Investigation of bias in the automated assessment of school violence.

OBJECTIVES: Natural language processing and machine learning have the potential to lead to biased predictions. We designed a novel Automated RIsk Assessment (ARIA) machine learning algorithm that assesses risk of violence and aggression in adolescents using natural language processing of transcribed student interviews. This work evaluated the possible sources of bias in the study design and the algorithm, tested how much of a prediction was explained by demographic covariates, and investigated the misclassifications based on demographic variables. METHODS: We recruited students 10-18 years of age and enrolled in middle or high schools in Ohio, Kentucky, Indiana, and Tennessee. The reference standard outcome was determined by a forensic psychiatrist as either a "high" or "low" risk level. ARIA used L2-regularized logistic regression to predict a risk level for each student using contextual and semantic features. We conducted three analyses: a PROBAST analysis of risk in study design; analysis of demographic variables as covariates; and a prediction analysis. Covariates were included in the linear regression analyses and comprised of race, sex, ethnicity, household education, annual household income, age at the time of visit, and utilization of public assistance. RESULTS: We recruited 412 students from 204 schools. ARIA performed with an AUC of 0.92, sensitivity of 71%, NPV of 77%, and specificity of 95%. Of these, 387 students with complete demographic information were included in the analysis. Individual linear regressions resulted in a coefficient of determination less than 0.08 across all demographic variables. When using all demographic variables to predict ARIA's risk assessment score, the multiple linear regression model resulted in a coefficient of determination of 0.189. ARIA performed with a lower False Negative Rate (FNR) of 15.2% (CI [0 - 40]) for the Black subgroup and 12.7%, CI [0 - 41.4] for Other races, compared to an FNR of 26.1% (CI [14.1 - 41.8]) in the White subgroup. CONCLUSIONS: Bias assessment is needed to address shortcomings within machine learning. In our work, student race, ethnicity, sex, use of public assistance, and annual household income did not explain ARIA's risk assessment score of students. ARIA will continue to be evaluated regularly with increased subject recruitment.

Management of Emotion Dysregulation and Outbursts in Children and Adolescents.

PURPOSE OF REVIEW: Emotion dysregulation and outbursts are very common reasons for referral to child and adolescent mental health services and a frequent cause of admission to hospitals and residential programs. Symptoms of emotion dysregulation and outburst are transdiagnostic, associated with many disorders, have the potential to cause severe impairment and their management presents a major challenge in clinical practice. RECENT FINDINGS: There are an increasing number of psychosocial interventions that demonstrate promise in improving emotion dysregulation and outbursts. Acute care systems to manage the most severely ill patients have limited best practice guidelines but program advancements indicate opportunities to improve care models. Pharmacotherapy may be of assistance to psychosocial interventions but must be used with caution due to potential adverse effects. Much remains to be discovered however evidence informed, targeted treatments for specific populations show potential for future improvements in outcomes.

A Large Naturalistic Study on the BRACHA: Confirmation of the Predictive Validity.

Aggression is a major challenge on child/adolescent inpatient psychiatric units. A screening instrument to accurately identify risk is urgently needed. To determine the predictive validity of the Brief Rating of Aggression by Children and Adolescents (BRACHA). Prospective cohort study. BRACHA is administered by clinical staff in the emergency department (ED) prior to inpatient psychiatric admission. A consecutive sample of 10,054 admitted patients from 2010-2021. No patients refused screening nor were excluded. BRACHA administered to patients in the ED prior to admission at Cincinnati Children's Hospital Medical Center (CCHMC). Patient behavioral outcomes measured by Overt Aggression Scale (OAS), categorizing aggression as verbal or physical, then as towards self, others, or objects. Female patients comprised 53.6% (n = 5,386) of the sample. Most patients were white (n = 6,556, 65.2%). Patients ranged in age from 4 to 18 years, with a mean age of 13.6 ± 3.1 years. A single biological parent (n = 5,317, 52.9%) was the predominant living arrangement among patients. The Area Under the Curve (AUC), as an assessment of predictive validity across all possible cut-offs of BRACHA scores ranged from 0.640 (aggression to self) to 0.758 (physical aggression towards others). Our findings support the BRACHA as a useful predictive instrument for aggression in inpatient psychiatric admissions from ED regardless of length of stay. Treating staff are then able to immediately classify risk level and inform care plans for all lengths of hospitalization. Applies to potential risk for aggression, except for self-aggression. Future data analyses will evaluate demographic factors to determine which improve predictive power of the BRACHA and can be used to create a BRACHA calculator. To our knowledge, this naturalistic outcomes study is one of the largest in psychiatry. The BRACHA will continue to be studied to evaluate risk for aggression on inpatient units and aim to assist in keeping unit staff and patients safe.

