Background: Adverse childhood experiences (ACEs) have been recognized as a significant contributor to poor mental health outcomes in adulthood. However, studies on the effect of ACEs on psychological distress across different racial and ethnic groups are limited. Moreover, the role of positive childhood experiences (PCEs) in moderating the impact of ACEs on psychological distress remains largely unexplored. In this study, we investigate whether ACEs have varying effects on psychological distress across racial and ethnic groups and if PCEs moderate these relationships.

Methods: Data on Non-Hispanic White (NHW) (n=22737), Non-Hispanic Black (n=2210), Hispanic (n=11746), and Non-Hispanic Asian American (n=7196) adults were analyzed from the 2021-2022 California Health Interview Survey. ACEs were categorized as 0 ACEs, 1-3 ACEs, and ≥4 ACEs from the 11-item Behavioral Risk Factor Surveillance System ACE module. PCEs were categorized into 0-2 PCEs, 3-5 PCEs, and 6-7 PCEs from the 7-item Child and Youth Resilience Measure–28. Moderate (K6 score ≥ 5) and serious (K6 score ≥ 13) psychological distress was measured using the Kessler Psychological Distress Scale. Multivariable logistic regression controlled for sociodemographics and was stratified by race and ethnicity.

Results: Black adults had the highest prevalence of ≥4 ACEs (27.8%), while Asian American adults had the lowest (13.9%). Hispanic adults had the lowest prevalence of 6-7 PCEs (24.4%). The prevalence of moderate and serious psychological distress was similar across all racial and ethnic groups. In stratified analyses, compared with those reporting 0 ACEs, Black adults reporting ≥4 ACEs had the highest odds of experiencing moderate psychological distress (OR: 3.34, 95% CI: 2.19-5.08) compared to other racial groups. We found a significant interaction between ACEs and PCEs (p-interaction<0.05) where PCEs modified the association between ACEs and serious psychological distress across all racial and ethnic groups.

Conclusion: There is heterogeneity in ACEs’ impact on psychological distress across racial and ethnic groups. Targeted upstream interventions promoting PCEs are crucial to reducing the impact of ACEs on psychological distress in later life.

Background: Racial and ethnic minorities disproportionately experience depression and anxiety symptoms. Yet, the impact of within-group racial and ethnic heterogeneity on these symptoms is unclear. Machine learning can address this gap by assessing changes in model performance based on various categorizations of racial and ethnic groups. This study investigated the extent to which within-group racial and ethnic heterogeneity, alongside sociodemographic factors, affect prediction of symptoms of depression and anxiety among young adults.

Methods: Data from n=2340 young adults (ages 18–25 years, 59% female, 38% Hispanic, 29% low SES) from the Fall 2021 cross-section of the Texas Adolescent Tobacco Marketing and Surveillance study (TATAMS) were analyzed. Six machine learning models were used to predict depression (PHQ-9) and anxiety (GAD-7) symptoms, both as binary variables. Random forest models were chosen to evaluate model performance and identify salient sociodemographic features for symptom classification. Models included a) various categorizations of racial and ethnic features only and b) various categorizations of racial and ethnic features and other sociodemographic features (i.e., socioeconomic status (SES), age, sex, county of residence). Sensitivity (Sn) and F1 scores were prioritized in evaluating model prediction while mean decrease Gini indices evaluated feature importance.

Results: Models solely including racial and ethnic features poorly predicted both depression and anxiety symptoms. Including sociodemographic features, predictive ability remained modest for symptoms of depression (Sn range: 0.17 – 0.58, F1 range: 0.22 – 0.42) and anxiety (Sn range: 0.21 – 0.54, F1 range: 0.28 – 0.40). While prediction remained consistent for models with racial and ethnic features only for both variables, models including sociodemographic features demonstrated worse predictive performance for increasingly disaggregated racial and ethnic categorizations. Across all model variations, SES remained the most influential feature.

