Abstract
Objective.
Access to wraparound care coordination within systems of care (SOC) is increasing nationwide for youth with emotional and behavioral disorders and their families. Though wraparound care coordination serves a broad population of youth who experience a variety of complex needs, less is known about the impact of wraparound services based on these specified needs. Using latent class analysis, the current study aimed to first identify classes of youth based on their presenting needs and then examine the impact of class membership on treatment experiences and outcomes at baseline and six-month follow-up.
Method.
Caregiver-reported data from 1,243 youth enrolled in wraparound care coordination services within a statewide SOC were used. Latent class analysis was used to determine classes of youth needs and regression results examined whether baseline characteristics, child and family team meeting characteristics, mental health outcomes, and perceptions of care differed based on identified classes.
Results.
Results revealed five distinct classes of youth needs: Behavioral Needs, ADHD-related Behavior Needs, Educational Needs, Mental Health Needs, and Multi-Needs. Overall participants saw improvement in all follow-up outcomes. Significant between-class differences were also found in all outcome categories measured.
Conclusion.
The current results further solidify the benefits of comprehensive wraparound care within a SOC. Additionally, understanding youth’s needs and their impact on treatment services allows for more targeted care for youth and their families.
Keywords: wraparound, care coordination, youth presenting needs, outcomes
Introduction
Wraparound care coordination services for youth with serious emotional and behavioral disorders are being implemented in all 50 states in the U.S. (Sather & Bruns, 2016). Often implemented in communities, wraparound care coordination is a structured effort that brings together a child and family team (CFT) consisting of youth, caregivers, natural supports (e.g., family members, community relationships), and a trained care coordinator (Bruns et al., 2004). Together, this team develops and implements a plan of care for youth and their family to meet their needs and ultimately transition them to receiving sustainable care in their own communities (Bruns et al., 2004). In contrast to an often-fragmented network of services for these youth (McKay & Bannon Jr, 2004), wraparound care coordination is often provided within a system of care (SOC) which brings together stakeholders and services from a variety of sectors (e.g., education, juvenile justice, child welfare; Olson et al., 2021; Suter & Bruns, 2009). Central to wraparound’s theory of change is the collaborative process of selecting and providing family-driven, strength-based services that are individualized to each family’s needs (Bruns et al., 2004; Coldiron et al., 2019; Olson et al., 2021). Thus, wraparound care coordination can, and is frequently used as a strategy to aid youth and families with multi-system involvement and/or presenting concerns (Suter & Bruns, 2009).
Research regarding outcomes of wraparound care coordination within SOCs has expanded rapidly in recent years and finds largely positive results (Coldiron et al., 2017; Olson et al., 2021; Suter & Bruns, 2009). In their 2021 meta-analysis of outcomes, Olson and colleagues found that overall, wraparound care coordination resulted in positive changes in residential outcomes, school outcomes, mental health symptoms, and functioning. Previous literature has highlighted similar positive benefits (Coldiron et al., 2017; Schreier, Horwitz, et al., 2019; Suter & Bruns, 2009) and recent studies have further solidified the benefits of wraparound care coordination for youth across varying income levels (Yohannan et al., 2017), gender (Chesney-Lind et al., 2008) and race and ethnicity (Madeline R. Stenersen et al.; Yohannan et al., 2017). It is often presumed that wraparound care coordination’s process of identifying and tailoring services to each family’s unique needs contributes significantly to its success (Olson et al., 2021).
Despite often serving a broad population of youth who experience a variety of complex needs including behavioral, mental health, and/or educational (Suter & Bruns, 2009), few studies have examined the impact and effectiveness of wraparound care coordination based on youth’s identified impairment and/or needs at intake into services (Bruns et al., 2015; Cox et al., 2009; Schreier et al., 2020). When investigating multiple predictors of outcomes in wraparound care, Cox and colleagues (Cox et al., 2009) found that lower levels of impairment at intake were associated with preserving home placement and more positive outcomes in care. Additional research on SOC and wraparound care coordination has revealed the impact of youth trauma (Schreier et al., 2020), behavioral problems, and substance use (Schreier, Champine, et al., 2019) on wraparound service dosage, characteristics, and outcomes. Specifically, Schreier and colleagues (Schreier, Champine, et al., 2019) found that children presenting to wraparound care with more behavioral problems received more services and child exposure to trauma predicted both recommendations for child welfare and mental health services. Though literature on wraparound care coordination within SOCs continues to expand, little research to date has examined the impact of youth’s specific areas of impairment on service characteristics and outcomes. Given the significant and varied needs of youth receiving wraparound care coordination within SOCs, increased understanding of how outcomes differ based on patterns of needs would extend the fields understanding of targeted implementation of wraparound care coordination for subgroups of youth.
The Current Study
To better understand the experiences and outcomes of youth with varying needs in wraparound care coordination, the current exploratory study has two primary aims. First, we aimed to identify and characterize patterns of youth needs at intake to wraparound care coordination using a latent class analysis (LCA). Second, we examined the association between class membership and 1) sample baseline characteristics, 2) characteristics of CFT meetings, 3) mental health outcomes at six month follow up, and 4) perceptions of care at six month follow up.
Methods
Participants
The current study utilized data from a total of 1,243 families who were enrolled and discharged from care coordination services through a statewide SOC. The initial dataset included 2,053 families and among those families, 1,243 were included in the current analyses as they completed at least six-months of services and a six-month follow-up assessment. All 1,243 of these families had information regarding the reasons for youth referral to services and were included in the current analysis. Among youth with known demographic data, youth ranged in age from 3 to 18.34 (Mage = 11.53, SD = 3.73) at the time of intake. Approximately 64% (n = 788) of youth identify as Male and 36.2% (n = 450) as Female. The majority of youth identified as White (n = 675, 54.3%) and not Hispanic/Latino (n = 668, 53.7%). Approximately 25% (n = 315) of youth were Black/African American. Full demographic information is outlined in Table 1.
