Abstract
Objective
To ascertain the prevalence of psychiatric symptoms and substance use disorders among adolescents with a lifetime history of foster care placement, using data from a nationally representative sample of US adolescents.
Methods
We studied adolescents aged 12–17 in the public use file of the 2000 National Household on Drug Abuse (N = 19,430, including 464 adolescents with history of foster care placement). Psychiatric symptoms and substance use disorders were ascertained through direct interviewing of adolescents. Logistic regression analyses were used to estimate the odds of past-year psychiatric symptoms and substance use disorders among adolescents involved with foster care, as compared to those without a lifetime history of foster care placement (comparison group)
Results
Adolescents involved with foster care had more past-year psychiatric symptoms, and especially more conduct symptoms, and past-year substance use disorders than those never placed in foster care. Adolescents involved with foster care were about four times more likely to have attempted suicide in the preceding 12 months (AOR, 3.95; 95% CI 2.78, 5.61), and about five times more likely to receive a drug dependence diagnosis in the same period (AOR, 4.81; 95% CI, 3.22, 7.18).
Conclusions
Adolescents involved with foster care have a higher prevalence of psychiatric symptoms and drug use disorders than those never placed in foster care. Additionally, the results of this study suggest that they may be at elevated risk for suicide attempts
INTRODUCTION
The U.S. child welfare system has been considered a de facto behavioral care system because of the large number of children and adolescents who have clinically significant emotional or behavioral problems (1). For example, a recent survey revealed that nearly half of youth involved with child welfare aged 2 to 14 years (N=3,803) had significant emotional or behavioral problems (1). The child welfare system includes, among other services, protective services, foster care, and adoption services. Over half a million children are currently placed in foster care in the U.S. (2). In order to understand why these children are so prone to develop emotional and behavioral problems and disorders, it is important to know who enters the foster care system. Even though a small number of orphans are placed in foster care, nowadays most are children of drug addicted parents who may also be mentally ill (3). Thus, both nature, i.e. the genetic loading associated with parental impairment, and nurture, i.e. abandonment, parental rejection, and trauma, conspire against these children.
Most published studies focusing on foster children have used small samples, and estimated the prevalence of psychopathological symptoms by reviewing records or using the parent-rated Child Behavior Checklist (CBCL) (4). Typically, foster children, and other children involved with the child welfare system, have a higher prevalence of most symptoms assessed than community norms or comparison groups (1, 3, 5–9). In several studies the prevalence of conduct symptoms was more elevated than the prevalence of other psychiatric symptoms. For example, a review of the records of 585 foster children (mean age of 13.5 years when first served by the child welfare agency studied) revealed a high prevalence of “conduct problems” (10), which are analogous to the DSM-IV diagnoses of oppositional and conduct disorders. Heflinger et al. (8) examined CBCL scores for 254 children in foster care aged 2 to 18 years. The greatest difficulties were found in the aggressive, delinquent, and withdrawn behavior scales. Similarly, a Canadian study of 248 children in foster care revealed that conduct disorder was the most common disorder (11). Furthermore, several follow-up studies of adults who had been in foster care as children found that a high proportion had criminal records in adulthood (10). Although these follow-up studies have significant methodological limitations, including the use of agency records and the absence of a comparison group, they add to the evidence suggesting that conduct symptoms may be the most prevalent psychiatric symptoms in the foster care population, and may have a long term impact.
Other than a recent observation suggesting that lifetime involvement with foster care was associated with the abuse of inhalants in adolescence (12), we are not aware of any published reports focusing on substance use disorders among children in foster care or with a lifetime history of foster care placement. Since foster children have a high prevalence of conduct symptoms (8,10,11), and these symptoms are often associated with substance abuse (e.g., 13–15), we expected that adolescents involved with foster care would have a higher prevalence of substance use disorders than those never placed in foster care.
Most published studies dealing with children and adolescents in foster care, including large well-designed studies (e.g., 1), provide information about youth involved with foster care or child welfare, without a comparison group. Our study is the first to include a nationally representative sample of adolescents with and without a history of involvement with foster care. Additionally, while most prior studies obtained information solely from caregivers using the CBCL (3, 8), the present study is based on direct interviews of adolescents. Parents and adolescents may underestimate internalizing and externalizing symptoms, respectively. Overall, there is more agreement for reports of externalizing symptoms (16, 17). Thus, our study may provide a more accurate report of internalizing symptoms, or at least a different perspective. The perspective of all informants is helpful and none can provide all the necessary information (18).
