Embed Size (px)
Citation preview
Lessons learned from LONGSCANLessons learned from LONGSCAN
Presented by
Diana English, PhDChild Welfare Research Group
University of Washington, School of Social Work
Desmond K. Runyan, MD, DrPHUniversity of North Carolina at Chapel Hill
1
Acknowledgments• U.S. Department of Health and Human
Services • Administration for Children, Youth, and
Families • Children's Bureau • Office on Child Abuse and Neglect • National Institutes of Health • National Institute of Child Health and Human Development
2
LONGSCAN Background• 1990 NCCAN funded consortium• Focus on child abuse & neglect
Need for theory-based longitudinal studies Address CAUSES & CONSEQUENCES of abuse
and neglect Implications for…
• Preventing maltreatment (initial or reoccurrence)
• Preventing negative effects of maltreatment • Promote recovery
3
Overview of LONGSCAN• LONGitudinal Studies of Child Abuse & Neglect• 5 distinct studies (East, South, Midwest, Northwest , & Southwest) • Measurement & data coordinated at UNC
coordinating center • Common measures, coding, training, data entry• Consortium governance agreement• Committees for governance, measurement, analysis, and publications/dissemination• For more information, see Runyan et al. 1998
4
Current Status• Data collection on-going
- Youth now 14 - 22 years old• Data summarized & updated 4 times a year
- Data are distributed to sites twice a year • Data archived with the National Data Archive
on Child Abuse and Neglect (NDACAN)- Age 4, 6, 8, and 12 interviews- Contact interviews through age 11- Includes CPS record reviews
5
6
Data Collected (Baseline to age 18)
Interviews Baseline* 4 6 8 12 14+ 16+ 18+
Child -- 1166 1176 1074 895 872 721 452
Caregiver -- 1247 1225 1130 956 925 752 283
Child or Caregiver
1354 1250 1236 1140 976 947 789 472
7
Notes. * Baseline refers to data at age 4 or age 6. + Data collection on-going at age 14, 16, and 18.
Data Collected (Baseline to age 18)Interviews Baseline* 4 6 8 12 14+ 16+ 18+
Site
East 282 237 255 237 190 197 162 101
Midwest 245 223 225 216 181 173 91 --
South 243 221 222 190 177 176 130 174
Southwest 330 319 299 274 236 216 217 106
Northwest 254 250 235 223 192 185 189 91
8
Notes. * Baseline refers to data at age 4 or age 6. + Data collection on-going at age 14, 16, and 18.
Sample Demographics (Baseline to Age 16)
Child Demographics Baseline 4 6 8 12 14+ 16+
% Male 48.5 48.8 48.7 48.1 49.8 48.9 48.2
% Caucasian 26.2 27.1 25.7 25.6 25.4 25.5 26.6
% African American 53.3 51.8 53.9 54.5 54.7 55.5 53.1
% Other Race 20.5 21.1 20.4 19.9 19.9 19.0 20.3
Mean Age -- 4.6 6.4 8.3 12.4 14.4 16.3
9
Notes. * Baseline refers to data at age 4 or age 6. + Data collection on-going at age 14 and 16.
Caregiver Demographics (Baseline to Age 16)
Caregiver Demographics
4 6 8 12 14+ 16+
% Married 33.1 33.0 35.5 38.3 37.9 39.4
% Single 44.6 42.6 41.2 34.4 34.1 30.9
% Separated 7.9 8.0 8.6 8.1 8.2 7.6
% Divorced 12.8 13.7 11.8 15.0 15.2 16.2
% Widowed 1.6 2.8 2.9 4.2 4.6 5.9
Mean Income 15k-20k
15k-20k
20k-25k
25k-30k
25k-30k
30k-35k
Mean Education H. S. H. S. H. S. H. S. H. S. H. S.
10
Notes. + Data collection on-going at age 14 and 16.
Measurement• Guided by Social-Developmental-Ecological
Theory (NRC, 1993; Bronfenbrenner, 1989; Hawkins & Catalano, 1996).
• Domains assessed: Child/Youth: Characteristics, functioning Caregiver: Characteristic, functioning Family microsystem: Home environment, functioning Macrosystem: Neighborhood, school,
support
11
Measurement • Multiple sources & methods
• Reports/ratings/questionnaires (Child/Youth, Caregiver, and Teacher) Performance (Child/Youth) Situational tests/samples Official records (CPS)
• Presentation of measures Interview & Audio-Computer Assisted Self
Interview (A-CASI)
12
LONGSCAN Publications to Date
13
Note. Publications as of February ’09.