Table 1.
Sample statistics
| Total (n = 1243) |
Class 1 (n = 346) |
Class 2 (n = 119) |
Class 3 (n = 248) |
Class 4 (n = 159) |
Class 5 (n = 371) |
|
|---|---|---|---|---|---|---|
|
| ||||||
| Variable | n(%) | n(%) | n(%) | n(%) | n(%) | n(%) |
|
| ||||||
| Youth Age (M(SD)) | 11.53 (3.73) | 10.56 (3.45) | 9.38 (3.13) | 12.28 (3.54) | 13.79 (2.95) | 11.64 (3.91) |
|
| ||||||
| Youth Gender | ||||||
| Male | 788 (63.4) | 239 (69.1) | 93 (78.2) | 142 (57.3) | 71 (44.7) | 243 (65.5) |
| Female | 450 (36.2) | 105 (30.3) | 26 (21.8) | 106 (42.7) | 88 (55.3) | 125 (33.7) |
|
| ||||||
| Youth Race | ||||||
| American Indian | 9 (0.7) | 4 (1.2) | 1 (0.8) | 0 (0) | 1 (0.6) | 3 (0.8) |
| Asian | 9 (0.7) | 0 (0) | 0 (0) | 2 (0.8) | 2 (1.3) | 5 (1.3) |
| Black/African American | 315 (25.3) | 94 (27.2) | 34 (28.6) | 58 (23.4) | 30 (18.9) | 99 (26.7) |
| Native Hawaiian | 2 (0.2) | 1 (0.3) | 0 (0) | 1 (0.4) | 0 (0) | 0 (0) |
| White | 675 (54.3) | 194 (56.1) | 60 (50.4) | 127 (51.2) | 91 (57.2) | 203 (54.7) |
| Unknown | 274 (22.0) | 68 (19.7) | 27 (22.7) | 65 (26.2) | 41 (25.8) | 73 (19.7) |
|
| ||||||
| Youth Ethnicity | ||||||
| Hispanic/Latino | 548 (44.1) | 139 (40.2) | 60 (50.4) | 119 (48.0) | 79 (49.7) | 151 (40.7) |
| Not Hispanic/Latino | 668 (53.7) | 195 (56.4) | 55 (46.2) | 126 (50.8) | 78 (49.1) | 214 (57.7) |
| Unknown | 27 (2.2) | 12 (3.5) | 4 (3.4) | 3 (1.2) | 2 (1.3) | 6 (1.6) |
|
| ||||||
| Youth Needs | ||||||
| Behavioral Problems | 466 (37.5) | |||||
| Developmental Disability | 112 (9.0) | |||||
| Delinquency | 122 (9.8) | |||||
| Educational Problems | 332 (26.7) | |||||
| Depression/Self-Injury | 250 (20.1) | |||||
| Anxiety | 360 (29.0) | |||||
| Adjustment | 109 (8.8) | |||||
| Hyperactivity/Attention Disorders | 322 (25.9) | |||||
Note: Class 1 = Behavior Problems/Needs, Class 2 = Behavioral Problems/Needs and ADHD, Class 3 = Educational Needs, Class 4 = Depression/Anxiety, Class 5 = Complex Needs.
From the initial 2,053 families, 810 families were excluded from the current study as they completed less than six months of services and were discharged prior to their six-month assessment. To examine any possible differences in needs between these families and those included in the current analysis, a post-hoc analysis was conducted. Youth included in the current analysis were more likely to be Male compared to those who completed less than six months of services (X2 (1) = 13.706, p = .003). There were no differences found regarding youth age, gender, or race/ethnicity.
Procedures
The Connecticut Network of Care Transformation (CONNECT) is a statewide SOC that provides wraparound services for youth and families in the state of Connecticut. Beginning in 2013, CONNECT began collecting and evaluating comprehensive data regarding the baseline functioning and outcomes of youth and families enrolled in CONNECT. The current study used care coordination data collected at the time of enrollment, throughout service provision (i.e., CFT meeting characteristics) and at six-month follow-up. All data in the current study is reported by primary caregivers with the exception of CFT characteristics data, which is reported by care coordinators. Care coordination is a key component to wraparound services and includes regular Child and Family Team (CFT) meetings with care coordinators, families, and natural supports to review youth and family needs and progress.
Measures
Reasons for Referral (Needs).
Youth needs were identified at enrollment by caregivers who were asked to identify “what led to the child being referred to services?” from a list of 22 reasons. Nine reasons were excluded from the current analysis due to low endorsement (Crying Tantrums(n = 1), Separation(n = 3), Noncompliance(n = 7), Maltreatment(n = 1), Current Home unable to fulfill needs(n = 8), Sleeping Problems(n = 1), Eating Disorder(n = 23), Psychotic Behaviors(n = 11), Suicide-Related Thoughts/Behaviors(n = 4)). Four sets of categories with significant overlap were condensed for analysis. In total, 8 categories of reasons for referrals were included in the current analysis: Behavioral Problems, Developmental Disability, Delinquency, Educational Needs, Depression/Self-Injury, Anxiety, Adjustment, and Hyperactivity/Attention Disorders.
CFT Meeting Characteristics.
