Prevalence of Mental Health Diagnoses among Adolescents before and during Foster Care

Abstract

Objective:

To characterize the rates and types of diagnosed mental health (MH) disorders among children and adolescents before and during foster care (FC) overall and by race and ethnicity.

Methods:

We used population-based linked administrative data of medical assistance (public insurance) claims records and child protective services data from a cohort of early adolescents who entered FC at 10–14 years old. MH diagnoses were coded according to the International Classification of Diseases, Ninth and tenth Revisions, Clinical Modification (ICD-9, ICD-10) and included adjustment disorders, disruptive disorders, ADHD, anxiety disorders, mood disorders, attachment disorders, autism, and other disorders.

Results:

Before FC entry, 41% of children and adolescents had at least one MH diagnosis. ADHD (25%), mood disorders (18%), and disruptive disorders (15%) were the most common pre-entry diagnoses. Among early adolescents entering FC with no previous diagnosis, 52% were later diagnosed with adjustment disorder (accounting for 73% of all youth with a new diagnosis during FC). White early adolescents had higher rates of diagnosed MH disorders before FC, whereas racial/ethnic minority early adolescents were more likely to receive a MH diagnosis during FC. Black early adolescents were more likely than White and Hispanic early adolescents to be diagnosed with disruptive disorders and less likely to be diagnosed with anxiety or adjustment disorders during FC.

Conclusions:

Results highlight the high rates of mental health needs among early adolescents prior to entry into FC, whether detected prior or during FC. Results also illustrate disparities in pre-entry MH care between racial/ethnic minority and White early adolescents with minority youth less likely to be receiving services prior to entry.

Children in foster care (FC) have high rates of mental health (MH) diagnoses and symptoms compared with the general population.^1–3 Rates are especially high among adolescents in FC, of whom an estimated 61% have a MH diagnosis and 37% received a MH diagnosis in the past year.¹ In contrast, an estimated 21% of adolescents in the general population have a current MH diagnosis.⁴ Yet, few studies have assessed whether MH diagnoses received in FC are new or reflect the continued treatment of a previously diagnosed MH disorder, or how the types of MH diagnoses shift after entry to FC.

Children who experience FC have high rates of MH conditions due to the deleterious effects of child maltreatment, which include increased risk for both internalizing and externalizing psychopathology⁵, as well as genetic factors, given that heritable substance dependency and mental illness are common among parents who are reported for maltreatment or lose custody of their children.⁶ In addition, older children (adolescents) who experience FC are likely to undergo frequent transitions in care and inconsistent access to mental health services.⁷ Given research that suggests that children involved with the child welfare system have similar rates of mental health symptoms³ and outcomes, such as completed suicides,⁸ regardless of whether they remain in their homes or are placed in FC, heightened rates of mental health issues in this population may be attributed to factors outside of the foster care experience itself. However, FC entry may lead to new MH diagnoses in three ways. First, FC may heighten detection of underlying MH conditions because of state requirements for physical and MH evaluation,⁹ an emphasis on trauma and behavioral health in foster parent training, and a general increase in the number of professionals with whom a child has contact. Second, FC may increase formal labeling of MH symptoms because a diagnosis is required for insurance reimbursement for MH services, and states are responsible for addressing the MH needs of adolescents in FC.¹⁰ Third, children may exhibit new MH symptoms in response to the uncertain and stressful experience of adjusting to a new environment in FC,¹¹ particularly for those who experience unstable placements (mostly adolescents).¹² To ensure that children can access therapeutic services quickly upon entry into FC, they may be diagnosed with an adjustment disorder, a diagnosis that refers to emotional responses to stressful life events or transitions occurring within the last 3 months.¹³ Adjustment disorder is a non-stigmatizing diagnosis (by attributing symptoms to an event or experience) with relatively broad symptoms¹³; thus, an adjustment disorder diagnosis may be used to facilitate access to supportive care while providers continue to assess and monitor symptoms to determine whether another MH diagnosis is more accurate. Taken together this literature suggests that a proportion of children who enter FC without a prior MH diagnosis may receive one once in FC, and adjustment disorders may account for a large proportion of initial diagnoses. However, children, especially adolescents, in FC are also disproportionately diagnosed with disorders characterized by externalizing behavior problems, particularly ADHD and disruptive disorders (e.g., conduct disorder and oppositional defiant disorder),^14,15 and it is not clear whether those precede FC entry or are newly assigned within FC.