Exposure of LONGSCAN To Maltreatment
Maltreatment History• Expand definitions & methods• Look at maltreatment
Physical & sexual abuse, neglect, emotional, witnessed violence (home and community)
Beyond just the occurrence: yes/no Multidimensional (focus on degree or
extent - CAN special issue May, 2005) Beyond CPS report
• Age 12 youth self-report
15
# of Maltreatment Records/Referrals per Child (birth through age 14)
16
1 Record (16%)
2 Records (12%)
3 Records (9%)
4 Records (7%)5 Records (5%)
8-22 Records (12%)
7 Records (3%)
6 Records (4%)
Total N = 1354
Age of First Record/Referral (birth through age 14)
17
# o
f Chi
ldre
n
Age of First Record/Referral
Descriptive Stats (0-14):Total N of kids with record = 916Mean Age = 2.2 years (SD = 2.9)Range of Ages = 0 – 14.4Median Age = 1.2
# of Allegations by Maltreatment Type (birth through age 14)
18
# of
Alle
gatio
ns
Age
Based on Baseline Sample (N = 1354)
# of Substantiations by Maltreatment Type (birth through age 14)
19
# of
Sub
stan
tiatio
ns
Age
Descriptive Stats (0-14):Total # of substantiations (0-14) = 2282Total # of physical abuse substantiations (0-14) = 369Total # of sexual Abuse substantiations (0-14) = 99Total # of neglect substantiations (0-14) = 1456Total # of emotional abuse substantiations (0-14) = 358
Frequencies (0-14):49% have 1 or more substantiations14% have 1 or more physical abuse substantiations6% have 1 or more sexual abuse substantiations41% have 1 or more neglect substantiations17% have 1 or more emotional abuse substantiations
Based on Baseline Sample (N = 1354)
Findings: Maltreatment Dimensions – Substantiation Status• Are there outcome differences for children
reported & substantiated compared to those who are reported and NOT substantiated?
• No differences on 10 developmental, social, emotional and behavioral outcomes for substantiated vs. not substantiated (Hussey et al., 2005).
20
Maltreatment Allegations by Severity (birth through age 14)
Allegations
Level of Severity Low (percents of rows) High
Overall M (SD)
1 % (n)
2 % (n)
3 % (n)
4 % (n)
5 % (n)
Physical Abuse 33.3 (158) 26.2 (124) 31.0 (147) 8.2 (39) 1.3 (6) 2.2 (1.0)
Sexual Abuse 9.6 (21) 3.2 (7) 33.9 (74) 46.8 (102) 6.4 (14) 3.4 (1.0)
Neglect 8.6 (69) 7.5 (60) 22.6 (180) 30.8 (246) 30.3 (242) 3.7 (1.2)
Overall 5.9 (51) 7.5 (66) 24.2 (213) 33.0 (291) 29.4 (259) 3.7 (1.1)
21
Note. Only includes participants who have an allegation.
Findings: Maltreatment Dimensions - Severity• Severity construction:
Maximum severity within type Maximum severity across type Total or sum of maximum severity Mean severity
• Maximum severity by type best predictor of outcomes including adaptive functioning and anger (Litrownik et al., 2005).
22
Chronicity of Maltreatment(birth through age 14)
Situational: referrals occur in only onedevelopmental period
Limited Episodic: referrals occur in morethan one developmental period – but notconsecutive periods
Limited Continuous: referrals occur in ONLYtwo consecutive developmental periods
Extended Episodic: referrals occur in twoconsecutive periods with one interveningperiod where no referrals occurred, followedby a period when one or more referralsoccurredExtended Continuous: referrals occur in threeor more consecutive periods
23
34.5%
15.7%13.1%
20.3%
16.4%
N = 903
Findings: Maltreatment Dimensions - Chronicity• Developmental vs. Calendar to examine
consequences over time.• The developmental approach across a
child’s lifespan was the most sensitive – as well as looking at extent (across developmental stages) and continuity (whether there are any gaps) (English et al., 2005).
24
Findings: Maltreatment Dimensions -Type• Hierarchical Type – prioritized abuse vs. neglect• Severity/Frequency Type – based on highest
severity or frequency • Expanded Hierarchical – differentiated multiple
vs. single sub-types• All three types predicted child social/emotional
and behavioral functioning, however, differentiating multiple vs. single sub-types was the more outcomes (Lau et al., 2005).
