Saahoon Hong
- Ph.D.
- Assistant Research Professor
-
IU Indianapolis Child Welfare Education and Training Partnership
Contact
-
(317) 232-7800
- saahong@iu.edu
About
Dr. Hong is an assistant research professor at the Indiana University School of Social Work. Dr. Hong's projects include utilizing the Indiana Family & Social Services Administration's integrative data system for health equity and the Division of Mental Health and Accition's quality improvement initiatives.
Education
PhD
Emphasis/Major: Educational Psychology/Learning Cognition2005 - University of Minnesota
M.Ed.
Emphasis/Major: Special Education1998 - University of Idaho
BA
Emphasis/Major: Special Education1996 - Kangnam University
Research Interests
Cognitive Psychology, Human Development, Cross-system data Analysis, Positive Behavioral Interventions and Supports, Mental Health and Education Well-Being, Adverse Childhood Experience & Child Maltreatment, Special Education, Program Evaluation, and Quantitative Research Methods.
Teaching Interests
Cognitive Psychology, Human Development, Cross-System Research, Applied Behavior Analysis, Intellectual/Developmental Disabilities, Emotional & Behavioral Disorder/Serious Mental Illness, Disparity and Disproportionality in Health and Education, Program Evaluation, Evidence-Based Practice, Machine Learning Algorithms, and Research Methods in Social Work & Education.
Awards and Honors
- The editorial board member
2023 - Journal of Special Education Apprenticeship - TCOM System Champion Award
2023 - Praed Foundation's 19th Annual TCOM Conference. - President Elect of the Korean American Educational Researchers Association (K-AERA) in 2021
2021 - K-AERA - The editorial board member
2021 - Educational Research for Tomorrow - The editorial board member
2020 - Journal of Behavior Analysis and Support (JBAS) - Appointed member in the KABA's Board of Directors
2019 - Korean Association for Behavior Analysis (KABA)
Publications
-
Hong, S., Walton, B. A., Kim, H.-W., Kaboi, M., Moynihan, S. S., & Rhee, G. (2023). Exploring Disparities in Behavioral Health Service Use in the Early Stages of the COVID-19 Pandemic. International Journal of Behavioral Medicine. Published. https://doi.org/https://doi.org/10.1007/s12529-023-10192-z
-
Kyere, E., Hong, S., & Gentle-Genitty, C. S. (2023). Mediational Effect of Teacher-based Discrimination on Academic Performance: An Intersectional Analysis of Race, Gender, and Income. Education Sciences, 13(4), 387-400. https://doi.org/https://doi.org/10.3390/educsci13040387
-
Hong, S., Walton, B. A., Kim, H.-W., & Rhee, G. (2023). Predicting Behavioral Health Needs of Asian Americans in Publicly-Funded Statewide Behavioral Health Services: A classification tree approach. Administration and Policy in Mental Health and Mental Health Services Research, 50(4), 630-643.
-
Kyere, E., Hong, S., & Gentle-Genitty, C. S. (2023). Effect of Racial Discrimination Activated at Teacher-Student Interaction Context on Academic Self-Efficacy: A Latent Transition Analysis. Families in Society: The Journal of Contemporary Social Services, 1-16. https://doi.org/https://doi.org/10.1177/10443894231163968
-
Hong, S., Walton, B. A., Kim, H.-W., Lee, S., & Rhee, T. G. (2023). Longitudinal Patterns of Strengths among Youth with Psychiatric Disorders: A Latent Profile Transition Analysis. Child Psychiatry and Human Development, 1-8.
A better understanding of variability in the strengths of youth with psychiatric disorders is critical as a strength-based approach can lead to recovery. This study aimed to identify subgroups of strengths among youth with mental disorders and determine whether subgroups changes were associated with mental health recovery. Youth with mental disorders(N=2,228) from a statewide database were identified in the state fiscal year of 2019. Using the latest profile analysis and latent transition analysis, we identified three strength profiles that emerged at the beginning and the end of services. The findings suggest that subgroups of strengths may be a promising source to track youth’s progress and guide clinicians to allocate community-based resources more efficiently to improve specific strength areas.
-
Park, S., Hong, S., & Yoon, C. (2022). Analysis of Mothers’ Perceptions Affecting Eating Habits of Young Children with/without Disabilities: A Machine Learning Method. Journal of Early Childhood Special Education, 22(1), 1-25. View Publication For Analysis of Mothers’ Perceptions Affecting Eating Habits of Young Children with/without Disabilities: A Machine Learning Method.
-
Hong, S., Kim, E., Sung, J., & Kwoon, O. (2021). Exploring Ecological Factors Associated with At-Risk Students with Suicidal Ideation: A Decision Tree Algorithm . Journal of Behavior Analysis and Supports, 8(3), 17-35.
-
Foldes, S. S., Long, K. H., Piescher, K., Warburton, K., & Hong, S. (2021). Does a drop-in and case management model improve outcomes for young adults experiencing homelessness: A case study of YouthLink. Center for Advanced Studies in Child Welfare, University of Minnesota. View Publication For Does a drop-in and case management model improve outcomes for young adults experiencing homelessness: A case study of YouthLink
This study used two approaches to examine YouthLink as an example of a drop-in and case management model for working with youth experiencing homelessness. These approaches investigated the same group of 1,229 unaccompanied youth, ages 16 to 24 and overwhelmingly Black, who voluntarily visited or received services from YouthLink in 2011. Both approaches looked at the same metrics of success over the same time period, 2011 to 2016. One approach— Study Aim 1—examined the drop-in and case management model overall, asking whether YouthLink’s service model resulted in better outcomes. It compared a YouthLink cohort with a group of highly similar youth who did not visit YouthLink but may have received similar services elsewhere. A second approach—Study Aim 2—investigated within the YouthLink cohort the ways in which YouthLink’s drop-in and case-management approach worked toward achieving the desired outcomes.
-
Walton, B. A., Hong, S., & , . (2021). Utilization of Child Mental Health Wraparound (CMHW) services: Eligible versus enrolled youth (Vols. 2020, Issue May).
To estimate the need (identify children and youth with complex behavioral health problems), the DMHA Youth Services Team and community service providers requested a report comparing eligibility for CMHW and CMHI services with CMHW program enrollment. Comparing the percentages of eligible and enrolled children/youth of color, 29.6% and 24.3%, respectively, summarize program utilization differences as compared with non-Hispanic White youth. The percentage of eligible Hispanic youth (7.5%), similar to the Indiana Hispanic population (7.1%; US Census, 2019), was higher than of enrolled Hispanic youth (4.2%). The percentage of eligible African American/Black only youth (15.8%) and enrollment group (13.2%) exceeded the state’s Black only population (9.8%; US Census), indicating a disproportionate percent of African American children and youth qualifying for intensive community-based services. Most youth010020030040050060070067891011121314151617EnrolledEligible8identifying more than one race were non-Hispanic Black and White races. Results point to the need to monitor differences in CMHW program utilization by race and ethnicity to consider implications for outreach and culturally appropriate services.
-
Lee, M., Hong, S., & Merighi, J. (2021). The Association Between Fatalism and Mammography Use in Korean American Immigrant Women. Health Education & Behavior, 1-10. https://doi.org/10.1177/10901981211029253
Fatalism is reported as a salient cultural belief that influences cancer screening disparities in racial and ethnic minority groups. Previous studies provide a range of measures and descriptions of cancer fatalism, but no studies to our knowledge have analyzed how fatalistic views cluster together within subgroups to form distinct profiles, and how these profiles can be predicted. This study identified subgroups of Korean American immigrants with similar fatalistic beliefs toward cancer and examined the influence of fatalism, health belief variables, and health literacy on mammography use. A cross-sectional survey design was used to obtain a convenience sample of 240 Korean American immigrant women in Los Angeles, California. Latent class analysis was used to identify unobserved subgroups of fatalism. Hierarchical logistic regression models were used to identify predisposing, enabling, and need factors associated with recent mammography use. The latent class analysis model identified three cancer fatalism subgroups: high fatalism (17.8%), moderate fatalism (36.7%), and low fatalism (45.5%). Women in the high fatalism subgroup were more likely to have had a mammogram within the past 2 years than women in the low fatalism subgroup. Regression analysis revealed three facilitators of recent mammogram use: level of fatalism, perceived barriers to mammogram, and family history of cancer. Although cultural beliefs can have a powerful influence on healthseeking behavior, it is important to weigh individual and contextual factors that may weaken or mediate the relationship between fatalism and engaging in preventive care such as having a mammogram.