In the present study, we hypothesized that adolescents involved with foster care, i.e. those with a lifetime history of foster care placement, would have more psychiatric symptoms, use more illicit drugs, and have more substance use disorders than those without such a history; and that the most frequent type of psychiatric symptoms among children involved with foster care would be conduct symptoms. We were able to test these hypotheses using a nationally representative sample of U.S. adolescents aged 12–17 years to compare past-year prevalences of substance used disorders and psychiatric symptoms among adolescents with and without a history of foster care involvement.
METHODS
Data Source
This study is based on data from the public use file of the 2000 National Household on Drug Abuse (NHSDA) (19). This study was declared exempt from the RTI International institutional review board because it used an existing public use data file. No information or identifiers on the data file can be associated with any survey respondent. The annual NHSDA is designed to provide national estimates of the use of a variety of licit and illicit substances by Americans aged 12 or older. A secondary focus of the survey is mental health and mental health treatment service utilization. Foster care placement and other childhood experiences were included in the mental health section of the survey. Civilian, noninstitutionalized populations residing within the United States and the District of Columbia were selected for participating in the study via multistage area probability sampling methods (20). They include residents of noninstitutional group quarters (e.g., shelters, rooming houses, dormitories, and group homes), residents of Alaska and Hawaii, and civilians residing on military bases. Persons who had no fixed household address at the time of the survey (e.g., homeless transients not in shelters) and residents of institutional group quarters (e.g., jails and hospitals) were excluded from the sampling frame (19).
To increase the accuracy of self-reports of drug use and other behaviors of a sensitive nature (21), the 2000 survey utilized a combination of computer-assisted personal interviewing (CAPI) and audio computer-assisted self-interviewing (ACASI) methodologies. ACASI interviewing was used for questions of a sensitive nature. Respondents either read the questions on the computer screen, or the questions were read to them through headphones, and the respondents entered their responses directly into the computer.
Respondents were interviewed at their place of residence for about an hour. Interviewers requested the respondents to identify a private area in the home away from other household members to conduct the interview. A total of 71,764 individuals completed the 2000 survey. The weighted interview response rate was 74% in 2000 and was higher among adolescents aged 12–17 (83%) than among adults (72–77%). Analysis weights were developed to adjust for variation in household selection, nonresponse, and poststratification of the selected sample to the census data. The NHSDA sample was representative of the U.S. general population aged 12 or older. The NHSDA design and data collection procedures have been reported in detail elsewhere (19, 20)
Study Sample
We studied adolescents aged 12–17 in the public use file of the 2000 survey. In the sample (N = 19,430 ), 49% were females, 33% were aged 12–13, 35% were members of nonwhite minority groups, 26% reported an annual family income less than $40,000, and 23% resided in nonmetropolitan areas.
Study Variables
All psychiatric symptoms, including substance use, abuse, and dependence refer to the past 12 months. Foster care status (yes/no) refers to a lifetime history of foster care placement.
Foster Care Status
Adolescents who responded affirmatively to the question “Have you ever been in foster care?” were considered to have a lifetime history of foster care placement.
Psychiatric Symptoms
All questions in the “Adolescent Mental Health” section of the interview except one (“During the past 12 months have you taken medication for being overactive, being hyperactive, or having trouble paying attention?”) were included in this analysis. The question about medication for hyperactivity was excluded because it refers to treatment—not to a symptom. All questions in this section refer to the last 12 months. These questions can be found in the SAMSHA web site (http://oas.samhsa.gov/nhsda/methods.cfm#2k). They address symptoms of anxiety disorders (31 items, Cronbach alpha=0.9), eating problems (3 items, Cronbach alpha=0.8), major depression (9 items, Cronbach alpha=0.8), mania (3 items, Cronbach alpha=0.6), enuresis (2 items, Cronbach alpha=0.6), encopresis (1 item), and disruptive behavior disorders (24 items, Cronbach alpha=0.8). Disruptive behavior disorders include attention deficit hyperactivity disorder, oppositional defiant disorder, and conduct disorder. Two items included among the nine associated with major depression (suicidal ideation and suicide attempts) were also considered separately because of their public health importance.
Suicidal ideation and suicide attempts
Ideation and attempts were considered present when adolescents responded affirmatively to “During the past 12 months has there been a time when you thought seriously about killing yourself?” and to “During the past 12 months have you tried to kill yourself?”, respectively.
Substance Use and Substance Use Disorders
We examined past year use of alcohol and drugs (cocaine/crack, hallucinogens, heroin, inhalants, marijuana/hashish, pain relievers, sedatives, stimulants and tranquilizers; prescription drugs were included only when not used for medical purposes), and past year alcohol and drug use disorders (12, 22). Past year alcohol and other drug abuse and dependence were assessed using DSM-IV criteria (19, 23). Diagnostic assessments were based on the questions used in the National Comorbidity Survey (NCS) (24) and were revised to meet DSM-IV criteria.