In addition, there are documented concerns regarding disproportionate and over-identification of highly stigmatized conditions, such as conduct disorder among Black adolescents in the general population.¹⁶ This may result from characterizing normative adolescent behaviors as disobedient or threatening at a clinical level when exhibited by Black children,^16,17 or inadequate consideration of differential diagnoses that explain the cause of behaviors, such as post-traumatic stress disorder. Although the status of foster care may prime medical providers to better assess for trauma, earlier research (largely from the 2000s) found that racial and ethnic minority youth in foster care were less likely than White youth to receive MH services.^7,18,19 This gap in MH care utilization is also found in the general population, likely driven by myriad factors including cultural variation in help-seeking behaviors or MH stigma, disparities in access, lack of culturally-responsive services, and/or bias by medical providers in treatment referrals or provision.¹⁸ Because most studies of MH in FC only track children after they have entered FC, it is not clear from extent research whether FC maintains, narrows, or widens the MH care gap.

The Present Study

The present study examines a cohort of 2,317 10 to 14-year-olds (hereafter, early adolescents) entering FC in Wisconsin between June 2009 and December 2016, tracked from six months before entering FC until exiting care or three years post-entry (whichever is earlier). We first characterize the rate and types of known (diagnosed) MH disorders of early adolescents before FC entry. We then assess the rates and types of new MH diagnoses that early adolescents received while in FC. We also analyze MH diagnosis patterns by race/ethnicity.

Methods

Data used for this study were derived from the Wisconsin Administrative Data Core which is a linked longitudinal administrative data set, including public insurance (Medicaid and CHIP) claims records and child welfare system data at the Institute for Research and Poverty at the University of Wisconsin-Madison. Data for this study were deidentified and thus deemed exempted by the University’s internal review board. The target population was all children who entered Wisconsin FC between June 2009 and December 2016 and were in early adolescence (aged 10 to 14 years) at the time of entry (N=3,762). We restricted our analytic sample to early adolescents who were consecutively covered by public insurance for the six months prior to FC entry (N=2,998; 80% of target population) and who remained in FC for at least three months (N=2,317). The latter restriction was to ensure that early adolescents were in FC long enough to have seen a health care provider if needed, given that Wisconsin policy is to schedule a visit with a primary health care provider within 30 days of entering care. It is likely then that referred visits with specialized mental health care providers would potentially take longer to occur, making three months the likely minimum amount of time for early adolescents to receive a diagnosis. The cohort was followed using FC and claims records for the earlier of either three years post-entry or until the date of exit from FC.

MH diagnoses were coded according to the International Classification of Diseases, Ninth and tenth Revisions, Clinical Modification (ICD-9, ICD-10) in the medical claims data. MH diagnoses included were adjustment disorders, disruptive disorders, ADHD, anxiety disorders, mood disorders, attachment disorders, autism, and other (e.g., eating disorders, psychotic disorders, and personality disorders; see Table 2). We present percentages for each type of MH diagnosis pre-entry and during FC, and report (within-type) continuity and discontinuity of diagnoses. Significance tests were not provided given the use of population level data and the lack of need to make inferences about a sample.

Table 2.

International Classification of Diseases, Ninth and tenth Revisions, Clinical Modification (ICD-9; ICD-10) Codes Used to Group Mental Health Diagnoses