-
Gentle-Genitty, C. S., Kyere, E., & Hong, S. (2021). Teacher Role in Absenteeism: Discrimination, Identity, and Intersectionality-A person-in-environment analysis (pp. 99-106). https://doi.org/https://doi.org/10.7912/ew9x-et58
As PARENT S P L AY an integral role in students’ absenteeism, so do teachers. A large part of the day, for students, is spent in schools and with teachers. In this chapter, the focus is on discrimination in teacher-student interactions and its direct influence on minority students regarding their school attendance problems. The data used, literature findings, results, and recommendations are shared from a person-in-environment perspective. The authors recommend exploring discrimination in teacher-student interactions as one mechanism to respond to absenteeism.
-
Walton, B. A., Kaboi, M., & Hong, S. (2020). Child Mental Health Wraparound (CMHW) Program Evaluation 2020.
This evaluation utilized administrative data in a series of logistic regressions to examine functional improvement predictors for youth participating in the CMHW program during SFY2019 and SFY2020. Total CMHW Medicaid service cost during their enrollment, initial complexity and severity of youth’s needs, Gender Identity, and Age predicted the likelihood of improved Youth Needs. Higher CMHW program costs, which reflected engagement, intensity, and duration of program services over time, were significantly related to improved needs and strength development. Youth with more complex and dangerous or disabling needs at the beginning of the program were more likely to improve. However, female children and adolescents enrolled in CMHW were less likely to experience functional improvement or to build strengths in the program or in community-based mental health services. Older youth who received Inpatient, Community-based, and CMHW services were more likely than younger youth to benefit (experience improvement in Youth Needs). While reflected as part of significant Total CMHW costs, the impact of specific ancillary program services (Habilitation, Training and Support of Caregivers, and Respite) was indeterminate in this study. Additional research (e.g., analysis of detailed claims data over the course of each youth’s enrollment, adding fidelity information, and examination of factors related to youth who did not improve) is needed.
-
Hong, S. (2020). Mental Health Statistics Improvement Program (MHSIP) survey study (Vols. 2020, Issue 1).
-
Hong, S. (2020). National suicide prevention month: Suicidality in SFY 2019 (Vols. 2020, Issues 3).
-
Hong, S. (2020). Indiana performance measures: SFY 2021 domain Reliable Change indices (RCI) (Vols. 2020, Issue September).
When measuring change for an individual with CANS or ANSA tools, any change in ratings is meaningful. For groups of children, youth, or adults, change can be measured utilizing a suite of metrics. One overall measurement of change is referred to as Reliable Change in at Least One CANS or ANSA Domain. For this metric, a domain reliable change index (RCI) specifies how much change within each CANS or ANSA domain is enough to be significant, not related to chance. The domain RCI is the amount of change that would be difficult to explain by measurement error.
-
Park, S., Baek, J., Hong, S., Lee, J., & Baek, S. (2020). The support needs of the parent on strengthen the compulsory education right for young children with special education neeeds. Journal of Special Education for Curriculum and Instruction, 13(3), 249-267. https://doi.org/10.24005/seci.2020.13.3.249
The purpose of this study was to investigate the measures to strengthen the right ofthe compulsory education for early childhood special education by identifying parents’ demands for supports. Seven hundreds and fifty-seven parents of children with special needs were recruited from 17 provinces in Korea. The survey were conducted to collect data and the frequency data analysis were used. The findings of the study were as follows. First, in terms of parents perception on the care settings and the education settings, the parents preferred to be enrolled their child into public educational settings rather than care settings to provide inclusive education to their child. Second, parents were aware of the operation requisite for compulsory education that teachers should have teaching certification issued by the Ministry of Education, that the chief of the organization should have education director certification issued by the Ministry of education, and that their children should be enrolled in public preschool. To improve compulsory educational practices, special teachers should be placed at preschool where special class were setup, after school caregivers who majored special education should be placed, and therapy support service providers should be hired at Supporting Center for Special Education. Based on the results of this study, the revision of the law and the direction of policy establishment related to the compulsory education support for young children with special educational needs were discussed and suggested.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: Salvation Army.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: Amethyst.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: Headwater Counseling.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY2019: Families First Indiana.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY2019: Emberwood.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: YWCA Central Indiana.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: St. Joseph Hospital and Health Center.
-
Hong, S. (2020). Outcomes for Adults with Substance Abuse Disorder in SFY 2019: Life Treatment Centers.
-
Hong, S., Ryoo, J., Lee, M., Noh, J., & Shin, J. (2020). The mediation effect of preservice teacher attitude toward inclusion for students with autism in South Korea: a structural equation modelling approach. International Journal of Inclusive Education , 24(1), 15-32. https://doi.org/https://doi.org/10.1080/13603116.2018.1449021
Giving preservice teachers experience in working with students with autism is essential for successful teacher preparation. Structural equation modelling was employed to identify the factors affecting preservice teachers’ attitudes toward the use of inclusive settings for students with autism. The results indicated that their coursework influenced attitudes toward full inclusion both directly and indirectly, but had no effect on their perceptions of partial inclusion and self-contained classroom settings. Preservice teachers’ beliefs toward special education exerted a mediation effect between special education coursework and full inclusion. Similarly, preservice teacher attitudes toward students with autism presented a mediation effect between field experience and inclusion.
-
Hong, S., Choi, W., Piescher, K. N., Zhang, Y., Rhee, T. G., & , . (2020). Does open enrollment policy improve academic performance among students involved with child protective service? Findings from Minnesota-linking information for kids. 108(108), 1-6. https://doi.org/https://doi.org/10.1016/j.childyouth.2019.104653
-
Kyere, E., Hong, S., & Gentle-Genitty, C. S. (2020). Mediational Effect of Teacher-based Discrimination on Academic Performance: An Intersectional Analysis of Race, Gender, and Income. Social Identities. Published.
Drawing on prior research this current study applies an intersectional framework to investigate how teacher-based discrimination operates to influence students’ academic outcomes inclusive of self-efficacy, school attendance, and GPA. We applied structural equation modeling and a cross-sectional sample of the Maryland and AdolescentDevelopment in Context Study (MADICS), Youth Self-Administered (YSA)questionnaires at Wave 3 for this analysis. A total of 1182 students completed the survey, of whom 704 were selected for this study. Findings show teacher discrimination as a mechanism to uncover some of the ways race and gender simultaneously intersect with patterns of students’ academic outcomes. The current study confirms and extends prior work establishing associations among race, gender, income, and teacher discrimination and academic outcomes among African American youth. African-American students, regardless of income levels, may benefit directly—evidenced invisible academic performance—from more positive and race-conscious interactions with teachers.
-
Hong, S., Ryoo, J., Lee, M., Noh, J., & Shin, J. (2020). The mediation effect of preservice teacher attitude toward inclusion for students with autism in South Korea: a structural equation modeling approach. International Journal of Inclusive Education, 24(1), 15-32.
Giving preservice teachers experience in working with students with autism is essential for successful teacher preparation. Structural equation modelling was employed to identify the factors affecting pre-service teachers’ attitudes toward the use of inclusive settings for students with autism. The results indicated that their coursework influenced attitudes toward full inclusion both directly and indirectly, but had no effect on their perceptions of partial inclusion and self-contained classroom settings. Preservice teachers’ beliefs toward special education exerted a mediation effect between special education coursework and full inclusion. Similarly, preservice teacher attitudes toward students with autism presented a mediation effect between field experience and inclusion.
-
Hong, S. (2020). Youth Services Survey for Families (YSS-F) study (Vols. 2020, Issues 2).
-
Hong, S., Choi, W. S., Piescher, K. N., Zhang, Y., & Rhee, T. G. (2020). Does open enrollment policy improve academic performance among students involved with child protective service? Findings from Minnesota-linking information for kids. Children and Youth Services Review, 108, 1-6.
-
Cho, M., Haight, W., Choi, W., Hong, S., & Piescher, K. (2019). A prospective, longitudinal study of risk factors for early onset of delinquency among maltreated youth. Children and Youth Services Review, 102, 222-230.