Data Analysis
Because of the use of multistage probability sampling methods in the NHSDA, data were weighted to adjust for variations in household selection and non-response and to reflect the representativeness of the NHSDA sample. We used SUDAAN software (25) to analyze the data, which applies a Taylor series linearization method to account for the effects of the complex NHSDA design features (weighting). All percentages reported in this paper are weighted estimates, while sample sizes are unweighted. First, we examined demographic characteristics of the sample in relation to foster care placement status (Table 1). Second, we examined the bivariate association between a history of foster care placement and each category of psychiatric symptoms via the chi square test for binary variables and the t test for continuous variables (Table 2). With the exception of encopresis and bulimia, which were addressed in the NHSDA interview using a single question, symptoms of psychiatric disorders were summed separately for each disorder. We then calculated the mean number of symptoms for adolescents with and without a lifetime history of foster care.
Table 1.
Selected sample characteristics of youth aged 12–17 in 2000, by history of foster care placement (N = 19,430)
| History of foster care placement
|
||||
|---|---|---|---|---|
| Demographic characteristics | All youth aged 12–17 (N=19,430) | Yes (N= 464) | No (N= 18,966) | χ2 test (df) P-values |
| % | % | % | ||
| Age in years | ||||
| 12–13 | 33.0 | 28.9 | 33.2 | |
| 14–15 | 34.6 | 35.4 | 34.6 | 3.28 (2) |
| 16–17 | 32.4 | 35.7 | 32.3 | p=0.1950 |
| Gender | ||||
| Male | 51.2 | 47.7 | 51.3 | 1.73 (1) |
| Female | 48.8 | 52.4 | 48.7 | p=0.1893 |
| Race/Ethnicity | ||||
| White | 65.5 | 54.4 | 65.8 | |
| Black | 14.3 | 23.3 | 14.1 | |
| Hispanic | 14.2 | 16.5 | 14.1 | 18.07 (3) |
| Other | 6.0 | 5.9 | 6.0 | P <.001 |
| Family Income | ||||
| $0 – $19,999 | 17.9 | 36.9 | 17.4 | |
| $20,000 – $39,999 | 25.8 | 26.8 | 25.8 | |
| $40,000 – $74,000 | 33.9 | 26.4 | 34.1 | 59.24 (3) |
| $75,000+ | 22.4 | 10.0 | 22.7 | P < .001 |
| Population Density | ||||
| Large metropolitan | 43.6 | 46.7 | 43.5 | |
| Small metropolitan | 33.4 | 29.0 | 33.5 | 2.98 (2) |
| Non metropolitan | 23.1 | 24.3 | 23.0 | P= 0.2255 |
Table 2.
Mean number (standard errors) or proportion of past year psychiatric symptoms among youth aged 12–17 according to foster care status (N = 19,430)
| Lifetime History of Foster Care
|
|||
|---|---|---|---|
| Symptoms 1 | Yes | No | χ2 test (df) p-values or t test p-values |
| Anxiety symptoms | |||
| Social phobia (2) | 0.87 (0.04) | 0.65 (0.01) | t=5.54, p<0.0001 |
| Separation anxiety disorders (7) | 1.79 (0.11) | 0.99 (0.01) | t=7.23, p<0.0001 |
| Agoraphobia2 (4) | 0.74 (0.07) | 0.37 (0.01) | t=5.68, p<0.0001 |
| Panic attacks3(2) | 0.39 (0.04) | 0.21 (0.00) | t=4.30, p=0.0001 |
| Generalized anxiety disorder4 (4) | 1.22 (0.08) | 0.89 (0.01) | t=4.24, p<0.0001 |
| Specific phobias (7) | 1.54 (0.11) | 0.95 (0.01) | t=5.57, p<0.0001 |
| Obsessive compulsive disorder (5) | 1.10 (0.09) | 0.72 (0.01) | t=4.06, p=0.0001 |
| Any anxiety symptoms (31) | 7.65 (0.42) | 4.78 (0.04) | t=6.76, p<0.0001 |
| Eating Disorders symptoms | |||
| Anorexia nervosa (2) | 0.27 (0.04) | 0.14 (0.00) | t=3.56, p=0.0004 |
| Bulimia nervosa (1) | 9.5% | 4.1% | χ2= 8.52 (1), p=0.0036 |
| Elimination disorders symptoms | |||
| Enuresis (2) | 0.11 (0.02) | 0.04 (0.00) | t=3.10, p=0.005 |
| Encopresis (1) | 3.8% | 0.3% | χ2= 6.35 (1), p=0.0119 |
| Mood symptoms | |||
| Major depression (9) | 3.53 (0.15) | 2.60 (0.02) | t=6.24, p<0.0001 |
| Mania (3) | 0.97 (0.05) | 0.74 (0.01) | t=4.16, p<0.0001 |
| DBD5 symptoms | |||
| ADHD (5) | 2.08 (0.09) | 1.40 (0.01) | t=7.34, p<0.0001 |
| Oppositional defiant disorder (6) | 2.64 (0.11) | 1.89 (0.02) | t=6.99, p<0.0001 |
| Conduct disorder (9) | 1.54 (0.14) | 0.62 (0.01) | t=6.39, p<0.0001 |
| Any DBD5 symptoms (24) | 6.44 (0.30) | 3.97 (0.03) | t=8.33, p<0.0001 |
| Suicidality | |||
| Suicidal ideation (1) | 26.8% | 11.4% | χ2=38.47 (1), p<0.0001 |
| Suicidal attempts (1) | 15.3% | 4.2% | χ2=27.55 (1), p<0.0001 |
The numbers shown in parenthesis indicate the total number of symptoms included in the interview. For example, the interview includes 2 social phobia and 7 separation anxiety symptoms.