25
Findings: Maltreatment Dimensions - Combined• Type, severity, chronicity and age at first
report.• Individual maltreatment dimensions have
distinct effects on child functioning.• Type (based on maximum severity of each
type) most consistent predictor across outcomes, however different types predicted different outcomes (English et al., 2005).
26
Findings: Maltreatment Dimensions - Combined• A comprehensive assessment of a child’s
maltreatment experience (not just the presenting incident) including type and severity.
• Age of onset• Pattern (extent and continuity)• All are important in understanding the outcome of
maltreatment on children’s growth and development.
(see Child Abuse & Neglect Volume 29: 2005)
27
Self Report of Abuse (birth through age 12)
28
* Indicators are NOT mutually exclusive
(N = 881) (N = 874) (N = 883)
Comparing Substantiations to Child Self-Report (birth through age 12)
Psychological Abuse(N = 883)
4.3%
18.6%
7.8%80.9%
14.2%
3.2%
51.2%
7.7%
33.0%
8.1%
29
Concordance between Child Self-Report and CPS Substantiations• Classifying children/youth with profiles
improves ability to identify antecedents and consequences of maltreatment: No PA/SA High PA, Low SA No PA/Moderate SA High PA, High SA (35% Youth did not endorse CPS reports)
• Agreement between CPS report and self-report not high (Everson et al., 2008).
30
Concordance between Child Self-Report and CPS Substantiations• A-CASI interview rates of abuse 4-6 X higher
than in CPS records (Everson et al., 2008). • Adolescent psychological adjustment
(measured by TSCC & YSR) more strongly associated with self-report than with CPS determination (Everson et al., 2008).
31
Discussion
• Implications of LONGSCAN findings on maltreatment for policy and practice
Foster Care in LONGSCAN
Foster Care: First 18 months• Examined number of placements during first 18
months following removal (N = 415) 1-15 placements : Mean = 4.23 placements Child behavior problems at time of initial
placement predicted subsequent # placements. # prior placements predicted later behavior
problems for those who did not have problems initially (Newton et al., 2000).
Long-Term Placement and Violence Exposure• Children placed in out of home care before 3.5
At age 6 Reunified children: Had more exposure to family violence Experienced more adverse life events Evidenced increased internalizing problems Received fewer mental health services However, felt less isolated
Parents of adoptive kids used more minor violence than foster parents for discipline
Adoptive kids report witnessing less violence in the home (Lau et al., 2003; Litrownik et al., 2003).
Long-Term Stability of Early Foster Care Sample• Examine caregiver stability for children age 6 to 8
who were removed prior to age 3.5 (N = 285). 1 in 7 unstable Adoption most stable (compared to reunified, kin
care, and non-kin care) Other predictors of stability: father involvement;
child intellectual functioning Predictors of instability: Child behavior;
expressive family (Randazzo et al., under review).
Foster Care – Discipline Practices and Child Aggression, Kin & Non-Kin• Kin use more harsh disciplinary
practice (N = 80).• Parents who use aggressive problem solving
strategies have more aggressive youth (DeRobertis & Litrownik, 2004).
Discussion
• Implications of foster care findings on policy/practice.
Family Context
• Caregiver violence exposure as a child/adult• Current and/or recent domestic
violence/family conflict• Caregiver depressive symptomatology• Parenting
39
Neglect – A Different Conceptualization• Definition of neglect as psychological and
physical safety and security.• Unsafe or dirty home & untreated
behavioral problems predicted language impairments & developmental problems (Dubowitz et al., 2005).
• A stimulating home environment predicted less impairment in cognitive development (Dubowitz et al., 2005).
40
Neglect – A Different Conceptualization• Multiple changes in residence predicted
externalizing behavior problems (Dubowitz et al., 2005).
• Exposure to verbally aggressive discipline predicted more behavior problems (Dubowitz et al., 2005).
• Conclusion: Important to conceptualize neglect in terms of child’s developmental needs (safety and security).
41
Neglect• Examined individual and cumulative
relationship among physical, psychological and environmental neglect and development (Age 3-5, N =136 - FTT & HIV Risk).
• Psychological neglect associated with increased internalizing/externalizing problems.
• Cumulative neglect index associated with internalizing problems
42
Influence of Early and Later Maltreatment on Childhood Aggression at ages 4, 6 & 8
Predictor Estimate (S.E.) t P
Early Neglect 1.29 (0.46) 2.80 <.01
Early Abuse 0.66 (0.68) 0.97 0.33
Later Neglect 0.14 (0.34) 0.42 0.68
Later Abuse 0.53 (0.39) 1.34 0.18
Notes. From Kotch et al., 2008.
