child functional impairment in the adjusted models. The estimated variance (0.74) and standard error (0.11) of the household intercept (that is, random effects) in the fully adjusted model was significant at the P < 0.001 level, confirming that observations from children from the same household were positively correlated.
Discussion
Novel findings from this research indicate that the distribution of child and household characteristics vary according to the specific child maltreatment type. For example, the prevalence of physical abuse, sexual abuse, and emotional maltreatment increased with older age of the child. However, the prevalence of neglect remained fairly consistent across all 3 age categories. Notably, high prevalence of single-parent homes, social assistance, regularly running out of money, and unsafe housing were found for substantiated cases of neglect. High prevalence of household overcrowding was noted among cases of neglect and emotional maltreatment. In addition, the prevalence of regularly running out of money, household overcrowding, and unsafe housing was the highest in families where children experienced more than one type of child maltreatment, compared with some of the other single child maltreatment types. Descriptively, these data provide insight into child and household vulnerabilities that may be more likely among families at risk for child maltreatment and that certain household characteristics, as noted above, may be more common among certain child maltreatment types. Other main findings from this research indicate that in adjusted models, physical abuse, sexual abuse, neglect, and more than one type of abuse relative to all other types of substantiated maltreatment combined were associated with 320 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
increased odds of child functional impairment. Exposure to IPV relative to all other single types of child maltreatment combined was associated with decreased odds of child functional impairment. Child characteristics associated with increased odds of functional impairment included being a male and being older. Household characteristics associated with increased odds of child functional impairment included being in a single-parent home, household overcrowding, moving 2 or more times in the past year, and being in a household that regularly runs out of money. Greater number of children in the home was associated with significantly decreased odds of child functional impairment. The finding that exposure to IPV is related to decreased odds of child functional impairment relative to all other types of child maltreatment needs to be considered in the context of the study design. When interpreting this finding, it is important to note that the comparison is exposure to IPV, compared with other single types of substantiated child maltreatment, and not that exposure to IPV is protective against child functional impairment. It may be that exposure to IPV may not have the same association with child functional impairment relative to other types of child maltreatment that are directed specifically at the child. Regardless, it would be incorrect to interpret this finding as exposure to IPV is protective against child functional impairment as the comparison group was other child maltreatment and not a no child maltreatment group. Male sex was associated with increased likelihood of functional impairment. When interpreting this finding, it seems that understanding the functional impairments assessed in the study are important. The types of functional impairments that are often more common among males are more externalizing in nature and likely more readily assessed by the child welfare worker. It may not be that maltreated males, compared with maltreated females, inherently have www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008
Table 3 Child and household characteristics by child functional impairment among substantiated reports of child maltreatment accounting for the positive correlation among family members as a random effect in a generalized mixed model Child functional impairment
Impairment
Impairment n (%)
No impairment n (%)
AOR-1 (95% CI)
AOR-2 (95% CI)
Female
1268 (49.1)
1317 (50.9)
1.00
1.00
Male
1626 (59.4)
1112 (40.6)
1.76 (1.50–2.07)
0 to 4
492 (31.9)
5 to 9
941 (55.1)
10 to 15
Child or household characteristic Child sex
a
1.83 (1.55–2.16)a
1052 (68.1)
1.00
1.00
767 (44.9)
3.35 (2.69–4.16)a
3.11 (2.49–3.88)a
1461 (70.5)
610 (29.5)
7.67 (6.15–9.56)a
6.51 (5.19–8.17)a