With or without panic attacks.
With or without agoraphobia.
Includes two somatization symptoms (“a lot of headaches” and “a lot of other aches and pains”). These are symptoms of “overanxious disorder”, a DSMIII-R diagnosis, and they are not included in the criteria for generalized anxiety disorder in DSM-IV.
Disruptive behavior disorders.
In order to estimate the odds of psychiatric symptoms (Table 3), we created a 5-category variable representing no symptoms and 4 levels of symptoms (i.e., grouping those reporting symptoms in quartiles). This variable was created separately for symptoms of anxiety, mood, and disruptive behavior disorders. Except for suicidal ideation and attempts, the odd ratios shown in Table 3 were estimated using multinominal logistic regression analysis (26). We compared each different symptom level of a psychiatric category to the absence of symptoms. The reference group is youth without a lifetime history of foster care. For binary outcome variables (i.e., presence or absence of suicidal behaviors, substance use, abuse, or dependence), we conducted (binary) logistic regression analyses (26) to estimate the strength of its association with a history of foster care placement. Each regression model controlled for the potential confounding influences of age, gender, race/ethnicity, family income, and population density on the association between the outcome variables and foster care placement.
Table 3.
Odds of past year psychiatric symptoms and suicidal behaviors among children with a lifetime history of foster care placement (N = 19,430)
| Symptoms1 | Unadjusted OR2 (95% confidence interval) | Adjusted OR2,3 (95% confidence interval) |
|---|---|---|
| Anxiety (31) | ||
| 1–2 vs. no symptoms | 1.36 (0.86 – 2.17) | 1.37 (0.86 – 2.17) |
| 3–4 vs. no symptoms | 1.87 (1.19 – 2.93)** | 1.83 (1.16 – 2.87)** |
| 5–8 vs. no symptoms | 2.10 (1.35 – 3.26)** | 2.04 (1.31 – 3.17)** |
| >8 vs. no symptoms | 3.66 (2.44 – 5.47)*** | 3.36 (2.19 – 5.14)*** |
| Mood (12) | ||
| 1–2 vs. no symptoms | 1.18 (0.81 – 1.72) | 1.27 (0.87 – 1.85) |
| 3–4 vs. no symptoms | 1.51 (1.05 – 2.18)* | 1.64 (1.13 – 2.37)** |
| 5–6 vs. no symptoms | 1.52 (1.05 – 2.20)* | 1.58 (1.09 – 2.30)* |
| >6 vs. no symptoms | 2.63 (1.87 – 3.69)*** | 2.75 (1.93 – 3.92)*** |
| Disruptive Behavior Disorders (24) | ||
| 1–2 vs. no symptoms | 1.37 (0.92 – 2.06) | 1.40 (0.94 – 2.10) |
| 3–5 vs. no symptoms | 1.89 (1.30 – 2.73)*** | 1.96 (1.35 – 2.84)*** |
| 6–7 vs. no symptoms | 2.14 (1.43 – 3.20)*** | 2.20 (1.47 – 3.29)*** |
| >7 vs. no symptoms | 4.34 (3.06 – 6.16)*** | 4.26 (2.99 – 6.06)*** |
| Suicidal ideation (1) | 2.86 (2.24 – 3.65)*** | 2.78 (2.14 – 3.60)*** |
| Suicide attempts (1) | 4.15 (3.01 – 5.72)*** | 3.95 (2.78 – 5.61)*** |
The number in parenthesis indicates the number of symptoms included in each category (e.g., 31 anxiety symptoms are included). Each row within a symptom category represents a quartile, e.g., children with more than eight anxiety symptoms represent the upper quartile (or the top 25%) for that symptom category.