N = 1,318; Model Chi-square = 858.49 (p < .0001).
Model includes child gender, child age, child race /ethnicity, caregiver marital status, caregiver education, income, caregiver depression, neighborhood safety, early neglect X early abuse, later neglect X later abuse, study site.
43
Influence of Early and Later Maltreatment on Childhood Aggression at ages 4, 6 & 8
• This study suggests that early neglect may be a more important precursor of youth violence than is physical abuse (Kotch et al., 2008).
• Limitations: not a representative probability sample the possibility of false negatives cannot be
ruled out site is a potential confounder
44
Caregiver’s History of Violence Exposures as a Child and/or Adult (age 4)
45
Of those assaulted (n = 608):
• 52% experienced some form of physical assault/abuse as a child or teenager.
• 44% experienced some form of sexual assault /abuse before age 13.
• 36% experienced some form of sexual assault/abuse as a teen.
• 75% experienced some form of physical assault as an adult.
• 22% experienced some form of sexual assault as an adult.Caregiver Report at Age 4 Interview
(N = 923)
Caregiver’s History of Violence Exposures as a Child and/or Adult• Maternal child victimization more powerful
indicator of child behavior problems at age 4 than caregiver adult victimization (Thompson, 2007).
• Caregivers victimized during childhood and adulthood had worse outcomes than if victimized either as an adult or child, or no victimization (Dubowitz et al., 2001).
46
Caregiver’s History of Violence Exposure as a Child and/or as an Adult
• Chronic victimization in childhood and adulthood more common than victimization in childhood alone or as an adult alone (Dubowitz et al., 2001).
• Women who were physically and/or sexually abused had worse outcomes than those with only one type of abuse (Dubowitz et al., 2001).
47
Caregiver’s History of Violence as a Child and/or as an Adult• Caregivers victimized in childhood and
adulthood had: More depressive symptomatology; Used harsher parenting behaviors; Children had higher internalizing and
externalizing scores on the CBCL (Dubowitz et al., 2001).
48
Caregiver’s History of Violence Exposure as a Child and/or as an Adult
• Caregivers with a history of CSA at increased likelihood of a violent adult relationship (English et al., 2003).
• Caregivers with a history of child physical abuse at increased risk of a violent adult relationship (English et al., 2003).
49
Caregiver’s History of Violence Exposures as a Child and/or as an Adult
• Caregivers with a history of CSA more depressed (Thompson, 2006).
• Adult victimization and depression predicted maltreatment (Thompson, 2006).
• Adult sexual victimization predicted problem drinking but not maltreatment (Thompson, 2006).
• Maternal adult physical victimization predicted a 2-3 fold increase in risk of maltreatment (Thompson, 2006).
50
Discussion
• Implications of Caregiver History of Abuse/Neglect on policy/practice.
Background – “Domestic Violence”Background – “Domestic Violence”• 2007 - 3.3 million reports of CA/N
• Co-occurrence of CA/N and adult domestic violence reported at 30-60% in clinical populations
• Estimates of 47% (2,000/35,000 random sample) of investigated CPS referrals - based on 1 item in assessment matrix and includes history & current
• 1999/2000 US national probability sample of 3,931 CPS cases - 14% current DV, 19% history of DV
• Between 990,000 – 1,410,000 estimated DV issues (current and/or past) present in CPS referrals
BackgroundBackground• Lack of clarity in definition of DV/IPAV/IPC – “woman
battering vs. common couple violence”• Research on special populations or community
populations may not be applicable to at-risk or CPS referred populations
• Limited examination of female IPAV/IPC and child maltreatment and child outcomes
Aggressive and Violent Behavior
Conflict Tactics Scale (Partner to Partner)
Violence Items (English et al., in press)
Mean (SD)Male-OnlyAV Perps.(N = 24)
Mean (SD)Males in Bi-Lateral AV
Couples(N = 351)
Mann-Whitney
Mean (SD)Female-
OnlyAV Perps.(N = 77)
Mean (SD)Females in Bi-Lateral
AV Couples(N = 351)
Mann-Whitne
y
Minor Violence Scale (# times) 0.21 (0.83) 0.68 (1.83) ns 0.16 (0.54) 1.1 (2.24) p < .001
…person threw something at other person 0.04 (0.20) 0.19 (0.61) ns 0.06 (0.37) 0.42 (0.97) p < .001