Conclusions: Findings suggest that race and ethnicity are poorly predictive of symptoms of depression and anxiety while SES is critical for symptom classification. Policymakers may consider sociodemographic factors, particularly SES, in addressing racial and ethnic differences in mental health among young adults at the population level. Additional machine learning techniques can parse effects of race, ethnicity, and other sociodemographic factors on measurement and surveillance of symptoms of depression and anxiety.

Background: Recent studies have shown increasing mental health illness among the Hispanic/Latino populations in the United States(US), with many experiencing a major depressive episode annually. General health status(GHS) has been linked with depression symptoms(DepS) in recent years. We investigated the differences, temporal prevalence trends, and the association between GHS and moderate/severe DepS among Hispanics/Latinos in the US.

Methods: Data used in this study was obtained from the Health Information National Trends Survey (2011-2019) involving 1,201 Hispanic/Latino adults aged ≥18 years in the US. The sampled data were weighted, accounting for complex survey designs to minimize Type I error and ensuring that estimates are nationally representative. Weighted average percent change(WAPC) in prevalence rate over the data durations was estimated using the semi-elasticity index. Bivariate Chi-square tests and multivariate logistic regression analyses were conducted to examine the prevalence trend and odds of GHS (poor/fair and good/very good/excellent) with DepS (from Patient Health Questionnaire-4(PHQ-4); categorized as low/none(0–2), mild(3–5), moderate(6–8) and severe(9–12)), adjusting for covariates (sociodemographic: age, sex, marital status, US census region; socioeconomic: education, income, health insurance, quality care, linguistically isolated household; and health behavior: body mass index, physical activity, smoking status).

Results: The weighted average prevalence trend (2011-2019) of DepS among Hispanics/Latinos was 55.4%. The weighted prevalence rate of DepS among those with fair/poor GHS was 78.6%. Latinos with poor/fair GHS continuously had a higher prevalence of DepS, having a 1.2% increased WAPC. However, those with excellent/very good/good GHS had a 55.9% increased WAPC of DepS. Hispanics/Latinos with fair/poor GHS were at a higher risk of moderate/severe DepS than those with excellent/very good/good GHS after adjusting for the sociodemographic (adjusted odds ratio[AOR]=3.99, 95% confidence interval[CI]=2.53-6.30), socioeconomic (AOR=4.00, 95% CI=2.49-6.43), and health behavior risk factors (AOR=3.85, 95% CI=2.31-6.43), respectively. There was an increasing relative risk of severity in DepS among Hispanics/Latinos with fair/poor GHS relative to those with excellent/very good/good GHS.

Conclusion: Our study shows fair/poor GHS is significantly associated with increased DepS among Hispanics/Latinos in the US. Therefore, we recommend that health policies/interventions to prevent/reduce DepS among Hispanics/Latinos should target improving the GHS of individuals with fair/poor GHS.

Background: Depression and obesity are two major health problems among women in the US. The prevalence of depression and obesity varies by income and race/ethnicity; yet, it is unclear how race/ethnicity and income may influence depression rates among obese women.

Objective: This study sought to understand how race and income may act as drivers of depression among obese women by examining data from NHANES between 2005 and 2018.

Methods: We used the Patient Health Questionnaire (PHQ-9) and the Poverty to Income Ratio (PIR) as a proxy for income to measure the presence of depressive symptoms. We used Negative Binomial Regression in a sample of 13,164 women to estimate how depressive symptom scores change by income inequality and between White non-Hispanics (WNH), Black non-Hispanics (BNH), and Hispanics.