The reference group is youth without a history of foster care
Each separate logistic regression adjusted for age, gender, race/ethnicity, family income, and population density.
p ≤ 0.05;
p ≤ 0.01;
p ≤ 0.001
RESULTS
In this section, we refer to adolescents with a lifetime history of foster care placement as “FC adolescents” and to those without such a history as the comparison group.
Demographics
There were no significant differences in the age and gender distribution of FC and comparison adolescents (Table 1). African-American and adolescents from low-income families were disproportionally represented among those ever placed in foster care.
Psychiatric Symptoms
FC adolescents had more symptoms than those in the comparison group in every category assessed. They had more than twice the number of conduct symptoms (1.54 vs. 0.62 mean number of conduct symptoms), and were significantly more likely to report suicide attempts and ideation than comparison adolescents (Table 2).
As shown in Table 3, FC adolescents were about three times more likely to present with 8 or more past-year symptoms of anxiety (AOR=3.36; 95% CI 2.19, 5.14), and were about four times more likely to present with 7 or more past-year symptoms of disruptive behavior disorders (AOR=4.26; 95% CI 2.99, 6.06) than comparison adolescents. They were also about four times more likely to have a history of suicide attempts (AOR=3.95; 95% CI 2.78, 5.61) in the preceding 12 months.
Substance Use and Substance Use Disorders
Compared to adolescents without a lifetime history of foster care, FC adolescents were slightly more likely to use alcohol, about two times more likely to engage in illicit drug use (AOR = 2.41, 95%CI 1.87, 3.11), about five time more likely to be drug-dependent (AOR = 4.81, 95%CI 3.22, 7.18), and about two to four times more likely to have other substance use disorders (Table 4).
Table 4.
Proportion (%) of youth aged 12–17 engaged in past year substance use, abuse, and dependence, and adjusted odds of substance use, abuse, and dependence according to foster care status (N = 19,430).
| Foster care
|
Unadjusted Odds1 (95% confidence interval) | Adjusted Odds1,2 (95% confidence interval) | |||
|---|---|---|---|---|---|
| Outcome: past year substance use or disorder | All youth | Yes | No | ||
| N=19,430 | N = 464 | N =18,966 | |||
| % | % | % | |||
| Any alcohol use | 33.1 | 40.0 | 32.9 | 1.36 (1.07 ─ 1.73)* | 1.45 (1.12 ─ 1.87)** |
| Alcohol abuse | 3.3 | 5.9 | 3.2 | 1.89 (1.11 ─ 3.23) * | 1.97 (1.15 ─ 3.39)* |
| Alcohol dependence | 1.9 | 6.5 | 1.8 | 3.91 (2.45 ─ 6.24)*** | 3.81 (2.30─ 6.30)*** |
| Any drug use | 18.5 | 34.2 | 18.1 | 2.36 (1.87 ─ 2.98)*** | 2.41 (1.87 ─ 3.11)*** |
| Drug abuse | 2.0 | 7.1 | 1.9 | 3.94 (2.53 ─ 6.15)*** | 3.78 (2.38 ─ 6.01)*** |
| Drug dependence | 2.4 | 9.8 | 2.2 | 4.90 (3.35 ─ 7.16)*** | 4.81 (3.22 ─ 7.18)*** |
The reference group is youth without a lifetime history of foster care.
Each separate logistic regression adjusted for age, gender, race/ethnicity, family income, and population density.
p ≤ 0.05;
p ≤ 0.01;
p ≤ 0.001.
DISCUSSION
Adolescents involved with foster care had more psychiatric symptoms of every type assessed than adolescents in the comparison group. This finding concurs with the extant literature (1, 3, 5, 6, 27, 28). Compared to adolescents never placed in foster care, the past year prevalence of conduct and anxiety symptoms was higher among adolescents involved with foster care. The preponderance of conduct symptoms among foster children has been reported by numerous investigators in the U.S. and elsewhere (e.g., 8,10,11,29).
Adolescents involved with foster care were also more likely to have significant anxiety and mood symptoms than those without foster care involvement. Previous studies reported a similarly high prevalence of internalizing symptoms, a CBCL category that includes depression, anxiety, withdrawal, and somatic complaints. For example, Heflinger et al. (8) reported that 23% and 19% of the foster children in their study, had symptoms in the clinical range of the externalizing and internalizing scale, respectively. Scores in the clinical range of the CBCL significantly discriminate between children referred for mental health and special educational services for behavioral or emotional problems and demographically similar children who are not referred (4).
