Prevention Science[JOURNAL]
Sun
200 papers
State-Level Extreme-Risk Protection Order Policies, Mental Health, and Gun Carrying: Demographic and Racial Disparities in U.S. Youth.
Gao X
Prev Sci · 2026 Apr · PMID 41964699 · Full text
Gun carrying among U.S. high school students increases risks for violence and injury. Poor mental health is linked to carrying guns, but less is known about how this link varies across demographic and racialized groups....
Gun carrying among U.S. high school students increases risks for violence and injury. Poor mental health is linked to carrying guns, but less is known about how this link varies across demographic and racialized groups. Extreme-risk protection orders (ERPOs) allow temporary removal of firearms from individuals at high risk of harming themselves or others, including during a mental health crisis. Their potential to reduce disparities in youth gun carrying is not well studied. This study assessed whether ERPOs modify the association between poor mental health and gun carrying in high school students overall and subgroups. Data came from the 2023 pooled state Youth Risk Behavior Surveillance System. Gun carrying was the outcome, poor mental health was the predictor, and ERPOs were the moderator, categorized as EROP (less restrictive), Ex parte (moderately restrictive), and Ex parte expanded (more restrictive). Sex, grade, and race were included. Multilevel mixed-effects logistic regression models tested interactions between ERPOs and poor mental health. Demographic and racialized differences in gun carrying were identified. Males had higher odds of carrying guns than females. Poor mental health was associated with higher odds of gun carrying overall and in subgroups, with stronger associations among males, 12th graders, African American students, and Hispanic students. Across the sample and subgroups, ERPOs were linked to reduced gun carrying among students reporting poor mental health. Findings suggest that ERPOs may lower the risk of gun carrying linked to poor mental health and may also reduce demographic and racialized disparities in youth firearm harm.
Engaging Latino Families in a Sibling-Focused Family Prevention Program.
Cahill KM, Updegraff KA, Umaña-Taylor AJ … +1 more , Feinberg ME
Prev Sci · 2026 Mar · PMID 41954835 · Publisher ↗
Enhancing the lives of Latino youth and families through evidence-based prevention programs designed to promote positive adjustment and reduce risk is of high public health significance. An important component of evidenc...
Enhancing the lives of Latino youth and families through evidence-based prevention programs designed to promote positive adjustment and reduce risk is of high public health significance. An important component of evidence-based prevention is successfully engaging participants. The goals of this study were twofold: (a) to describe children's and their caregivers' engagement in a novel family-focused prevention program targeting sibling relationships and parenting of siblings in Latino families and (b) to investigate caregiver and family cultural factors as predictors of program engagement from a socio-ecological perspective. Participants were 158 Latino families who were randomly assigned to the intervention condition and participated in pre- and post-program data collection. Attendance was high, as children averaged 9.82 of 12 sessions (SD = 2.90) and children and at least one caregiver averaged 2.28 of 3 family nights (SD = 0.92). Furthermore, caregiver ratings of program experiences indicated high satisfaction and enjoyment, and qualitative responses illustrated parents' perceived program benefits. Economic hardship was positively associated with family night attendance and program expectations for fathers, and with less clear program expectations for mothers. Mothers' familial ethnic socialization and fathers' familism values were associated with higher program satisfaction and clearer program expectations. In addition, fathers' familism was associated with higher family night attendance. Higher Anglo cultural orientations were related to mothers' lower attendance and fathers' clearer program expectations. Discussion considers factors associated with variability in program engagement in sibling-focused family-based prevention with Latinos and offers future directions. Registered at ClinicalTrials.gov, trial registration number NCT03706014, Intervention study start date 2018-09-29 and study completion date 2025-09-30.
The Utility of Machine Learning-Enhanced Developmental Cascade Models in Prevention Science.
Morales V, Cardozo F, Balise RR … +2 more , St George SM, Feaster DJ
Prev Sci · 2026 Apr · PMID 41926047 · Publisher ↗
Developmental cascade models provide a valuable framework for understanding how risk and protective factors interact over time to shape health and behavioral outcomes. Traditional statistical methods, such as logistic re...
Developmental cascade models provide a valuable framework for understanding how risk and protective factors interact over time to shape health and behavioral outcomes. Traditional statistical methods, such as logistic regression and structural equation modeling, have been instrumental in uncovering developmental pathways within prevention science. However, these methods often impose constraints on model complexity and face limitations in capturing the non-linear and interdependent nature of developmental processes. Machine learning (ML) offers complementary advantages, such as the ability to incorporate high-dimensional data, detect complex interactions, and enhance predictive accuracy. These capabilities can improve identification of at-risk individuals, support the timing of interventions across developmental stages, and refine theory-driven models. By integrating ML with developmental cascade models, researchers can more effectively identify when and how which risk accumulates and protective factors exert influence, thereby improving the tailoring and efficiency of prevention strategies. This conceptual paper outlines how ML can extend traditional analytic approaches in developmental cascade research, discusses key practical considerations for researchers including data requirements, software selection, and model validation, and highlights its potential to advance prevention science across the life course.
