Exploring Adherence to Client Treatment Recommendations in the Neurosequential Model of Therapeutics

Abstract

Purpose: Treatment adherence is a central component of implementation fidelity and key to our understanding of client outcomes. This study examines treatment adherence in the Neurosequential Model of Therapeutics (NMT) and four functional outcomes among a sample of adopted youth: sensory integration, self-regulation, and relational and cognitive functioning. Methods: The sample includes 178 youth receiving supportive services from a U.S. adoption service provider. Regression analysis was used to examine clinician adherence to NMT recommendations and child outcomes. Results: Most (61.24%) of the essential or therapeutic treatment recommendations were carried out with majority adherence. A key finding is the predictive relationship between adherence to treatment recommendations and improvement in child outcomes at T2, controlling for all other variables in the model. Conclusions: Findings illustrate the importance of treatment adherence in promoting positive outcomes for children engaged in NMT. Implications are discussed regarding clinician training and reducing barriers to treatment adherence.

Keywords

Neurosequential Model of Therapeutics adopted youth treatment adherence implementation fidelity intervention research

The Neurosequential Model of Therapeutics (NMT; Perry, 2009) is an innovative approach to clinical assessment and treatment planning and delivery with children and youth, which incorporates key principles of neurodevelopment, traumatology, and attachment theory. Exposure to a range of adversities during development—especially during early childhood—can alter core regulatory networks that mediate stress response capabilities, and alter a range of functions, including self-regulation and sensory integration (Perry, 2009). Recognizing this foundational principle, the NMT process addresses the sequential and use-dependent nature of brain development; the timing, duration, and severity of adverse childhood experiences; and the role of relational supports (including interpersonal, community, and cultural) across a child's lifespan (see Perry, 2009, 2020). A central component of the NMT assessment and treatment planning process involves providing individualized therapeutic treatment recommendations designed to improve children's functioning across four domains: cognitive, regulatory, relational, and sensory integration. The NMT is not a discreet, time-limited intervention; rather, it is a holistic clinical assessment and treatment planning approach informed by principles of neurobiology that provides an overall lens through which clinical work is conducted (Perry, 2020). The NMT is used widely across multidisciplinary settings that serve children and youth exposed to trauma, including therapeutic preschools (Barfield et al., 2012), residential programs (Hambrick et al., 2018a), juvenile justice agencies (Perry et al., 2018), and child welfare (Cox et al., 2021; Frederico et al., 2019).

As an emerging therapeutic model, the body of published empirical literature on the NMT has primarily focused on descriptive studies of NMT implementation (de Nooyer & Lingard, 2017; Frederico et al., 2019; Freedle & Slagle, 2018) and child outcomes (Barfield et al., 2012; Hambrick et al., 2018a; Keegan et al., 2020). For example, previous studies have described implementation of the NMT and improved youth behavior outcomes in nature-based behavioral health centers (de Nooyer & Lingard, 2017), in the adolescent education and employment training program Empowering Youth to Thrive (Keegan et al., 2020), and in therapeutic preschool settings with children diagnosed with severe emotional and/or behavioral disturbances (Barfield et al., 2012). Less is known about implementation fidelity and treatment adherence with the NMT, which are core components of implementation frameworks endorsed by the National Implementation Research Network (NIRN; Metz et al., 2021). Examination of implementation fidelity and treatment adherence can enrich our understanding of factors influencing child outcomes, particularly within the context of new and emerging models of practice. Thus, this article seeks to provide an overview of the NMT and the strategy for monitoring NMT treatment plan adherence. NMT treatment plan adherence—and associated child outcomes—will then be further explored through the example of one specific agency setting: a nonprofit adoption service provider that provides supportive services for adopted youth and their families.

The Neurosequential Model of Therapeutics

To initiate use of the NMT, clinicians draw from multiple sources of information—including interviews with caregivers and children—to complete detailed assessments of children's developmental, interpersonal, and trauma histories, their past and current relational supports, and their current functioning (e.g., cognitive, interpersonal, physiological, socioemotional, etc.). Results from these assessments are entered into the Neurosequential Network online portal. The NMT metrics use the assessment data to guide the NMT-trained clinician in the selection and sequencing of educational, enrichment, and therapeutic interventions that are client specific. Drawing from neurobiological principles on the sequential and use-dependent development of the brain, the metric tool outputs a clinical summary and series of recommendations based on the specific history and experiences of the child. The NMT-trained clinicians, in consultation with the family, develop individualized treatment plans that include individual, family, and community-level treatment recommendations for the child. Emphasis is placed on sequencing intervention activities in a way that mirrors the “bottom-up” sequential development of the brain (Perry, 2006). That is, initial treatment activities tend to focus on symptoms that are mediated by lower regions of the brain responsible for basic functions such as self-regulation and impulsivity. As improvements are maintained, intervention activities progress sequentially through higher-order regions of brain functioning, eventually progressing to insight-based and cognitive-based therapies only after self-regulatory and relational capacities have been established (Perry, 2006, 2009). Additional details about the NMT assessment process and clinical summary and recommendations are provided below.

Child Assessment: The NMT Metric Scoring Tool

The NMT online metric scoring tool features assessment questions focused on a child's: (1) history of exposure to trauma and adversity, (2) current and past relational health, and (3) current functioning. Relational health refers to the consistency, depth, and quality of supportive relationships in a child's life (Hambrick et al., 2019). Current functioning refers to a child's functioning across several developmental domains, which are organized by regions of the brain that mediate each cluster of functions, including the brainstem, cerebellum/diencephalon, limbic, cortex, and frontal cortex (Perry, 2020). Clinicians review and report the child's history of adverse experiences and relational health across each of six developmental periods (prenatal, perinatal, infancy, early childhood, childhood, and adolescence), and estimate the timing and severity of developmental risk that may have influenced brain development (see Perry, 2009, 2020). Response options are on a 12-point Likert scale for each assessment question, with lower scores indicating lower levels of risk/adversity and lower levels of relational health, and higher scores indicating higher levels of risk/adversity and higher levels of relational health. Assessment information may be drawn from prior physical or mental health records, school records, caregivers, clinicians, and/or other people who have information or contact with the child. When historical information is incomplete, clinicians use their own clinical judgment to reconstruct histories and estimate child scores.

NMT Assessment Output: Clinical Summary and Treatment Recommendations

NMT Clinical Summary

Once all scores for the NMT metric have been entered for a child, the online scoring tool generates a report that quantifies the nature, timing, and severity of the child's adverse experiences and relational health across all stages of development. The report includes the following: (1) graphs indicating a child's developmental risk and relational health across stages of development; (2) a functional “brain map,” which estimates a current picture of the child's brain organization, functioning, and brain-mediated strengths and weaknesses relative to what would be developmentally appropriate; (3) a summary of the child's current functioning and therapeutic recommendations across four focal domains: sensory integration, self-regulation, relational, and cognitive (Perry, 2009, 2020). For each domain of functioning, a child's score is contextualized to indicate how their functioning compares to age-typical children. Specifically, the NMT clinical summary indicates if a child's score is below 65% of age-typical functioning, between 65% and 85% of age-typical functioning, or above 85% of age-typical functioning.