As part of this SOC’s assessment of adherence to the National Wraparound Initiative (NWI) practice standards (Coldiron et al., 2019), administrative data was collected by care coordinators at every CFT meeting with families throughout their enrollment in services. The current study utilized data regarding time between intake and six-month follow-up including 1) the total number of CFT meetings, 2) number of days from intake to the first CFT meeting, 3) the average size of CFT meetings (number of people present), 4) the average length of CFT meetings in minutes, 5) percentage of CFT meetings where a natural support was present, 6) percentage of CFT meetings in which the target youth attended, 7) percentage of CFT meetings that were facilitated by the caregiver, 8) percentage of CFT meetings in which the plan of care (POC) was reviewed and 9) percentage of CFT meetings that took place in the home of the youth.
Mental Health Outcomes.
Mental health outcomes were assessed using several different measures.
Ohio Scales for Youth (Parent Form, OSP).
The Ohio Scales for Youth (Parent Form, OSP) includes scales for level of Problem Behavior and Functioning among youth (Ogles et al., 2000). Each scale is 20 items. Items on the Problem Behavior scale ask caregivers to identify their youth’s frequency of problem behaviors based on a 6-point Likert scale from not at all (0) to all the time (5). The Functioning scale includes items inquiring about the youth’s level of functioning on a 5-point Likert scale from extreme troubles (0) to doing very well (4). Each scale is summed with higher scores indicating higher levels of either problem behaviors or a higher level of functioning. The current study exhibited good to excellent reliability at baseline and follow-up for the Problem Behavior (α = .884 & .989) and Functioning (α = .909 & .941) scales respectively.
Caregiver Strain Questionnaire (CGSQ).
The CGSQ was developed by Brannan and colleagues (1997) to assess the amount of strain experienced by caregivers of youth. The original 21-item measure was modified in the current study to a 13-item scale as recommended by the national evaluation team for SAMSHA SOC grantees. The modified CGSQ includes 3 subscales: Objective Strain scale (OS), Subjective Internalizing Strain scale (SIS) and Subjective Externalizing Strain scale (SES). Additionally, a Global Strain score is calculated by summing the means of each subscale. Higher scores on each of the CGSQ scales indicates a higher level of strain. The current sample revealed good (α = .883) and excellent (α = .918) alpha coefficients at baseline and follow-up respectively.
Child Trauma Screen (CTS).
The CTS (Lang & Connell, 2017) is a caregiver-reported measures that assesses youth exposure to and symptoms related to traumatic events. The CTS includes four dichotomous questions related to trauma exposure and additional six questions assessing the presence and frequency of trauma symptoms (e.g. avoidance, negative affect, difficulty sleeping etc.) on a four-point Likert scale rating from never/rarely(0) to 3+ times per week(3). The current study utilized only questions related to trauma symptoms which were summed to create an overall symptom score and demonstrated acceptable reliability at baseline (α = .789) and good reliability at follow up (α = .809).
Perceptions of Care.
To assess caregiver’s perceptions of the services they and their child received, caregivers completed eight questions from the SAMHSA National Outcome Measures at the six- month follow-up assessment (Administration, 2009). Questions inquired about the caregiver’s perception of treatment by service staff (e.g., staff respected my religious beliefs), inclusion in their child’s services (e.g., I helped choose my child’s treatment) and satisfaction with services (e.g., overall, I am satisfied with the services my child received).
Data Analysis
Preliminary descriptive analyses were conducted to determine the sample size of eligible families and the numbers of families who endorsed each youth need. As outlined above, some categories were then condensed due to significant overlap. Next, a Latent Class Analysis (LCA) was conducted using MPlus 8’s DE3STEP approach (Asparouhov & Muthén, 2014; Muthén & Muthén, 1998–2017). Specifically, a series of LCAs were run with different class sizes to estimate the most optimal number of classes for the current data based on five relevant fit indices (e.g., Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Sample-Size Adjusted BIC (SSABIC), Lo-Mendell-Rubin Adjusted Likelihood Ratio Test (LRT) and Bootstrapped Likelihood Ratio Test (BLRT)). Entropy values were also examined to understand the distinctness of each class from one another in each model. After deciding on the best-fitting model (Step 1), probabilities of each family to belong in each class was obtained (Step 2). Finally, the relationship between class membership and distal outcomes (e.g., CFT Characteristics, Mental Health Outcomes, Perceptions of Care) was assessed (Step 3).
Results
Latent Class Analysis (LCA)
Latent class analyses were conducted to determine the presence of participant classes based on caregiver-reported youth needs. As outlined in Table 2, six latent class models were tested examining the fit of two to seven classes and the five-class model was chosen as optimal. Given their high entropy and minimal differences in fit indices, the four, five, and seven class models all appeared statistically viable. The five-class model was chosen as it 1) exhibited smaller fit indices (i.e., AIC, BIC, SSABIC) compared to the four-class model, 2) showed greater distinction between classes based on youth need and 3) fit the data best conceptually. Probability of each youth need across the five-class model is visualized in Figure 1.
Table 2.
Fit Indices for the Latent Class Models with Two to Seven Classes based on Youth Needs
| Model | Log-likelihood | AIC | BIC | SSABIC | Entropy | LMR-LRT | Bootstrapped LRT |
|---|---|---|---|---|---|---|---|
|
| |||||||
| 2 | −4655.396 | 9344.791 | 9431.921 | 9377.922 | .915 | 228.494 (p < .001) |
p <.001 |
| 3 | −4579.578 | 9211.155 | 9344.412 | 9261.825 | .906 | 149.308 (p = .019) |
p <.001 |
| 4 | −4516.456 | 9102.911 | 9282.296 | 9171.121 | .978 | 124.305 (p < .001) |
p <.001 |
| 5 | −4468.244 | 9024.488 | 9250.001 | 9110.237 | .936 | 94.942 (p < .001) |
p <.001 |
| 6 | −4435.177 | 8976.353 | 9247.993 | 9079.642 | .902 | 65.347 (p = .002) |
p <.001 |
| 7 | −4401.731 | 8927.462 | 9245.229 | 9048.290 | .958 | 69.863 (p < .001) |
p <.001 |
Figure 1.