Neighborhood Opportunity and Genetic Literacy in a Representative Sample of US Adults.
Bather JR, Goodman MS, Kaphingst KA
Prev Sci · 2026 Apr · PMID 41910920 · Publisher ↗
Research shows that genetic literacy varies as a function of individual-level factors, but these factors may not account for all observed differences in genetic literacy. We tested the hypothesis that neighborhood opport...
Research shows that genetic literacy varies as a function of individual-level factors, but these factors may not account for all observed differences in genetic literacy. We tested the hypothesis that neighborhood opportunity-a structural factor-is associated with genetic literacy. We analyzed nationally representative cross-sectional data on a weighted sample of 606 US adults from the 2024 Measurement of Genetic Literacy Survey. The Genetic Literacy and Comprehension measure assessed genetic literacy ( = 0.87). The Childhood Opportunity Index 3.0 measured overall neighborhood opportunity and three domains (Education, Health and Environment, Social and Economic resources). Unadjusted and adjusted weighted linear regression models quantified the associations between neighborhood opportunity and genetic literacy. Among the weighted sample (mean age = 48, SD = 18), 52% were female, and 61% were as non-Hispanic White. Very low overall neighborhood opportunity was significantly associated with lower genetic literacy (β = - 0.70, 95% CI: - 1.40 to - 0.04, p = 0.037), adjusting for demographic characteristics, health-related factors, and receipt of genetic testing. We observed a similar pattern for exposure to very low social and economic resources (β = - 0.95, 95% CI: - 1.60 to - 0.31, p = 0.004). There was no evidence of a statistically significant association between the Health and Environment domain and genetic literacy in the final model (β = - 0.13, 95% CI: - 0.64 to - 0.38, p = 0.62). Findings indicate that neighborhood opportunity is associated with genetic literacy. These results reinforce the importance of assessing structural factors along with individual-level characteristics in genetic literacy research.
Multi-level Risk and Protective Factors for Vaping Onset and Escalation Among Youth: a Focus on LGBTQ + Disparities.
Anjorin O, Asghari-Kamrani A, Lindley L … +3 more , Mandell CJ, Nakkash R, Griffin KW
Prev Sci · 2026 Mar · PMID 41904763 · Full text
Youth vaping remains a major public health concern, with sexual and gender minority youth experiencing disproportionately higher rates of e-cigarette use compared to their non-LGBTQ + peers. These disparities are not att...
Youth vaping remains a major public health concern, with sexual and gender minority youth experiencing disproportionately higher rates of e-cigarette use compared to their non-LGBTQ + peers. These disparities are not attributable to identity itself, but to social and structural conditions that shape exposure to risk and access to protection. This narrative review synthesizes evidence on shared and LGBTQ + -specific risk and protective factors for vaping initiation and escalation among youth aged 10-24 years. Guided by the Social Ecological Model, findings are organized across individual, interpersonal, organizational/community, and societal levels, and interpreted through two complementary mechanisms: a minority stress-related coping pathway and a socialization and identity pathway. Across levels, discrimination and stigma emerge as cross-cutting risk processes that intensify psychological distress, normalize vaping within social networks, and undermine protective environments. Protective factors, including identity affirmation, supportive relationships, affirming institutions, and inclusive policies, buffer these pathways and promote resilience. This review highlights limitations of one-size-fits-all prevention approaches and underscores the need for selective, mechanism-informed interventions that address both shared developmental risks and the structural drivers of LGBTQ + vaping disparities.
Diffusion of an Effective Social and Behavioral Change Intervention to Prevent Intimate Partner Violence and Enhance Gender-Equitable Norms in Nepal.
Clark CJ, Hadd AR, Shervinskie A … +4 more , Subedi S, Ferguson G, Shrestha B, Baker HS
Prev Sci · 2026 Mar · PMID 41904361 · Publisher ↗
Social norms approaches to intimate partner violence (IPV) prevention are a growing evidence-based practice, but assessment of impact diffusion beyond direct beneficiaries is still relatively rare. We examined the sustai...