NMT Treatment Recommendations

For each domain of functioning (sensory integration, self-regulation, relational, and cognitive), clinicians work collaboratively with the child's caregiver(s) to articulate specific therapeutic recommendations across levels of the child's ecology, based on the child's assessment results. Therapeutic tasks are designated across three contexts: the individual child, the child's family, and the child's broader therapeutic web, including relevant relationships across school, community, recreational, cultural, and/or spiritual groups. A hallmark of the NMT is the prioritization of this “outside-in” approach to assessment and intervention planning, which emphasizes supports and resources in a child's family and broader web of relationships, rather than just intervening with the individual child. Examples of treatment recommendations at all three levels can be found in Table 1. Individual-level therapeutic recommendations are ranked and prioritized by the NMT metric report as follows: essential treatment recommendations (child is below 65% of age-typical functioning), therapeutic treatment recommendations (child is between 65% and 85% of age-typical functioning), or enrichment treatment recommendations (child is at or above 85% of age-typical functioning). This ranking of treatment recommendations based on a child's level of functioning is a core component of the NMT's sequential approach to treatment planning, and illuminates a child's highest-priority needs. Family and therapeutic web treatment recommendations are not ranked by the NMT metric report; rather, clinicians assign these rankings based on their clinical judgment.

Table 1.

Examples of Treatment Recommendations.

Treatment Domain	Type or Target	Recommendation	Frequency	Ranking
Individual	Primary somatosensory	Mother will use brushing w/ client	3x daily, 5–7 min	Therapeutic
	Sleep hygiene	Client will create and implement structured daytime and bedtime routine	5x/week (school nights)	Therapeutic
	Music – movement	Client will use music/rhythm throughout day	3x daily, 5–7 min	Therapeutic
	Breathing exercises	Client will engage in mindfulness/breathing activities	3x daily, 5–7 min	Therapeutic
	Parallel play – dyadic adult	Caregiver will engage client in relational activities	2–3x per week	Essential
	Parallel play – dyadic adult	Caregiver will engage client in child-led activity	2–3x per week	Enriching
Family	Mother	Provide mother w/ education about attachment-based parenting and regulation strategies	Biweekly	Enriching
Family	Father	Provide father w/ education about attachment-based parenting and regulation strategies	As needed	Therapeutic
Therapeutic web	School	Provide school with NMT and help to identify in class needs	As needed	Enriching
Therapeutic web	Culture / community of faith	Client will continue to attend church and extracurricular activities	Weekly	Enriching

NMT: Neurosequential Model of Therapeutics.

Implementation Fidelity: The Role of Treatment Adherence

Implementation science is focused on the translation of evidence-based research into real-world practice settings, with the ultimate goal of achieving positive outcomes for those being served (Metz et al., 2021). One component of successful intervention implementation is implementation fidelity, or the degree to which a program, intervention, or practice approach is delivered as it was intended by the developers (Carroll et al., 2007; Dusenbury et al., 2003; Slaughter et al., 2015). Implementation fidelity is an essential factor in determining whether the intervention itself caused change in the client, or whether the client's change—or lack thereof—was a result of extraneous factors unrelated to the intervention (Dusenbury et al., 2003; Proctor et al., 2013). Carroll et al. (2007) offer a framework for implementation fidelity that prioritizes intervention adherence as the primary means of measuring implementation fidelity. Adherence refers to the comprehensive and accurate delivery of program content, and the frequency and duration (e.g., dose) with which the intervention is applied. An intervention that is implemented with consistent, comprehensive, and accurate content, and for the specified length and frequency prescribed by the developers, is said to have high adherence, and therefore high implementation fidelity (Carroll et al., 2007). When an intervention is implemented with high fidelity, clinicians can draw sound conclusions about whether the intervention is linked with a change in client outcomes.

Continuous monitoring and measurement of treatment adherence has been identified as a key step in implementation science, and serves to bridge the “implementation gap,” or the difference between “the evidence of what works in theory and what is delivered in practice” (Burke et al., 2012, p. 2). The NIRN recognizes such adherence measurement as an essential component of its “Ongoing Improvement” domain among the core competencies for implementation support practitioners (Metz et al., 2021). The NIRN indicates the necessity of ongoing data collection and analysis when implementing and evaluating the effectiveness of a therapeutic model, and subsequent use of such data to inform agency decision-making, staff training, and organizational policy. Bertram et al. (2015) build on the NIRN implementation framework with the addition of implementation drivers, which represent components of an organization's infrastructure that are necessary for successful implementation fidelity. Measurement of intervention adherence is noted as an essential component of the competency driver, and can provide insight into clinicians’ preparedness and training on the intervention model, as well as the degree to which an agency's organizational culture and infrastructure support high-fidelity implementation of the model (Bertram et al., 2015).

Current Study

Treatment adherence remains a central component of implementation fidelity, and thus a key element in our understanding of the association between the NMT and client outcomes. However, to the authors’ knowledge, there are no NMT outcome studies that evaluate implementation fidelity and adherence. Furthermore, although treatment recommendations are a central component of the NMT, there is a need to empirically study this component of the model, and how treatment adherence may vary depending on the priority level (i.e., ranking) and/or domain of treatment recommendation.

The current study sought to describe rates and patterns of adherence to NMT treatment planning and recommendations in a sample of adopted youth receiving mental health and supportive services from an adoption service provider in a southeastern US state. The following research questions were examined: (1) What level of adherence to NMT treatment recommendations are evidenced in this sample? (2) What are rates of adherence by recommendation rating (e.g., essential, therapeutic, and enriching) and recommendation domain (individual, family, and therapeutic web)? (3) Does adherence to treatment recommendations predict changes in NMT functional outcomes between pretreatment assessment and follow up (T2)?

Method

Agency Context: Harmony Family Center

This study involves secondary data analysis, and was approved by Case Western Reserve University Institutional Review Board. The data used was collected by Harmony Family Center, a Tennessee-based private nonprofit organization specializing in pre-adoption and post-adoption services. Since 2004, Harmony Family Center has administered the state-funded Adoption Support and Preservation Program (ASAP). All adoptive families in Tennessee are eligible to receive services from ASAP. Services are available at no cost or low cost to any state resident who adopts privately, domestically, or internationally (Tennessee Department of Children Services Annual Progress and Services Report, 2015).