Results: 33% of obese women experienced some level of depression compared to 23% of non-obese women, with higher prevalence among low and medium PIR groups. PHQ-9 scales were highest in the obese group with 4.2% vs 3% in the non-obese group. PHQ-9 scales were highest in the obese group for moderate (7.9% vs 4.8%), moderate-severe (4.0% vs 1.9%), and severe (1.3% vs 0.7%) categories. Women in lower-PIR had a higher obesity incidence-rate ratio and a higher rate of depression (IRR: 1.22, CI: 1.15-1.29). BNH and Hispanic obese women had lower rates of depressive symptoms (IRR: 0.93, CI: 0.87-0.99 and IRR: 0.92, CI: 0.85-1.00 respectively) compared to WNH obese women.

Conclusion: In treating depressive symptoms, it is important to prioritize obese women of all racial groups in low-PIR. Policy also needs to target WNH obese women with higher rates of depressive symptoms.

Background: Behavioral health (BH) integration in primary care is a promising practice for improving access to care, service utilization, and health outcomes for patients in safety-net health settings. Ensuring that BH integration considers cultural contexts, systems, and workflows that are responsive to diverse patient populations those historically underserved. Assessing equity in integration requires stratifying and monitoring outcome measures for priority populations.

Methods: The Advancing Behavioral Health Equity in Primary Care Learning Collaborative (ABHE-PC) provided multi-disciplinary teams from 14 California community health centers (CHCs) a 20-month engagement to improve equity-supporting capabilities. To support quality improvement and evaluation, the study team developed a Universal Measure Set (UMS) tracking BH care need, access, utilization, coordination, and patient BH and physical health outcomes. Researchers aggregated potential measures through review of federal reporting requirements. Co-design sessions with primary care associations, QI directors, and academic and applied researchers winnowed the measures to a parsimonious set for consideration. The 14 CHC teams participating in ABHE-PC then reviewed the proposed UMS against their reporting capacities. Eleven measures available within the majority of CHC electronic health record systems remained. All measures were stratified by race, ethnicity, language, gender, sexual orientation, and gender identity (SOGI), and submitted three times over the course of the program (2021, 2023) and one-year post-program (2024). Each submission captures a 6-month assessment window. Frequencies and chi-square tests for linear trend will characterize change over time and between strata.

Results: Preliminary results show improvement in stratified outcome reporting, though only half could stratify care coordination measures at endline. Poor SOGI data completeness limits some CHC capacity to track patient outcomes. Staffing ratios have improved due to an increase in licensed BH professionals. The proportion of adult patients receiving any mental health service increased. However, the average number of services received per patient decreased. Reduced disparities in depression screening within some race/ethnicities are present. Full results including data from all three timepoints will confirm trends.

Conclusions: ABHE-PC helped spotlight for teams the importance of tracking stratified outcome measures and strengthened the internal communication channels supporting reporting. However, stratification of key measures remains challenging.

Background: Health disparities are typically studied independently, such that one disparate factor is studied at a time, but the field of public health is increasingly recognizing the importance of intersectionality on an individual’s health and treatment outcomes. This study aims to determine how compounding disparate factors (ie. race, gender, disability status, housing stability) impact health outcomes and client’s perceptions of providers’ cultural competency of CCBHC clients.

Methods: Self-report phone surveys were conducted (2023-2024) on a random sample of clients (n=155) receiving services at three CCBHCs. Baseline and six-month reassessment data was collected using the National Outcome Measures, GAD-7, PHQ-8, and Doctor Patient Relationship Survey. Dependent variables included perception of care, cultural competency, daily functioning, physical health, mental health, and substance use. The independent variable was operationalized as the number of disparate factors endorsed between zero and nine. SPSS was used to conduct independent t-tests and bivariate correlations.

Results: At baseline, mental (p=.005) and physical health (p<.001) were worse for those with multiple disparities. After six months, this discrepancy disappeared for physical health (p=.484), but not for mental health (p=.036). Physical health improvement was greater for individuals facing multiple disparities (p<.001) and individuals who improved in physical health reported more disparities than those who did not improve (p=.003). Additionally, there was a significant correlation between physical health improvement and disparity score (X²=.304, p<.001). Though no relationships were found between disparity score and composite perception of care or cultural competency, there was a small correlation between disparity score and client ratings of providers’ cultural awareness (X²=.168, p=.049).