Even though, as shown in this study, children involved with foster care have a higher prevalence of both internalizing and externalizing symptoms, the accumulating evidence suggests an association between foster care placement and conduct symptoms (3, 10, 11). Is this relationship cause and effect? Most children enter foster care after they have been abused or neglected, and often after their behavior becomes problematic (3). Child abuse is often associated with aggressive behavior (30, 31), and a seminal study found that adult criminal activity was associated with child maltreatment, not with living in a foster home (32).
There is an extensive literature suggesting that attachment problems and traumatic separations in early childhood interfere with children’s ability to learn that parents can be essentially benevolent. According to Katz (33) such children are unable to tolerate limits and frustrations that are part of the socialization process. Willock (34) suggests that children of unavailable parents feel unloved and disregarded. Lacking a conviction that somebody cares for them, they fell alone, threatened and frightened. They “must therefore constantly be ready to fend for themselves” (34, p. 63). According to this formulation, by being aggressive or defiant, these children disregard others and devalue them, thus reversing their conviction about themselves, i.e. that they are of little value to others.
Compared to those without a history of foster care placement, the prevalence of past year suicidal ideation (26.8% vs. 11.4%) and attempts (15.3% vs. 4.2%) among adolescents involved with foster care was markedly elevated. There is consistent evidence that child maltreatment is associated with suicide attempts, and less consistent evidence of an association with suicidal ideation (35). Since different types of abuse experiences can have different outcomes (30) it is difficult to identify the mechanism that may underlie this association. Wagner (35) has suggested that child maltreatment is associated with a sense of expendability. There are some reports suggesting that living away from both parents increases the risk of suicide attempts (e.g. 36, 37), but it is unclear whether separation from both parents or its antecedents, which may include overt parental rejection, account for the association with suicide attempts.
Adolescents involved with foster care had a higher prevalence of substance use disorders, and were about five times more likely to have a past year diagnosis of drug dependence than adolescents in the comparison group, after adjusting for gender and age. To the best of our knowledge, there are no published reports of the prevalence of substance use disorders among adolescents in foster care or with a history of foster care placement. The literature, however, suggests that childhood maltreatment may be a frequent antecedent of adult substance abuse (e.g., 38). This literature is subject to recall bias because it is largely based on retrospective reports of childhood maltreatment by adult substance abusers.
Adolescents involved with foster care have numerous risk factors that precede their entry into foster care, and therefore we view foster care as a marker of adversity, rather than a cause of psychopathology. Nevertheless, foster care often constitutes another stressful experience for youth with a prior history of multiple stressors, such as parental addiction, neglect, abuse, and other environmental conditions often associated with poverty in the U.S., e.g., dilapidated schools and lack of recreational facilities. Former foster children frequently have multiple negative adult outcomes, including criminality and incarceration (e.g., 10), homelessness (39), and selected adult psychiatric symptoms (40).
Involvement with foster care is often a marker of prior adversities, and an antecedent of negative adult outcomes. Most of these outcomes are associated with the aforementioned childhood adversities, and these adversities may have a greater impact on adult outcomes than foster care per se. Thus, involvement with foster care should be seen by clinicians as a possible marker—not necessarily a cause—of high risk for psychopathology and substance use disorders in childhood and adolescence, and as a risk factor for the above mentioned negative adult outcomes.
Limitations
This study relies solely on information provided by the adolescents. As discussed above, adolescents may under-report externalizing symptoms, and the most complete picture is obtained when all informants (parents, teachers, and adolescents) participate in the evaluation of the adolescent. Another limitation is that the data is cross-sectional, and therefore causal attributions are not warranted. For example, we do not know whether symptoms arose before or after placement in foster care. Nevertheless, the preponderance of conduct symptoms, observed both in this study and in prior studies of foster children, suggests that children in the present study had similar symptoms to those noted in prior research. As in most large epidemiological studies, the diagnoses of substance use disorders are not based on clinician-administered interviews. Last, the survey does not include information about the duration of foster care placement.
Acknowledgments
This work was partly supported by the National Institute on Alcohol Abuse and Alcoholism (R21AA013255) and the National Institute on Drug Abuse (R21DA015938). The Substance Abuse and Mental Health Data Archive and the Interuniversity Consortium for Political and Social Research provide the public use data files of the National Household Survey on Drug Abuse, which is sponsored by Office of Applied Studies, Substance Abuse and Mental Health Services Administration.