Maltreated youth tend to enter into the juvenile justice system at younger ages than their non-maltreated counterparts. Early involvement in the juvenile justice system compounds maltreated youth’s risk of adverse developmental outcomes, including serious and continued offending. This study prospectively examined risk factors of first time delinquency for maltreated youth between ages 9 and 14. Using integrated statewide administrative data from Minnesota, this study followed 5002 students with maltreatment histories in 3rd grade for their first adjudication of delinquency over a 6-year period. Cox proportional hazard regression was employed to model time to youth’s first time delinquency, and to identify risk factors. Approximately 7% of maltreated youth (n = 332) were adjudicated as delinquent during this period. The results indicated significant risk factors for early onset of delinquency in maltreated youth: being male (HR = 1.87, 95% CI 1.45, 2.40), belonging to particular racial minority groups, especially Black (HR = 1.80, 95% CI 1.36, 2.39), Native American (HR = 2.34, 95% CI 1.61, 3.39, and Hispanic (HR = 1.73, 95% CI 1.10, 2.71), diagnoses of emotional/behavioral disabilities (HR = 1.96, 95% CI 1.30, 2.93), receiving an out-of-school suspension (HR = 1.53, 95% CI 1.04, 2.25), and experiencing more than three previous maltreatment incidents (HR = 2.02, 95% CI 1.54, 2.64). Solutions for maltreated youth who cross over to the juvenile justice system clearly require interventions that are developmentally sensitive and simultaneously address risk factors across multiple ecological levels.
Presentations
-
Walton, B. A., & Hong, S. (2024). Exploring cultural identity from a strengths perspective: Implications for social work practice.
Following a brief discussion of the strengths perspective’s benefits in recovery (Hong et al., 2021; Rapp & Sullivan, 2014; Rapp & Goscha, 2012), we will explore cultural identity as a strength. Reflecting membership in and support from a social group with specific values, practices and norms (Heersmink, 2021), cultural identity includes, but is not limited to, age, generation, religion/spirituality, gender identity and expression, race and ethnicity, language, profession, class, nationality, rural/urban residence (Heersmink, 2021; Harris et al., 2024). Therapeutic approaches have evolved to a multi-cultural approach, focused on professional development and cultural humility. As social workers, being aware of one’s own cultural identity and curious about/respectful of others’ is consistent with ethical principles (NASW, 2021), enhances relationships, and supports effective practice. In this interactive workshop, participants will have the opportunity to discuss the role of strengths in recovery, explore their own multi-cultural identity, and discuss how cultural identity is related to social work practice.
-
Hong, S., Walton, B. A., & Kim, H.-W. (2024). The Mediating Roles of Psycho-Social Strengths to Substance Use Disorder Treatment Completion.
Background: Research on substance use disorder (SUD) treatment dropout has highlighted the adverse impact it can have on clients, including legal and financial issues, relapse, and unfavorable physical and behavioral health conditions. Previous research has identified factors associated with higher treatment completion rates for adults with SUD, such as substance type, education, employment status, health insurance, and referral source. However, little research has examined the interrelationships, including direct and indirect/mediation effects, among legal system involvement, psycho-social factors related to personal strengths, improved behavioral health needs, and SUD treatment completion. Therefore, the primary purpose of this study is to examine the direct and indirect effects of measured and latent variables on SUD treatment completion.
Methods: Using a sample from one Midwestern state, we identified 3,627 young adults (aged 18-25; 83.2% White; 11.7% Black; 4.7% Hispanic; 52% female; 62.5% lived in urban areas) receiving publicly funded behavioral health services. The court/justice system referred thirty percent. We compared the treatment completion of these young adults to those referred by sources other than the court/justice system (e.g., individual/self-referral, child protective services, health care, and other community referrals). Thirty-three percent of the sample used opioids as a primary, secondary, or tertiary substance in SFY 2021. We conducted a logistic regression model to examine the unique associations of gender, race/ethnicity, urban/rural residence, psycho-social strengths, court/justice system involvement, opioid use disorder, improved behavioral health needs, and SUD treatment completion. We then applied structural equation modeling (SEM) to detect the mediating role of psycho-social strengths in behavioral health needs improvement that led to SUD treatment completion.
Results: Using MLE in AMOS 28, the logistic regression model identified unique associations of legal system involvement, strengths, and improvement in behavioral health needs with SUD treatment completion. The SEM estimation terminated normally within the default convergent criterion (Chi-square = 95.203, df = 12, p = 0.000; RMSEA = 0.044, CI = (0.036, 0.052); CFI = 0.97; TLI = 0.923), and these model fit values exceed the criteria recommended by Hu and Bentler (1999), confirming that positive attributes in strengths mediated the effect of court/justice involvement on improvement in behavioral health needs that led to SUD treatment completion.
Discussion: While there were significant differences in completing SUD treatment between White and Black participants, the SEM model confirms that legal system involvement and the mediating role of psycho-social strengths were critically important in detecting improvement in behavioral health needs and SUD treatment completion, regardless of race. Psycho-social factors related to strengths include social connectedness, optimism, job history, and resilience. Based on these results, implications for social work practice and future research are discussed.
-
Lipsey, A., Hong, S., Walton, B. A., & , . (2023). A Machine Learning-Informed Young Adults Substance Use Treatment Outcomes by Race and Ethnicity. Praed Foundation.
-
Hong, S., Walton, B. A., Kim, H.-W., & Moynihan, S. (2023). Young Adults with the Mental Health and Criminal Justice System Involvement: A Preliminary Study.
Background: The Bureau of Justice Statistics indicated that 37% to 44% of persons in the criminal justice system had mental health issues (Bronson & Berzofsky, 2017). It is well documented that a lack of support for mental health needs would increase the likelihood of further involvement in the justice system (Sung et al., 2011; Williams, 2015). In one Midwestern state, the justice system was the second-highest referral source of young adults (ages 18-25). However, after dual system involvement, little is known about their needs, challenges, and success. This study examined the intersection of characteristics, behavioral health needs, and strengths for young adults with dual involvement in the mental health and criminal justice systems.
Methods. Utilizing a statewide publicly funded behavioral health administrative dataset that included demographic, assessment, and national outcome measures, participants were identified by referral source and legal system involvement (Adult Needs and Strength Assessment, Lyons, 2009). Young adults with past or current legal systems who participated in behavioral health services in CY2019 were identified (n = 8, 170). Four thousand six hundred seventy-nine young adults were never involved in the justice system, while 2,048 and 1,443 had current and historic dual system involvement, respectively.Analysis.This study focused on two groups (current dual system involvement vs. mental health system only), the ANSA’s six domains (i.e., strengths, life functioning, cultural factors, caregiver needs and resources, behavioral health needs, risk behaviors), and their actionable ratings at the last assessment in 2019, and demographic information (i.e., age, gender, race, and ethnicity). The intersection of demographic information, behavioral health needs, and strengths was examined by a machine learning decision tree model, chi-square automatic interaction detection (CHAID) (Milanović & Stamenković, 2016). In addition, hierarchical logistic regression models were implemented to confirm the findings from the CHAID analysis.
Results. Findings predicted dual system involvement with the following ANSA items: 1) substance use; 2) gender; 3) depression; 4)anxiety; 5) volunteering (strength); 6) developmental; 7) impulse control; 8) residential stability; 9) parental/caregiver role, and10) anger control. The most significant predictor associated with the dual system involvement, differentiating from the non-dual system involvement, was substance use, followed by gender and depression. More young men than young women had substance use needs. Young adults with dual system involvement presented higher rates of actionable ratings on depression and impulse control than their counterparts. Of individuals without actionable substance use needs, there was a higher rate of anxiety than youngfemales with dual system involvementThe overall model accuracy was .81, indicating that the model distinguished well betweenindividuals with dual system and non-dual system involvement.
Conclusion and Implications. The findings present well-posed variables to build the prediction model of young adults with dual system involvement. The model could lead us to what areas of need and support we should consider for dual-system youth in mental health recovery. It could eventually be a foundation for another in-depth analysis of behavioral health services, such as effective programs to divert young adults with behavioral health needs from the criminal justice system.
-
Hong, S. (2023). The Intersection of Disability, School Climate, and School Violence in Inclusive Settings.
Background. School violence, including bullying and other forms of aggression, has been a common topic of national concern throughout the last decades across countries. Literature on school violence highlighted the importance of school climate in reducing violence and fostering a safe school environment (Espelage & Hong, 2019). In South Korea, nearly 55 percent of students reported that their friends became victims of school violence at least once during the previous month of the survey (SeoulEducation Longitudinal Study, 2016). However, little is known about school violence and its school climate in inclusive settings that students with disabilities experience. This study examined the intersection of disability, psychosocial characteristics, school violence, friendship, and teacher roles in reviewing the effect of school violence and school climate on self-efficacy among students with disabilities.
Methods. The study utilized the third year of data collected from the Seoul Education Longitudinal Study (SELS, 2016). The student sample was interesting and comprised students with/without disabilities who were 9th graders in middle schools (total n=4,056; of students with disabilities=171). In addition, demographic information and psychosocial characteristics were included to identify intersections of disability, gender, self-efficacy, school violence, friendship, and teacher effectiveness.