Social norms approaches to intimate partner violence (IPV) prevention are a growing evidence-based practice, but assessment of impact diffusion beyond direct beneficiaries is still relatively rare. We examined the sustained impact and diffusion of the Change Starts at Home (Change) intervention with a sample of 1181 married adults in Nepal's Gandaki province. Timepoints included baseline, midline, endline, and 6-month follow-up. Difference-in-difference analyses assessed change in IPV, social norms, and secondary outcomes for two intervention conditions-(1) Listening, Discussion, and Action Group members (LDAG condition) and (2) residents of the community in which LDAGs were present (community condition)-relative to adjacent community residents (condition 3) where no core intervention activities took place. LDAG and community women reported sustained decreases in IPV (27% and 21%, respectively) at follow-up, with significant improvements in descriptive norms, gender equitable attitudes, relationship quality, and communication (for both groups) and injunctive norms, leadership, anti-violence advocacy, in-law violence, financial and sexual decision-making, and diffusion (for LDAGs). The intervention showed little impact on LDAG and community men, in part due to improvement across all conditions. Exposure to other programming unrelated to Change was also detectable, especially among men and to a lesser degree women, in the adjacent and community groups. Change may be effective at reducing IPV, improving associated norms, and extending programming benefits beyond those intentionally targeted. The general lack of intervention impact among men requires further investigation, but with benefits among women persisting at least 6 months post-intervention, replication, and cost-effective analyses are warranted.
Measuring Indigenous Cultural Strengths: a Systematic Review of a Decade of Approaches.
Walls ML, Brown DL, O'Keefe VM … +12 more , White EJ, Gonzalez M, Blackmore I, Maudrie TL, Richardson M, Werwie TR, Rakena HGT, Medley A, Desjardins MM, Sevillano L, Wilson G, Lewis ME
Prev Sci · 2026 Mar · PMID 41893983 · Full text
Indigenous Peoples experience severe health inequities rooted in experiences of colonization. Calls for strengths-based, culturally grounded research and practice to redress these harms continue to grow. This systematic...
Indigenous Peoples experience severe health inequities rooted in experiences of colonization. Calls for strengths-based, culturally grounded research and practice to redress these harms continue to grow. This systematic review assesses how Indigenous "culture" has been quantified in research across the globe over a 10-year span. Following PRISMA guidelines, we searched four databases (2013-2023) for peer-reviewed, English-language articles featuring quantitative measures of Indigenous culture as a positive/protective factor. Data were extracted via a double coding system to assess cultural domains, operationalization strategies, and sociodemographic characteristics of reviewed studies. We reviewed 279 records and identified 461 measures, 289 of which were unique. The most frequently coded cultural domains were connectivity/belonging (47%) and traditional medicine/spirituality (47%), followed by cultural identity (43%) and Indigenous language (36%). Measures most often operationalized culture through behaviors (61%) and beliefs/values (41%). Only 56% of studies reported any psychometric evaluation. Most studies centered on North American contexts and human health outcomes. Efforts to measure Indigenous culture are increasing but remain fragmented, lacking consensus in domain definitions and operational approaches. Results suggest a need for resources to appropriately grapple with and build momentum around ethically and logistically complicated issues surrounding Indigenous cultural measures.
Comparing AI-Assisted Coding and Traditional Qualitative Analysis: a Study Examining Differences in Methods and Results of AI-Assisted Coding and Traditional Content Coding Using Community Engagement Data Collected During the Development of a Municipal Food Plan.
Poulos N, Price H, Bell L … +3 more , Byrd-Williams C, Torres-Peralta S, Marty E
Prev Sci · 2026 Mar · PMID 41886214 · Publisher ↗
Qualitative data poses a challenge for prevention science and public health, as it is critical to explain the context of communities, health, and behavior, yet collecting and analyzing qualitative data using traditional...
Qualitative data poses a challenge for prevention science and public health, as it is critical to explain the context of communities, health, and behavior, yet collecting and analyzing qualitative data using traditional methods is time-intensive and requires extensive training. As artificial intelligence (AI) models have improved, there is a growing interest in using AI to code qualitative data quickly and reliably. This study compares the similarities and differences in methods and results of artificial intelligence (AI)-assisted qualitative analysis to traditional qualitative content analysis using data collected during the development of a city and county-based food plan. In total, 2820 community comments were collected across 43 community events in 27 zip codes across the region between March 2023 and January 2024. AI-assisted analysis was completed using a combination of a transcription app (Post-It), GPT4 Plus, and GPT for Sheets with oversight from a public health practitioner. Traditional qualitative content analysis was completed with two trained coders who completed codebook development, reliability analysis, and full content coding. Both methods used deductive codes to represent key aspects of the food system and generated inductive codes to represent areas not included by the deductive food system codes. Results found that AI-assisted methods and traditional content analysis produced similar deductive coding results, while inductive coding results were less comparable across methods. Given that qualitative data has become a central part of prevention science, we believe with careful considerations, AI-assisted methods with intentional oversight have the potential to strengthen our ability to process large amounts of qualitative data.
Can 1st Grade Data that Schools Can or Do Routinely Collect Predict Suicide Attempts Among Black Youth?
Singletary BD, Ialongo NS, Hung IT … +2 more , Lindsey MA, Rabinowitz JA
Prev Sci · 2026 Mar · PMID 41879912 · Publisher ↗
Suicide rates among Black youth have shown disproportionately increasing trends in recent years. This study examined whether teacher-reported behaviors in early childhood were associated with suicide attempts in young ad...