…person push-shove-grab other person 0.17 (0.63) 0.34 (0.89) ns 0.09 (0.40) 0.49 (1.05) p < .001
…person slapped other person 0.00 (0.00) 0.14 (0.60) ns 0.00 (0.00) 0.21 (0.68) p < .005
54
Aggressive and Violent Behavior
Conflict Tactics Scale (Partner to Partner)
Violence Items (English et al., in press)
Mean (SD)Male-OnlyAV Perps.(N = 24)
Mean (SD)Males in Bi-Lateral AV
Couples(N = 351)
Mann-Whitney
Mean (SD)Female-
OnlyAV Perps.(N = 77)
Mean (SD)Females in Bi-Lateral
AV Couples(N = 351)
Mann-Whitne
y
Severe Violence Scale (# times) 0.25 (1.20) 0.38 (1.59) ns 0.15 (0.80) 0.78 (2.29) p < .005
…person kick-bit-hit other person with fist 0.08 (0.41) 0.11 (0.52) ns 0.05 (0.46) 0.24 (0.78) p < .05
…person hit other person with something 0.04 (0.20) 0.12 (0.51) ns 0.03 (0.23) 0.29 (0.81) p < .005
…person beat other person up 0.12 (0.61) 0.09 (0.51) ns 0.03 (0.23) 0.14 (0.59) ns
…person choked other person 0.00 (0.00) 0.04 (0.28) ns 0.05 (0.36) 0.04 (0.34) ns
…person threatened other with knife-gun 0.00 (0.00) 0.03 (0.24) ns 0.00 (0.00) 0.08 (0.44) ns
…person used knife-gun on other person 0.00 (0.00) 0.01 (0.09) ns 0.00 (0.00) 0.01 (0.13) ns
55
Results – CTS P-to-PResults – CTS P-to-P(Overall prevalence)(Overall prevalence)
Male Female Significance
Verbal Aggression 82.7% 93.6% p<.001
Minor Aggression 15.5% 24.6% p<.005
Severe Aggression 7.7% 15.3% p<.001
Children Exposed to IPAV
• Significant child behavior problems for all types of IPAV (verbal, minor and severe) and for male and/or female perpetrators (English et al., in press).
• Conclusion: Need for comprehensive assessment of IPAV when assessing risk, safety and harm issues for children.
DiscussionDiscussion• Majority of IPAV in this sample is bilateral, and
women report more IPAV than their partners.
• No difference in rates or type of IPAV in unilateral IPAV homes (males/ females), however, females may sustain greater injury
• The majority of reported IPAV is verbal – but not necessarily benign for children
DiscussionDiscussion• Child outcomes are best in homes where no
maltreatment/no IPAV is reported • From the maltreated child’s perspective the impact of
witnessing IPAV appears not to vary based on gender of perpetrator
• From the non-maltreated child’s perspective, witnessing female-only IPAV does have an impact, as does witnessing bilateral IPAV (both compared to homes without IPAV)
Intimate Partner Violence/Aggression and Child Outcomes
• Conclusion: Need for comprehensive assessment of IPAV when assessing risk, safety and harm issues for children in homes where IPAV reported as an issue.
Domestic Violence (DV)• DV associated with higher alcohol use by
female caregiver (English et al., 2003).• DV associated with poor family health and
competence, poor family cohesion & high family conflict (SFI) (English et al., 2003).
• DV was not related to child outcome when other factors taken into account,
• The effect on children through mother’s depression and reduced capacity to parent (English et al., 2003).
Domestic Violence (DV)
• Poor quality mother/father relationships predicted higher child internalizing problems (Black, Papas et al., 2002).
• Black/Papas findings consistent with Katz & Gottman – children exhibit externalizing problems with mutually hostile caregiver(s) and internalizing when father figure is angry (Katz & Gottman).
Witnessing Violence• Caregivers report higher youth internalizing
scores than youth, and internalizing scores increase with exposure to violence (Lewis et al., in preparation).
• Child maltreatment associated with higher externalizing scores and higher caregiver depression (Lewis et al., in preparation).
• Higher witnessed violence scores associated with higher externalizing scores (Lewis et al., in preparation).
Domestic Violence (DV)
• Poor quality mother/father relationships predicted higher child internalizing problems (Black, Papas et al., 2002).
• Black/Papas findings consistent with Katz & Gottman – children exhibit externalizing problems when caregiver(s) mutually hostile and internalizing when father figure is angry (Katz & Gottman).
64
Domestic Violence (DV)
• In homes with DV, young age of caregiver, low education level, low income, & low religious involvement > risk of maltreatment (Cox et al., 2003).