Analysis. This study focused on two groups (Disability vs. Non-Disability): the SELS’s six domains (i.e., school climate, school violence, attitudes toward friends/teachers/parents, self-efficacy), gender, and household income. The balance node in IBM SPSSModeler was administered to distribute the participants better. Then, the intersection of disability and school violence was examined by a machine learning decision tree model, chi-square automatic interaction detection (CHAID) (Milanović &Stamenković, 2016). In addition, hierarchical linear regression models were implemented to confirm the findings from the analysis.
Results. Findings presented that students with disabilities were more likely to experience school violence, and their self-efficacy was negatively associated with school violence. The CHAID indicated that the most significant predictor of a disability was income, followed by parents, school climate, school violence, friends, teachers, and gender. The overall model accuracy was .82, indicating that the model distinguished well between disability and non-disability.
Conclusion and Implications: The findings present well-posed variables to build the prediction model of students with disabilities in inclusive settings. The prediction model describes, in reverse, what areas of need and support we should consider for students with disabilities in inclusive settings. It could eventually be a foundation for another in-depth analysis of the best practice model for successful inclusion, such as effective programs to divert students with disabilities from exclusion, isolation, and school violence.
-
Hong, S., Yi, E., Walton, B. A., & Kim, H.-W. (2023). Behavioral Health Needs of Older Adults Living in Poverty: Machine Learning-Based Predictive Models.
Background and Purpose: Growing evidence revealed behavioral health (BH) service needs for older adults (WHO, 2020; Webb, 2020; Novotney, 2019). However, the BH service needs and use of older adults living under 200% of poverty (42%; Cubanski et al., 2021) remained largely understudied. To develop contextually sensitive and effective services for older adults in poverty, this study aimed to identify the characteristics and patterns of older adults’ BH service needs compared to those of middle-aged adults.
Methods: This study used one Midwestern state’s 2019 publicly funded BH administrative data. A sample of adults aged 55 and older were identified (total n=8,071; middle-aged adults (aged 55 to 64): n=6,313; older adults (aged 65 to 74): n=1,758). For each individual, the Adult Needs and Strengths Assessment (ANSA; Lyons, 2009) was completed to facilitate care planning and to monitor outcomes in State Fiscal Year 2019, and at the moment, their initial assessments were done within three years.
A balance node provided by SPSS Modeler was used to correct imbalances between the two groups (aged 55 to 64 [N= 6,313] and aged 65 to 74 [N=6,307]). Adult Needs and Strengths Assessment (ANSA; Lyons, 2009) was completed to facilitate care planning and to monitor outcomes in the State Fiscal Year 2019, and initial assessments were completed within three years (2017 through 2019). A chi-square automatic interaction detection (CHAID) model was used to detect the relationship between Behavioral Health needs (BH) and age group. The CHAID analysis divided data into training (80% out of the balanced sample) and testing (20%) datasets to test the model.
Forty-five ANSA items included (a) life functioning (e.g., medical/physical, independent living, sleep, social functioning, recreational, residential stability, developmental/intellectual, family functioning, employment, self-care, decision making, involvement in recovery, transportation, medication involvement, parental/caregiver role, and legal involvement); (b) BH needs (e.g., psychosis, impulse control, depression, anxiety, interpersonal program, antisocial behavior, anger control, eating disturbance, adjustment to trauma, and substance use); (c) cultural factors (e.g., language, cultural identity, tradition/rituals, and cultural stress), (d) risk behaviors (e.g., suicide risk, self-harm, and exploitation), and € caregiver needs (e.g., physical/behavioral health, involvement with care, knowledge, social resources, family stress, and safety). Variables were coded as binary (0=no action required; 1= = intervention needed). Demographic information (i.e., age, gender, disability, co-occurring disorders, length in the system, and race/ethnicity) was included.
Results: Employment, adjustment to trauma, independent living, and legal system involvement were the four most essential predictors distinguished by age groups, followed by sleep, disability, transportation, social, self-care, and anxiety. For middle-aged adults, the actionable ratings on employment (70%) were significantly higher than for older adults (30%). Additionally, for adults with employment challenges, history/current legal system involvement, independent living, recreational, sleep, interpersonal skills, anxiety, impulse control, depression, disability, and medical/physical issues, significantly distinguished age groups. Finally, among individuals without any employment issues, legal system involvement was also an essential factor in differentiating functional needs by age group. In this group, many older adults had disability conditions that required assistance (62%), while the middle-aged group showed 38% for disability conditions. Older adults without employment and legal system involvement issues experienced high rates of challenges in social functioning (59%) and transportation (66%). The overall model accuracy was acceptable (AUC = .71).
Conclusion and Implications: Different characteristics and patterns of BH and functional needs emerged between middle-aged and older adults. Not surprisingly, employment, adjustment to trauma, independent living, and legal system involvement issues were identified as discernable behavioral health needs that were associated with different age groups. These factors shed light on the need for contextual knowledge of behavioral health needs between middle-aged and older adult groups. Notably, older adults with disabilities experienced higher rates of need related to social functioning, transportation, self-care, and decision-making. Behavioral health needs among older adults without any disability were also associated with medical/physical, anxiety, and independent living. Middle-aged adults with legal system involvement showed higher rates of recreational activities, impulse control, depression, interpersonal skills, and stress than their counterparts. Along with the importance of different BH needs for older people, this study demonstrates the need for age-appropriateage-group-sensitive BH treatment services and support.
-
Hong, S., Walton, B. A., & Kim, H.-W. (2023). Combining ANSA with other administrative data to predict Treatment Completion for Young Adults Who Use Opioids: A CHAID Analysis. Praed Foundation, Lexington, KY United States.
Despite extensive research and policy initiatives, the opioid use crisis persists, affecting approximately 3% of the US adult population with opioid misuse (SAMHSA, 2022). In response to this ongoing challenge, service systems increasingly employ data-driven approaches to identify risks and enhance intervention strategies (Brandt, 2020). To strengthen our comprehension of substance use disorder (SUD) treatment dynamics, it becomes crucial to define progress not only in terms of service completion but also through improved functioning. However, the utilization of assessment data in related studies remains limited. To contribute to this gap in knowledge, we focused on investigating outpatient treatment completion among young adults with opioid use, incorporating the influence of protective factors that contribute to successful treatment outcomes.
Using a statewide sample in one midwestern state, we identified 2,286 young adults (aged 18-25) who used opioids as a primary, secondary, or tertiary substance in SFY 2021.
We compared the treatment completion of these young adults to young adults who used problem substances other than opioid use (N=6,714), including alcohol, cocaine, marijuana, methamphetamine, benzodiazepines, or other drugs.
All analyses were conducted with IBM SPSS Modeler and SPSS Statistics. The chi-square automatic interaction detection (CHAID) approach was used to predict treatment completion. The independent variables were ANSA strengths, National Outcome Measures, TEDS, and demographic data. Preliminary findings found association patterns of treatment completion with resourcefulness, optimism, social connectedness, community involvement, family strengths, source of referral, natural supports, talents/interests, job history, and employment status.
During the session, participants will be engaged by using a digital app (e.g., kahoot.com) to identify ANSA strengths and other factors that could be related to treatment completion. In addition to being introduced to CHAID decision tree analysis and findings, they will discuss the implication of the findings for direct practice, program development, and policy. Their views on the likely relationship between treatment completion and improved functioning will be solicited. We will encourage feedback about the analytical approach.
-
Hong, S., Walton, B. A., & Kim, H.-W. (2023). The impact of faith-driven substance use disorder treatment on treatment completion success.
Introduction:
Substance use disorders (SUDs) are a primary public health concern with significant impacts on individuals, families, and communities. Treatment for SUDs is critical for recovery, but completion rates for SUD treatment programs vary widely. Identifying factors influencing treatment completion can inform the development of targeted interventions to improve outcomes. In recent years, there has been growing interest in utilizing prediction models, mainly statistical and machine learning approaches, to identify factors that may predict SUD treatment completion. These prediction models have the potential to enhance treatment planning and interventions by identifying individuals who may be at higher risk of dropping out of treatment and providing tailored interventions to improve completion rates. In this paper, we will review the existing literature on the development of prediction models for SUD treatment completion, including the predictors used, the statistical and machine learning techniques employed, and the potential implications of these models for improving SUD treatment outcomes.