Group	Diagnosis	ICD9 Code	ICD10 Code
Adjustment Disorders	Adjustment Disorder	309	F43.2
	Acute Reaction to Stress	308	F43.0, F43.8, F43.9
Attention Deficit Disorders	ADD/ADHD	314	F90
Disruptive Disorders	Conduct Disorder	312	F91.0 F91.1 F91.2, F91.8 F91.9
	Oppositional Defiant Disorder	313.81	F91.3
Anxiety Disorders	Agoraphobia	300.21, 300.22	F40.0
	Obsessive Compulsive Disorder	300.3	F42
	Panic Disorder	300.01	F41.0
	Phobias	300.20, 300.23, 300.29	F40
	Generalized anxiety disorder	300.02	F41.1
	Other anxiety states/disorders	300.00, 300.09	F41.3, F41.8, F41.9
	Overanxious disorder specific to childhood and adolescence	313
PTSD	PTSD	309.81	F43.1
Mood Disorders	Bipolar Disorder	296.0, 296.4–296.8	F31
	Cyclothymic Disorder	301.13	F34.0
	Depressive Disorders	296.2, 296.3, 311	F32, F33
	Dysthymic Disorder	300.4	F34.1
	Manic episode	296.1	F30
	Other persistent mood disorders		F34.8, F34.9
	Other and unspecified episodic mood disorder	296.9	F39
Attachment Disorders	Reactive attachment disorder of childhood	313.89	F94.1
	Disinhibited attachment disorder of childhood	313.89	F94.2
Autism	Autism spectrum diagnoses	299	F84
Other	Tourette Syndrome	307.23	F95.2
	Tic disorders (other than Tourette)	307.20, 307.21, 307.22	F95.0, F95.1 F95.8, F95.9
	Somatoform disorders	300.8, 300.7	F45
	Delusional Disorder	297	F22
	Psychotic Disorder	290–294	F23
	Schizophrenia	295.0–295.3, 295.5, 295.6, 295.8, 295.9	F20
	Schizophreniform Disorder	295.4	F20.81
	Schizoaffective Disorder	295.7	F25
	Antisocial Personality Disorder	301.7	F60.2
	Avoidant Personality Disorder	301.82	F60.6
	Borderline Personality Disorder	301.83	F60.3
	Dependent Personality Disorder	301.6	F60.7
	Histrionic Personality Disorder	301.5	F60.4
	Obsessive Compulsive Personality Disorder	301.4	F60.5
	Paranoid Personality Disorder	301	F60.0
	Schizoid Personality Disorder	301.2	F60.1
	Schizotypal Personality Disorder	301.22
	Explosive personality disorder	301.3	F60.3 or F63.81?
	Affective personality disorder	301.1	F34.0, F34.1, F60.89,
	Narcissistic personality disorders	301.81	F60.81
	Passive-aggressive personality	301.84	F60.89
	Unspecified personality disorders	301.9	F60.81
	Anorexia Nervosa	307.1	F50.0
	Bulimia	307.51	F50.2
	Other/unspecified eating disorders	307.50, 307.59	F50.8, F50.9
	Dissociative, conversion, and factitious disorders	300.1	F44, F68.1, F68.A
	Other nonpsychotic mental disorders	300.9; V40.2	F48
	Sleep disorders (not due to physical condition or substance)	307.4	F51
	Unspecific childhood emotional disorder	313.9	F93.9
	Disturbance of emotions specific to childhood and adolescence (if not spec elsewhere)	313.2; 313.82; 313.83	94.9
	Misery and unhappiness disorder specific to childhood and adolescence	313.1	F93.8
	Selective mutism	313.23	F94.0
	Other behavioral and emotional disorders with onset usually occurring in childhood and adolescence	307.3; 307.6; 307.7; 307.9; 307.52; 307.53; 307.54	94.8

Results

The cohort was 51% male and 49% female; 41% was White, 27% Black, and 15% Hispanic. Forty-one percent of the cohort received health care services for at least one MH diagnosis in the six months prior to FC entry. The most common pre-entry diagnosis types were ADHD (25%), mood disorders (18%), and disruptive disorders (15%). See Table 1.

Table 1.

Sample Characteristics

	Total Num (%)
	2,317 (100)
Race/Ethnicity
White	943 (41)
Black	629 (27)
Hispanic	359 (15)
American Indian	81 (4)
Asian or Pacific Islander	23 (1)
Multiple races	282 (12)
Sex
Female	1,177 (51)
Male	1,139 (49)
Pre-Entry Diagnosis Category
Any	947 (41)
Adjustment	224 (10)
Mood	409 (18)
ADHD	568 (25)
PTSD	112 (5)
Disruptive	341 (15)
Anxiety	136 (6)
Other	188 (8)
Attachment	44 (2)
Autism	42 (3)

Of the 1,370 early adolescents who entered FC with no MH diagnosis in the 6 months prior to entry, 72% of early adolescents received a diagnosis during care. The most common new diagnoses were adjustment disorders (52%), mood disorders (26%), ADHD (23%) and PTSD (18%). Of all early adolescents who had no pre-entry diagnosis but received a diagnosis during FC, 73% were diagnosed with adjustment disorder (see Figure 1).

Figure 1.