References
- 1.Burns BJ, Phillips SD, Wagner HR, et al. Mental health need and access to mental health services by youths involved with child welfare: A national survey. Journal American Acad Child Adolesc Psychiatry. 2004;43:960–970. doi: 10.1097/01.chi.0000127590.95585.65. [DOI] [PubMed] [Google Scholar]
- 2.Curtis PA, Dale G The foster Care Crisis: Translating Research into Policy and Practice. Washington, DC, Child Welfare League of America, 2001
- 3.Rosenfeld AA, Pilowsky DJ, Fine P, et al. Foster Care: An update. J Am Acad Child Adolesc Psychiatry. 1997;36:448–457. doi: 10.1097/00004583-199704000-00006. [DOI] [PubMed] [Google Scholar]
- 4.Achenbach TM, Rescorla LA Manual for the ASEBA school-age forms and profiles. Burlington, VT, University of Vermont, Research Center for Children, Youth, & Families, 2001
- 5.Clausen JM, Landsverk J, William G, et al. Mental health problems of children in foster care. Journal of Child & Family Studies. 1998;7:283–296. [Google Scholar]
- 6.Dubowitz H, Feigelman S, Harrington D, et al. Children in kinship care: how do they fare? Child Youth Serv Rev. 1994;16:85–106. [Google Scholar]
- 7.Halfon N, Mendonca A, Berkowitz G. Health status of children in foster care. The experience of the Center for the Vulnerable Child. Arch Pediatr Adolesc Med. 1995;149:386–392. doi: 10.1001/archpedi.1995.02170160040006. [DOI] [PubMed] [Google Scholar]
- 8.Heflinger CA, Simpkins CG, Combs-Orme T. Using the CBCL to determine the clinical status of children in state custody. Child Youth Serv Rev. 2000;22:55–73. [Google Scholar]
- 9.Urquiza AJ, Wirtz SJ, Peterson MS, et al. Screening and evaluating abused and neglected children entering protective custody. Child Welfare. 1994;73:155–171. [PubMed] [Google Scholar]
- 10.Fanshel D, Finch SJ, Grundy, JF Foster children in a life course perspective. New York, NY, Columbia University Press, 1990
- 11.Stein E, Rae-Grant N, Ackland S, et al. Psychiatric disorders in children in care: Methodology and demographic correlates. Can J Psychiatry. 1994;39:341–347. doi: 10.1177/070674379403900605. [DOI] [PubMed] [Google Scholar]
- 12.Wu LT, Pilowsky DJ, Schlenger WE. Inhalant abuse and dependence among adolescents in the United States. J Am Acad Child Adolesc Psychiatry. 2004;43:1206–1214. doi: 10.1097/01.chi.0000134491.42807.a3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kellam SG, Rebok GW, Ialongo N, et al. The course and malleability of aggressive behavior from early first grade into middle school; results of a developmental epidemiologically-based preventive trial. J Child Psychol Psychiatry. 1994;35:259–281. doi: 10.1111/j.1469-7610.1994.tb01161.x. [DOI] [PubMed] [Google Scholar]
- 14.Crowley TJ, Mikulich SK, MacDonald M, et al. Substance-dependent conduct-disordered adolescent males: severity of diagnosis predicts two year outcome. Drug Alcohol Depend. 1998;49:225–237. doi: 10.1016/s0376-8716(98)00016-7. [DOI] [PubMed] [Google Scholar]
- 15.Tarter R, Vanyukov M, Giancola P, et al. Etiology of early age onset substance use disorder: a maturational perspective. Dev Psychopathol. 1999;11:657–83. doi: 10.1017/s0954579499002266. [DOI] [PubMed] [Google Scholar]
- 16.Kashani JH, Orvaschel H, Burk JP, et al. Informant variance: The issue of parent-child disagreement. J Am Acad Child Psychiatry. 1985;24:437–441. doi: 10.1016/s0002-7138(09)60561-6. [DOI] [PubMed] [Google Scholar]
- 17.Randazzo KVD, Landsverk J, Ganger W. The informants’ reports of child behavior: parents teachers and foster parents. J Am Acad Child Adolesc Psychiatry. 2003;42:1343–1350. doi: 10.1097/01.chi.0000085753.71002.da. [DOI] [PubMed] [Google Scholar]
- 18.Stanger C, Lewis M. Agreement among parents teachers and children on internalizing and externalizing behavior problems. J Clin Child Psychol. 1993;22:107–115. [Google Scholar]
- 19.Substance Abuse and Mental Health Services Administration. Summary of Findings from the 2000 National Household Survey on Drug Abuse. Office of Applied Studies, NHSDA Series H-13, DHHS Publication No. (SMA) 01-3549. Rockville, MD, 2001.