Profile Plot of the Indicators for the 5 Class Model
Class 1 (27.8%, n = 346) is represented by youth whose caregivers endorsed Behavioral Problems as well as low levels of youth needs across multiple other areas including Educational Problems, Depression/Self-Injury, Anxiety, Adjustment, Developmental Disability and Delinquency. This class had a very low level of endorsement of Hyperactivity/Attention Disorders. Class 1 is termed Behavioral Problems/Needs. Class 2 (9.6%, n = 119) is represented by youth whose caregivers endorsed both Behavior Problems and Hyperactivity/Attention Disorders and no other needs. Class 2 is termed Behavior Problems/Needs and ADHD. Class 3 (20.0%, n = 248) represents youth whose caregivers endorsed Educational Needs at intake as well as low levels of additional needs (e.g., Anxiety, Depression/Self-Injury etc.). Class 3 is termed Educational Needs. Class 4 (12.8%, n = 159) includes youth whose caregivers identified high levels of youth Depression/Self-Injury and Anxiety and low levels of other needs. Class 4 is termed Mental Health Needs. The largest class is Class 5 (29.8%, n = 371) which includes youth whose caregivers endorsed moderate levels of Anxiety, Adjustment, and Hyperactivity/Attention Disorders; low levels of endorsement on Developmental Disability, Delinquency, and Depression/Self-Injury; and no needs related to Behavioral Problems or Educational Problems. Class 5 is termed Multi-Needs.
Distal Outcomes
Youth Needs and Mental Health Baseline.
Exemplified in Figure 2, Classes 1 (Behavior Problems/Needs) and 2 (Behavior Problems/Needs and ADHD) entered treatment with higher baseline scores on almost all mental health outcome measures (excluding trauma symptoms) compared to all other classes. Specifically, caregivers of youth with Behavioral Problems/Needs (Class 1, m = 31.04) and Behavior Problems/Needs and ADHD (Class 2, m = 30.84) reported their child had significantly higher levels of problem behaviors (X2 (3) = 51.443, p <.001) and lower levels of functioning (X2 (3) = 26.742, p <.001) at baseline compared to almost all other classes of youth needs.
Figure 2.
Change in Youth Outcomes from Baseline to 6 Month Follow-Up by Youth Class
Further, global (X2 (3) = 32.951, p <.001), subjective externalizing (SES; X2 (3) = 33.482, p <.001) and objective (X2 (3) = 40.425, p <.001) caregiver strain for youth in Classes 1 (Behavioral Problems/Needs, m = 23.98 & 11.68 respectively) and 2 (Behavior Problems/Needs and ADHD, m = 24.24 & 11.74 respectively) were significantly higher at baseline compared to those in almost all other classes. Baseline CGSQ SIS scores were highest for caregivers of youth in Class 4 (Mental Health Needs, m = 9.81) and significantly higher than Class 2 (Behavior Problems/Needs and ADHD, m = 8.66) and Class 3 (Educational Needs, m = 8.13; X2 (3) = 11.759, p =.019).
Regarding trauma symptoms, caregivers of youth with Mental Health Needs (Class 4, m = 7.67) reported the highest level of youth trauma symptoms at baseline compared to all other classes. The next highest was youth in Class 5 (Multi-Needs, m = 6.36) whose caregivers reported significantly higher youth trauma symptoms at baseline compared to youth in Classes 1 (Behavioral Problems/Needs, m = 5.44) and 2 (Behavior Problems/Needs and ADHD, m = 5.34; X2 (3) = 26.415, p <.001).
Youth Needs and CFT Characteristics.
When examining CFT characteristics, there were several significant differences based on youth needs classes. Overall, care coordinators of youth in Classes 4 (Mental Health Needs) and 5 (Multi-Needs Needs) reported significantly fewer meetings compared to youth in Classes 1 (Behavioral Problems/Needs) and 3 (Educational Needs). Youth with Multi-Needs (Class 5) reported longer CFT meetings and waited longer for their first CFT meeting compared to youth with Behavior Problems/Needs and ADHD (Class 2). Youth with Mental Health Needs (Class 4) attended the highest percentage of CFT meetings compared to all other classes.
Conversely, youth with Educational Needs (Class 3) attended the lowest percentage of CFT meetings, and significantly less than those with Behavioral Problems/Needs (Class 1). Youth with Educational Needs (Class 3) also had the smallest percentage of CFT meetings in which a natural support (e.g. family member or community relationship) was present and fewer CFT meetings where their plan of care was reviewed. Finally, youth with Educational Needs (Class 3) had the smallest percentage of their CFT meetings in their home and significantly less than youth in Classes 1 (Behavioral Problems/Needs) and 5 (Multi-Needs). Full results related to CFT characteristics are outlined in Table 3.
Table 3.