New Diagnoses Received During Foster Care among Adolescents without a Pre-Entry Diagnosis

A total of 947 early adolescents entered FC with one or more prior MH diagnoses. Figure 2 shows stacked bars that aggregate to reflect the total percentage of early adolescents who received each type of diagnosis (adjustment, mood, etc.) either before or during FC. The bars are comprised of three components: (1) New diagnosis: early adolescents receiving a specific diagnosis type during FC but not prior to FC entry (indicating that either their prior diagnosis changed or an additional diagnosis was given); (2) Continuing diagnosis: early adolescents with a specific diagnosis type before FC that was also diagnosed during FC; and (3) Discontinued diagnosis: early adolescents with a specific diagnosis type before FC but not during FC. Most early adolescents continued to receive billable services related to MH diagnoses that they had prior to FC. For example, among early adolescents with at least one pre-entry diagnosis, 60% had a pre-entry ADHD diagnosis, with 53% continuing to receive health care services related to ADHD versus 7% receiving no Medicaid-billed ADHD-related services in FC. Only 12% were newly diagnosed with ADHD during FC. MH diagnoses characterized by symptoms that may be environmental or stress-induced – anxiety disorders, adjustment disorders, and PTSD – were more likely than other types of diagnoses to be new (versus continuing). Discontinuation of prior diagnosis was most common for disruptive disorders (12% discontinued versus 24% continued) and anxiety disorders (7% discontinued versus 8% continued).

Figure 2.

Diagnosis Continuity and Change for Adolescents with One or More Diagnoses Prior to Foster Care Entry (N=947)

Noticeable differences by race/ethnicity were identified both before and during FC (see Figure 3). Overall, 87% of White early adolescents had a diagnosis at some point during the study time frame (before or during FC), compared to 77% of Black early adolescents, and 78% of Hispanic early adolescents. Yet, 47% of the White early adolescents had a continuing diagnosis, versus 32% of Black and 36% of Hispanic early adolescents. In short, Black and Hispanic early adolescents were more likely to be newly diagnosed with a MH disorder in FC, whereas White early adolescents were more likely to already have a diagnosis (though their exact diagnoses may still have changed following entry to FC). For the diagnosis groups, we see that White early adolescents had higher rates for all types of continuing diagnoses. However, by the end of the observation period, Black and Hispanic early adolescents had similar rates to White early adolescents in several diagnosis groups. Black early adolescents continued to have lower rates of diagnosis for anxiety, adjustment, and autism disorders, and ultimately had higher rates of diagnosis for disruptive and mood disorders. Hispanic early adolescents had lower rates of ADHD diagnosis.

Figure 3.

Percentage of adolescents with a new versus continuing diagnosis, by race and ethnicity

Discussion

Using a population-level cohort of Wisconsin early adolescents on public (Medicaid or CHIP) insurance who entered FC at 10 to 14 years between June 2009 and December 2016, this study provides estimates of MH diagnoses in the six months preceding FC through FC exit or three years post-entry. Our analysis confirms that a substantial proportion (41%) of early adolescents had a known MH condition for which they received some health care services in the months leading up to FC entry, the vast majority of whom continued to receive MH-related services during FC. Given the known effects of child maltreatment and related adversities on MH,⁵ and that many who enter FC have experienced repeated adversities throughout childhood, this is likely a significant underestimate of the prevalence of pre-existing MH concerns and may indicate unmet MH needs.

By the end of the study observation period, most of the cohort had received one or more MH diagnoses. Adjustment disorders were the most common diagnoses received in FC, especially for early adolescents who did not have a pre-entry MH diagnosis. This is consistent with state and federal requirements that agencies ensure therapeutic support services for early adolescents during the difficult process of separation from their familial environments, even when providers may lack adequate medical and social history needed to evaluate or rule out more specific MH diagnoses.

Among early adolescents entering FC with one or more MH diagnoses, new diagnoses were most common for conditions that require or may be substantially impacted by exposure to traumatic events or severe deprivation, including PTSD and anxiety, attachment, and adjustment disorders. It is possible that the status of being in FC primes health care providers to more closely consider environmental factors in symptom etiology. In turn, receiving new diagnoses within FC that consider trauma and adversity as etiological factors may allow for more appropriate treatment recommendations. Where childhood trauma is present, typical pharmacological treatments may be less effective, potentially indicative of heterogeneous causal pathways for various MH conditions.^20,21 Similarly, behavioral treatments that do not address trauma as a proximal cause of symptoms may be less effective (e.g., traditional cognitive behavioral therapy (CBT) versus trauma-focused CBT).