Lessons Learned to Building Stakeholder Engagement during the Initial Stages of Pragmatic Research Development and Implementation.

Research does not occur in a vacuum. Effective stakeholder engagement occurs on several levels, including outside influence and cooperation inside the institution. Little guidance around designing and implementing pragmatic mental health research exists. The following paper outlines lessons learned during the initial stages of research design and implementation for a project focused on mental health treatment outcomes.

Automated Risk Assessment for School Violence: a Pilot Study.

School violence has increased over the past ten years. This study evaluated students using a more standard and sensitive method to help identify students who are at high risk for school violence. 103 participants were recruited through Cincinnati Children's Hospital Medical Center (CCHMC) from psychiatry outpatient clinics, the inpatient units, and the emergency department. Participants (ages 12-18) were active students in 74 traditional schools (i.e. non-online education). Collateral information was gathered from guardians before participants were evaluated. School risk evaluations were performed with each participant, and audio recordings from the evaluations were later transcribed and manually annotated. The BRACHA (School Version) and the School Safety Scale (SSS), both 14-item scales, were used. A template of open-ended questions was also used. This analysis included 103 participants who were recruited from 74 different schools. Of the 103 students evaluated, 55 were found to be moderate to high risk and 48 were found to be low risk based on the paper risk assessments including the BRACHA and SSS. Both the BRACHA and the SSS were highly correlated with risk of violence to others (Pearson correlations>0.82). There were significant differences in BRACHA and SSS total scores between low risk and high risk to others groups (p-values <0.001 under unpaired t-test). In particular, there were significant differences in individual SSS items between the two groups (p-value <0.001). Of these items, Previous Violent Behavior (Pearson Correlation = 0.80), Impulsivity (0.69), School Problems (0.64), and Negative Attitudes (0.61) were positively correlated with risk to others. The novel machine learning algorithm achieved an AUC of 91.02% when using the interview content to predict risk of school violence, and the AUC increased to 91.45% when demographic and socioeconomic data were added. Our study indicates that the BRACHA and SSS are clinically useful for assessing risk for school violence. The machine learning algorithm was highly accurate in assessing school violence risk.

Suicidality in psychiatrically hospitalized children and adolescents: Demographics, treatment, and outcome.

BACKGROUND: Despite the high prevalence of suicidality in psychiatrically hospitalized youth, its risk factors and impact on inpatient psychopharmacologic treatment are unknown. We identified characteristics associated with suicidality in psychiatrically hospitalized youth and determined the association of suicidality with subsequent psychopharmacologic interventions. METHODS: Medical records from consecutive psychiatric admissions to a large, acute care, urban, pediatric hospital were analyzed retrospectively (N = 1,309). Demographic, clinical, and treatment-related features of suicidal and nonsuicidal youth were characterized. Logistic regression identified predictors of suicidality, and multiple comparison analyses evaluated the association between suicidality and changes to antidepressant prescribing during inpatient course. RESULTS: Compared with nonsuicidal patients, inpatients who were suicidal were more likely to have a mood disorder or posttraumatic stress disorder, as well as Cannabis and alcohol use, were more commonly girls, and at least 13 years of age (all P ≤ .05). Hospitalization was shorter for suicidal patients, was more likely to be associated with antidepressant treatment (P ≤ .001), and among suicidal patients prescribed antidepressants at the time of admission, was associated with a greater likelihood of changing antidepressant treatment compared with nonsuicidal inpatients (P ≤ .05). CONCLUSIONS: These findings reveal differences between suicidal and nonsuicidal psychiatrically hospitalized youth and suggest that suicidality is associated with specific pharmacologic treatment approaches within this population.

A Pilot Study on Developing a Standardized and Sensitive School Violence Risk Assessment with Manual Annotation.