- 20.Bowman, KR, Chromy JR, Odom DM, et al. 2000 National Household Survey on Drug Abuse: Sample design report. NC, Research Triangle Institute, 2002. Also available at http://oas.samhsa.gov/NHSDA/2kmrb/00SampleDesignRptW.pdf
- 21.Turner CF, Ku L, Rogers SM, et al. Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science. 1998;280:867–873. doi: 10.1126/science.280.5365.867. [DOI] [PubMed] [Google Scholar]
- 22.Wu LT, Pilowsky DJ, Schlenger WE. High prevalence of substance use disorders among adolescents who used marijuana and inhalants. Drug Alcohol Depend. 2005;78:23–32. doi: 10.1016/j.drugalcdep.2004.08.025. [DOI] [PubMed] [Google Scholar]
- 23.American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association, Washington, DC, 1994
- 24.Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey Arch Gen Psychiatry. 1994;51:8–19. doi: 10.1001/archpsyc.1994.03950010008002. [DOI] [PubMed] [Google Scholar]
- 25.Research Triangle Institute: SUDAAN User’s Manual: Release 8.0., Research Triangle Park, NC: Research Triangle Institute, 2002
- 26.Hosmer DW, Lemeshow S: Applied Logistic Regression. New York, NY, John Wiley & Sons, 2000
- 27.Leslie LK, Hurlburt MS, Landsverk J, et al. Outpatient mental health services for children in foster care: a national perspective. Child Abuse Negl. 2004;28:699–714. doi: 10.1016/j.chiabu.2004.01.004. [DOI] [PubMed] [Google Scholar]
- 28.Pilowsky D. Psychopathology among children placed in family foster care. Psychiatr Serv. 1995;46:906–910. doi: 10.1176/ps.46.9.906. [DOI] [PubMed] [Google Scholar]
- 29.Berridge D, Cleaver H: Foster home breakdown. Oxford, England, Basil Blackwell, 1987
- 30.Shields AM, Cicchetti D, Ryan RM. The development of emotional and behavioral self-regulation and social competence among maltreated school-age children. Dev Psychopathol. 1994;6:57–75. doi: 10.1017/S0954579400005885. [DOI] [PubMed] [Google Scholar]
- 31.Shields A, Cicchetti D. Reactive aggression among maltreated children: the contributions of attention and emotion dysregulation . J Clin Child Psychol. 1998;27:381–385. doi: 10.1207/s15374424jccp2704_2. [DOI] [PubMed] [Google Scholar]
- 32.Widom CS. The cycle of violence. Science. 1989;244:160–166. doi: 10.1126/science.2704995. [DOI] [PubMed] [Google Scholar]
- 33.Katz CL Aggressive children. In: O’Brien JD, Pilowsky DJ and Lewis OW, eds. Psychotherapies with children and adolescents: Adapting the psychodynamic process. Northvale, New Jersey: Jason Aronson, 2000:91–108
- 34.Willock B. Narcissistic vulnerability in the hyperaggressive child: The disregarded (unloved, uncared-for) self. Psychoanalytic Psychology. 1986;3:59–80. [Google Scholar]
- 35.Wagner BM. Family Risk Factors for child and adolescent suicidal behavior. Psychol Bull. 1997;121:246–298. doi: 10.1037/0033-2909.121.2.246. [DOI] [PubMed] [Google Scholar]
- 36.Rossow I, Wichstrom L. Parasuicide and use of intoxicants among Norwegian adolescents. Suicide Life Threat Behav. 1994;24:174–183. [PubMed] [Google Scholar]
- 37.Wagner BM, Cole RE, Schwartzman P. Psychosocial correlates of suicide attempts among junior and senior high school youth. Suicide Life Threat Behav. 1995;25:358–372. [PubMed] [Google Scholar]
- 38.Marcenko MO, Kemp SP, Larson NC. Childhood experiences of abuse, later substance use, and parenting outcomes among low-income mothers. Am J Orthopsychiatry. 2000;70:316–26. doi: 10.1037/h0087853. [DOI] [PubMed] [Google Scholar]
- 39.Herman DB, Susser ES, Struening EL, et al. Adverse childhood experiences: Are they risk factors for homelessness? Am J Public Health. 1997;87:249–255. doi: 10.2105/ajph.87.2.249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Herman DB, Susser ES, Struening E. Childhood out-of-home care and current depressive symptoms among homeless adults. Am J Public Health. 1994;84:1849–1851. doi: 10.2105/ajph.84.11.1849. [DOI] [PMC free article] [PubMed] [Google Scholar]