The NMT purveyor, the Neurosequential Network, has developed training materials, supervised training experiences, and clinical practice tools to train and certify clinicians in the NMT model. Complete NMT certification involves a multiyear process which includes a Train-the-Trainer and Reflective Supervision component to ensure sustainability. This will typically take an organization five or more years to complete. In the present study, all clinicians at Harmony Family Center underwent the Phase I NMT certification process focused on understanding the impact of trauma and maltreatment on child development and neurobiology. This 120+ hour certification process involves a series of recorded lectures and “case staffing” calls, live monthly training sessions and study groups, and reading materials focused on topics such as basic child development, neurobiology, traumatology, attachment theory, usage of NMT online metric tools, and interpretation of the NMT's functional brain map (Perry, 2020). Participants in the Phase I certification process must have a Master's degree in a social science field or equivalent, and hold a current practice license in their field (e.g., licensed professional counselor, licensed social worker, etc.).

Sample

Between October 2016 and May 2020, a total of 319 children were served by Harmony Family Center. Of those, 255 children had a completed NMT metric and 178 (70% of 255) had both a completed metric and NMT recommendations. These 178 children were the analytical sample used in this study. Reasons children did not have complete data were not specified, but it is possible that children and families discontinued services, or there were other changes in their engagement with Harmony that resulted in incomplete data. Comparisons between the full and analytical sample found no significant differences on most demographic characteristics; the only statistically significant difference between the two samples was that children in the analytical sample were more likely to have been publicly adopted (rather than private/intercountry adoption) than children in the full sample (82.02% vs. 65.25% respectively; χ² = 11.69, p = .001; V = 0.013, 95% confidence interval [CI] [0.170, 0.616]).

NMT Database

A database of NMT treatment recommendations and adherence was created from cases listed in the NMT online portal. Frequencies of all treatment and adherence variables were calculated. Specifically, we examined the following: (1) the number of recommendations by ranking (essential, therapeutic, and enriching) by child; (2), the frequency of adherence ratings by recommendation ranking; (3) the frequency of adherence ratings by recommendation domain; and (4) recommendation ranking and adherence rating (medium or high vs. low) by recommendation domain (individual, family, and community). Harmony Center staff were consulted throughout the analysis process to understand how they implemented the NMT and how they assigned adherence ratings.

Measures

Treatment Recommendations by Ranking and Domain

Clinicians specified treatment recommendations in the following three domains: the individual child, the child's family, and the child's broader therapeutic web (e.g., school and religious community). Individual, family, and therapeutic web treatment recommendations were assigned to one of the three priority levels based on cut-offs suggested by intervention developers: essential treatment recommendations (child is below 65% of age-typical functioning), therapeutic treatment recommendations (child is between 65% and 85% of age-typical functioning), or enrichment treatment recommendations (child is at or above 85% of age-typical functioning). Individual treatment recommendations are assigned a ranking based on output received from the NMT metric report, while the family and therapeutic web recommendations are ranked based on clinician judgment (see Table 1 for examples of treatment recommendations). In discussion with Harmony staff, we found that there was a substantive difference between essential or therapeutic recommendations and enrichment recommendations. Enriching recommendations were often a lower priority and not as clinically significant as essential or therapeutic recommendations. As such, in the regression models we included only cases with essential or therapeutic recommendations.

Majority Adherence to Treatment Recommendations

The NMT assesses treatment adherence as the degree to which clients, families, and clinicians implement the treatment recommendations as they were intended (see Table 1). A system for monitoring adherence to the treatment recommendations was developed by Harmony Family Center staff, in consultation with the purveyor, and was implemented with NMT cases closing in 2018 or later. For each therapeutic recommendation that appears in the child's treatment plan, the clinician assigned to the child's case rated whether the recommendation was completed with low, medium, or high adherence. Adherence levels and follow through on treatment recommendations are discussed and determined through informal discussion between the clinician and client/family on a weekly or monthly basis, depending on the case and the client's regular appointment schedule. Harmony Family Center standardized their adherence as follows: treatment recommendations adhered to 0% to 33% of the time were considered “low” adherence, those completed 34% to 67% of the time were considered “medium” adherence, and those completed 68% to 100% of the time were rated as “high” adherence.

Adherence data were condensed to create a dichotomous variable to capture an overall rating of majority adherence. For children where the majority (50% or more) of the essential or therapeutic recommendations were rated with medium or high levels of adherence (coded as 1), we refer to this as majority adherence. For families where <50% of essential or therapeutic recommendations were implemented with high or medium level of adherence, or adherence was not assessed, this was coded as 0, and we refer to this as low adherence. Those children who did not have any recommendations at the essential or therapeutic levels were coded as missing for adherence. These decisions regarding dichotomization of the adherence variable were informed by both the overall NMT model treatment recommendation ratings and adherence calculations, and conversations with Harmony Center staff. Discussions with the staff revealed ambiguity in differentiating between the medium and high ratings. Thus, from a feasibility standpoint, clinicians indicated that adherence was operationalized in practice as high/medium or low. Based on this feedback from clinicians, we operationalized adherence ratings to reflect the ways in which medium or high adherence ratings were substantively different from low levels of adherence.

NMT Outcomes

This study examined the four primary domains of functioning assessed by the NMT: cognitive functioning, relational functioning, sensory integration and self-regulation. Cognitive functioning refers to a range of executive functions, speech, language, planning for the future, moral reasoning, and abstract thinking. Relational functioning refers to capacity for attachment, empathy, attunement, and mutual connection with others. Sensory integration involves the capacity to integrate and act on sensory input from outside and/or inside the body. Lastly, self-regulation involves the capacity to flexibly respond and modulate one's feelings and reactions to stressors or demands. These outcomes were measured using Part A of the NMT online scoring tool that assesses the child's current neurodevelopment in regard to these four functional domains. In each of the four domains, scores provided by clinicians are auto-compared by the metric tool to age-typical children (Perry, 2009, 2020). For example, a score of .85 on cognitive functioning means that a participant's cognitive functioning score is 85% of a predicted age-typical child's cognitive functioning score. Each of the four NMT outcomes was rated at pretreatment assessment and T2 (time two) based on information entered by the clinician following interviews and observations with caregivers and children. Clinicians are also instructed to use what they have learned from any rating scales completed by the family at intake to inform their metric ratings. Pretreatment assessment NMT scores were included as a covariate in all linear models.

