Prevention Science最新文献

Artificial intelligence and machine learning (AI/ML) in prevention science may improve or perpetuate health inequities. Community engagement is one proposed strategy thought to empirically mitigate bias in AI/ML tools. We outline how to incorporate community engagement at every stage of the model development and implementation. Borrowing from a framework for phases of prevention research, we describe the value and application of engaging communities to help shape more rigorous and relevant applications of AI/ML for prevention science. We provide concrete examples from real-world applications, including efforts in suicide prevention with Indigenous communities, on chronic disease prevention for Hispanic and Latino populations, and a community-driven effort to leverage AI/ML to improve allocation of resources focused on social determinants of health for Native Hawaiians. This work aims to provide applied examples of how community-engagement has been incorporated into AI/ML development and implementation, with the goal of encouraging those in the prevention science field to consider the voices of the community as the use of such tools grows. Engaging with the community around AI/ML is critical to ensure these tools reach populations in need and advance health equity for all.

This systematic review synthesizes empirical research on organizational strategies that support the sustainability and scale-up of community-based interventions designed to promote youth psychological well-being. While research has established the effectiveness of youth mental health interventions in community contexts, less is understood about the processes that ensure their long-term sustainability and scale-up. A search across seven databases yielded 27 eligible empirical studies (2003-2025) including peer-reviewed and grey literature. The Synthesis without Meta-analysis (SWiM) and the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) standards provided a framework for conducting and documenting the review. Evaluations of methodological quality were carried out using Strengthening the Reporting of Observational Studies in Epidemiology (STROBE), Consolidated Standards of Reporting Trials (CONSORT), and the Critical Appraisal Skills Programme (CASP). Two dominant strategies for intervention retention and effectiveness were identified: training and technical assistance (12 studies, 44%) and capacity building with implementation support (12 studies, 44%). The main barriers to sustainability and scale-up were reported as staffing as well as other resource limitations (financial, human, and technical), while facilitators included strong communication and leadership, sufficient funding and partnerships, and access to training and technology. Eleven studies (41%) sustained interventions beyond the initial funding period, with key factors including intervention fidelity, agency incentives, organizational support, therapist retention, and stakeholder engagement. The findings highlight the potential benefits of embedding sustainability and scale-up strategies in the pre-implementation phase, anticipating foreseeable barriers, and considering intervention fidelity, workforce stability, and organizational readiness as important precursors to sustainability and scale-up.

Pre-exposure prophylaxis (PrEP) is a highly effective biomedical prevention tool for HIV yet remains underutilized among key populations, particularly among young sexual and gender minorities (SGM). Recognizing the popularity of specific dating and social media apps among SGM young adults, we leveraged user data from these platforms to build a machine learning (ML) model that could inform targeted, data-driven interventions aimed at improving PrEP uptake and adherence. We adapted eWellness, an Android mobile app, to passively collect data from research participants capturing mobile app usage, keystroke patterns and logs, and GPS location data between 2021 and 2024. These data were used to train a ML model to predict self-reported PrEP use. Model accuracy was evaluated through F1 scores across different data types and feature combinations. Study protocols were developed in collaboration with community partners and adhered to strict ethical and privacy standards. A total of 82 SGM young adults participated, with 46 (56%) reporting PrEP use at baseline. Our machine learning model demonstrated good predictive accuracy for predicting PrEP use and non-use, achieving an F1 score of 0.84 (PrEP use) and 0.82 (non-use) outcomes when incorporating data from all mobile apps, including messaging, dating, and social media mobile apps. By contrast, predictions based solely on social media mobile app usage, language associated with sexual behavior and substance use risk, or location monitoring demonstrated worse accuracy (F1 scores of 0.79/0.75, 0.70/0.57, and 0.70/0.52, respectively). Additional feature extraction methods, as well as various combinations of these features, were also tested. However, none achieved predictive accuracy as well as the model incorporating all mobile app usage data combined. This study demonstrates the potential of machine learning to accurately predict PrEP use status among SGM young adults. The findings offer a foundation for developing more personalized PrEP promotion strategies, particularly among SGM young adults who use social media and dating apps. Future research should assess the model's adaptability across diverse SGM subgroups to further inform intervention development. Registry: ClinicalTrials.gov, ID: NCT04710901, November 9, 2020.