Substance use disorder (SUD) is a severe public health concern affecting individuals, families, and communities worldwide. Treatment completion, defined as successfully finishing an entire course of SUD treatment, is a critical outcome in the recovery process. Understanding the factors that influence SUD treatment completion is essential for improving treatment outcomes and reducing the burden of SUD on individuals and society.
Numerous studies have investigated the factors associated with SUD treatment completion, including individual, environmental, and treatment-related factors. For example, previous research has found that factors such as age, gender, race/ethnicity, employment status, social support, motivation for treatment, and type of treatment received can impact treatment completion rates (Smith et al., 2017; Johnson & Brown, 2018). Additionally, characteristics of the treatment setting, such as the availability of evidence-based practices, the quality of care, and the presence of barriers to treatment, can also influence treatment completion rates (National Institute on Drug Abuse, 2020).
Despite the existing literature on SUD treatment completion, there is still a need for further research in this area. Many studies have focused on specific populations or settings, and findings may not be generalizable to other contexts. Moreover, there may be gaps in our understanding of the most effective strategies to improve treatment completion rates and reduce disparities in access and outcomes for diverse populations.
The present study aims to address these gaps by examining the factors associated with SUD treatment completion in a diverse sample of individuals seeking treatment for SUD in a community-based setting. This study aims to provide a comprehensive understanding of the factors that contribute to SUD treatment completion by investigating a broad range of factors, including individual, environmental, and treatment-related factors. The findings of this study may have important implications for informing treatment interventions, policies, and practices aimed at improving SUD treatment outcomes.
-
Lee, J., Hong, S., & Brodzinsky, D. (2022). Parental Gender Identity and Its Relationships to Racial Socialization Self-Efficacy, Beliefs, and Perceptions in Transracial Adoption: Using a Machine Learning Decision Tree Model. SSWR, Washington, DC United States.
Background and Purpose:A growing number of non-heterosexual parents have adopted a child of a different race. However, there is limited understanding about their needs in racial socialization practices, which are critical in transracial adoptees’ healthy ethnic and racial identity development. This study used a machine learning decision tree model to identify racial socialization self-efficacy, beliefs, and perceptions associated with parental gender identity among transracial adoptive parents.
Methods: Data were drawn from the Modern Adoptive Families Study, which included 1,616 parents in foster care, private domestic adoption, or international adoption. Parents were mostly Caucasians (90%), mothers (87%), married (71%), and college educated (70%). Approximately 11% were lesbian mothers or gay fathers.
The 29-item Transracial Adoptive Parenting Scale (TAPS) was used to measure racial socialization beliefs and perceptions. Responses were on a 6-point Likert-type scale from 1 (strongly disagree) to 6 (strongly agree), with higher scores including stronger endorsement of racial socialization. The 7-item Racial Socialization Self-Efficacy Scale (RSSES) was used to assess parental feelings of self-efficacy in enacting racial socialization practices. Responses were on a 5-point Likert-type scale from 1 (not at all confident) to 5 (highly confident), with higher scores indicating a higher level of parental confidence in their ability to enact racial socialization practices. The TAPS and RSSE items and married status were examined to identify their associations with parents’ gender identity by a machine learning decision tree model, chi-square automatic interaction detection (CHAID).
Results: Findings showed the intersectionality of adoptive parental gender identity with adoption type, married status, RSSES items, and TAPS items. The most significant predictor associated with adoptive parents’ gender identity was adoption type, followed by variables named “Helping my child feel a sense of belonging within a community of people from his or her birth culture makes me a better parent,” “I believe that discussions of racial differences with my child may do more harm than good,” and “Talk about my feelings about racism and discrimination with my child.” These predictors emerged as critical variables intersected with adoptive parents’ gender identity upon repeated decision tree constructions. The overall accuracy, a percentage of correct predictions for the machine learning model, was 83%.
Conclusion and Implications: The findings suggested that adoptive lesbian parents were more likely to utilize international adoption; the majority of them were not married but partnered. Among the foster care or private domestic adoptive parents, gay parents were less likely to discuss racial differences with their child. The machine learning approach to identifying adoptive parents’ needs could be a promising way to detect the intersection of gender identity, racial socialization self-efficacy, and racial socialization beliefs and practices. This approach eventually be a basis for developing a supporting model for adoptive parents who are either lesbian mothers or gay fathers. Further research is needed to explore the relationship between the identified intersection and social and emotional support needs of lesbian mothers and gay fathers. Such a relationship study would inform development of an efficient supporting model for children whose adoptive parents are not heterosexual.
-
Hong, S. (2022). Young Adults with Behavioral Health Services and Justice System Involvement. Praed Foundation, New Orleans, LA United States.
Program evaluations, informed by relevant literature and combined with demographic, assessment, educational, and service information, can identify key factors for managing change. An example follows.
Background. Transition-aged youth (TAY), ages 18 to 26, have higher rates of Substance Use Disorders (SUD) than adolescents or adults over 26. In 2019, 17% experienced a major depressive disorder, with 12.1% having severe impairments. Overall, 30.6% experienced mental illness, and 9.7% had a severe mental illness. Although TAY reported somewhat lower recovery levels than older adults, predictors of behavioral health recovery for TAY have seldom been explored.
Indiana’s Division of Mental Health and Addiction (DMHA) has funded TAY services since 2019 to positively impact this population.
Methods. Qualitative information from seven currently DMHA-funded programs will be used to support and supplement data analysis. An FFY21 Midwestern sample (n=2557) of treated young adults (ages 18-26) included 700 People of Color (POC; 12% of the sample were Black only, 0.04% Native American only, 0.06% Asian only, 5% other race only, 3% Multiracial, and 6% Hispanic); .73% were White only. Half were female. All youth had substance and mental health disorders. The Adult Needs and Strengths Assessment (ANSA) identified needs that interfered with functioning and strengths. Transportation, depression, anxiety, and psychosis were identified early in care. Other need and strength items reflected status when treatment ended. In a secondary analysis of state-level data, a hierarchical linear regression predicted recovery, the rate of improved Total Actionable Items (Resolved/Ever identified needs). Predictive variables were directly entered into four blocks: 1 (race, gender, transportation, Medicaid status), 2 (social functioning, recreation, employment, resiliency, natural supports, optimism), 3 (suicide risk, psychosis, depression, anxiety, SUD), and 4 (involvement in recovery, SUD Recovery Supports, Duration, Motivational Enhancement Therapy [MET). Race was converted to POC and gender to ‘female.’
Results. Each step of the regression model documented significant contributions of added variables (R2s =.008, .180, .237, .298). POC was less likely to improve than white individuals. Individuals with actionable employment, transportation, suicide risks, and danger to others at the beginning of treatment were more likely to improve. Not being actively involved in recovery, inadequate SUD recovery support, no positive leisure activities, and problematic social relationships at the end of treatment were associated with worse outcomes. Resiliency predicted recovery. Higher levels of active involvement in recovery, resiliency, optimism, longer service duration, and MET were related to higher recovery rates. Gender, Medicaid status, psychosis, and depression were not significant predictors.
Discussion. In addition to addressing SUD and mental health concerns, young adults’ recovery is related to developmental tasks (employment, recreation, and social relationships), supporting involvement in managing one’s health and developing resiliency. Attention to social determinants of health, such as transportation, is necessary to access services and support. Service adaptations for POC to increase involvement in recovery and equitable outcomes require consideration and study. Managing change for TAY involves attention to developmental, cultural, and behavioral health needs, the concurrent utilization/development of strengths, and monitoring progress.
-
Hong, S., Walton, B. A., & , . (2022). Managing Recovery with Adults Involved in Behavioral Health and Criminal Justice Systems. Praed Foundation, New Orleans, LA United States.
Background. The Bureau of Justice Statistics reported that 37% to 44% of persons in the criminal justice system had mental health issues (Bronson & Berzofsky, 2017). Young adults with mental health needs represent the increased criminal behaviors, peaking at 16-25 years, because they are in the middle of the developmental transition to emotional maturity (Stolzenberg & D’Alessio, 2007; Pullmann, 2010). Lack of support for young adults’ behavioral health needs would increase the likelihood of further involvement in the justice system (Sung et al., 2011; Williams, 2015). In one Midwestern state, the third highest referral source for adult behavioral health services was the criminal justice system, following self and health care referrals. Little is known about individuals’ needs, challenges, and successes after they were involved in dual justice and behavioral health systems.
This study examined the characteristics and behavioral needs among individuals with dual system involvement in mental health and criminal justice systems, aged 18 and above, in order to identify their characteristics and behavioral health needs and determine critical factors associated with their dual involvement. To narrow the knowledge gap in understanding adults in dual system and to better support individuals through behavioral health services, there were two evaluation aims: 1) To identify the prevalence rate of dual system adults, characteristics, behavioral health needs, and critical factors associated with their mental health recovery; 2) To provide a prediction model for dual system involvement and examine implications for practice for adults in legal and behavioral health systems.