Additional Child-Level and Family-Level Covariates

Child-level covariates were reported by clinicians at pretreatment assessment. Child race was coded as Black, Indigenous, or People of Color (BIPOC) (1) or White (0). Race and ethnicity was recorded by the agency in one variable. One race or ethnicity category was reported for each person. The research team created the BIPOC category from this variable and includes the following groups: American Indian, Asian, bi-racial or multi-racial, Black or African American, Hispanic or Latino, and other. Child sex was recorded by the agency and coded as female (1) or male (0). Child age of adoption was reported in months and years of age. Family-level covariates were also reported by clinicians at pretreatment assessment. Kinship placement was coded as child biologically related to adoptive parent (1) or child not biologically related to adoptive parent (0). Adoptive parent marital status was coded as currently married/partnered (1) or not currently married/partnered (0). Lastly, adoptive parent educational status was coded as education level greater than high school or education level high school or less (0).

Analytic Approach

All analyses were conducted using SAS version 9. We examined our research questions using descriptive analyses and four regression models. Given the sampling frame, only cases with complete data on the focal predictors and outcomes were utilized in these analyses.

Missing Data

In the analytical sample, 149 of the 178 (84%) participants had essential or therapeutic recommendations. Of the 149 participants who had essential or therapeutic recommendations, 116 had both pretreatment assessment and T2 scores. Demographic data was missing for six participants, resulting in a dataset of 110 for the regression models. Listwise deletion was used to eliminate participants with missing data from the regression models.

Regression Models

Using four regression models, change from pretreatment assessment to T2 on the four NMT outcomes was examined (cognitive functioning, relational health, sensory integration, and self-regulation). Four separate models were run for each NMT outcome. For each model, the focal predictors were the number of essential or therapeutic recommendations and majority adherence to treatment recommendations. The following covariates were included in each model: pretreatment assessment NMT outcome score for the given outcome, child race, child sex assigned at birth, child age at adoption, kinship placement, adoptive parent marital status, and adoptive parent education status. Additional caregiver characteristics (race, sex assigned at birth, level of education, and age at the time of adoption) and child characteristics (public vs. private or domestic adoption) were omitted from the models for more parsimonious models; when included in the models, these characteristics did not improve model fit, or add significantly to our understanding of the findings.

Results

Sample Characteristics

Of the 178 children in the sample, 25.29% were BIPOC, 82.02% were publicly adopted, and a little over half the sample were female (51.72%; see Table 2). Few children (6.90%) were biologically related to at least one adoptive parent. The mean age of children when they were adopted was 6.02 years (range = birth to 16 years). Most of the primary caregivers were White (93.10%), female (70.11%), married or partnered (85.88%), and had received education beyond high school (83.13%). The mean age of primary caregivers at pretreatment assessment was 42.04 years (range = 26.93–76.44 years). The mean days from pretreatment assessment to T2 was just over 7 months (M = 219.48 days, SD = 75.71).

Table 2.

Participant Demographics (N = 178).

	N (%)	M (Range)
Child characteristics
Child is BIPOC	44 (25.29)	–
Child assigned female at birth	90 (51.72)	–
Public adoption (private, domestic, or intercountry)	146 (82.02)	–
Kinship placement (biologically related to at least one adoptive parent)	12 (6.90)	–
Mean age at the time of adoption (years)	–	6.02 (0 −16)
Parent characteristics
Caregiver is BIPOC	12 (6.90)	–
Primary caregiver assigned female at birth	122 (70.11)	–
Married or partnered	146 (85.88)	–
Level of education, at least some college	133 (83.13)	–
Mean age at the time of adoption (years)	–	42.04 (26.93 −76.44)
Number of days between pretreatment assessment and T2		219.48 (23–407)

Note. Children reported as BIPOC included the following: 20 bi-racial or multi-racial, 10 Black or African American, 7 Asian, 3 Hispanic or Latino, 3 other, and 1 American Indian. Caregivers reported as BIPOC included the following: 8 Black or African American, 2 other, 1 bi-racial or multi-racial, and 1 Hispanic or Latino. BIPOC: Black, Indigenous, or People of Color.

As reported in Table 3, pretreatment assessment cognitive functioning was, on average, 85% of age-typical functioning (M = 0.85, SD = 0.14) and increased to 86% at T2 (M = 0.86, SD = 0.14), d = 0.05, p = .03). Similarly, pretreatment assessment relational functioning was, on average, 84% of age-typical functioning (M = 0.84, SD = 0.13) and increased to 86% at T2 (M = 0.86, SD = 0.12), d = 0.21, p < .0001. Pretreatment assessment functioning was, on average, highest in sensory integration, 89% (M = 0.89, SD = 0.12) and increased, on average, to 91% (M = 0.91, SD = 0.12), d = 0.15, p < .0001. Pretreatment assessment functioning was lowest for self-regulation, an average of 80% (M = 0.80, SD = 0.13) and increased to 83% at T2 (M = 0.83, SD = 0.13), d = 0.23, p < .0001.

Table 3.

NMT Outcomes: Pretreatment Assessment and Time Two (T2) Paired t-Tests and Effect Sizes.

	Pretreatment Assessment			Time Two (T2)							95% CI
	N	Mean	Standard Deviation	N	Mean	Standard Deviation	Pairedt-Test^a	DF	Pr > \|t\|	Effect Size^b	LL	UL
Cognitive	178	0.85	0.14	138	0.86	0.14	2.20	137	0.0293	0.046	0.046	0.269
Relational	178	0.84	0.13	138	0.86	0.12	5.07	137	<.0001	0.210	0.013	0.432
Sensory integration	178	0.89	0.12	138	0.91	0.12	6.00	137	<.0001	0.148	0.075	0.370
Self-regulation	178	0.80	0.13	138	0.83	0.13	6.56	137	<.0001	0.232	0.009	0.455

t-Value for paired t-tests comparing scores from pretreatment assessment and T2.

Cohen's d effect sizes based on comparisons between pretreatment assessment and T2.

CI: confidence interval; NMT: Neurosequential Model of Therapeutics; DF: degrees of freedom; LL: lower limit; UL: upper limit.

Majority Adherence by Treatment Recommendation Ranking and Domain

A total of 2,633 recommendations were prescribed to the 178 participants. Overall, most of the treatment recommendations were carried out with majority adherence. Of the 410 essential recommendations, 253 (61.71%) were rated with majority adherence; of the 1,232 therapeutic recommendations, 769 (62.42%) were rated majority adherence; and 732 (73.86%) of the 991 enriching recommendations were rated with majority adherence (see Table 4). The average number of essential or therapeutic recommendations was 9.22 (SD = 6.75) per participant, and on average 64.77% (SD = 30.47%) of the essential or therapeutic recommendations were implemented with majority adherence. These summary variables were used in the regression models.

Table 4.

Recommendation Adherence by Ranking (N = 2,633).