No
1673 (50.5)
1637 (49.5)
1.00
1.00
Yes
1221 (60.7)
792 (39.3)
1.34 (1.11–1.62)b
1.22 (1.01–1.49)c
1
729 (57.3)
543 (42.7)
1.00
1.00
2
1065 (54.8)
880 (45.2)
0.80 (0.64–1.00)d
0.84 (0.67–1.06)e
3
582 (49.4)
595 (50.6)
0.59 (0.46–0.76)a
0.59 (0.45–0.76)a
518 (55.8)
411 (44.2)
0.73 (0.55–0.98)c
0.74 (0.55–1.00)f
No
1915 (53.0)
1701 (47.0)
1.00
1.00
Yes
860 (57.9)
625 (42.1)
1.13 (0.90–1.41)
No
1832 (49.7)
Yes
611 (66.3)
No
2487 (52.9)
2215 (47.1)
1.00
1.00
Yes
320 (65.4)
169 (34.6)
1.47 (1.05–2.05)c
1.43 (1.03–2.01)c
0
1482 (54.3)
1249 (45.7)
1.00
1.00
1
559 (51.7)
2 ≥3
Child age, years
Single-parent home
Total children in home
≥4 Social assistance
d
g
1.09 (0.87–1.37)h
1853 (50.3)
1.00
1.00
310 (33.7)
1.78 (1.38–2.29)a
1.67 (1.29–2.16)a
Household regularly runs out of money
Household overcrowding
Number of moves in past year 523 (48.3)
0.92 (0.75–1.13)
i
0.95 (0.77–1.17)j
222 (66.3)
113 (33.7)
1.47 (1.02–2.12)c
1.47 (1.02–2.13)c
161 (67.6)
77 (32.4)
1.62 (1.04–2.52)
c
1.63 (1.04–2.56)c
No
1844 (48.9)
1926 (51.1)
1.00
1.00
Yes
476 (64.9)
258 (35.1)
1.57 (1.21–2.04)a
1.23 (0.94–1.60)k
Unsafe housing
e
The cross-tabulations do not adjust for the extra variation owing to the clustering effects of children from the same family. No functional impairment is the reference category with an odds ratio of 1.00. a P ≤ 0.001; b P ≤ 0.01; c P ≤ 0.05; d P = 0.053; e P = 0.13; f P = 0.052; g P = 0.29; h P = 0.44; i P = 0.41; j P = 0.62; k P = 0.14 d
Social assistance or no income, compared with employed full-time, part-time, multiple jobs, seasonal, employment insurance, or other benefits.
e
Unsafe housing includes the following: guns and (or) weapons in the home; needles, legal and (or) illegal drugs in the home; home is a drug or narcotics laboratory, grow operation, or crack house; industrial chemicals or solvents in the home; poisons, fire implements, or electrical hazards in the home; or quality of the home poses a health risk to a child.
AOR-1 = adjusted odds ratio (adjusted for all other child and household characteristics in model); AOR-2 = adjusted odds ratio (adjusted for child and household characteristics and types of maltreatment)
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 321
Original Research
more functional impairment, but rather the many types of functional impairment assessed in our study may be both more common among males and easier for a child welfare worker to assess. Further research is needed on the specific functional impairment among males and females. Increasing age of the child was associated with more functional impairment in our study, which is contradictory to many studies that indicate that maltreatment occurring in the younger age categories is more damaging.22,23 However, it may not be that this finding is inconsistent, but rather assessing a different phenomenon. It may be that the older age of the child is an indicator of longer exposure of maltreatment, as maltreatment is likely to be ongoing in these reported substantiated cases. As well, older children are easier to interview, which may make the recognition of impairment easier. It may also be that it takes time for the impairment to develop. Further, many of the functional impairments assessed are somewhat age-specific and biased toward the older age categories (that is, academic difficulties, running away from home, suicidal thoughts, and alcohol or drug abuse). Therefore, it may be that functional impairment is more likely in the older age categories (5 to 9 years and 10 to 15 years) relative to the youngest age category (0 to 4 years) simply owing to the nature of the types of child functional impairment assessed. Single-parent homes, household overcrowding, and regularly running out of money were 3 household characteristics that were associated with increased child functional impairment. These characteristics may be indicators of limited resources that can have an impact on the health and well-being of a child. Interestingly, social assistance was not associated with child functional impairment. It may be that families involved with social service agencies receive some additional funds and may qualify for more services and opportunities to help families. The findings from our study should be considered in light of important limitations. First, the data from the study were cross-sectional in nature, which precludes inferences regarding causation. We are unable to conclude that the child maltreatment caused the child functional impairment. Second, the data