• Social support moderates relationship between DV & maltreatment (Cox et al., 2003).
65
Witnessing Violence• Caregivers report higher youth internalizing
scores than youth, • internalizing scores increase when there is any
exposure to violence (Lewis et al., in prep.).• Child maltreatment associated with higher
externalizing scores and higher caregiver depression (Lewis et al., in preparation).
• Greater witnessed violence associated with higher externalizing scores (Lewis et al., in preparation).
66
Caregiver Depression (4, 6, 8, 12, and 14)
67
Note. + At ages 4, 6, 12, and 14 the CES-D was utilized. A score of16 or higher has been used extensively as the cut-off point for high depressive symptoms however, false positives on the order of 15% to 20% have resulted from use of this cut-off point, leading some researchers to suggest that a higher cut-off point be used (Boyd, Weissman, Thompson, & Myers, 1982; Zich, Attkisson, & Greenfield, 1990). * At age 8, the Brief Symptom inventory (BSI) was utilized. A cut-off of 63 or higher was used to determine clinical depression for the BSI.
31%
69% 70% 91% 73% 73%
30%
9%
27% 27%
Discussion
• Implication for findings on intimate partner aggression/violence, child exposure on policy/practice?
Depression and Child Maltreatment• If Caregiver is victim of child abuse or neglect as
child: More depressed (Dubowitz et al., 2001)
69
Depression & Child MaltreatmentIf Caregiver depressed:
Perceive partner relationship more negatively (Black, Papas et al., 2002).
Children are 1.6 times more likely to be maltreated (Radhakrishna et al., 2006; Thompson, 2006).
Not related to neglect in adolescence (Black, Papas et al., 2002).
70
Perceive child needs services but child not necessarily involved in services (Thompson et al., 2007; Dubowitz et al., 2001).
Perceive more behavioral problems but effect of maltreatment on child outcome not mediated by caregiver depression (Thompson et al., 2007; Black, Papas et al., 2002).
Depression & Child Maltreatment
Maternal Depressive Symptoms
• Depression symptoms did not predict re-referral when controlling for type of maltreatment and whether or not maltreatment was substantiated (Thompson & Wiley, 2009).
72
Quality of Parenting
• Parental warmth (empathy) moderates caregiver’s use of physical discipline and externalizing behavior (Lau et al., 2006).
• Physical discipline not effective with children predisposed to impulsive, aggressive or non-compliant behavior (Lau et al., 2006).
73
Quality of Parenting• If behavior already problem at age 4, parental
use of physical discipline predicts behavior problems at older ages (Lau et al., 2006).
• Harsh discipline predicts child’s internalizing and externalizing problems (Dubowitz et al., 2001).
• Caregivers with violent & aggressive behaviors toward each other more likely to use verbal aggression and minor violence with child (English et al., 2003).
74
Discussion
• Caregiver depression and quality of parenting – implications for policy/practice?
Maltreatment outcomes
% of Kids in the Normal/Borderline/ Clinical Range for Externalizing Behaviors (ages 4-16)
77
Child Behavior Checklist Teacher Report Form
Youth Self Report Form
Highlights of LS Findings Examining Externalizing Behaviors
• About 20% of LONGSCAN children in clinical range across ages
• Teachers report more clinical problems than parents
• Children self-report fewer behavior problems
• By any source, 60% of LONGSCAN children are in the NORMAL range on externalizing behavior
78
% of Kids in the Normal/Borderline/ Clinical Range for Internalizing Behaviors (ages 4-16)
79
Child Behavior Checklist Teacher Report Form
Youth Self Report Form
Highlights of LS Findings Examining Internalizing Behaviors• Caregivers report more problems as
children age• At 8, teachers noted more internalizing
problems than did caregivers• Youth self report had lowest sense of
problems• By any reporter, over 70% of the children
were normal on internalizing behaviors!
80
% of Kids in the Normal/Borderline/ Clinical Range for Total Problems (ages 4-16)
81
Child Behavior Checklist Teacher Report Form
Youth Self Report Form
Highlights of LS Findings Examining Total Problems• Children looked worse at age 8• No overall pattern of doing better or worse• Teachers more likely to note problem
behaviors• By any source of report 60% of children
were in the normal range on total problems!