School violence has increased over the past decade and innovative, sensitive, and standardized approaches to assess school violence risk are needed. In our current feasibility study, we initialized a standardized, sensitive, and rapid school violence risk approach with manual annotation. Manual annotation is the process of analyzing a student's transcribed interview to extract relevant information (e.g., key words) to school violence risk levels that are associated with students' behaviors, attitudes, feelings, use of technology (social media and video games), and other activities. In this feasibility study, we first implemented school violence risk assessments to evaluate risk levels by interviewing the student and parent separately at the school or the hospital to complete our novel school safety scales. We completed 25 risk assessments, resulting in 25 transcribed interviews of 12-18 year olds from 15 schools in Ohio and Kentucky. We then analyzed structured professional judgments, language, and patterns associated with school violence risk levels by using manual annotation and statistical methodology. To analyze the student interviews, we initiated the development of an annotation guideline to extract key information that is associated with students' behaviors, attitudes, feelings, use of technology and other activities. Statistical analysis was applied to associate the significant categories with students' risk levels to identify key factors which will help with developing action steps to reduce risk. In a future study, we plan to recruit more subjects in order to fully develop the manual annotation which will result in a more standardized and sensitive approach to school violence assessments.

Emotionally troubled teens' help-seeking behaviors: an evaluation of surviving the Teens® suicide prevention and depression awareness program.

Many school-based suicide prevention programs do not show a positive impact on help-seeking behaviors among emotionally troubled teens despite their being at high risk for suicide. This study is a secondary analysis of the Surviving the Teens(®) program evaluation to determine its effect on help-seeking behaviors among troubled youth. Results showed significant increases in mean scores of the Behavioral Intent to Communicate with Important Others Regarding Emotional Health Issues subscale (p < .0005) from pretest to 3-month follow-up. There was a significant increase (p = .006) in mean scores of the Behavioral Intent Regarding Help-Seeking Behaviors when Suicidal subscale from pretest to posttest, but not at 3-month follow-up. Also, there was a significant increase (p = .016) in mean scores in the item "I would tell an adult if I was suicidal" from pretest to 3-month follow-up. These findings suggest that the Surviving the Teens program has a positive effect on help-seeking behaviors in troubled youth.

The association between salivary hormone levels and children's inpatient aggression: a pilot study.

Aggression is a common management problem for child psychiatry hospital units. We describe an exploratory study with the primary objective of establishing the feasibility of linking salivary concentrations of three hormones (testosterone, dehydroepiandrosterone [DHEA], and cortisol) with aggression. Between May 2011 and November 2011, we recruited 17 psychiatrically hospitalized boys (age 7-9 years). We administered the Brief Rating of Aggression by Children and Adolescents (BRACHA) and Predatory-Affective Aggression Scale (PAAS) upon admission. Saliva samples were collected from the participants during a 24-h period shortly after admission: immediately upon awakening, 30 min later, and again between 3:45 and 7:45 P.M. Nursing staff recorded Overt Aggression Scale ratings twice a day during hospitalization to quantify aggressive behavior. The salivary cortisol concentrations obtained from aggressive boys 30 min after awakening trended higher than levels from the non-aggressive boys (p = 0.06), were correlated with the number of aggressive incidents (p = 0.04), and trended toward correlation with BRACHA scores (p = 0.06). The aggressive boys also showed greater morning-to-evening declines in cortisol levels (p = 0.05). Awakening levels of DHEA and testosterone were correlated with the severity of the nearest aggressive incident (p < 0.05 for both). The BRACHA scores of the aggressive boys were significantly higher than scores of the non-aggressive boys (p < 0.001). Our data demonstrate the feasibility of collecting saliva from children on an inpatient psychiatric unit, affirm the utility of the BRACHA in predicting aggressive behavior, and suggest links between salivary hormones and aggression by children who undergo psychiatric hospitalization.

Appraising risk factors of aggression in children and adolescents at psychiatric inpatient units.

This study examined predictors of aggression and assessed whether different subgroups of children and young people (CYP) display varying risks of aggressive incidents during hospitalization. Data from 10,090 children admitted to the psychiatric inpatient units of Cincinnati Children's Hospital between April 2010 and June 2021 were analysed. Multivariable logistic regression models were used to determine significant predictors associated with aggression, followed by average marginal effects and cluster analyses to rank and establish clusters by the order of predictor importance. About 32.5% reported positive history of an aggressive incident. The mean BRACHA score was doubled compared to those without a prior history. The primary analysis showed that both younger and male CYPs had higher odds of aggressive incidents. We also found that CYP with an African descent, not being able to live with both biological parents, those who reported positive history of psychiatric hospitalisation, and prior externalising behaviours had higher odds of aggressive incidents. These findings have important clinical and public health implications, as they provide valuable knowledge for healthcare professionals to improve prevention strategies for aggression amongst this vulnerable population.