Conclusions: Clearly, a strong relationship exists between disparity scores and physical health outcomes. Because of the CCBHC model’s emphasis on integrated care, this may be evidence of the continuity of care between physical and mental healthcare providers. Furthermore, those with multiple disparities may have more room for growth, though this does not explain why mental health did not similarly improve. Additional research is needed to understand the nuanced impact of intersectionality on health and healthcare experiences within CCBHCs and to identify steps clinics can take to address the complex needs of those encountering multiple disparities.

Objectives: National data of the impact of COVID-19 on mental health and access to mental health care among children is limited in United States. Using a national representative sample, we assessed the impact of COVID-19 on mental health and access to mental health care access among children.

Methods: Data were from the 2019 and 2022 National Survey of Children’s Health (NSCH). All data were self-reported by parents/guardians. The two outcomes were (1) mental health problems (yes/no) in the past 12 months, and (2) mental healthcare access (yes/no) in the past 12 months. The analytical sample included 71,447 children ages 3-17 years. We ran separate multiple logistic regression models to assess the changes in mental health problems and access to mental health care from 2019 (pre-COVID-19) to 2022 (during COVID-19). All analyses accounted for the survey design of NSCH.

Results: In the mental health model, children in 2022 were more likely to have mental health problems than in 2019 (AOR=1.27, 95% CI: 1.18-1.37); children of minority groups were less likely to have mental health problems than non-Hispanic white children (all p<.001); children whose parents with higher education levels were more likely to have mental problems (all p<.001); and children from safe communities (AOR=0.65, 95% CI: 0.54-0.79) and supportive communities (AOR=0.75, 95% CI:0.70-0.82) were less likely to have mental health problems. In the access to mental healthcare model, children in 2022 were less likely to have access to mental healthcare than in 2019 (AOR=0.78, 95% CI: 0.63-0.97); non-Hispanic black children (AOR=0.52, 95% CI: 0.35-0.77) were less likely than non-Hispanic white children to have access to mental healthcare; and children with health insurance were more likely to have access to mental healthcare (all p<.01).

Conclusions: We found the prevalence of mental health problems in children increased while the prevalence of access to mental health care decreased from 2019 to 2022 due to the COVID-19 pandemic. This trend should be closely monitored. And special efforts are needed to help those children who are affected.

Mental illness is associated with a number of negative outcomes, particularly suicide, and became more prominent during the COVID-19 pandemic. The purpose of this study was to examine trends over time among adults in a number of U.S. cities as well as explore the influence of four demographic factors.

Methods

Poor mental health among those 18 years of age and greater was measured as respondents to the Behavioral Risk Factor Surveillance System (BRFSS) saying they had experienced 14 or more days of poor mental health in the past 30 days. We used a subset of BRFSS data called the Selected Metropolitan / Micropolitan Area Risk Trends (SMART), in order to examine specific cities. We combined data for 2012-2021, restricting to MSAs which had data for all ten years and focused on four demographic factors which had very little missing data. The study sample (n=1,812,931) represented 83 MSAs. Survey-weighted cross-tabulation and multivariable regression was done using SAS 9.4. Seattle, WA was selected as reference for regression due to its prevalence of poor mental health being closest to the overall sample average.

Results

The sample represented an estimated annual population of 118.5 million adults. Overall prevalence of self-reported poor mental health in the selected cities increased from 11.2% in 2012 to 14.1% in 2021. During those 10 years rates increased most for women (12.8 to 16.4%), those 18-24 years of age (12.3 to 22.1%), those with a college degree (6.4 to 10.3%), and those never married (13.2 to 19.8%). Not adjusting for socio-demographics and time, there were 34 cities having higher odds of poor mental health compared to Seattle. With covariates, there were 18 cities having higher odds. In both models, the city with highest rate was Charleston, WV (odds ratios 1.6 and 1.5, respectively) and lowest rate was Sioux Falls, SD (OR 0.7 and 0.7).