Despite well-studied benefits of preventive parenting programs for early child development, various real-world barriers may impede families from engaging in those programs. The current study aims to provide new insights into family engagement by examining enrollment, retention, and involvement and their predictors in an evidence-based universal pediatric primary care parenting program for families with young children. Data (n = 204) were from an ongoing longitudinal randomized controlled trial of PlayReadVIP in Flint, MI. Families (66% Black, 35% White, low socioeconomic status) had high enrollment and retention in the program. As exploratory analyses, random forest models, a machine learning method, identified a multitude of sociodemographic, psychosocial, and contextual predictors of retention and involvement in PlayReadVIP across the first 9 months. As confirmatory analyses, multiple regressions showed that COVID-19 significantly hindered retention (odds ratio = .04; b =  - .30) and involvement (b =  - .31) and that higher parenting self-efficacy was associated with lower retention (odds ratio = .76). Furthermore, the association between COVID-19 and family engagement was moderated by household income, suggesting that families with the highest economic risks were less likely to attend and be actively involved in sessions during the pandemic. This study addresses important research gaps by focusing on multiple aspects of family engagement in a pediatric program during infancy, assessing whether experiencing contextual adversity hinders or motivates engagement, and employing a machine learning method. These findings have crucial implications for designing and implementing early childhood prevention parenting programs to more effectively engage families with higher needs.

Vaccination played a critical role in controlling the 2022 mpox outbreak in the United States, yet only a quarter of eligible individuals are estimated to be fully vaccinated. Using mixed methods, we analyzed cross-sectional data collected from 60 sexual and gender minority people with and without HIV (PWH and PWOH) in August 2022. We investigated willingness to receive an mpox vaccine and to participate in mpox vaccine research, as well as actual vaccine uptake and experiences, and compared results by HIV status. Two-thirds of participants (66.7%) had received at least one dose of the mpox vaccine, including 61.5% of PWH and 70.6% of PWOH. PWH and PWOH had similarly high levels of willingness to receive an mpox vaccine. In open-ended responses, participants described personal and societal benefits of vaccination and had positive attitudes towards vaccines. Despite high vaccine acceptance and uptake, satisfaction with the vaccine rollout was low. A majority of unvaccinated participants who had tried to get the vaccine reported that they were unable to access it. Willingness to participate in mpox vaccine research was lower than willingness to receive a vaccine, though still moderately high. Some participants expressed uncertainty about research participation, concerns about safety, and negative attitudes toward research in general. PWH were more willing to participate in an mpox vaccine trial than were PWOH; however, this difference was not statistically significant. These results can inform future mpox research, vaccine rollout, and community engagement for vaccine trial recruitment.

Pediatric firearm injuries represent a leading cause of morbidity and mortality in the USA, yet clinical care models for youth with firearm injuries remain fragmented and unstandardized. Unlike other serious pediatric conditions such as asthma and cancer, which benefit from established interdisciplinary care pathways, firearm injuries are often treated as isolated events without structured follow-up or coordinated support systems. This absence of standardized clinical guidelines reflects broader systemic biases in how these injuries are perceived and managed within health care. Drawing on the Injury Equity Framework and established chronic disease management models, we propose comprehensive, trauma-informed clinical guidelines for pediatric firearm injury care spanning primary, secondary, and tertiary prevention. These guidelines emphasize universal firearm safety screening integrated into routine pediatric visits, standardized protocols for acute trauma care with automatic interdisciplinary team activation, and structured longitudinal follow-up addressing medical, behavioral health, and social needs. By comparing current fragmented approaches to evidence-based care pathways used in asthma and cancer management, we identify critical gaps in continuity and comprehensiveness of firearm injury care and offer specific, actionable recommendations to close these gaps.