Suicide rates among Black youth have shown disproportionately increasing trends in recent years. This study examined whether teacher-reported behaviors in early childhood were associated with suicide attempts in young adulthood among Black youth. Participants were drawn from a randomized controlled trial of school-based interventions conducted in an urban Mid-Atlantic school district; intervention status was included as a covariate. Of the original 2311 participants, 1516 identified as Black. Analyses were estimated using full information maximum likelihood (FIML), under a missing-at-random assumption, yielding an analytic sample of 1510 Black youth (51.7% female). First-grade predictors included teacher-rated authority acceptance, attention-concentration, social interaction, and school-recorded absenteeism. Participants self-reported lifetime suicide attempts during follow-ups in young adulthood at ages ~ 20, ~ 21, and ~ 30. Analyses tested associations between early behavioral indicators and suicide attempts, stratified by sex. Among males, the lower the level of teacher-rated social interaction, the greater the odds of suicide attempts (aOR = 1.29, 95% CI 1.01-1.66). Among females, higher attention-concentration problems predicted greater odds of suicide attempts (aOR = 1.14, 95% CI 1.03-1.28). No significant associations were observed for authority acceptance or absenteeism. These findings underscore the potential utility of early school-based screening to identify Black children at elevated risk for suicide attempts later in life.
Building and Validating an Explainable Machine Learning Model for Predicting Health-Promoting Behaviors in Older Adults: A Multicenter Study.
Zhang P, Hou Y, Zhai Y … +6 more , Tian Y, Yang Y, Li T, Lu D, Zhou L, Wu T
Prev Sci · 2026 Apr · PMID 41874782 · Publisher ↗
Enhancing health-promoting behaviors (HPBs) in older adults is crucial for chronic disease management and healthy aging in the context of population aging. Accurate assessment of individual HPB levels can facilitate the...
Enhancing health-promoting behaviors (HPBs) in older adults is crucial for chronic disease management and healthy aging in the context of population aging. Accurate assessment of individual HPB levels can facilitate the development of personalized interventions. This study aimed to identify factors influencing HPBs in older adults using multicenter data and to develop and validate an interpretable machine learning (ML) model for prediction. We conducted a multicenter cross-sectional study among 781 older adults in Shanghai, Jiangsu, and Shandong from June 2024 to September 2025. The collected data included sociodemographic characteristics, health status, community sports facility conditions, mobile phone proficiency, and internet skills. Data from the Shanghai (n = 319) and Shandong (n = 228) centers formed the training set, and data from the Jiangsu center (n = 234) constituted the independent external test set. Model discrimination was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, specificity, positive and negative predictive value (PPV, NPV), recall, and F1-score. Calibration was assessed with the Hosmer-Lemeshow test and Brier score, and clinical utility was evaluated via decision curve analysis (DCA). The mean age of participants was 61.79 ± 11.54 years. Based on HPB levels, 436 (55.8%) participants were categorized into the HPB group and 345 (44.2%) into the no HPB group. On the external test set, the Stochastic Gradient Boosting Trees (SGBT) model demonstrated optimal performance, with an area under the curve (AUC) of 0.891 (95% CI, 0.848-0.951), excellent calibration (Brier score = 0.103), and a calibration curve closely aligned with the ideal line. Additional metrics included accuracy (0.895), specificity (0.867), PPV (0.897), NPV (0.892), recall (0.917), and F1-score (0.907). DCA indicated a high net clinical benefit across a wide probability threshold range (0-0.6). SHAP analysis elucidated the contribution of each feature, and a user-friendly online prediction platform was deployed. We developed a high-performance, interpretable ML model to predict HPBs in older adults, and systematically identified key predictors such as internet proficiency, educational level, and functional independence. This tool can assist healthcare professionals in rapidly assessing HPB levels, facilitating the precise delivery of health information and services.
Intervention Mapping to Develop "Creating Peace:" Translating a Youth Activism Intervention from Global Settings to the United States Urban Context.
Fields AD, Ragonese C, Friedman FS … +4 more , Garth J, Lauro G, Slegh H, Miller E
Prev Sci · 2026 Mar · PMID 41866459 · Full text
Creating Peace is a youth violence prevention program adapted for implementation in Pittsburgh, Pennsylvania, from prior work in conflict-affected global settings. This manuscript outlines the adaptation process, emphasi...