Despite a shift in the nature of diagnoses before and during foster care, we reiterate that substantial proportions of early adolescents – both before and during FC – were labeled with ADHD or other conditions characterized by disruptive behavior, such as conduct or oppositional defiant disorder. These MH diagnoses – while perhaps fitting patterns of children’s’ behavior – do not require an antecedent adverse event (e.g., abuse, neglect, or other adversities) in the etiology or presentation of symptoms. The overdiagnosis of ADHD, a problem found in the general population of adolescents,²² is likely disproportionately impacting children exposed to trauma due to overlapping symptom profiles with PTSD.²³ We note that our study period was largely before the DSM-V was in effect, which altered the diagnostic criteria for PTSD to be more sensitive to symptom presentation in children and adolescents.²⁴ (Of note, there is also some evidence that behavioral problems symptomatic of ADHD or other conditions precede and result in exposure to abuse and neglect²⁵, so heightened rates of these diagnoses in the FC population is not itself proof of over- or misdiagnosis.)

The study also found proportionally large increases in diagnoses for conditions that emerge in early childhood, such as attachment disorder (for which symptoms must onset early in life) and ADHD (for which the average age of first diagnosis in the general population is seven years²⁶), especially for early adolescents with no pre-entry diagnosis. This may signal that FC increases formal diagnosis of symptoms that onset earlier in childhood. Alternatively, normative challenging behaviors may, when exhibited by children in foster care, be subject to stigma-related pathologizing that result in (over)diagnosis.

Lastly, with regard to racial/ethnic differences, White early adolescents had significantly higher rates of MH diagnoses before FC compared to Black and Hispanic early adolescents and this gap was partially closed by the end of the observation period. In addition to possible population differences in the prevalence of MH conditions, pre-entry diagnosis gaps may reflect disparities in access to and uptake of MH services. Care for Black children may be impeded by accessibility problems related to transportation, insurance type, or lack of local providers, uptake barriers related to poor quality, culturally inappropriate, or discriminatory treatment by healthcare providers and systems, , and cultural factors such as MH stigma.^18,19,27,28 Furthermore, within FC, it is possible – though we are aware of no research on this subject –that there are systemic differences in quality of treatment by caregivers, caseworkers, or agencies that result in disproportionately low rates of MH referrals or disparate characterization or reports to MH providers about the nature or severity of children’s symptoms. We note that Black early adolescents had higher rates of diagnosis for disruptive disorders and lower rates for anxiety and adjustment disorders. Given the stigma of disruptive disorder diagnoses – and their potential to limit children’s options for family-based foster care placement -- future research is needed on whether these patterns are found in other states and to assess for bias in the interpretation or characterization of symptoms^16,17 by individuals within the child welfare system (foster or kinship caregivers, congregate care providers) and/or the health care system.

Limitations

This is a single-state study and should be replicated elsewhere. Our analysis was restricted to early adolescents were enrolled in public insurance in the 6 months prior to removal, which excluded 20% of adolescents in the target population. In addition, our use of claims data means that we only capture MH diagnoses for which early adolescents received billable health care services. We note, however, that typically early adolescents taking psychotropic medication for a MH diagnosis should see a provider bi-annually,²⁹ and other outpatient MH services are typically bi-monthly or weekly. Of course, not all MH professionals, especially non-medical services, accept public insurance; thus, our estimates of MH diagnoses are conservative or lower-bound. We further note that we cannot ascertain the timing of onset of symptoms, nor can we independently confirm the accuracy of the MH diagnoses. However, because MH diagnoses are typically based on self- or caregiver-reported symptoms – which may be inconsistent or inaccurate³⁰ -- this limitation exists for all common measures of MH conditions, not only medical claims records.

Conclusion

The present study underscores the high rates of MH issues among early adolescents entering FC, whether detected prior to or upon entry. Given that the Family First Prevention Services Act seeks to limit the use of congregate settings and reduce the use of FC overall,³¹ findings point to a need to ensure that the child welfare system is adequately resourced and equipped to assess, diagnose, and treat early adolescents with MH problems. If the new diagnoses received in FC reflect unmet MH needs that preceded FC, MH assessment and treatment referrals upon initial contact with the child welfare system, regardless of whether they are placed in FC, may improve the timeliness of diagnosis and treatment. Moreover, effective MH interventions reduce child maltreatment, especially physical abuse, and thus may lessen the need for FC.³² Yet, such interventions – though eligible for Medicaid reimbursement –remain inaccessible to many due to lack of MH providers and other barriers. Within FC, placement instability both inhibits consistent MH treatment and exacerbates MH problems¹², pointing to the urgent need to recruit and retain family foster and kinship homes that are committed to providing safe and stable care for children with MH challenges – a consistent challenge for which there are few evidence-based solutions.^33,34 Lastly, we note the need for further research on disparities in MH diagnosis rates and types for racial and ethnic minority children in FC, including potential under-diagnosis of trauma-related MH conditions and potential overdiagnosis of behavioral disorders.