Methods. A statewide administrative behavioral health data system was utilized to identify behavioral health needs and strengths, and predict behavioral health recovery for individuals with dual system involvement. The database was integrated with episode, encounter, National Outcome Measures, demographics, diagnostics, and assessment data to identify age, gender, race/ethnicity, co-occurring disorders, disabilities, days in the system, source of referral, and behavioral health needs. Dual system adults were identified by the ANSA Legal Functioning (criminal justice system involvement) item and the legal basis of referral. For state-funded adult behavioral health services, 49% of 69,416 participants in state-funded behavioral health services had either a significant history or current justice system involvement. For this study, the sample was limited to young adults, aged 18 through 25 with either open or completed episodes of care during Calendar Year 2019. Improvement rate (resolved items divided by Ever item) was as a dependent variable for the hierarchical linear regression. And dual system involvement was a target variable in machine learning. The most predictive of strengths’ items, behavioral health needs, and sociodemographical/clinical variables for dual system involvement were identified.
Conclusion. The findings suggest that dual system adults experienced higher numbers of actionable needs and lower rates of clinical improvement. Their sociodemographic and clinical variables were significantly intersected with mental health recovery. The identified variables and their intersections were well-posed to predict adults with dual system involvement. It could eventually be a foundation for further in-depth analysis of dual-system involvement and cross-system collaborative efforts to support infrastructure to better manage behavioral health needs.
-
Lee, M., Hong, S., & Merighi, J. (2022). Using Machine Learning to Identify Factors Associated with Mammography Adherence in Korean American Immigrant Women. SSWR, Washington, DC United States.
Background and Purpose: The incidence of breast cancer is rising among Korean American women (Tuan et al., 2020). However, these women report lower rates of breast cancer screening than other racial and ethnic groups (Lee et al., 2018). To promote breast cancer screening in this population, researchers have identified various factors associated with mammogram use such as sociodemographic characteristics, accessibility to health care, and cultural health beliefs (Oh et al., 2017). The purpose of this study is to expand this research by using machine learning to model additional factors associated with mammography uptake in Korean American immigrant women based on American Cancer Society (ACS) screening guidelines.
Methods: A cross-sectional survey was administered to 538 Korean immigrant women in North Carolina in 2019. The study participants were recruited using a study flyer and snowball sampling strategies at various community-based sites (e.g., churches and grocery stores). The survey was either self-administered or conducted face-to-face by the researcher. The participants’ mean age was 57.4 years old (SD=8.3) and their average length of time in the United States was 25.2 years (SD=11.9). About 89% completed undergraduate or graduate education, and 40% reported their income was less than $50,000 per year. A machine learning decision tree model, using chi-square automatic interaction detection (CHAID), was performed to identify factors associated with adherence to ACS mammography screening guidelines.
Results: Approximately 91% of the participants had a mammogram at some point in their lifetime, and among them, 65% adhered to ACS guidelines. The most significant factor associated with adherence to ACS guidelines was having a regular medical check-up. Other significant factors included: primary care physician’s recommendation, social support, alcohol use, health information from Korean newspapers and magazines, mammogram frequency, family history of cancer, the country of their first mammogram, follow-up test experience after mammogram, and having heard about another woman’s mammogram experience. The overall accuracy of correct predictions for the machine learning model was .84.
Conclusions and Implications: The study findings highlight the importance of access to routine primary care services and screening referrals for Korean American immigrant women. In addition, education about risk factors for breast cancer (alcohol use, family cancer history) and the need for follow-up tests after receiving abnormal screening results are warranted. One mechanism to promote breast cancer screening in this population is to use Korean newspapers and magazine to provide public health messaging about the primacy and benefits of having a mammogram. Interestingly, women who had their first mammogram in Korea tended to adhere to ACS screening guidelines as compared to those who had their first mammogram in the US. Future research needs to explore the motivational factors associated with breast cancer screening in Korean American immigrant women.
-
Kyere, E., Hong, S., & Gentle-Genitty, C. S. (2022). Understanding Academic Outcomes: Mediational Effect of Teacher-Based Discrimination. Indiana University, Washington, DC United States.
Background: The extant literature paints the US educational canvas as racialized, gendered, and stratified on income, particularly for minority backgrounds. African American students primarily—perhaps most situated because of their social position—experience significant bouts of discrimination due to the joint effects of these structural factors. Research examining the independent associations among race, gender, income, and teacher-based discrimination has very minimally examined the intersectionality of these variables. Although race, gender, and income are structural forces, they are activated through social interactions. For many students, their school experiences serve as the activator through teacher-student discrimination. Studies where differences in race and gender and income/social class have been examined suggest associations. However, such independent associations provide limited insights into the relational nuances operating to influence student outcomes. The current study moves beyond independent associations to intersectional analysis. We examine the mechanism by which race, gender, and income may simultaneously intersect with teacher-based discrimination to affect students’ academic outcomes, including self-efficacy, attendance, and GPA.Method: Participants were chosen from the Maryland and Adolescent Development in Context Studies (MADICS). The analysis focused on eight-graders. Of the 1182 students who completed the survey, 704 were selected for this study. The mean age was 14.30 (SD=0.42). Students were labeled either African American (65%) or White (35%). Forty-seven percent was male (n=333). Most of the families’ income levels were less than $49,999 (49%), followed by income levels between $50,000 and $74,999 (33.9%), and more than $75,000 (17%). We applied structural equation modeling to answer our research hypotheses.
Result: Race and gender negatively affected GPA: those students who were either African American or male were more likely to have lower GPA scores. Race and gender had significant effects on GPA with/without the presence of teacher-based discrimination. However, race and gender did not have a significant effect on GPA with the presence of self-efficacy but showed a significant effect without the presence of academic self-efficacy. Therefore, unlike academic self-efficacy, teacher-based discrimination plays a role as a mediator in the relationship between gender, race, and academic performance as measured by grade point average (GPA), but not between income/class and GPA. Similarly, absence/attendance showed a partial mediation effect between GPA and income/class.
Conclusions and Implications: Stereotypical narratives shaping race and gender and teacher-student relationships affect academic performance. Teacher-based discrimination must be included in the discourse around poor academic performance. African American students, regardless of income/class, can benefit from consistent positive race-conscious interactions with their teachers. Poverty is another critical factor to inform policies and practices in around school attendance problems including truancy and absenteeism. Poverty, which is strongly related to race, affects students’ school attendance and performance. Implications will be discussed, including ways to create a school culture that fosters positive and race-conscious relationships for bolstering students’ school engagement and performance
-
Hong, S., Walton, B. A., & Kim, H.-W. (2022). The Intersectionality of Gambling Addiction Recovery and Mental Illness: A Machine Learning Approach. SSWR, Washington, DC United States.
Background and Purpose: Given that a variety of gaming activities (i.e., casinos, TV and instant scratch lotteries, sports bettings) are growing, COVID-19 related stress is presenting a severe threat to worsen addictive behaviors (Håkansson’s et al., 2020). Since certain forms of gambling, like internet-based and other forms of gambling activities, could remain unchangeably available to these adults in the COVID-19 related confinement, special attention to the gambling addiction as consequences of the COVID-19 pandemic is needed. In this regard, the purpose of this study is to examine and identify intersectionality for gambling addiction recovery with mental health needs in a behavioral health system.
Methods: The sample of adults aged 18 and above who participated in Midwestern state-funded mental health and addiction services in 2019 and 2020 was selected. Their initial and most recent assessments among adults with the need for problem gambling treatment at the initial assessment were analyzed for the study (N=654). All participants were taken the Adult Needs and Strengths Assessment (ANSA; Lyons, 2009) as the last assessment in either 2019 or 2020, including six domains: (1) strengths, (2) life functioning, (3) cultural factors, (4) caregiver needs, and resources, (5) behavioral health needs, and (6) risk behaviors. This study focused on the ANSA 57 items and four demographic information (i.e., age, gender, race/ethnicity, calendar year). Each ANSA item was rated on a four-point scale, ranging from 0 (non-actionable) to 3 (immediate action required). These ratings were changed into non-actionable (0) and actionable (1) and were examined by a machine learning decision tree model, chi-square automatic interaction detection (CHAID).