Creating Peace is a youth violence prevention program adapted for implementation in Pittsburgh, Pennsylvania, from prior work in conflict-affected global settings. This manuscript outlines the adaptation process, emphasizing integration of healing-centered, gender-transformative, and community-driven approaches to address violence and structural inequities. An intervention mapping adaption process involved facilitators from youth-serving organizations to co-develop a curriculum for youth ages 14-19 years. Adaptations included optimizing facilitator training to discuss United States history of racism and gender inequity, addressing concerns around law enforcement engagement, and promoting youth-led activism as a mechanism for change. The adaptation process highlighted the importance of local contextualization, ensuring the curriculum was responsive to community needs while maintaining core programmatic goals. Facilitator training and ongoing support as well as iterative feedback from youth were critical for optimizing curriculum implementation. This development of Creating Peace, using a systematic intervention mapping adaptation approach, demonstrates the feasibility of translating global lessons into locally relevant programming. Future efforts should include assessing long-term impact, establishing best practices for adapting gender-transformative violence prevention programs in diverse contexts, and involvement of youth and adult community members in sustainability and scalability of such community-based prevention programming.
Balancing In-Person and Virtual Home Visits: a Descriptive Analysis of Family-Level Outcomes.
Estes K, Kilburn MR, Cruz TH
Prev Sci · 2026 Mar · PMID 41863588 · Publisher ↗
Home visiting programs have traditionally provided their services in person. Due to COVID-19, programs were forced to change this prior practice and began offering services virtually. Prior literature has documented the...
Home visiting programs have traditionally provided their services in person. Due to COVID-19, programs were forced to change this prior practice and began offering services virtually. Prior literature has documented the successes and challenges of offering home visiting services virtually, but little is known about today's practice of offering hybrid (combining in-person and virtual) services. The purpose of this descriptive analysis is to examine the relationship between family-level home visiting outcomes and the percentage of home visits a family completes in person. We examine administrative data from home visiting programs funded by the state of New Mexico during a period of hybrid home visiting. A total of 3180 families who received a mix of virtual and in-person visits are included in the analysis. Family-level home visiting outcomes measuring program dosage, program engagement, screenings, and referrals for outside services are examined. We find the percentage of visits conducted in person was not associated with the total number of visits received. However, a higher proportion of in-person visits was associated with greater total home visiting hours, higher likelihood of completing the home visiting program, increased screenings, and a greater number of referrals to outside services. These findings suggest that while hybrid home visiting may maintain visit frequency, in-person visits may facilitate the home visiting curriculum and goals.
Evaluating A Brief Telehealth Positive Parenting Intervention: A Randomized Controlled Trial.
Ricker BT, Cooley JL, Dennis VE … +4 more , Streicher BE, Mitchell TB, Cummings C, Singer J
Prev Sci · 2026 Mar · PMID 41863587 · Full text
Additional work is needed to increase the accessibility and scalability of parenting interventions. Selected Positive Parenting Program (i.e., Triple P) is a brief, evidence-based parenting intervention, but the efficacy...
Additional work is needed to increase the accessibility and scalability of parenting interventions. Selected Positive Parenting Program (i.e., Triple P) is a brief, evidence-based parenting intervention, but the efficacy of telehealth delivery has not yet been tested. This three-arm randomized controlled trial evaluated the feasibility, acceptability, and preliminary efficacy of Selected Triple P delivered as a universal intervention through telehealth. We also investigated whether an active discussion component administered by mental health providers improved outcomes. This clinical trial was retrospectively registered on 3/3/2025 (NCT06865183). Participants included 97 parents/caregivers of children ages 2-12 randomly assigned to one of three conditions: intervention-as-usual (90-min condition; n = 32), intervention without an active discussion (60-min condition; n = 24), or waitlist control (n = 41). Parents completed assessments at baseline and 2-months post-intervention. No baseline differences were found between conditions on any demographic or outcome variables. Parents in both intervention conditions reported high levels of satisfaction and acceptability, with no significant differences observed between groups. Results indicated that, compared to the control group, parents in the 90-min condition reported significant increases in positive reinforcement and positive parenting, along with reductions in parental hostility; decreases in negative parenting also approached significance. Although changes were not statistically significant for parents in the 60-min condition relative to the control group, there were no significant differences between intervention conditions across these outcomes. Current findings have important implications for the scalability and dissemination of universal parenting interventions, as well as strategies for increasing access to mental health services among families. ClinicalTrials.gov 3/3/2025; Registration Number: NCT06865183. Trial Registration.
Designing Health Care Policy and Systems to Reach and Retain Those with Severe Mental Illness in HIV Prevention and Care Efforts: Insights from Ten US States.
Beg W, Koester KA, Walkup J … +4 more , Cournos F, Crystal S, Mangurian C, Arnold EA
Prev Sci · 2026 Mar · PMID 41860685 · Publisher ↗
People with severe mental illness (SMI) have a higher risk of HIV infection than the general US population. Despite this increased prevalence, robust efforts to engage this population in routine HIV testing, timely linka...