Conclusion

The rise in poor mental health during COVID-19 was most notable among women, young adults, and those never married. A handful of socio-demographic factors explained the increase in 16 cities; but not in 18 other cities. Additional study is needed to identify city-specific factors.

Objective: Recent evidence suggests that the COVID-19 pandemic has exacerbated negative mental health outcomes among colleges students. Although subjective social status (SSS) and hopelessness have previously been associated with anxiety and depression, this work is limited among college students. We examined whether SSS and hopelessness are associated with anxiety and depression.

Methods: Data obtained from a 2021 cross-sectional survey of undergraduate students (N=1602) in the Eastern part of North Carolina included outcome measures: anxiety (GAD-7) and depressive (PHQ-9) symptoms. Higher scores indicate greater perceived depression and anxiety. Independent measures included the MacArthur Scale of SSS and hopelessness (Brief-H-Neg); higher scores represent greater perceived SSS and hopelessness. Linear regression models were used to test the interactive effect of hopelessness and SSS on anxiety and depression.

Results: Most students reported moderate to severe depressive (81.44%) and anxiety (62.88%) symptoms. Hopelessness was positively associated with anxiety (r=0.61, p<.001) and depression (r=0.53, p<.001); and social status was negatively associated with anxiety (r=-0.29, p<.001) and depression (r=-0.29, p<.001). While there was no significant interaction effect, SSS (OR: 1.35, 95% CI: 1.21, 1.50) and hopelessness (β: -0.25, CI: -0.37, -0.13) were independently associated with anxiety, after adjusting for age, sex, race/ethnicity, income and college class. Similarly, SSS (β: 1.65, 95% CI: 1.49, 1.81) and hopelessness (β: -0.25, CI: -0.37, -0.14) were associated with depression.

Conclusion: The impact of SSS and hopelessness on depression and anxiety suggests a need to develop relevant college-based public health programs and interventions that addresses determinants of poor mental health outcomes among students.

Background: The COVID-19 pandemic has raised significant concerns about the mental health of children and adolescents, with reports suggesting alarming increases in the need for mental health treatment among these age groups. Despite the growing need, only half of children with mental health problems receive treatment from mental health professionals, and just a quarter of this population receives advanced mental health care. Adverse Childhood Experiences (ACEs) can negatively impact mental health care access through their association with lower socioeconomic status and risky health behaviors, potentially exacerbating the mental health care gap within already vulnerable populations. However, little is known about the association between ACEs and mental health care needs and utilization.

Methods: A cross-sectional analysis of the 2018-2019 National Survey of Children’s Health was conducted, focusing on children aged 6-17 years. The dependent variables were mental health care needs defined by the presence of any of the 10 mental health conditions, and mental health care utilization (i.e., received services in the past year vs. not received). Independent variables were reported ACEs (categorized into no ACE, one ACE, two to three ACEs, and four or more ACEs). The covariates for this study were age, sex, race and ethnicity, insurance status, and household income. Two models of multivariate logistic regression model were used: (i) without covariates and (ii) covariates were used to estimate the associations between ACEs, mental health care needs, and mental health care utilization.

Results: The study sample included 43,213 children and adolescents. Results showed that 28% reported the need for mental health care, and 46% reported one or more ACEs. Children reporting two or more ACEs showed a two-fold increase in mental health care needs (aOR = 2.53; 95% CI: 2.21-2.90) and a two-fold greater likelihood of mental health care utilization in the past year (aOR = 2.11; 95% CI: 1.66-2.67) compared to no ACEs.

Conclusion(s): In this nationally representative sample, ACEs were found to be important predictors of mental health care needs and mental health care utilization among children and adolescents. Children and adolescents with ACEs should be encouraged and facilitated to access mental health care services.