Results: Upon repeated decision tree constructions, the intersectionality of the gambling recovery with sexuality, criminal behavior, adjustment to trauma, resiliency, eating disturbance, employment, sleep, legal, residential stability, and substance use was found. The most significant predictor for gambling addiction recovery was substance use. Specifically, it means that adults were more likely to present the gambling addiction recovery when they stayed clean from substance use and did not struggle with impulse control than their peers. In contrast, the current difficulties of substance use, impulse control, decision making, and residential stability were the major barriers to the gambling addiction recovery. The overall accuracy was .8, which indicated that the model was at good distinguishing between gambling addiction recovery and sustaining gambling addiction.
Conclusion and Implications: The findings suggest that staying clean from substance use and impulse control were primary predictors that led to gambling addiction recovery, regardless of the COVID-19 pandemic. The machine learning-based gambling addiction recovery model could be a promising approach to detect the intersection of race/ethnicity and behavioral health challenges and their recovery. It could eventually be a basis for developing a gambling addiction recovery model for adults with needs for gambling addiction treatment at the initial assessment. Further research is also needed to explore the relationship between the identified intersection and other mental health illnesses. Such a relationship study will support the development of an efficient mental health and gambling recovery model.
-
Hong, S., Walton, B. A., Rhee, T. G., Kim, H.-W., & , . (2021). Patterns of Child and Adolescent Needs and Strengths Among Youth at Risk for Suicide Attempt. Society for Social Work and Research.
Understanding the relationship of strengths to mental health disorders is critical as the strength-based approach supports mental health recovery (Gable & Haidt, 2005; Xie, 2013). Although recovery occurs among many adolescents, mental health problems remain common and suicide rates continue to grow (Hedegaard, Curtin, & Warner, 2018). A few studies were conducted in identifying patterns of strengths, classifying similar groups, or tracking them using latent classes (Leon & Dickson, 2018). This study aimed to (1) identify subgroups of strengths and (2) determine how subgroups are predicted based on demographic characteristics (e.g., gender and race).
-
Lee, M., & Hong, S. (2021). Breast Cancer Screening Behaviors in Korean American Immigrant Women: Does Fatalism Matter?. Society for Social Work and Research.
Fatalism has been reported as a noteworthy cultural health belief that influences cancer screening participation (Jun & Oh, 2013), preventive behaviors (Niederdeppe & Levy, 2007), and treatment-seeking behaviors (Drew & Schoenberg, 2011).
Although some research has examined fatalistic beliefs in ethnic minority groups compared to Whites at a national level (Niederdeppe & Levy, 2007), Korean American immigrant women’s (KAIW) fatalistic views on cancer and their link to breast cancer screening behaviors have been understudied. This investigation examined the role of fatalism as a mediator between regular health check-ups, health insurance and family history of cancer and three types of breast cancer screening behaviors (breast self-exam, clinical breast exam, and mammogram) in KAIW. Given that a fatalistic belief in cancer is one of the frequently cited barriers to health care access in KAIW, the aim of this study was to expand our understanding of how cultural health beliefs influence KAIW’s screening behaviors.
-
Hong, S., Walton, B. A., Kim, H.-W., Rhee, T. G., & , . (2021). A Structural Equation Model of Risk Behaviors and the Length of Behavioral Health Services Among Youth with Mental Health Disorders. Society for Social Work and Research .
Understanding the association of strengths- and needs-based approach with the length of stay is critical to guide outcome-oriented interventions. A few studies primarily conducted in congregate care settings predicted a shorter length of stay by the severity of disorders and awareness of treatment need (Megna et al. 2015; Pavkov, Goerge, & Czapkowicz, 1997; Stewart, Kam, & Baiden, 2014). A few studies considered that there is a substantial effect of strengths on behavioral health treatment outcomes (Quiroga & Walton, 2013; Radigan & Wang, 2013). This study aimed to (1) identify the direct and indirect impacts of risks and behavioral/emotional needs, and (2) determine to what extent life functioning and strength domains were mediated.
-
Kim, H.-W., Walton, B. A., & Hong, S. (2021). Utilizing the Adult Needs and Strengths Assessment (ANSA) with Young Adults: Exploratory and Confirmatory Factor Analysis. Society for Social Work and Research.
The Adult Needs and Strengths Assessment (ANSA) is a structured, holistic assessment tool that is used to identify needs and strengths, develop intervention plans, and monitor progress. Despite a wide implementation of the tool across several states and behavioral health service organizations, limited research exists on the ANSA. With an increased awareness of continued developmental tasks and of various challenges faced by young adults with behavioral health needs, it is important to examine the psychometric properties of the ANSA. The purpose of this study is to examine and confirm the factor structure of the ANSA with a sample of young adults who have behavioral health disorders.
-
Kyere, E., & Hong, S. (2021). Examination of the Associations Among Depression, Academic-Self-Concept, and Racial Stress in African American Middle Scholars. Implications for School Social Work. Society for Social Work and Research.
Research has shown that depression is prevalent among youth as they transition into secondary school. For African American youth, depression is detrimental and can significantly contribute to their educational and social impairment. Additionally, depression can lead to increased substance misuse, risky sexual behaviors, and suicidal ideation and contributes to health disparities among African Americans. Thus it is critical that social workers understand predicting and mitigating factors to depression among African American youth. Drawing on emerging research that suggests an inverse relationship between depression and academic self-concept, the current study investigated the contributions of academic self-concept (perception of learning and academic subjects) and racial stress in predicting depression among African American middle scholars. In addition, we investigated whether academic self-concept interacts with racial stress to predict depression.
-
Walton, B. A., Kim, H.-W., & Hong, S. (2021). Once TCOM Is Running-Consider Program Evaluation. Praed Foundation.
Once TCOM implementation is running, ANSA or CANS information can be used for program management and evaluation. This workshop will illustrate different approaches to program evaluation using ANSA and CANS data and related information. Whether in response to system-level requests, a call for papers, or a timely issue, program evaluation can demonstrate the usefulness of TCOM tools in managing programs. Building upon existing research, two examples illustrate alternative strategies. For adults with co-occurring mental health and substance use disorders, social determinants of health, recovery, and service factors were examined as potential predictors of functional change (total actionable needs) using hierarchical linear regression. For children and youth with complex needs, the Medicaid wraparound program routinely used patterns of CANS data to predict program eligibility, as compared with program participation. Used to support hiring decisions and manage state programs, the evaluation was enhanced to compare differences between eligible, but not enrolled youth and program participants. Mixed methods included quantitative and qualitative approaches. For enrolled youth, predictors of improvement were informed by demographics, service duration, Medicaid claims, and CANS data.
To engage participants, depending upon group size, brief introductions with questions about program evaluation experience will open the workshop. They will be given a worksheet on which to note program evaluation ideas. Brief group discussion will follow to explore program evaluation opportunities for participants.
-
Hong, S., Walton, B. A., & Kim, H. (2021). Examining the intersection of mental illness and suicidal risk in the shadow of a pandemic: A Machine Learning Approach. Praed Foundation.
To develop the suicidal recovery model for adults with mental illness during the pandemic and better serve them in the mental health system, it is necessary to ensure that we can identify the intersection of mental illness and suicidal risk. In Artificial Intelligence-based psychiatry, the progressive use of machine learning has been utilized to detect and monitor different mental health states and identify between the target response and a set of input features of interest (i.e., Garcia-Ceja et al., 2018; Tate et al., 2020). In this regard, we used machine learning to examine the intersection of mental illness and suicide aged 17 years old and above adults in the Mideastern state-funded mental health service (n=31,138) during the calendar years of 2019 and 2020. Classification, regression tree analyses, and chi-square automatic interaction detection (CHAID) were used to identify the intersection of mental illness and suicidal risk and determine their classification accuracy.
Findings included the convergence of several associations that included depression, trauma, age, interpersonal, psychosis, impulse control, anxiety, anger control, pandemics, and substance use. Both depression and trauma emerged as the most important variables intersected with suicidal risk during the COVID-19 pandemic. The machine learning approach provides a better understanding of the associations between mental illness and suicidal risk. It could eventually be a basis for the suicidal recovery model among adults with mental illness.
-
Hong, S., Walton, B. A., & Kim, H.-W. (2020). Latent transition analysis of the CANS strengths domain membership changes among youths at risk for a suicide attempt. Praed Foundation.
Understanding variability in strengths of youths at risk for suicide ideation is critical as a strength-based approach leads to mental health recovery. This study aimed to identify subgroups of strengths among youth at risk for suicide attempts and to determine whether subgroups were changed during the mental health recovery process. Child and Adolescent Needs and Strengths assessment measures included eleven strength areas. Only seven percent of youth experienced positive strength development over time, especially for those rated as buildable strengths at the baseline. The implications were discussed in a direction for future research and utilization in practice.