People with severe mental illness (SMI) have a higher risk of HIV infection than the general US population. Despite this increased prevalence, robust efforts to engage this population in routine HIV testing, timely linkage to and ongoing engagement in care, and treatment adherence have been challenging. We sought to identify solutions at the health care delivery system level, as well as policies at the state and local level, that might affect HIV-related health outcomes for this key population, particularly those that featured integrated care models. We recruited key stakeholders in 10 states within the United States, based on HIV prevalence and approaches to HIV and mental health services. From June 2018 to May 2021, we conducted 64 in-depth interviews with key informants which lasted 45-90 minutes. Interviews were recorded and transcribed. Informants included state and county-level public administrators in HIV and/or behavioral health; HIV and behavioral health clinic administrators, medical providers, and case managers; service providers at non-profit agencies serving those with SMI and/or HIV; and academics. Interview topics included HIV testing policies and systems, HIV linkage and treatment policies and organization of SMI and HIV health care systems, funding streams, recent care integration efforts, and electronic data sharing. Transcripts were coded for broad themes, and segments were further reduced for key content. Key factors leading to improvements in HIV-related outcomes for those with SMI include state and local efforts to integrate physical and mental health services, funding requirements for grantees to perform HIV testing, shared electronic health records, ability to combine funding streams, expansion of telehealth services to deliver psychiatric and behavioral health care, and charismatic leaders at the state or local level championing care integration. Fragmented and decentralized physical and behavioral health care systems that disincentivized care integration, policies that limited data sharing, stigma, a depleted workforce, and a lack of collegial working relationships between behavioral health and infectious disease leaders were perceived barriers to improved HIV outcomes for those with SMI. Scaling up facilitators of implementation and addressing the present challenges has the potential to promote successful implementation of integrated care models and positively impact the health and well-being of individuals living with HIV and SMI.
Pathways to Prosociality: How Classroom Strategies That Support Basic Psychological Needs Foster Prosocial Behavior in Children.
Sun Y, van Loon AWG, Kaufman TML
Prev Sci · 2026 Mar · PMID 41851565 · Full text
This study investigated how specific components of a classroom-based intervention promote prosocial behavior in primary school children, guided by Basic Psychological Needs Theory (BPNT). Drawing on data from a large-sca...
This study investigated how specific components of a classroom-based intervention promote prosocial behavior in primary school children, guided by Basic Psychological Needs Theory (BPNT). Drawing on data from a large-scale intervention (N = 1132, 43.3% girls, M = 9.96, SD = 1.22), we examined two sequential mediation pathways: (1) the quantity pathway: whether the presence of intervention components (meaningful roles, peer compliments, and democratic classroom meetings) predicted prosocial behavior through increased general need fulfilment (autonomy, competence, relatedness) and prosocial motivation; (2) the quality pathway: whether the specific basic psychological needs fulfilment within each component predicted prosocial behavior through increased general need fulfilment and prosocial motivation. Results supported both mediation pathways. For the quantity pathway, the presence of compliments and democratic classroom meetings showed indirect effects through general autonomy and relatedness, whereas meaningful roles exerted a primarily direct influence. For the quality pathway, component-specific relatedness emerged as the most consistent predictor of prosocial behavior, followed by more selective effects for component-specific autonomy, namely, only for meaningful roles. This suggests that everyday classroom strategies and experiences can incrementally contribute to general need fulfillment, thereby promoting prosocial behavior. These findings underscore the importance of identifying specific mechanisms of interventions and highlight how need-supportive strategies can foster prosocial development in school settings.
Disseminating Health Prevention Programs Using AI-Crafted, Culturally Tailored Ads for Somali American Adolescents.
Simenec T, Ibrahim S, Ferguson G
Prev Sci · 2026 Mar · PMID 41851564 · Publisher ↗
Recent advancements in artificial intelligence (AI) and digital health messaging present opportunities to bridge service health gaps in underserved communities. Adolescents are highly engaged with technology; however, th...