collected were based on the knowledge of the child welfare worker at the time of the investigation. It is possible that some types of functional impairment may have been underreported owing to lack of knowledge of the child welfare worker. Also, some types of functional impairment (externalizing in nature) may have been easier to evaluate and report, compared with other types of functional impairment (internalizing in nature). Third, these data consisted of only reported cases of child maltreatment without a comparison group of children without child maltreatment histories. This should be considered in interpreting the findings. Finally, the data were reported cases of child maltreatment in Canada. Many incidents of child maltreatment go unreported and it is often the more severe and chronic cases that are reported to authorities. Therefore, these data are not representative of all child maltreatment in Canada. 322 W La Revue canadienne de psychiatrie, vol 60, no 7, juillet 2015
Our study provides insight into child and household factors that may be associated with child maltreatment reported to child welfare agencies and child functional impairment. Importantly, when thinking of our study finding, in relation to better child functioning, being female, younger child age, being in a 2-parent household, not regularly running out of money, living in an uncrowded home, and not moving multiple times in 1 year were associated with increased odds of better child functioning. These findings should be considered when developing prevention policies and programs aimed at reducing child maltreatment in Canada. There especially seems to be an opportunity to intervene with young children before they are exposed to maltreatment and develop functional impairment, including mental illness across the lifespan. Prevention activities, such as the Nurse Family Partnership, adapted for Canada, is presently being evaluated in a randomized controlled trial in British Columbia.24 More work is warranted to determine if providing particular resources for single-parent families, financial counselling, and facilitating adequate and stable housing for families with child maltreatment histories or at risk for child maltreatment could be effective for improving child functional outcomes.
Acknowledgements
Preparation of this article was supported by a Canadian Institutes of Health Research New Investigator Award (Dr Afifi), a Research Manitoba (formerly Manitoba Health Research Council) Operating and Establishment Award (Dr Afifi), and a Research Manitoba Chair Award (Dr Sareen). Data used in this publication are from the Canadian Incidence Study of Reported Child Abuse and Neglect, and are used with the permission of the Public Health Agency of Canada. The study was funded by the federal, provincial, and territorial governments of Canada; the Social Sciences and Humanities Research Council of Canada; and the Canadian Foundation for Innovation. The analyses and interpretations presented in this work do not necessarily reflect the opinions of the above-mentioned sponsors. The authors thank Sarah Turner for her assistance with the preparation of this manuscript.
References 1. Afifi TO, MacMillan HL, Boyle M, et al. Child abuse and mental disorders in Canada. CMAJ. 2014;186:E324–E332. 2. Afifi TO, Brownridge DA, Cox BJ, et al. Physical punishment, childhood abuse, and psychiatric disorders. Child Abuse Negl. 2006;30(10):1093–1103. 3. Afifi TO, Boman J, Fleisher W, et al. The relationship between child abuse, parental divorce, and lifetime mental disorders and suicidality in a nationally representative adult sample. Child Abuse Negl. 2009;33:139–147. 4. Afifi TO, Brownridge DA, MacMillan H, et al. The relationship of gambling to intimate partner violence and child maltreatment in a nationally representative sample. J Psychiatr Res. 2010;44:331–337. 5. Afifi TO, Mather A, Boman J, et al. Childhood adversity and personality disorders: results from a nationally representative population-based study. J Psychiatr Res. 2011;45:814–822. 6. Afifi TO, Mota NP, Dasiewicz P, et al. Physical punishment and mental disorders: results from a nationally representative US sample. Pediatrics. 2012;130:184–192. www.LaRCP.ca
Substantiated Reports of Child Maltreatment from the Canadian Incidence Study of Reported Child Abuse and Neglect 2008 7. Kessler RC, Davis CG, Kendler KS. Childhood adversity and adult psychiatric disorders in the US National Comorbidity Survey. Psychol Med. 1997;27(5):1101–1119.