At the Recommendation Level	Recommendation Ranking n (%)
At the Recommendation Level	Essential	Therapeutic	Enriching
Recommendation adherence
Low adherence	157 (38.29%)	463 (37.58%)	259 (26.14%)
Majority adherence^a	253 (61.71%)	769 (62.42%)	732 (73.86%)
Summary variables	N	M (SD)
Number of essential or therapeutic recommendations per participant	178	9.22 (6.75)
Percentage of essential or therapeutic recommendations implemented with majority adherence^a	149	64.77% (30.47%)

Majority adherence is indicated by 50% or more of the essential or therapeutic recommendations rated with medium or high levels of adherence. SD: standard deviation

Multivariate Models: Majority Adherence and NMT Outcomes

All four of the overall regression models were statistically significant: Model 1, cognitive (R2 = 0.77, F (8, 101) = 41.38, p < .0001); Model 2, relational (R2 = 0.65, F (8, 101) = 23.39, p < .0001); Model 3, sensory integration (R2 = 0.81, F (8, 101) = 59.09, p < .0001); and Model 4, self-regulation (R2 = 0.64, F (8, 101) = 24.70, p < .0001). The number of essential or therapeutic recommendations significantly predicted T2 sensory integration scores (b = 0.002, t(101) = 2.06, p < .05); higher numbers of recommendations were associated with higher sensory integration scores at T2, after controlling for pretreatment assessment sensory integration scores (see Table 5). Majority adherence was statistically significant for all four models; a majority adherence rating was associated with higher T2 functioning on each focal outcome (cognitive: b = 0.038, t = 2.297, p < .01; relational: b = 0.051, t = 3.14, p < .001; sensory integration: b = 0.029, t = 2.50, p < .05; self-regulation: b = 0.038, t = 2.22, p < .05), controlling for pretreatment assessment scores. Pretreatment assessment scores on all outcomes were significantly associated with T2 outcomes such that pretreatment assessment scores were associated with higher functioning at T2 (cognitive: b = 0.851, t = 16.05, p < .001; relational: b = 0.732, t = 11.28, p < .001; sensory integration: b = 0.912, t = 19.25, p < .001; self-regulation: b = 0.866, t = 11.28, p < .001). Regarding covariates, children who identified as female demonstrated higher cognitive functioning (b = 0.032, t = 2.28, p < .05) at T2 than children identified as male, controlling for all other variables in the model.

Table 5.

Recommendation Adherence and NMT Outcomes.

	Model 1: Cognitive				Model 2: Relational				Model 3: Sensory Integration				Model 4: Self-regulation
	b	SE b	t-Value	p-Value	b	SE b	t-Value	p-Value	b	SE b	t-value	p-Value	b	SE b	t-value	p-Value
Intercept	0.116	0.056	2.07	.042	0.235	0.063	3.70	.000	0.088	0.049	1.78	.079	0.118	0.072	1.64	.103
Child race (BIPOC = 1)	−0.012	0.016	−0.75	.456	−0.011	0.016	−0.67	.502	−0.006	0.012	−0.54	.588	−0.009	0.017	−0.54	.588
Child sex (female = 1)	0.032	0.014	2.28	.025	0.021	0.013	1.59	.116	0.011	0.010	1.09	.279	0.019	0.015	1.29	.202
Child adoption age (years)	0.001	0.002	0.81	.421	0.000	0.002	0.03	.979	−0.001	0.001	−0.88	.380	0.000	0.002	−0.10	.917
Kinship placement (yes = 1)	−0.019	0.034	−0.57	.568	−0.047	0.033	−1.41	.161	−0.035	0.024	−1.45	.149	−0.064	0.035	−1.79	.076
Caregiver married/partnered (yes = 1)	−0.038	0.020	−1.90	.060	−0.028	0.019	−1.43	.155	−0.020	0.014	−1.40	.165	−0.031	0.021	−1.50	.137
Pretreatment assessment functioning^a	0.851	0.053	16.05	<.0001	0.732	0.065	11.28	<.0001	0.912	0.047	19.25	<.0001	0.866	0.077	11.28	<.0001
Number of essential/therapeutic recs	0.001	0.001	0.49	.625	0.000	0.001	−0.10	.918	0.002	0.001	2.06	.042	0.003	0.002	1.67	.099
Majority adherence (yes = 1)	0.038	0.017	2.29	.024	0.051	0.016	3.14	.002	0.029	0.012	2.50	.014	0.038	0.017	2.22	.029
Model statistics	F value = 41.38, p < .0001; Adj R-Square = 0.77			F value = 23.39, p < .0001; Adj R-Square = 0.65			F value = 59.09, p < .0001; Adj R-Square = 0.81			F value = 24.70, p < .0001; Adj R-Square = 0.64

Pretreatment assessment functioning: Model 1 pretreatment assessment cognitive functioning; Model 2 pretreatment assessment relational functioning; Model 3 pretreatment assessment sensory integration functioning; and Model 4 pretreatment assessment self-regulation functioning.

Bold text = statistically significant at p = .05 level.

BIPOC: Black, Indigenous, or People of Color; SE: standard error.

Discussion and Applications to Practice

Although research on the NMT has grown exponentially in the past decade, the majority of studies have focused on setting-specific implementation strategies (Cox et al., 2021; de Nooyer & Lingard, 2017; Frederico et al., 2019; Freedle & Slagle, 2018) and assessment of youth outcomes (Barfield et al., 2012; Hambrick et al., 2018a, 2018b). The current study makes a novel contribution to this body of work by examining an understudied aspect of the NMT—adherence to treatment recommendations—and contextualizing client outcomes within this lens.

Overall, this study revealed higher participant functioning across all four functional domains at T2, as compared to pretreatment assessment scores. These findings support prior research that has found improved socioemotional and behavioral outcomes for children and youth following implementation of the NMT (Barfield et al., 2012; Cox et al., 2021; Keegan et al., 2020), while providing novel insight into this model's implementation with adopted youth in particular. It is not uncommon for adopted youth—particularly those adopted from foster care or international institutional care—to be at risk for clinical-level difficulties with interpersonal, executive, and/or regulatory functioning (Brodzinsky et al., 2022; Herzberg et al., 2018; Smith et al., 2022). When families first seek supportive services to navigate these concerns, their children's challenges may feel overwhelming or exceedingly difficult to manage (Vasquez & Stensland, 2016). Families and youth alike are often seeking concrete action steps that will help bring some relief to the day-to-day challenges that brought them to seek services in the first place (Saxe et al., 2015). The current study provides evidence of the NMT as one possible approach to clinical assessment and treatment planning that may enhance the well-being and functioning of adopted youth with mental health challenges.