Mean across the Latent Classes on Youth Distal Outcomes
| CFT Characteristics | Class 1 (Behavioral) | Class 2 (Behavioral & ADHD) | Class 3 (Educational) | Class 4 (Mental Health) | Class 5 (Multi-Need) | Chi-Square Tests | Class Comparisons (p ≤ .05) |
|---|---|---|---|---|---|---|---|
|
| |||||||
| Number of CFT Mtgs | 3.494 | 3.257 | 3.533 | 2.943 | 3.198 | 11.828* | 1 > 5 3 > 4 1 > 4 3 > 5 |
|
| |||||||
| Days to First CFT | 47.266 | 42.254 | 46.531 | 49.603 | 49.856 | 5.572 | 5 > 2 |
|
| |||||||
| Average Size of CFT Mtg | 5.875 | 5.721 | 5.803 | 5.799 | 5.819 | 0.512 | |
|
| |||||||
| Average Length of CFT Mtg (Minutes) | 68.134 | 65.927 | 68.416 | 67.809 | 70.412 | 6.403 | 5 > 2 |
|
| |||||||
| Percentage of Mtgs with Natural Support Present | 0.239 | 0.231 | 0.174 | 0.217 | 0.266 | 9.678* | 1 > 3 5 > 3 |
|
| |||||||
| Percentage of CFT Mtgs Target Child Attended | 0.555 | 0.575 | 0.029 | 0.745 | 0.026 | 17.899* | 1 > 3 4 > 2 4 > 3 4 > 5 4 > 1 |
|
| |||||||
| Percentage of CFT Mtgs Facilitated by Caregiver | 0.192 | 0.147 | 0.209 | 0.163 | 0.190 | 3.149 | |
|
| |||||||
| Percentage of CFT Mtgs where POC was Reviewed | 0.851 | 0.780 | 0.792 | 0.858 | 0.802 | 8.784 | 1 > 3 |
|
| |||||||
| Percentage of Mtgs that happened inhome | 0.477 | 0.480 | 0.382 | 0.456 | 0.478 | 7.100 | 1 > 3 5 > 3 |
|
| |||||||
| Youth Mental Health Outcomes | |||||||
|
| |||||||
| CG OSP Problem Behavior Change Score | −9.997 | −5.385 | −6.662 | −11.504 | −7.285 | 11.287* | 1 > 3 1 > 2 4 > 2 4 > 3 4 > 5 |
|
| |||||||
| CG OSP Function Change Score | 8.615 | 4.796 | 1.048 | 9.459 | 6.548 | 6.569 | 1 > 2 |
|
| |||||||
| CGSQ OS Change Score | −2.664 | −1.371 | −2.578 | −3.391 | −1.080 | 10.826* | 1 > 5 3 > 5 4 > 5 |
|
| |||||||
| CGSQ SES Change Score | −0.812 | −0.853 | −0.306 | −0.782 | −0.276 | 6.313 | 1 > 5 |
|
| |||||||
| CGSQ SIS Change Score | −1.420 | −1.374 | −1.142 | −2.657 | −1.822 | 5.607 | 4 > 3 |
|
| |||||||
| CGSQ G Change Score | −6.560 | −5.318 | −4.897 | −9.055 | −4.252 | 9.013 | 1 > 5 4 > 3 4 > 5 |
|
| |||||||
| CG CT Symptom Change Score | −1.083 | −0.269 | −1.243 | −2.696 | −1.794 | 12.556* | 4 > 1 4 > 2 4 > 3 5 > 2 |
|
| |||||||
| Perception of Care Variable | |||||||
|
| |||||||
| Treated With Respect | 4.759 | 4.731 | 4.738 | 4.797 | 4.683 | 3.287 | |
|
| |||||||
| Staff Respected Religious Beliefs | 4.666 | 4.719 | 4.679 | 4.773 | 4.617 | 5.424 | 4 > 5 |
|
| |||||||
| Staff Spoke In a Way Understood | 4.717 | 4.733 | 4.745 | 4.690 | 4.620 | 5.983 | 3 > 5 |
|
| |||||||
| Staff Sensitive to Culture | 4.649 | 4.719 | 4.677 | 4.755 | 4.604 | 5.770 | 4 > 5 |
|
| |||||||
| Choose Services | 4.578 | 4.539 | 4.583 | 4.537 | 4.422 | 7.299 | 1 > 5 3 > 5 |
|
| |||||||
| Choose Treatment | 4.556 | 4.571 | 4.540 | 4.599 | 4.482 | 3.035 | |
|
| |||||||
| CG Participated in Treatment | 4.626 | 4.679 | 4.620 | 4.700 | 4.537 | 7.343 | 2 > 5 4 > 5 |
|
| |||||||
| Satisfaction with Services | 4.509 | 4.665 | 4.559 | 4.577 | 4.468 | 8.374 | 2 > 1 2 > 5 |
Notes: CFT = Child Family Team, CG = Caregiver, CGSQ = Caregiver Strain Questionnaire, POC = Plan of Care, OSP = Ohio Scales for Youth Parent Form, OS = Objective Strain, SES = Subjective Externalized Strain, SIS= Subjective Internalized Strain, G = Global Strain, CT = Child Trauma.
Youth Needs and Mental Health Outcomes.
Overall, youth in all classes exhibited a decrease in mental health symptoms, an increase in functioning, and a decrease in all types of caregiver strain from intake to six-month follow up. There were several significant differences in the amount of change seen from intake to follow-up based on youth need classes. Youth with Mental Health Needs (Class 4) saw the greatest decrease in their problem behaviors and trauma symptom level compared to almost all other classes. Further youth with Behavioral Problems/Needs (Class 1) exhibited the next highest decrease in problem behaviors and a significantly higher increase in functioning compared to youth with Behavior Problems/Needs and ADHD (Class 2).
Based on response to the Caregiver Strain Questionnaire (CGSQ), caregivers of youth with Multi-Needs (Class 5) saw the smallest decrease in almost all aspects of caregiver strain compared to other classes. Specifically, caregivers of youth in Class 5 saw a significantly lower level of decrease in objective strain (OS) and Global Strain compared to caregivers of youth in Classes 1 (Behavioral Problems/Needs) and 4 (Mental Health Needs) and a lower decrease in subjective externalizing strain (SES) compared to caregivers of youth in Class 1. Full results regarding mental health outcomes change scores are outlined in Table 3 and visualized in Figure 2.