Narrative Review: Impairing Emotional Outbursts: What They Are and What We Should Do About Them.

OBJECTIVE: Impairing emotional outbursts, defined by extreme anger or distress in response to relatively ordinary frustrations and disappointments, impact all mental health care systems, emergency departments, schools, and juvenile justice programs. However, the prevalence, outcome, and impact of outbursts are difficult to quantify because they are transdiagnostic and not explicitly defined by current diagnostic nosology. Research variably addresses outbursts under the rubrics of tantrums, anger, irritability, aggression, rage attacks, or emotional and behavioral dysregulation. Consistent methods for identifying and assessing impairing emotional outbursts across development or systems of care are lacking. METHOD: The American Academy of Child and Adolescent Psychiatry Presidential Task Force (2019-2021) conducted a narrative review addressing impairing emotional outbursts within the limitations of the existing literature and independent of diagnosis. RESULTS: Extrapolating from the existing literature, best estimates suggest that outbursts occur in 4%-10% of community children (preschoolers through adolescents). Impairing emotional outbursts may respond to successful treatment of the primary disorder, especially for some children with attention-deficit/hyperactivity disorder whose medications have been optimized. However, outbursts are generally multi-determined and often represent maladaptive or deficient coping strategies and responses. CONCLUSION: Evidence-based strategies are necessary to address factors that trigger, reinforce, or excuse the behaviors and to enhance problem-solving skills. Currently available interventions yield only modest effect sizes for treatment effect. More specific definitions and measures are needed to track and quantify outbursts and to design and assess the effectiveness of interventions. Better treatments are clearly needed.

A Collaborative Psychiatric-Genetics Inpatient Care Delivery Model Improves Access to Clinical Genetic Evaluation, Testing, and Diagnosis for Patients With Neurodevelopmental Disorders.

Neurodevelopmental disorders including autism spectrum disorder, intellectual disability, and global developmental delay are among the most common indications for referral to clinical genetics evaluation; and clinical genetic testing is indicated for people with neurodevelopmental disorders. There are known barriers to care in accessing clinical genetics evaluation for this patient population. We created a collaborative psychiatric-genetics consultation service and psychiatric-genetics outpatient clinic with the goal to improve care delivery to patients with neurodevelopmental disorders. Two years after the launch of this pilot program, our data demonstrate improved access to genetics evaluation with shorter wait times and fewer patients lost to follow-up. Perhaps most importantly, new genetic diagnoses changed medical care for the majority of patients.

Treatment of Childhood Emotion Dysregulation in Inpatient and Residential Settings.

Children hospitalized in inpatient and residential treatment facilities often present with severe emotion dysregulation, which is the result of a wide range of psychiatric diagnoses. Emotion dysregulation is not a diagnosis but is a common but inconsistently described set of symptoms and behaviors. With no agreed upon way of measuring emotion dysregulation, the authors summarize the existing contemporary treatment focusing on proxy measures of emotion dysregulation in inpatient and residential settings. Interventions are summarized and categorized into individual- and systems-level interventions in addressing aggressive behaviors. Going forward, dysregulation will need to be operationalized in a standard way.

Predicting the time trend of first episodes of aggressive behaviors in pediatric psychiatric inpatient units.

The goal of the current study is to assess whether the scores of Brief Rating of Aggression by Children and Adolescents (BRACHA) at the emergency room (ER) can predict the aggressive incidents at pediatric psychiatric inpatient units. The study aims to identify predictors for two outcome measurements: 1) hazard rates for the first aggressive incident and 2) numbers of days between admission and the first aggressive incident, using the Cox regression model and Poisson regression model, respectively. The clinical records of a total of 5,610 adolescents admitted into the pediatric psychiatric inpatient units of Cincinnati Children's Hospital Medical Center were extracted for the analysis. The aggressive incident was defined as a score >0 from any category of the Overt Aggression Scale (OAS) and the high-aggressive incident was defined as a score ≥ 2 from any category of the OAS. The results indicate that the BRACHA score was not associated with high-aggressive incidents (hazard ratio: 0.98, p = 0.7543). Similarly, the BRACHA scores was only associated with the number of days from admission to the first aggressive incident (Poisson regression coefficient: 0.24, p < 0.0001) but not the number of days from admission to the first high-aggressive incident (Poisson regression coefficient: 0.03, p = 0.3994). Furthermore, the second peak of first aggressive incidents during the hospitalization highlights the importance of interventions at the end of the inpatient treatment course. To summarize, BRACHA scores based on initial assessments at the ER could correlate with first aggressive incidents, but not the first high-aggressive incidents.

A Framework-Based Approach to Assessing Mental Health Impacts of the COVID-19 Pandemic on Children and Adolescents.

The COVID-19 pandemic has yielded extensive impacts globally in the year of 2020. Although the mental health of children and adolescents may be particularly susceptible to stressors stemming from the pandemic and anti-contagion policies, most ongoing efforts are geared toward curbing the viral spread. In the current perspective, we have identified four domains of factors corresponding to an ecological framework that may directly or indirectly influence the mental health of children and adolescents during the pandemic. The evidence suggests that anti-contagion policies might trigger cascades that impact the mental health of children and their families through multiple different sectors that used to form a safety net for youths. Additionally, children with neuropsychiatric disorders could experience exacerbated symptoms during the pandemic. Furthermore, the risk of domestic violence has surged during the pandemic, which further compounds the imminent mental health crisis. A mental health pandemic could be inevitable if no proactive prevention strategies were in place. Therefore, we recommend understanding each individual mental health risk pathway via the ecological framework in order to develop integrative prevention and intervention strategies.

Psychiatric Readmission of Children and Adolescents: A Systematic Review and Meta-Analysis.

OBJECTIVE: To investigate predictors of psychiatric hospital readmission of children and adolescents, a systematic review and meta-analysis was conducted. METHODS: Following PRISMA statement guidelines, a systematic literature search of articles published between 1997 and 2018 was conducted in PubMed/MEDLINE, Google Scholar, and PsycINFO for original peer-reviewed articles investigating predictors of psychiatric hospital readmission among youths (<18 years old). Effect sizes were extracted and combined by using random-effects meta-analysis. Covariates were investigated with meta-regression and subgroup analyses. RESULTS: Thirty-three studies met inclusion criteria, containing information on 83,361 children and adolescents, of which raw counts of readmitted vs. non-readmitted youths were available for 76,219. Of these youths, 13.2% (N=10,076) were readmitted. The mean±SD study follow-up was 15.9±15.0 months, and time to readmission was 13.1±12.8 months. Readmission was associated with, but not limited to, suicidal ideation at index hospitalization (pooled odds ratio [ORpooled]=2.35, 95% confidence interval [CI]=1.64-3.37), psychotic disorders (ORpooled=1.87, 95% CI=1.53-2.28), prior hospitalization (ORpooled=2.51, 95% CI=1.76-3.57), and discharge to residential treatment (ORpooled=1.84, 95% CI=1.07-3.16). There was evidence of moderate study bias. Prior investigations were methodologically and substantively heterogeneous, particularly for measurement of family-level factors. CONCLUSIONS: Interventions to reduce child psychiatric readmissions should place priority on youths with indicators of high clinical severity, particularly with a history of suicidality, psychiatric comorbidity, prior hospitalization, and discharge to residential treatment. Standardization of methods to determine prevalence rates of readmissions and their predictors is needed to mitigate potential biases and inform a national strategy to reduce repeated child psychiatric hospital readmissions.

Finding warning markers: Leveraging natural language processing and machine learning technologies to detect risk of school violence.