Black youth in New Orleans experience some of the highest rates of violence in the US. Although Youth Participatory Action Research (YPAR) is recognized as an effective tool for empowering young people and fostering social change, its application in violence prevention remains an area worthy of further application. This case study presents a YPAR-based program designed to address structural issues, including racism, discrimination, and economic disinvestment, that contribute to youth violence. The Enrichment to Empowerment (E2E) program provides a platform for youth to advocate for systemic change in their communities. This paper offers actionable insights into equity-centered approaches for violence prevention, highlighting the potential for YPAR programs as a strategy for addressing complex, community-driven issues through youth-led initiatives.

Cognitive-behavioral treatments (CBTs) for adolescents with ADHD demonstrate promise of long-term effects on outcome. However, their implementation in routine care community clinics faces barriers that impact quantity, efficiency, and quality of delivery, as well as client outcomes. This study is a randomized controlled trial designed to evaluate the impact of an AI-assisted service delivery model on therapist implementation of Supporting Teens' Autonomy Daily (STAND), a CBT blended with Motivational Interviewing (MI) for adolescents with ADHD. Adolescents with ADHD (N = 51), who were clients at three community mental health agencies, received treatment from 23 therapists. There was randomization of adolescents and therapists to AI-assisted or standard implementation supports. In addition to standard supports (i.e., training, standard facilitation resources, technical assistance, case supervision), AI-assisted support package included digitized facilitation resources housed in a clinical dashboard (Care4), feedback on content fidelity, and AI-generated feedback on MI implementation quality. The AI-assisted group was associated with more efficient treatment delivery and lower number of appointments attended by the adolescent. There was also a significant decrement in MI quality over time in the AI-assisted group compared to the standard support group. Feedback in focus groups indicated that therapists perceived a task-oriented mindset to be associated with receipt of the AI-assisted support package, leading therapists to prioritize efficiency over relational aspects of therapy. Following the results of this trial, a future, larger RCT should examine the impact of the AI-assisted implementation model on mental health outcomes and cost savings to organizations, third party payers, and clients. Trial registration number: NCT05135065; https://www.clinicaltrials.gov ; Registered September 2021.

In 2020, firearm violence became the leading cause of death for American children and teens, a critical datapoint informing the 2024 U.S. Surgeon General's advisory on firearm violence. However, firearm violence has been a leading cause of death for Black youth-particularly, Black males-for decades, disproportionately impacting their morbidity and mortality. As the rights of Black youth to experience safety from firearm violence converge with the interests of white youth now increasingly impacted by it, it is imperative that prevention scientists critically interrogate what contributes to the national willfulness to see Black youth as perpetrators of violence deserving of punishment and a national reluctance to see Black youth as victims of violence deserving of healing and prevention? Guided by the Cycle of Dehumanization framework, we contend this pattern is symptomatic of racial dehumanization, a root cause of structural racism and violence. We argue that to successfully prevent youth firearm violence using a structural approach, we must disrupt dehumanizing narratives about Black male criminality and offer a viable solution through our visual storytelling and narrative change campaign, In All Ways Human. Using an adapted community engaged participatory action research approach and qualitative interviewing, our multimodal narrative change project captured 50 strategically disseminated (mural, billboards, kiosks, exhibits, digital galleries) portraits and stories that construct a counter-narrative with the power to prevent youth violence by transforming the ways in which Black males are seen, see one another, and see themselves. The impact and future directions of our narrative change effort are discussed.

Mediation analysis is essential for understanding causal mechanisms and designing effective policy interventions. This paper explains a comprehensive framework for implementing VanderWeele's (2009) three-way decomposition using natural effect models (NEM), focusing on its application in health equity research. While three-way decomposition helps social epidemiologists understand how social exposures influence health outcomes through differential mediator effects, its implementation through NEM has not been fully utilized. Using the R package medflex to implement NEM, we demonstrate the methodology through two case analyses: healthcare expenditure disparities between Medicaid beneficiaries and health insurance enrollees using continuous outcomes, and racial/ethnic disparities in preterm birth using binary outcomes. We provide detailed interpretations of NEM estimates in terms of three-way decomposition and discuss methodological considerations for researchers designing health policies to reduce disparities.