Youth Needs and Perceptions of Care.
Though results revealed several significant differences in perceptions of care based on youth need classes, caregivers of youth across classes expressed overall agreement in feeling satisfied with their care, respected by staff, and involved in their child’s care. This finding is indicated by the fact that the mean scores of all classes across perceptions of care variables never fell below 4 (Agree) on the 5-point Likert scale.
Between classes, caregivers of youth with Multi-Needs (Class 5) reported the lowest level of satisfaction in their care compared to all other classes. With regard to interactions with staff, caregivers of youth with Multi-Needs (Class 5) reported feeling less respect for their religious beliefs, and less sensitivity to their culture compared to caregivers of youth with Mental Health Needs (Class 4). Caregivers with youth in Class 5 also noted significantly less agreement that staff spoke in a way they understood compared to their counterparts in Class 3 (Educational Needs).
Regarding perceptions of involvement in their child’s services, caregivers of youth in Class 5 (Multi-Needs) reported significantly less agreement that they participated in their child’s treatment and had a choice in the services their child received compared to almost all other classes. Finally, caregivers of youth with Behavior Problems/Needs and ADHD (Class 2) reported the highest level of overall satisfaction with services, significantly higher than caregivers of youth in Class 1 (Behavioral Problems/Needs) or Class 5 (Multi-Needs). Full results regarding perceptions of care are outlined in Table 3.
Discussion
The current exploratory study sought to 1) identify patterns of youth needs at intake to wraparound care coordination using latent class analysis and 2) understand the association between these classes of needs and care coordination outcomes. Presented in more detail below, results revealed five distinct classes of youth based on their presenting needs at intake to services. Findings also indicated several differences in a variety of service outcomes based on youth needs class.
Classes of Youth Needs
The results of the current study revealed five distinct classes of youths’ presenting needs. Apart from Class 5 (Multi-Needs), all needs classes were defined by an almost absolute presence of one youth need. Notably, Classes 1 (Behavioral Problems/Needs) and 2 (Behavior Problems/Needs and ADHD) both exhibited a very high probability of endorsing behavioral problems, however Class 2 also unanimously endorsed the presence of Hyperactivity/Attention Disorders whereas the endorsement of this need was almost nonexistent in Class 1. These findings highlight the importance of taking a multi-faceted view of youth needs at intake so services can be properly individualized to meet the needs of each youth and family.
Distal Outcomes
Taken together, results revealed overall positive improvements in outcomes across youth needs classes. Specifically, families across classes exhibited decreases in problem behaviors and trauma symptoms, increases in functioning, and decreases in caregiver strain from baseline to follow-up. Caregivers also reported that they felt respected by care coordination staff, participated in their child’s treatment, and were satisfied with the services they received. Consistent with previous literature (Coldiron et al., 2019; Olson et al., 2021), the current results further solidify the positive benefits of participation in wraparound care coordination within a SOC for youth and families of varying needs. Results also found several differences in distal outcomes based on youth needs class that can help inform youth’s individualized care coordination.
Collectively, youth whose caregivers identified behavioral needs entered treatment with higher caregiver-reported youth mental health symptoms on almost all baseline measures including higher problem behaviors and lower levels of functioning, and higher levels of caregiver strain compared to other classes of youth needs. Though higher levels of problem behaviors may be expected in a group with endorsed behavioral needs, lower levels of caregiver-reported youth functioning and higher caregiver strain at baseline may also be partially influenced by the externalizing nature of behavioral problems (Bussing et al., 2003; Vaughan et al., 2013). In particular, caregivers of children with externalizing symptoms and/or ADHD may experience higher levels of caregiver strain (Vaughan et al., 2013). These findings emphasize the importance of not only looking at the needs of youth, but also the impact of those youth needs on the caregivers and family unit at the outset of services. As is paramount in wraparound care, acknowledging, identifying, and supporting the needs of all family members in the care coordination process can have a positive impact on family retention, and youth outcomes (Accurso et al., 2015; Attride-Stirling et al., 2004).
Results also revealed multiple patterns across participation in care coordination and distal outcomes based on youth needs classes. First, caregivers of youth in the Multi-Needs class waited longer for their first CFT meeting, reported the smallest decrease in caregiver strain, and reported significantly lower levels of satisfaction with services compared to other classes. These findings are consistent with previous literature showing a relationship between the wait time until first CFT meeting and caregiver strain (Schreier, Horwitz, et al., 2019). Further, research regarding youth and families with multiple treatment needs has noted the possible burden treatment services has on a family unit, and particularly the caregiver (Gopalan et al., 2017; Staudt, 2007). In their 2017 study, Gopalan and colleagues found that caregivers reported feeling overwhelmed at times with the number of meetings and professionals involved in their child’s care. This was especially true when the caregiver felt they had less say in their child’s services and/or were not seeing improvements (Gopalan et al., 2017). As one of the goals of wrapround is to reduce this burden by having coordinated care, these findings suggest that families with multiple areas of needs may benefit from increased support from care coordinators in facilitating access to multiple services (VanDenBerg & Grealish, 1996).