Recent advancements in artificial intelligence (AI) and digital health messaging present opportunities to bridge service health gaps in underserved communities. Adolescents are highly engaged with technology; however, they lack the access to relevant health information and resources tailored to their specific needs. This may be especially true for Black migrant-background adolescents. Utilizing publicly available AI resources can support the dissemination of culturally responsive digital prevention programs. The present study developed four 1-min direct-to-consumer (DTC) video health advertisements that varied in cultural and developmental tailoring, using focus-group-generated digital health message preferences. The advertisements promoted a culturally adapted food-focused media literacy program, JUS Media? Global Classroom - Somali American. The four ads were experimentally tested among 230 Somali American adolescents (ages 11-19, M = 14.9; 67.5% female) who through age-stratified randomization viewed the DTC video ads. Surveys assessed personal relevance, elaboration (i.e., critical thinking after receiving the message), message liking, and intentions to use the advertised online health intervention. Hierarchical regression analyses revealed that cultural adaptation significantly increased three of four primary indicators of program effectiveness-personal relevance, liking, and usage intentions-whereas developmental adaptation did not significantly predict outcome variables. Insights from the study offer novel strategies in applying human-centered AI to co-develop tailored digital advertisements promoting digital prevention programs. Cultural tailoring of health messages has been criticized for requiring a substantial degree of human and financial capital. Publicly available AI tools can efficiently design tailored develop tailored direct-to-consumer (DTC) ads to disseminate digital health programs and overcome logistical barriers. This is important given delivering health interventions to youth, perhaps especially Black migrant-background youth, is often challenging due to institutional gatekeepers and lack of culturally responsive interventions and recruitment/dissemination strategies (Hill, L., Ndugga, N., & Artiga, S. (2024, June 11). Key data on health and health care by race and ethnicity. KFF. https://www.kff.org/key-data-on-health-and-health-care-by-race-and-ethnicity/). To propel accessibility and scalability of services to support Black migrant youth in the U.S., who are at risk for unhealthy eating due to disproportionate targeting of unhealthy food messages, prevention and intervention programs can capitalize on the very same tactics used by junk food advertisers to strategically develop and disseminate health services. By leveraging Artificial Intelligence (AI) to develop DTC marketing of digital prevention programs, service providers can effectively reach audiences most likely to benefit, offering a more cost-effective alternative to traditional recruitment strategies (Simenec et al., Journal of Child and Family Studies 32:1425-1437, 2023). Most research on tailored DTC strategies has been conducted with adults and only after intervention efficacy studies, delaying the reach/impact of targeted youth health programs. This experimental study investigates the effectiveness of culturally and developmentally tailored DTC video ads for a digital food-focused media literacy program for Somali American adolescents. Ads were developed with the support of AI tools and in partnership with the Somali American community.
Quasi-randomization to Cannabinoid Condition in Studies of US Legal Market Cannabis: Characteristics of Accepters Versus Decliners of Condition Assignment.
Skrzynski CJ, Bryan AD, Schmiege SJ
Prev Sci · 2026 Mar · PMID 41820717 · Full text
Although random assignment is the standard for drawing causal inferences within clinical trials, it is generally precluded in legal market cannabis research given its federal classification as a schedule 1 drug. Unfortun...
Although random assignment is the standard for drawing causal inferences within clinical trials, it is generally precluded in legal market cannabis research given its federal classification as a schedule 1 drug. Unfortunately, this may cause selection bias and compromise internal validity, and thus, alternative approaches are necessary. One such approach in this context, as well as more broadly, is quasi-random assignment whereby participants are randomly assigned to conditions but can accept or decline this assignment. This study explores whether those who accept or decline condition assignment differ in ways that impact study outcomes and informs best practices for other research areas where random assignment is not feasible or permitted. Data came from two studies examining cannabis, inflammation, and insulin sensitivity. The first included individuals who infrequently used cannabis; the second included regular users. Across studies, individuals were quasi-randomly assigned via dice roll to purchase and use either THC-dominant, CBD-dominant, or approximately equal THC:CBD ratio flower products for 1 (study 1) or 4 weeks (study 2). Demographics, cannabis use, health behaviors (e.g., exercise), and anthropometrics (i.e., body mass index [BMI]) were compared across individuals who accepted versus declined their assigned condition. Most participants accepted their assignment (83% and 63% for studies 1 and 2, respectively). Those who accepted did not differ from those who declined on any variable (ps > 0.11). While findings cannot rule out a selection process outside the variables assessed, results support use of this methodology in situations where true random assignment is not possible. Clinical trials: The larger project from which the current paper draws data was pre-registered on Clinicaltrials.gov (NCT04114903) on 09-06-2019.
Characterizing the Substance Use Prevention Funding Landscape in the United States: a Cross-Sectional Study of National Prevention Network Representatives and Practitioners.
Liu SS, Elek E, Blackburn N … +3 more , Wondimagegnehu F, Ballard PJ, Graham PW
Prev Sci · 2026 Mar · PMID 41803401 · Full text
Youth and adolescent substance use remains a persistent public health challenge in the United States; the delivery of evidence-based interventions (EBIs) is critical to improving related negative consequences. The Substa...