Although increases in functioning were observed across all domains in this study, it is also notable that children's pretreatment assessment scores were reasonably high, ranging from 80% (self-regulation) to 89% (sensory integration) of age-typical functioning. A key consideration when evaluating this finding is the inherently subjective nature of clinician ratings, and the strengths-based perspective from which the NMT operates. During the NMT certification training, clinicians are advised to approach the metric scoring process from a strength orientation, wherein each child's ratings begin at the age-appropriate score, and the clinician moderates the child's scores based on their observations, formalized assessment data (if available), and their own clinical judgment. It is possible that the subjective and strengths-focused nature of this process may lend itself to rating children with higher functional outcome scores than expected, but such conclusions are beyond the scope of this investigation. Exploring the perspectives and clinical reasoning strategies that NMT-trained clinicians use when completing metrics would offer an interesting direction for future research, and additional insight into NMT scoring procedures.

A central finding of this study is the predictive relationship between majority adherence to treatment recommendations and child outcomes across all NMT core domains of functioning: sensory integration, self-regulation, relational, and cognitive. These results align with the broader body of implementation science literature on fidelity and treatment adherence across psychotherapeutic models, which asserts that adherence is a critical element for intervention success and positive changes in child mental and behavioral outcomes (Collyer et al., 2020; Dvorsky et al., 2021; Fauskanger Bjaastad et al., 2018; Staudt, 2007). For example, a recent meta-analysis of 35 studies found that there was a significant positive relationship between clinician adherence to their treatment model of choice and reduced symptomatology in child and adolescent clients, with 83% of included studies reporting a significant adherence/child outcome relationship (Collyer et al., 2020). Thus, the current study adds to a growing body of evidence that underscores the importance of defining, monitoring, and evaluating treatment adherence in both clinical practice and intervention research.

This finding also has important implications for clinicians and researchers working specifically with the NMT. In particular, the strong relationship between adherence to treatment recommendations and client outcomes suggests that this should perhaps be a focal topic early in the NMT certification training process. Much of the current Phase I NMT certification training content is focused on establishing a solid base of knowledge on the neurobiology of trauma and the stress response system, and learning how to complete, interpret, and use the NMT metric and clinical practice tools (Perry, 2020). Translating such knowledge into clinical improvements for clients involves a variety of factors though, and the present study suggests that it may be prudent to begin sharing adherence/child outcome data with trainees, and strategies for defining, monitoring, and evaluating adherence in their work at the outset of the NMT certification process. After the completion of this project, additional levels of certification were developed, including the Phase II—Train-the-Trainer and Phase III—Mentor tracks, which focus on more advanced aspects of system implementation, staff support, and supervision. However, these certification levels have not been evaluated, and were not a component of the current study. Ultimately, the multiyear NMT certification process can present a challenge with this approach, as pursuit of additional certifications requires sustained time, effort, and financial commitment from clinicians and organizations.

The only significant covariates across models were pretreatment assessment functioning on each of the NMT outcomes and child age at adoption. Given the ways in which NMT is a neurobiologically informed assessment of child functioning (i.e., ways in which adversity and relational supports have each fundamentally shaped a child's development to date; Perry, 2009), we found that pretreatment assessment functioning in each focal outcome was predictive of T2 functioning. The significant effect of child age at adoption is also supported by previous literature; later age of adoption has consistently been found to be predictive of poorer psychosocial and behavioral outcomes (e.g., Julian, 2013). Later-adopted children may experience more placement instability prior to adoption resulting in disrupted social relationships with foster parents/siblings and poorer relational health (Julian, 2013). Alternatively, later-adopted children could have entered into the foster care or adoption system following more exposure to adversity, which could also contribute to lower functioning across all NMT focal outcomes (Hambrick et al., 2019; Julian, 2013).

While this study has several strengths, some study limitations warrant consideration. First, this was implemented in one state, in a long-standing program that has provided a wide array of services for adoptive families. The staff implementing the NMT approach were trained by the purveyor, and have also received training in providing an array of evidenced-based interventions. Service providers who are considering implementing the NMT approach should take this into consideration in their implementation planning. Regarding the NMT approach, some cautions and caveats are also in order. A strength of the NMT approach is the flexibility practitioners have in using the tools they have available. There is not, for example, a prescribed tool to use for understanding child behavioral issues, or trauma exposure or experiences. While this allows for the approach to be used in a variety of settings, it also means that while the training and specific metric items are the same at all sites, the sources of information that inform scoring will vary by location, and may limit comparability across sites. Another limitation is the subjective nature of using a single clinician's judgment when completing the metric and adherence ratings. Clinicians are trained to use their best judgment when conducting the assessment, and gathering information about the client, which also limits comparability across sites. This may be particularly challenging when working with adoptive families who may have limited knowledge about the child's early years prior to adoption. In addition, the priority level of the family and therapeutic web recommendations are a subjective clinician determination, yet the individual-level recommendations are determined by the metric, and the adherence rating is also a subjective clinician decision. The clinician training and clinician monitoring is designed to address the potential effect of clinician judgment on intervention fidelity, but this has not been rigorously tested. Lastly, another limitation of this data is the lack of context around the number of treatment recommendations each child has. It is possible that a higher number of treatment recommendations could be linked with a higher level of child vulnerability or need. Or, it could alternatively be an indicator of a particularly engaged family or clinician. Such nuance is not provided in this data, and could impact our understanding/interpretation of the number of treatment recommendations provided to each child.

Given the predictive power that adherence played in child outcomes in this study, a key practice implication from this study is the importance of addressing and mitigating barriers to treatment recommendation adherence. Barriers to adherence in child therapeutic treatment can include unrealistic or inappropriate expectations about the therapeutic process, client motivation, structural barriers (e.g., scheduling conflicts and transportation challenges), cultural factors, financial issues (e.g., insurance, inability to pay co-pays), and untreated family stressors that interfere with therapeutic engagement, such as parental substance abuse or mental health difficulties (Becker et al., 2015; Gearing et al., 2012). For example, Gearing et al. (2012) research found that one of the most common barriers to treatment adherence was caregiver misunderstanding about family involvement in the psychotherapeutic treatment process for child/youth clients, with many parents not expecting that their child's therapy would also require substantial involvement from the entire family.

Working with adoptive families at the beginning of treatment to clarify expectations about the therapeutic process and the NMT, identify possible barriers to treatment adherence, and problem-solve together are essential first steps. Becker et al. (2015) offer a number of evidence-based recommendations on the topic of barrier reduction and treatment adherence for children receiving mental health services. Examples include providing caregivers and children with psychoeducation about the treatment process, time commitment, and expectations for all members of the treatment team, administering a “barrier questionnaire” that explores any potential barriers to treatment adherence and collaboratively exploring solutions, and implementing “accessibility promotion” strategies at the agency, including providing transportation, childcare, and convenient access to treatment (Becker et al., 2015). Additionally, alignment between parental priorities for treatment and service providers’ treatment priorities may also shape adherence patterns. When parents, children, and service providers work together to collaboratively design treatment goals that are high priority for families—and have feasible treatment activities associated with them—adherence is more likely (Saxe et al., 2015). The NMT offers such flexibility in its approach; while metric scores and assessment data provide a general guide and direction for treatment goals, clinicians are encouraged to exercise their own clinical judgment to help families prioritize their most pressing needs, and reasonable treatment activities that align with the family's resources, capacities, and limitations.

Findings from this study also have implications for the NMT clinical training process. Incorporating data on treatment recommendation adherence and client outcomes into NMT Phase I clinical training modules could offer one step to continue building competence and efficacy in NMT-trained clinicians. Deepening clinicians’ understanding of how different elements of the NMT—such as adherence—directly link to client outcomes may enhance their capacity to implement the model with high fidelity. Beyond the initial training/certification modules, ongoing supervision and consultation related to treatment recommendation adherence is also recommended. Research on other models of practice, such as multisystemic family therapy, has shown that adherence to treatment protocols is predicted by clinicians’ ongoing, high-quality supervision and expert consultation about therapeutic implementation (Schoenwald et al., 2000, 2004). On a global scale, the NMT already has a sophisticated in-service training program in place for its certified clinicians, which includes a series of ongoing monthly virtual meetings that showcase the work of agencies and professionals implementing the model. Emphasis is often placed on factors such as agency-level implementation strategies, metric completion and interpretation, and client case studies. Requiring the presenting agencies and clinicians to report data on treatment recommendation adherence would help contextualize child outcomes at their agencies, while also reinforcing the necessity of incorporating adherence into discussions at all levels of NMT implementation (family and client discussions, case staffings, and agency evaluation/research).

Research on agency-level approaches to defining and monitoring adherence to NMT treatment recommendations is also called for. A significant focus of the NMT certification and maintenance process is clinicians’ fidelity to the NMT assessment approach, including how accurately they can use client case data to complete an NMT metric with precision, and how closely their assessment matches that of the purveyor. Less is known about how individual agencies and clinicians are defining treatment recommendation adherence in their specific settings, and how this may influence child outcomes. Research designed to build and test a consistent framework for measuring adherence within the NMT could advance child outcome research with this model, promote implementation fidelity, and add nuance to evaluation research that examines the effectiveness of the NMT.

Treatment adherence can provide an important context for interpreting the relationship between treatment activities and client outcomes. Ongoing measurement of client and provider adherence throughout the treatment process has been identified as a key step to bridging the “implementation gap” (Burke et al., 2012), while also demonstrating the importance of accurate delivery of treatment activities, in the frequency and duration they were prescribed. Findings from the present study illustrate the powerful role that adherence can play in shaping outcomes among adopted youth receiving supportive and mental health services. This study also calls attention to issues specific to clinicians and agencies working with the NMT, including addressing treatment adherence in NMT clinician training and reducing barriers to treatment adherence for families and youth. Future research should continue building our understanding of NMT recommendation rankings and domains, processes, and patterns of adherence, to enhance implementation fidelity for this growing approach to trauma assessment and treatment.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Kylie E. Evans

Kevin White

References

Barfield

Gaskill

Dobson

Perry

B. D.

(2012). Neurosequential Model of Therapeutics in a therapeutic preschool: Implications for work with children with complex neuropsychiatric problems. International Journal of Play Therapy, 21, 33–44. https://doi.org/10.1037/a0025955

Becker

K. D.

Lee

B. R.

Daleiden

E. L.

Lindsey

Brandt

N. E.

Chorpita

B. F.

(2015). The common elements of engagement in children's mental health services: Which elements for which outcomes? Journal of Clinical Child & Adolescent Psychology, 44(1), 30–43. https://doi.org/10.1080/15374416.2013.814543

Bertram

R. M.

Blase

K. A.

Fixsen

D. L.

(2015). Improving programs and outcomes: Implementation frameworks and organization change. Research on Social Work Practice, 25(4), 477–487. https://doi.org/10.1177/1049731514537687

Brodzinsky

Gunnar

Palacios

(2022). Adoption and trauma: Risks, recovery, and the lived experience of adoption. Child Abuse & Neglect, 130, 105309. https://doi.org/10.1016/j.chiabu.2021.105309

Burke

Morris

McGarrigle

(2012). An introductory guide to implementation: Terms, concepts and frameworks. Centre for Effective Services. https://hdl-handle-net.web.bisu.edu.cn/10147/306846

Carroll

Patterson

Wood

Booth

Rick

Balain

(2007). A conceptual framework for implementation fidelity. Implementation Science, 2(1), 1–9. https://doi.org/10.1186/1748-5908-2-40

Collyer

Eisler

Woolgar

(2020). Systematic literature review and meta-analysis of the relationship between adherence, competence and outcome in psychotherapy for children and adolescents. European Child & Adolescent Psychiatry, 29(4), 417–431. https://doi.org/10.1007/s00787-018-1265-2

Cox

Perry

B. D.

Frederico

(2021). Resourcing the system and enhancing relationships: Pathways to positive outcomes for children impacted by abuse and neglect. Child Welfare, 98(6), 177–202.

de Nooyer

K. M.

Lingard

M. W.

(2017). Applying principles of the neurosequential model of therapeutics across an adolescent day program and inpatient unit. Australian Psychiatry, 25(2), 150–153. https://doi.org/10.1177/1039856216658824

10.

Dusenbury

Brannigan

Falco

Hansen

W. B.

(2003). A review of research on fidelity of implementation: Implications for drug abuse prevention in school settings. Health Education Research, 18(2), 237–256. https://doi.org/10.1093/her/18.2.237

11.

Dvorsky

M. R.

Friedman

L. M.

Spiess

Pfiffner

L. J.

(2021). Patterns of parental adherence and the association to child and parenting outcomes following a multicomponent school-home intervention for youth with ADHD. Behavior Therapy, 52(3), 745–760. https://doi.org/10.1016/j.beth.2020.09.007

12.

Fauskanger Bjaastad

Henningsen Wergeland

G. J.

Mowatt Haugland

B. S.

Gjestad

Havik

O. E.

Heiervang

E. R.

Öst

L. G.

(2018). Do clinical experience, formal cognitive behavioural therapy training, adherence, and competence predict outcome in cognitive behavioural therapy for anxiety disorders in youth? Clinical Psychology & Psychotherapy, 25(6), 865–877. https://doi.org/10.1002/cpp.2321

13.

Frederico

Jackson

Black

Pawsey

Cox

(2019). Take two–implementing a therapeutic service for children who have experienced abuse and neglect: Beyond evidence-informed practice. Child Abuse Review, 28(3), 225–239. https://doi.org/10.1002/car.2563

14.