The current findings also highlighted the important connection between CFT meetings, natural supports, and mental health outcomes. Specifically, youth in the Educational need class exhibited the highest number of CFT meetings but were least likely to have a natural support attend. Consistent with previous literature (Schreier, Horwitz, et al., 2019), youth in this class also exhibited smaller changes on mental health outcomes compared to youth in other classes. Engaging natural supports, including those developed through youth school involvement, in wraparound care coordination has been a noted challenge (Bartlett & Freeze, 2019; Bruns, 2010). Families of youth with severe emotional and behavioral problems may experience isolation from potential natural supports (Bruns, 2010). However, the presence of natural supports remain integral to ensuring that youth and families have sufficient community-based support even after ceasing services (Bartlett & Freeze, 2019; Bruns, 2010; Yu et al., 2020). For youth with educational needs, it may be especially important for care coordinators to focus on building a natural support network for both youth and caregivers that are connected to the youth’s school. For example, Bartlett and Freeze (2019) found that difficulties in engaging natural supports may be helped by utilizing a community mobilization team of potential natural supports (e.g., volunteer community members, faith leaders) to discuss the ongoing needs of youth in their communities. In the current system of care family peer support specialists work as a team with care coordinators to assist families in identifying and engaging with natural supports.
Strengths and Limitations
Patterns of youth needs were influential in affecting youth and caregivers’ experiences and outcomes of wraparound care coordination. Expanding on the current results, future research would benefit from understanding how knowledge of these patterns can be utilized by care coordinators to enhance service recommendations for youth and families. Nevertheless, this current study has several limitations to consider. First, the current study uses caregiver- and care coordinator-reported measures in all analyses. Future research may benefit from examining youth needs classes and outcomes using data reported by the youth themselves. Additionally, as all participants in the current study participated in wraparound services, comparison between the outcomes of wraparound and other services or no services was not examined in the current study. Third, results of the current study reflect outcomes up to six months following entry into services. Research into longer term outcomes beyond six months would be beneficial. Finally, though statistically significant, examining the clinical significance of the differences between classes was beyond the scope of the current study and would be beneficial for future research to explore.
Conclusions
Though recent research has continued to demonstrate the benefits of wraparound care coordination within a SOC on youth with severe emotional and Behavioral Needs and their families, there remains a limited understanding of the impact of wraparound care coordination for youth based on their presenting needs. The current study utilized data from wraparound care coordination as part of a statewide SOC to examine the impact of services based on classes of youth’s needs at intake. Results revealed five distinct classes of youth based on their presenting needs and the overall positive impact of wraparound services across classes. Further, several patterns of service characteristics and outcomes were identified based on youth needs. Findings highlight the unique experiences of youth and families in SOC wraparound care coordination based on youth needs and bolster evidence for continuing to support an individualized experience for each family in treatment.
Highlights.
Wraparound care coordination serves a broad population of youth who experience a variety of complex needs, yet little is known about the impact of wraparound services based on these needs.
Understanding youth’s needs and their impact on treatment services allows for more targeted care for youth and their families.
LCA analysis revealed five distinct classes of youth based on their presenting needs and the overall positive impact of wraparound services across classes.
Several patterns of service characteristics and outcomes were identified based on classes of youth needs.
Acknowledgements:
The preparation of this article was supported, in part, by Drs. Stenersen and Schreier’s postdoctoral training fellowships funded by the National Institutes of Health (NIH) (T32DA019426-13) and by a grant from the Center for Mental Health Services of the Substance Abuse and Mental Health Services Administration (SAMHSA) (1 U79 SM061646-01). The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of NIH or SAMHSA.
Footnotes
The authors attest that we have no conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain
Contributor Information
Madeline R. Stenersen, Division of Prevention and Community Research, Department of Psychiatry, Yale University School of Medicine.
Alayna Schreier, Division of Prevention and Community Research, Department of Psychiatry, Yale University School of Medicine.
Michael J. Strambler, Division of Prevention and Community Research, Department of Psychiatry, Yale University School of Medicine.
Tim Marshall, Office of Community Based Mental Health Services, Connecticut Department of Children and Families.
Jeana Bracey, Child Health and Development Institute of Connecticut, Inc..
Joy S. Kaufman, Division of Prevention and Community Research, Department of Psychiatry, Yale University School of Medicine.
References
- Accurso EC, Garland AF, Haine-Schlagel R, Brookman-Frazee L, & Baker-Ericzén MJ (2015). Factors contributing to reduced caregiver strain in a publicly funded child mental health system. Journal of emotional and behavioral disorders, 23(3), 131–143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Administration S. A. a. M. H. S. (2009). National Outcome Measures (NOMs) for Mental Health System Transformation. [Google Scholar]
- Asparouhov T, & Muthén B (2014). Auxiliary Variables in Mixture Modeling: Three-Step Approaches Using Mplus. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 329–341. 10.1080/10705511.2014.915181 [DOI] [Google Scholar]