16. Rhodes AE, Boyle MH, Bethell J, et al. Child maltreatment and onset of emergency department presentations for suicide-related behaviors. Child Abuse Negl. 2012;36:542–551.
8. MacMillan HL, Boyle MH, Wong MYY, et al. Slapping and spanking in childhood and its association with lifetime prevalence of psychiatric disorders in a general population sample. CMAJ. 1999;161(7):805–809.
17. Katz LY, Au W, Singal D, et al. Suicide and suicide attempts in children and adolescents in the child welfare system. CMAJ. 2011;183:1977–1981.
9. MacMillan HL, Fleming JE, Streiner DL, et al. Childhood abuse and lifetime psychopathology in a community sample. Am J Psychiatry. 2001;158(11):1878–1883. 10. Scott KM, Smith DR, Ellis PM. Prospectively ascertained child maltreatment and its association with DSM-IV mental disorders in young adults. Arch Gen Psychiatry. 2010;67:712–719.
18. Tonmyr L, Ouimet C, Ugnat A-M. A review of findings from the Canadian Incidence Study of Reported Child Abuse and Neglect (CIS). Can J Public Health. 2012;103:103–112. 19. Tonmyr L, De Marco R, Hovdestad WE, et al. Policy makers’ perspectives on the utility of a national study of child maltreatment. Child Maltreat. 2004;9:304–308.
11. Afifi TO, Mota NP, MacMillan HL, et al. Harsh physical punishment in childhood and adult physical health. Pediatrics. 2013;132(2):e333–e340.
20. Public Health Agency of Canada. Canadian Incidence Study of Reported Child Abuse and Neglect—2008 (CIS-2008): user’s manual and codebook. Ottawa (ON): Injury and Child Maltreatment Section, Centre for Chronic Disease Prevention and Control; 2008.
12. Chartier MJ, Walker JR, Naimark B. Childhood abuse, adult health, and health care utilization: results from a representative community sample. Am J Epidemiol. 2007;165:1031–1038.
21. Knoke D, Trocmé N, MacLaurin B, et al. Reliability of the Canadian Incidence Study data collection instrument. Can J Program Eval. 2009;23:87–112.
13. Tonmyr L, Jamieson E, Mery L, et al. The relationship between childhood adverse experiences and disability due to physical health problems in a community sample of women. Women Health. 2005;41:23–35.
22. English DJ, Graham JC, Litrownik AJ, et al. Defining maltreatment chronicity: are there differences in child outcomes? Child Abuse Negl. 2005;29:575–595.
14. Afifi TO, Enns MW, Cox BJ, et al. Child abuse and health-related quality of life. J Nerv Ment Dis. 2007;195:797–804.
23. Kaplow JB, Widom CS. Age of onset of child maltreatment predicts long-term mental health outcomes. J Abnorm Psychol. 2007;116:176–187.
15. Enns MW, Cox BJ, Afifi TO, et al. Childhood adversities and risk for suicidal ideation and attempts: a longitudinal population-based study. Psychol Med. 2006;32:1769–1778.
24. Jack SM, Busser D, Sheehan D, et al. Adaption and implementation of the nurse–family partnership in Canada. Can J Public Health. 2012;103:542–548.
CPA
Membership Matters ●
Leading-edge professional development to support improved quality of care
APC
Nos membres, notre force ●
Perfectionnement professionnel de pointe pour soutenir une meilleure qualité de soins
A three-day annual conference
Full-day educational programs through the CPA CPD Institute
Un congrès annuel d’une durée de trois jours
Des programmes de formation d’une journée organisés par l’Institut APC PPC
●
National voice that speaks for psychiatry and mental health
●
Celebrates member excellence with awards, grants and fellowship designations
●
Une voix nationale porte-parole de la psychiatrie et de la santé mentale ● Célèbre l’excellence des membres par des prix, des bourses et des titres de Fellow.
INFORMATION: [email protected] www.cpa-apc.org
www.TheCJP.ca
The Canadian Journal of Psychiatry, Vol 60, No 7, July 2015 W 323