82
Trauma Symptom Checklist (ages 8, 12, 16)
83
Mea
n T
Scor
es
Age 8 Frequencies:- 9% Clinical on Anger T Score- 24% Clinical on Anxiety T Score- 18% Clinical on Depression T Score- 16% Clinical on PTSD T Score
Age 12 Frequencies:- 2% Clinical on Anger T Score- 2% Clinical on Anxiety T Score- 2% Clinical on Depression T Score- 2% Clinical on PTSD T Score
Age 16 Frequencies:- 2% Clinical on Anger T Score- 2% Clinical on Anxiety T Score- 2% Clinical on Depression T Score- 4% Clinical on PTSD T Score
Higher scores, indicate greater distress.
Highlights of LS Findings using Trauma Symptom Checklist as an Outcome• Stability of trauma symptoms in a high risk
sample under examination (Lewis et al., in preparation).
• Adolescent psychological adjustment (Trauma Symptom Checklist & Youth Self Report) more strongly associated with self-reports than with CPS determinations (Everson et al., 2008).
• Children appear to do better on psychological symptoms as they age!
84
Vineland Screener (ages 6, 8, 12)
85
Mea
n St
anda
rd S
core
s
Higher scores, indicate greater competence.
Highlights of LS Findings using Vineland Screener as an Outcome• While psych distress score are better for
older children• Daily living and socialization scores decline• Impact of other forces on these high risk
children? Education, neighborhoods?
86
WPPSI (age 6) and WRAT (age 12)
87
% o
f Sta
ndar
d Sc
ores
15%
16%
69%
N = 1144 N = 840
9%
12%
79%
Highlights of LS Findings using WPPSI/WRAT as Outcomes• Pattern again is that 60% of the children are
in the normal range!• Increase in 10% of sample with cognitive
problems.
88
Suicidality (age 8)
• More maltreatment was associated with an increased likelihood of suicidal ideation (OR 2.0) (Thompson et al., 2005).
• Witnessed violence also increased the risk of > suicidal ideation (OR 1.6) (Thompson et al., 2005).
89
Relative Contributions of Violence at ages 8 and 12
Sample• Children from five sites who had completed the
Youth Self Report (n=872) at age 12Variable N %
Male Gender 416 49
Minority Race 648 76
Self-report of abusePhysicalSexual
Psychological
199138346
231640
Analysis• Hierarchical Linear Regression will be presented
• We predict the outcomes: Youth Self-Report scores for total problems, internalizing & externalizing scales
• 1: we examine how race, gender, and site predict scores• 2: we add witnessing community violence• 3: we add witnessing family violence• 4: we add self-reported exposure to child abuse
• At each stage, we see how much better the model predicts child distress on Youth Self Report
Standardized Betas for Internalizing YSR Scores
Beta p Beta p Beta pFemale 0.05 NS 0.05 NS 0.03 NS
Black Race -0.04 NS -0.04 NS -0.04 NS
East site -0.09 NS -0.08 NS -0.04 NS
Central site -0.05 NS -0.04 NS -0.02 NS
South site -0.01 NS 0.00 NS 0.00 NS
SW site -0.05 NS -0.04 NS -0.05 NS
Witness violence 0.16 <.001 0.11 <.001 0.12 <.001
Witness IPV 0.13 <.001 0.11 <.01
Physical abuse 0.12 <.001
Sexual abuse 0.11 <.01
Psych abuse 0.17 <.001
R square 0.02 0.04 0.13
R square change - 0.02 0.09
Looking Back at Age 8• Exposure Data used Child Protective Services Records• Outcomes: Trauma Symptom Checklist & CBCL• For TSC Depression Score- 13% of variance explained
• Standardized Beta for neglect = 0• Standardized Beta for Sexual abuse =.02 • Standardized Beta for physical abuse=.04• Standardized Beta for emotional abuse = .04• Standardized Beta for violence in home = .15• Standardized Beta for neighborhood violence = .23
• At age 8, exposure to domestic violence and neighborhood violence were predictive of depression
• At age 8, abuse, neglect, & sexual abuse did not strongly predictor depression
• Age 12 self-reported psychological maltreatment predicted most of the child outcome
• CPS allegations containing reports of psychological maltreatment explain almost none of the child outcome
Abuse, Neglect and HIV Risky Behaviors
Trajectory Modeling of LONGSCAN Children
HIV Risk and Sexual Abuse
• 5000 teens diagnosed with HIV each year• 234 teens died of HIV in 2005• Acquired through risky sexual behavior &
substance abuse• 7% of teens sexually active before age 13• Extensive literature links CSA to risky behavior• Cross-sectional and retrospective• Need to understand the relative contribution
of CSA to risky behaviors
Methods & approach• Examine trajectories of sexual abuse, other forms of
maltreatment, & risky behavior• 725 subjects (all 5 sites) with Age 14 interview• Maltreatment exposure classified by MMCS• Witnessed violence by Coddington Life Events• Risky behaviors by DISC questions about drugs &
alcohol and separate sexual activity questionnaire• Most youth denied either behavior• “Proc Traj” groups children by longitudinal patterns of
exposure
Sexual Abuse & Physical Abuse Trajectories
2 4 6 8 1 0 1 2
0.0
0.1
0.2
0.3
0.4
0.5
0.6
F i g u r e 1
A g e
Sexual Abuse Report
1 4 . 4 p e r c e n t
8 5 . 6 p e r c e n t
2 4 6 8 1 0 1 2
0.0
0.1
0.2
0.3
0.4
0.5
0.6
F i g u r e 2
A g e
Physical Abuse Report
3 2 . 2 p e r c e n t
6 7 . 8 p e r c e n t
2 groups for SA: 14% who had a pattern of sexual abuse & most with no CSA
2 groups for PA: 32% with a pattern of physical abuse & most with no PA
Group Trajectories for Emotional Abuse, Neglect, & Witnessed Violence
2 4 6 8 1 0 1 2
0.0
0.2
0.4
0.6
0.8
1.0
F i g u r e 3
A g e
Emotional Abuse Report
9 . 3 p e r c e n t
2 3 . 5 p e r c e n t
6 7 . 2 p e r c e n t
2 4 6 8 1 0 1 2
0.0
0.2
0.4
0.6
0.8
1.0
F i g u r e 4
A g e
Neglect Report
2 8 . 6 p e r c e n t
2 8 . 6 p e r c e n t
4 2 . 8 p e r c e n t
5 6 7 8 9 1 0 1 1 1 2
0
1
2
3
4
F i g u r e 5
A g e
Witnessed Violence Report 2 4 . 1 p e r c e n t
5 8 . 4 p e r c e n t
1 7 . 5 p e r c e n t
3 groups for emotional abuse
3 group model of
neglect
3 groups on witnessed
violence
CSA, PA, EA, neglect & witnessed violence
explaining alcohol use, sexual activity or both.
Child Genderab ns S-
Sexual Abuse c S+ S+
Physical Abused ns S+
Emotional Abuse 1e ns ns
Emotional Abuse 2f ns S+
Neglect 1g ns ns
Neglect 2h ns ns
Witnessed Violence 1i ns ns
Witnessed Violence 2j ns ns
Model 1 Model 2 Model 3 Model 4 Model 5 SA Only SA & PA SA & EA SA & Neglect SA & WV
Either Both Either Both Either Both Either Both Either Both Abuse/Violence ______________________________________________________________________________
Summing it up!
102
• Largest Longitudinal study of child maltreatment to date
• Sample demographics have held pretty well (>50%AA)
• Attrition has been acceptable• Standard coding of maltreatment is important• Most of sample reported 0-2 times• Most maltreatment low to moderate severity• 50% of reported maltreatment in single
developmental period or limited - not chronic
Summing it up II• Substantiation versus allegation- little difference• Astounding differences between self-report and
what DSS knows• Behavior at placement predicted # foster care
placements• For children with few problems: # placements
predict problems later• Early Neglect, not physical abuse, associated with
aggression at ages 4,6, & 8
Summing it up III
• Maternal exposure to maltreatment has impact on next generation
• IPV has strong influences on children and risk of maltreatment recurrence
• About 30% of mothers has clinical depression• Harsh discipline predicted more behavior
problems• Majority of children not in clinical range on TSCC,
CBCL, YSR, etc.• At age 8, witnessing IPV more hazardous that SA or
PA• At age 12, psychological abuse greatest risk
Summing it up IV
• Among teens, HIV risk behaviors associated with sexual abuse and physical abuse, not so much CPS reported emotional abuse or neglect
• Maltreatment explains just a small portion of risky behavior
• Lots of work left to do with LONGSCAN data
LONGSCAN Website (http://www.iprc.unc.edu/longscan/)
Public website:• Provides background information, helpful links,
and contact information.• Access to publications, measures manuals,
presentations, policies, and research briefs (Vol. 1 - Hunter & Knight, 1998; Vol. 2 -The LONGSCAN Investigators, 2006).
Internal website:• Access to electronic publications, data
dictionaries, retrieval information, grant submissions, final reports, & manuscript proposals.
106