- Attride-Stirling J, Davis H, Farrell L, Groark C, & Day C (2004). Factors influencing parental engagement in a community child and adolescent mental health service: A qualitative comparison of completers and non-completers. Clinical child psychology and psychiatry, 9(3), 347–361. [Google Scholar]
- Bartlett NA, & Freeze TB (2019). Examining Wraparound Fidelity for Youth with Mental Health Needs: An Illustrative Example of Two Rural Canadian Schools. International Journal of Special Education, 33(4), 846–868. [Google Scholar]
- Bruns E (2010). Wraparound Fidelity Index, version 4: Summary of relevant psychometrics, reliability, and validity studies. Seattle: University of Washington, Division of Public Behavioral Health and Justice Policy. [Google Scholar]
- Bruns EJ, Pullmann MD, Sather A, Denby Brinson R, & Ramey M (2015). Effectiveness of wraparound versus case management for children and adolescents: results of a randomized study. Administration and policy in mental health, 42(3), 309–322. 10.1007/s10488-014-0571-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bruns EJ, Walker JS, Adams J, Miles P, Osher TW, Rast J, & VanDenBerg J (2004). Ten principles of the wraparound process. Portland, OR: National Wraparound Initiative, Research and Training Center on Family Support and Children’s Mental Health, Portland State University. [Google Scholar]
- Bussing R, Gary FA, Mason DM, Leon CE, Sinha K, & Garvan CW (2003). Child Temperament, ADHD, and Caregiver Strain: Exploring Relationships in an Epidemiological Sample. Journal of the American Academy of Child & Adolescent Psychiatry, 42(2), 184–192. 10.1097/00004583-200302000-00012 [DOI] [PubMed] [Google Scholar]
- Chesney-Lind M, Morash M, & Stevens T (2008). Girls troubles, girls’ delinquency, and gender responsive programming: A review. Australian & New Zealand Journal of Criminology, 41(1), 162–189. [Google Scholar]
- Coldiron JS, Bruns EJ, & Quick H (2017). A comprehensive review of wraparound care coordination research, 1986–2014. Journal of Child and Family Studies, 26(5), 1245–1265. [Google Scholar]
- Coldiron JS, Hensley SW, Parigoris RM, & Bruns EJ (2019). Randomized control trial findings of a wraparound program for dually involved youth. Journal of emotional and behavioral disorders, 27(4), 195–208. [Google Scholar]
- Cox K, Baker D, & Wong MA (2009). Wraparound Retrospective: Factors Predicting Positive Outcomes. Journal of emotional and behavioral disorders, 18(1), 3–13. 10.1177/1063426609336955 [DOI] [Google Scholar]
- Gopalan G, Horen MJ, Bruns E, Corey M, Meteyer S, Pardue M, Lane T, & Matarese M (2017). Caregiver perceptions of parent peer support services within the wraparound service delivery model. Journal of Child and Family Studies, 26(7), 1923–1935. [Google Scholar]
- Lang JM, & Connell CM (2017). Development and validation of a brief trauma screening measure for children: The Child Trauma Screen. Psychological trauma: theory, research, practice, and policy, 9(3), 390. [DOI] [PubMed] [Google Scholar]
- Stenersen Madeline R.. ,, Kelly Aleece. ,, Bracey Jeana. ,, Marshall Tim. ,, Cummins Mary. ,, Clark Kathryn. ,, & Kaufman Joy S.. Understanding Racial-Ethnic Disparities in Wraparound Care for Youths With Emotional and Behavioral Disorders. Psychiatric Services, 0(0), appi.ps.202100086. 10.1176/appi.ps.202100086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McKay MM, & Bannon WM Jr (2004). Engaging families in child mental health services. Child and Adolescent Psychiatric Clinics, 13(4), 905–921. [DOI] [PubMed] [Google Scholar]
- Muthén L, & Muthén B (1998–2017). Mplus User’s Guide Eight Edition. Muthén M. [Google Scholar]
- Ogles B, Melendez G, Davis D, & Lunnen K (2000). The Ohio youth problem, functioning, and satisfaction scales: Technical manual. Athens, OH: Ohio University. [Google Scholar]
- Olson JR, Benjamin PH, Azman AA, Kellogg MA, Pullmann MD, Suter JC, & Bruns EJ (2021). Systematic Review and Meta-analysis: Effectiveness of Wraparound Care Coordination for Children and Adolescents. Journal of the American Academy of Child & Adolescent Psychiatry. 10.1016/j.jaac.2021.02.022 [DOI] [PubMed] [Google Scholar]
- Sather A, & Bruns EJ (2016). National Trends in Implementing Wraparound: Results of the State Wraparound Survey, 2013. Journal of Child and Family Studies, 25(10), 3160–3172. 10.1007/s10826-016-0469-7 [DOI] [Google Scholar]
- Schreier A, Champine RB, & Kaufman JS (2019). Predictors of Service Dosage, Recommendation, and Usage in an Early Childhood System of Care. Journal of Child and Family Studies, 28(2), 599–608. 10.1007/s10826-018-1290-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schreier A, Horwitz M, Marshall T, Bracey J, Cummins M, & Kaufman JS (2019). Child and family team meeting characteristics and outcomes in a statewide system of care. American journal of community psychology, 63(3–4), 487–498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schreier A, Horwitz M, Marshall T, Bracey J, Cummins M, & Kaufman JS (2020). Trauma Symptoms and Relationship With Child and Family Team Meeting Characteristics and Outcomes in a Statewide System of Care. Journal of emotional and behavioral disorders, 1063426620980695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Staudt M (2007). Treatment engagement with caregivers of at-risk children: Gaps in research and conceptualization. Journal of Child and Family Studies, 16(2), 183–196. [Google Scholar]
- Suter JC, & Bruns EJ (2009). Effectiveness of the Wraparound Process for Children with Emotional and Behavioral Disorders: A Meta-Analysis. Clinical Child and Family Psychology Review, 12(4), 336. 10.1007/s10567-009-0059-y [DOI] [PubMed] [Google Scholar]
- VanDenBerg JE, & Grealish EM (1996). Individualized services and supports through the wraparound process: Philosophy and procedures. Journal of Child and Family Studies, 5(1), 7–21. [Google Scholar]
- Vaughan EL, Feinn R, Bernard S, Brereton M, & Kaufman JS (2013). Relationships between child emotional and behavioral symptoms and caregiver strain and parenting stress. Journal of Family Issues, 34(4), 534–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yohannan J, Carlson JS, Shepherd M, & Batsche-McKenzie K (2017). Exploring attrition, fidelity, and effectiveness of wraparound services among low-income youth of different racial backgrounds. Families, Systems, & Health, 35(4), 430–438. 10.1037/fsh0000313 [DOI] [PubMed] [Google Scholar]
- Yu R, Haddock A, & Womack T (2020). Integrating Supports for Students with Wraparound. Contemporary School Psychology. 10.1007/s40688-020-00284-y [DOI] [Google Scholar]