Freedle

L. R.

Slagle

(2018). Application of the Neurosequential Model of Therapeutics (NMT) in an integrative outdoor behavioral healthcare program for adolescents and young adults. In Perry

Griffin

Maikoetter

Graner

Rosenfelt

Perry

(Eds.), Proceedings of the 2nd International Neurosequential Model Symposium: Banff, Alberta, Canada (pp. 20–26). The ChildTrauma Academy Press.

15.

Gearing

R. E.

Schwalbe

C. S.

Short

K. D.

(2012). Adolescent adherence to psychosocial treatment: Mental health clinicians’ perspectives on barriers and promoters. Psychotherapy Research, 22(3), 317–326. https://doi.org/10.1080/10503307.2011.653996

16.

Hambrick

E. P.

Brawner

T. W.

Perry

B. D.

Brandt

Hofmeister

Collins

J. O.

(2019). Beyond the ACE score: Examining relationships between timing of developmental adversity, relational health and developmental outcomes in children. Archives of Psychiatric Nursing, 33(3), 238–247. https://doi.org/10.1016/j.apnu.2018.11.001

17.

Hambrick

E. P.

Brawner

T. W.

Perry

B. D.

(2018a). Examining developmental adversity and connectedness in child welfare-involved children. Children Australia, 43(2), 105–115. https://doi.org/10.1017/cha.2018.21

18.

Hambrick

E. P.

Brawner

T. W.

Perry

B. D.

Wang

E. Y.

Griffin

DeMarco

Capparelli

Grove

Maikoetter

O’Malley

Paxton

Freedle

Friedman

Mackenzie

Perry

K. M.

Cudney

Hartman

Kuh

Morris

Polales

Strother

(2018b). Restraint and critical incident reduction following introduction of the Neurosequential Model of Therapeutics (NMT). Residential Treatment for Children & Youth, 35(1), 2–23. https://doi.org/10.1080/0886571X.2018.1425651

19.

Herzberg

M. P.

Hodel

A. S.

Cowell

R. A.

Hunt

R. H.

Gunnar

M. R.

Thomas

K. M.

(2018). Risk taking, decision-making, and brain volume in youth adopted internationally from institutional care. Neuropsychologia, 119, 262–270. https://doi.org/10.1016/j.neuropsychologia.2018.08.022

20.

Julian

(2013). Age at adoption from institutional care as a window into the lasting effects of early experiences. Clinical Child and Family Psychology Review, 16(2), 101–145. https://doi.org/10.1007/s10567-013-0130-6

21.

Keegan

Cartmel

Harris

(2020). I’ve changed, I’m smarter: Empowering youth to thrive neurosequential approach to employment, education and training outcomes for youth. Children and Teenagers, 3(1), 16–32. https://doi.org/10.22158/ct.v3n1p16

22.

Metz

Albers

Burke

Bartley

Louison

Ward

Farley

(2021). Implementation practice in human service systems: Understanding the principles and competencies of professionals who support implementation. Human Service Organizations: Management, Leadership & Governance, 45(3), 1–22. https://doi.org/10.1080/23303131.2021.1895401

23.

Perry

B. D.

(2006). Applying principles of neurodevelopment to clinical work with maltreated and traumatized children: The Neurosequential Model of Therapeutics. In Webb

N. B.

(Ed.), Working with traumatized youth in child welfare (pp. 27–52). The Guilford Press. https://doi.org/10.1007/s10615-006-0072-2.

24.

Perry

B. D.

(2009). Examining child maltreatment through a neurodevelopmental lens: Clinical applications of the Neurosequential Model of Therapeutics. Journal of Loss and Trauma, 14(4), 240–255. https://doi.org/10.1080/15325020903004350

25.

Perry

B. D.

(2020). The Neurosequential Model: A developmentally sensitive, neuroscience-informed approach to clinical problem solving. In Mitchell

Tucci

Tronick

(Eds.), The handbook of therapeutic care for children: Evidence-informed approaches to working with traumatized children and adolescents in foster, kinship, and adoptive care (pp. 137–158). Jessica Kingsley Publishers.

26.

Perry

B. D.

Griffin

Davis

Perry

J. A.

Perry

R. D.

(2018). The impact of neglect, trauma, and maltreatment on neurodevelopment: Implications for juvenile justice practice, programs, and policy. In Beech

A. R.

Carter

A. J.

Mann

R. E.

Rotshtein

(Eds.), The Wiley Blackwell handbook of forensic neuroscience (pp. 813–835). John Wiley & Sons, Ltd. https://doi.org/10.1002/9781118650868.ch31.

27.

Proctor

E. K.

Powell

B. J.

McMillen

J. C.

(2013). Implementation strategies: Recommendations for specifying and reporting. Implementation Science, 8(1), 1–11. https://doi.org/10.1186/1748-5908-8-139

28.

Saxe

G. N.

Ellis

B. H.

Brown

A. D.

(2015). Trauma systems therapy for children and teens. Guilford Publications.

29.

Schoenwald

S. K.

Henggeler

S. W.

Brondino

M. J.

Rowland

M. D.

(2000). Multisystemic therapy: Monitoring treatment fidelity. Family Process, 39(1), 83–103. https://doi.org/10.1111/j.1545-5300.2000.39109.x

30.

Schoenwald

S. K.

Sheidow

A. J.

Letourneau

E. J.

(2004). Toward effective quality assurance in evidence-based practice: Links between expert consultation, therapist fidelity, and child outcomes. Journal of Clinical Child and Adolescent Psychology, 33(1), 94–104. https://doi.org/10.1207/s15374424jccp3301_10

31.

Slaughter

S. E.

Hill

J. N.

Snelgrove-Clarke

(2015). What is the extent and quality of documentation and reporting of fidelity to implementation strategies: A scoping review. Implementation Science, 10(1), 1–12. https://doi.org/10.1186/s13012-015-0320-3

32.

Smith

Durham

Beatty

Price-Cameron

Kartusch

Shlonsky

Browne

(2022). Trajectories of psychosocial functioning and attachment behaviors among children adopted in the Ontario child welfare system. Journal of Public Child Welfare, 16(3), 321–348. https://doi.org/10.1080/15548732.2021.1881690

33.

Staudt

(2007). Treatment engagement with caregivers of at-risk children: Gaps in research and conceptualization. Journal of Child and Family Studies, 16(2), 183–196. https://doi.org/10.1007/s10826-006-9077-2

34.

Tennessee Department of Children Services Annual Progress and Services Report (2015). https://www.tn.gov/content/dam/tn/dcs/documents/quality_improvement/cfsr/TN_DCS_APSR_FY_2015.pdf

35.

Vasquez

Stensland

(2016). Adopted children with reactive attachment disorder: A qualitative study on family processes. Clinical Social Work Journal, 44(3), 319–332. https://doi.org/10.1007/s10615-015-0560-3