INTRODUCTION: School violence has a far-reaching effect, impacting the entire school population including staff, students and their families. Among youth attending the most violent schools, studies have reported higher dropout rates, poor school attendance, and poor scholastic achievement. It was noted that the largest crime-prevention results occurred when youth at elevated risk were given an individualized prevention program. However, much work is needed to establish an effective approach to identify at-risk subjects. OBJECTIVE: In our earlier research, we developed a risk assessment program to interview subjects, identify risk and protective factors, and evaluate risk for school violence. This study focused on developing natural language processing (NLP) and machine learning technologies to automate the risk assessment process. MATERIAL AND METHODS: We prospectively recruited 131 students with or without behavioral concerns from 89 schools between 05/01/2015 and 04/30/2018. The subjects were interviewed with two risk assessment scales and a questionnaire, and their risk of violence were determined by pediatric psychiatrists based on clinical judgment. Using NLP technologies, different types of linguistic features were extracted from the interview content. Machine learning classifiers were then applied to predict risk of school violence for individual subjects. A two-stage feature selection was implemented to identify violence-related predictors. The performance was validated on the psychiatrist-generated reference standard of risk levels, where positive predictive value (PPV), sensitivity (SEN), negative predictive value (NPV), specificity (SPEC) and area under the ROC curve (AUC) were assessed. RESULTS: Compared to subjects' sociodemographic information, use of linguistic features significantly improved classifiers' predictive performance (P < 0.01). The best-performing classifier with n-gram features achieved 86.5 %/86.5 %/85.7 %/85.7 %/94.0 % (PPV/SEN/NPV/SPEC/AUC) on the cross-validation set and 83.3 %/93.8 %/91.7 %/78.6 %/94.6 % (PPV/SEN/NPV/SPEC/AUC) on the test data. The feature selection process identified a set of predictors covering the discussion of subjects' thoughts, perspectives, behaviors, individual characteristics, peers and family dynamics, and protective factors. CONCLUSIONS: By analyzing the content from subject interviews, the NLP and machine learning algorithms showed good capacity for detecting risk of school violence. The feature selection uncovered multiple warning markers that could deliver useful clinical insights to assist personalizing intervention. Consequently, the developed approach offered the promise of an accurate and scalable computerized screening service for preventing school violence.

A Machine Learning Approach to Identifying Changes in Suicidal Language.

OBJECTIVE: With early identification and intervention, many suicidal deaths are preventable. Tools that include machine learning methods have been able to identify suicidal language. This paper examines the persistence of this suicidal language up to 30 days after discharge from care. METHOD: In a multi-center study, 253 subjects were enrolled into either suicidal or control cohorts. Their responses to standardized instruments and interviews were analyzed using machine learning algorithms. Subjects were re-interviewed approximately 30 days later, and their language was compared to the original language to determine the presence of suicidal ideation. RESULTS: The results show that language characteristics used to classify suicidality at the initial encounter are still present in the speech 30 days later (AUC = 89% (95% CI: 85-95%), p < .0001) and that algorithms trained on the second interviews could also identify the subjects that produced the first interviews (AUC = 85% (95% CI: 81-90%), p < .0001). CONCLUSIONS: This approach explores the stability of suicidal language. When using advanced computational methods, the results show that a patient's language is similar 30 days after first captured, while responses to standard measures change. This can be useful when developing methods that identify the data-based phenotype of a subject.

The Importance of Second-Generation Antipsychotic-Related Weight Gain and Adherence Barriers in Youth with Bipolar Disorders: Patient, Parent, and Provider Perspectives.

Objectives: The objective of this research was to understand physician, patient, and parent perspectives on barriers to second-generation antipsychotic (SGA) medication adherence in youth with bipolar spectrum disorders, and attitudes toward treatment of SGA-related weight gain. Methods: Patients diagnosed with bipolar disorder before age 18, parents of children diagnosed before 18, and clinicians with experience prescribing SGAs for these patients completed surveys regarding SGA-related side effects, adherence barriers, and acceptability of weight management strategies. Results: Patients (n = 225), parents (n = 128), and clinicians (n = 54) reported weight gain as the most concerning SGA-related side effect (45.6%, 38.9%, and 70.4%, respectively). Weight gain was also the top adherence barrier for patients (35.9%), but was ranked fourth (41.8%) by parents. Patients (61.5%) were more likely "definitely" willing to co-initiate another medication to manage weight gain upon SGA initiation than parents (20.1%) or clinicians (1.9%). Conversely, parents (54.9%) and clinicians (84.9%) were "definitely" willing to accept/prescribe a second medication aiming to reverse weight gain of ≥10 lbs., and patients (61.1%) were willing to add another medication to reverse any weight gain. Conclusion: SGA-related weight gain impairs medication adherence in young patients with bipolar disorder. Many young patients would start pharmacologic treatment to mitigate SGA-related weight gain at treatment initiation, parents and clinicians are more hesitant. This research informs patient-centered perspectives on SGA adherence barriers and strategies to minimize potential side effects, which may improve adherence in this vulnerable patient population.