-
Hong, S. (2020). Community-Based Interagency Collaboration for Students with EBD. Korean Association for Behavior Analysis , Seoul, South Korea.
-
Hong, S. (2020). Establishing National Standards for Korean Behavior Analysis. Korean Association for Behavior Analysis , Seoul, South Korea.
-
Hong, S., & Walton, B. A. (2020). Strength-Based Approach Among Youth with Serious Emotional Disturbance: A Latent Profile Transition Analysis. Children’s Mental Health Network, Tampa, FL United States.
Background. Understanding diverse or similar strengths of youth with mental health disorders is critical in that the strength-based approach leads to mental health recovery (Gable & Haidt, 2005; Xie, 2013). However, very few studies have been conducted to identify patterns of strengths, classify similar groups, and track them in the latent classes (Leon & Dickson, 2018).
Study Aims. To identify unobserved subgroups in a population of youth ending behavioral health treatment. To examine profiles of youth strengths that move into categories that are more/less usable. To determine whether transition patterns among youth strengths classes are the same across the time points.
Conclusions. Given the dynamic nature of youth, this study extends the utilization of the CANS system by identifying youth strengths profiles and comparing transitions in youth strengths classes at two time points. The results of the current study point out the need for continued support for all the youth in the system. Specifically, the values along the diagonal, shown in Table 2, indicate that they were stable in strength status, and the off-diagonals describe a change in the class status. Interestingly, most youth, rated as either “buildable Strengths” or “medium,” transitioned into a class of youth with usable strengths. Surprisingly, 83% of youth in the “strength in place” latent class at Time 1 moved to “buildable Strength” at Time 2. Such transitions informed that prevention and intervention efforts should be made for youth at high risk and those with relatively low risk in the CANS system. Systematic utilization of strengths and needs assessment identified buildable strengths that frequently improved during treatment. Further research is needed regarding the effect of covariates (i.e., age, gender, race, days in the system) on observed transitions between two-time points.
-
Walton, B. A., Hong, S., Kim, H.-W., & Moynihan, S. (2020). Identifying and analyzing DARMHA disparity data. FSSAs Department of Mental Health and Addiction, Indianapolis, IN.
-
Hong, S. (2020). A Latent Class Analysis of Cancer Fatalism with Korean American Immigrant Women.
Contract Fellowship Grants
-
Measuring Mental Health Provider Accountability: Value-Added Models
Program Director (PD)/Principal Investigator (PI): Saahoon Hong
Co-PD/PI: Betty Walton
Co-PD/PI: Hea-Won Kim -
TCOM Training, Technical Assistance, & Research 2023 - 2024
Program Director (PD)/Principal Investigator (PI): Betty Walton
Supporting Personnel: Saahoon Hong
Supporting Personnel: Hea-Won Kim -
Predicting Young Adults’ Completion of Substance Use Treatment by Race and Ethnicity: A Machine Learning Approach
Program Director (PD)/Principal Investigator (PI): Saahoon Hong
Co-PD/PI: Betty Walton
Co-PD/PI: Hea-Won Kim
Co-PD/PI: Wendy Harrold -
TCOM Training, Technical Assistance, & Research 2021 - 2022
Program Director (PD)/Principal Investigator (PI): Betty Walton
Supporting Personnel: Saahoon Hong -
Racial and Ethnic Disparities in Prevalence, Severity, and Comorbidity of Mental Illnesses During the COVID-19 Pandemic
Program Director (PD)/Principal Investigator (PI): Saahoon Hong
Co-PD/PI: Betty Walton
Co-PD/PI: Hea-Won Kim
Co-PD/PI: Stephanie Moynihan
Co-PD/PI: Taeho Rhee -
TCOM Training, Technical Assistance, & Research 2019 - 2020
Program Director (PD)/Principal Investigator (PI): Betty Walton
Consultant: Saahoon Hong
Supporting Personnel: Stephanie Moynihan
Consultant: Hea-Won Kim
Institutional Services
-
Committee Member
2019 - IUSSW Diversity Committee
Media Appearance
-
Predicting the Behavioral Health Needs of Asian Americans | Research Briefs
2023 - The TCOM Channel
Read the Story Predicting the Behavioral Health Needs of Asian Americans | Research Briefs -
Mental Illness and Suicide Risk
2022 - The TCOM Channel
Read the Story Mental Illness and Suicide Risk -
Strengths Over Time and Mental Health
2021 - The TCOM Channel
Read the Story Strengths Over Time and Mental Health -
Studying mental illness and suicidal risk with machine learning
2021 - On Topic with IU
Read the Story Studying mental illness and suicidal risk with machine learning -
Racial & Ethnic Disparities and Its Relation to Mental Illness
2021 - IU SSW Commitment to Research
Read the Story Racial & Ethnic Disparities and Its Relation to Mental Illness -
Latent Transition Analysis of the CANS Strengths Domain among Youths At Risk for a Suicide Attempt
2021 - The TCOM Channel
Read the Story Latent Transition Analysis of the CANS Strengths Domain among Youths At Risk for a Suicide Attempt
Memberships
-
American Public Health Association
2023 to 2024 -
Menatal Health America
2023 to 2024 -
Society for Social Work and Research
2019 to 2024 -
Korean Association for Behavior Analysis
2019 to 2024 -
Korean American Educational Researchers Association
2010 to Present -
American Educational Research Association
2005 to 2024
Professional Services
-
Committee Chair
2024 to 2025 - US Korea ConferenceUS Korea Conference (UKC) will be held on August 21-24, 2024. UKC is a national conference sponsored by the Korean-American Scientists and Engineers Association (KSEA), the largest Korean-American professional organization promoting the application of science and technology for societal well-being. As the chair of the technical group symposium for D-1: Social Sciences, I oversee a symposium that presents a holistic view of advanced technology's intersections with education, mental health, and crime prevention. In this capacity, I will lead three distinct sessions: 1)AI-driven Methods in Education and Social Science; 2) Promoting Mental Health and Well-being in K-12 Education; 3) Navigating the Nexus of Technology and Crime Prevention in the Digital Age.
-
Officer, President/Elect/Past
2022 to 2023 - Korean-American Educational Researchers Association -
Board of Advisors
2021 to Present - The Pivot Attendance Solutions (PAS)The Pivot Attendance Solutions (PAS) is a local organization focused on advancing new and innovative attendance practices, policies, and solutions.
-
Board of Advisors
2021 to Present - Korean Alliance on Mental Health -
Editorial Review Board Member
2021 to 2023 - The Education Research for Tomorrow -
Officer, Vice President
2021 to 2022 - Korean-American Educational Researchers Association -
Committee Member
2020 to Present - DMHA Suicide Prevention Committee -
Committee Chair
2020 to 2021 - KAERA Outstanding Research Selection Committee -
Reviewer, Ad Hoc Reviewer
2020 to 2020 - Asia Pacific Education Review -
Reviewer, Ad Hoc Reviewer
2020 to 2020 - Journal of Behavior Analysis and Support -
Chairperson
2019 to 2022 - Korean American Educational Researchers AssociationDr. Hong has served as a chairperson of the outstanding research paper award committee member to select the most distinguished manuscript every academic year.
-
Editorial Review Board Member
2019 to 2024 - Journal of Behavior Analysis and SupportDr. Hong serves as an editorial review board member for the Journal of Behavior Analysis and Support.
-
Board of Directors
2019 to 2024 - Korean Association for Behavior AnalysisDr. Hong serves as a board of director member for the Korean Association for Behavior Analysis that contributes to the development of national standards for the Korean Behavior Analyst who work with students with special needs.
Public Services
-
President
2022 to 2023 - Korean American Educational Researchers Association -
Other
-
Advisory Board members
2021 to Present - Pivot Attendance SolutionsPivot Attendance Solutions, an Indiana-based education organization, bridges the gap between the community and entities serving school children. As an academic advisory board member, I consult on research-related activities.
-
Chair
2017 to 2020 - Minnesota Association for Korean AmericansDr. Hong served as a chairperson at the Minnesota Association for Korean Americans Scholarship Search Committee.
-
Board of Directors
2017 to 2020 - Minnesota Korean Service CenterDr. Hong served as a board of director member at the Korean Service Center, located in Minnesota. Korean Service Center has provided a full array of social services to Korean immigrants to help them maintain a full and happy life. It has also promoted mental wellness, physical well-being, and self-sufficiency by providing a variety of social, recreational, and cultural programs for Korean immigrants.