Youth and adolescent substance use remains a persistent public health challenge in the United States; the delivery of evidence-based interventions (EBIs) is critical to improving related negative consequences. The Substance Abuse and Mental Health Services Administration funds a large portion of the implemented substance use prevention interventions in the United States by supporting a funding infrastructure that plays an important role in the adoption and scaling of interventions. Prevention intervention developers and researchers need to understand this infrastructure and its influence on local practitioners to increase the adoption of their EBIs. This study sought to identify which agencies in each state and jurisdiction are involved in funding allocation, how they prioritize and distribute funding to intervention implementers, and, subsequently, how they guide the selection of EBIs. This study used a mixed-methods, cross-sectional design to understand the infrastructure of prevention funding that underlies EBI decision-making. In 2023, we conducted surveys with 40 National Prevention Network representatives (NPNs) and 222 community-level practitioners; in early 2024, we conducted qualitative interviews with a subset of 16 NPNs. NPNs' priorities were shaped by the agencies in which they were housed and the partners with whom they collaborated. Most were located within their state's or jurisdiction's department of health or behavioral health, and many engaged in partnerships with departments of public health or education. Most NPNs reported that they prioritized school and health settings and youth populations for prevention intervention delivery. Almost all NPNs directly distributed funding to intervention implementers (community, regional, or state entities); about half distributed some funds through an intermediary that then subcontracted another entity to implement interventions. More NPNs required or recommended that funded recipients select EBIs from lists or registries (75%) than required or recommended a specific strategy for at least some of their programs (53%). Many practitioners (47%) reported that they selected a recent strategy from a list of interventions provided by their funder, but 27% received no funder guidance on intervention selection. Prevention developers and researchers could increase the adoption of EBIs by focusing them on the priority areas for NPNs, including the health, behavioral health, and education sectors. Developers need to get their EBIs onto registries or intervention lists and increase the EBIs' wide-scale dissemination. Audiences for information about specific EBIs should include NPNs, regional entities, and their funded community practitioner recipients.
Using Machine Learning to Predict Features Within Substance Use Disorder Treatment Service Settings That Increase the Likelihood of Positive Treatment Outcomes.
Becker T, Gonzalez-Martinez A
Prev Sci · 2026 Mar · PMID 41774402 · Publisher ↗
Given the conceptual issues involved in defining and measuring recovery and accordingly substance use disorder (SUD) treatment outcomes, the role of each state's treatment system and social factors, the objective is to e...
Given the conceptual issues involved in defining and measuring recovery and accordingly substance use disorder (SUD) treatment outcomes, the role of each state's treatment system and social factors, the objective is to examine underlying and interrelated patterns within SUD treatment, outcomes, and recovery. Using a recovery-oriented framework, a Machine Learning Random Forest model was developed to analyze publicly funded SUD treatment services across the United States. The aim was to predict the 10 most important features that increase the likelihood of positive treatment outcomes, defined as less substance use (SU) or abstinence. Over 78% of SUD treatment services were provided to individuals either with Medicaid coverage or were uninsured. The most important feature identified was the number of days in treatment, regardless of setting. The second most important feature was the state and whether various treatment services were available. The third and fourth ranked features were the type of treatment at discharge and at admission, respectively. Housing status, SU self-help group participation, and employment were lower ranked. Referral source was the tenth ranked feature. The length of time in SUD treatment is consistent with the clinical perspective of the individual seeking treatment and continuing in care and recovery support. Individuals in Medicaid-funded treatment live in poverty, with peer support and community who have the least resources to support their recovery journey. States that prioritize behavioral health should coordinate to increase the availability of higher-cost, longer-duration treatment services across state lines, to states with low availability.
Leveraging Machine Learning to Understand the Link Between School Climate and Youth Substance Use: a Focus on Cannabis and Alcohol Use.
Ünlü A, Skrzypek C, Bradshaw CP
Prev Sci · 2026 Feb · PMID 41762379 · Publisher ↗
This paper focuses on school climate indicators, which have been previously linked with aspects of students' well-being and school-related success, to explore how they relate to alcohol and cannabis use. We used machine...
This paper focuses on school climate indicators, which have been previously linked with aspects of students' well-being and school-related success, to explore how they relate to alcohol and cannabis use. We used machine learning (ML) approaches and leveraged data from a diverse sample of 69,513 students (45.4% White, 23.9% Black, 8.9% Latine) across 111 middle and high schools, with 12% (n = 7783) reporting cannabis use and 18.8% (n = 12,220) reporting alcohol use in the past 30 days. We focused on 154 items related to school climate, student attitudes and behaviors, and demographics. We employed a two-stage feature selection method, initially reducing the 154 features to 31, and subsequently to 20, for both alcohol and cannabis use. Alcohol and cannabis use shared 15 common features and 5 distinct features, though some variation occurred across these two outcome variables. We identified both unique and shared factors that best classified current users vs. non-users. Specifically, gender, sense of pride in the school, weapon carrying, and bullying others were unique indicators that best classified alcohol use. In contrast, difficulties overcoming challenges, problems controlling temper, and becoming angry easily were more strongly associated with cannabis use. Shared indicators associated with both substances included gang membership, skipping school, violent behavior, school-parent and school-student engagement, and gambling. The inclusion of diverse classification factors underscored ML's ability to capture complex social and environmental factors that may be associated with substance use differently across student demographics. These features were tested in 12 classification models for both substances, achieving ROC-AUC scores up to 86% with fine-tuning of the best-performing models. The results highlight the utility of ML for examining complex, multidimensional indicators associated with substance use that complement traditional models.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe