Tania de Jesus Espinosa

Background: Economic instability may shape substance use (i.e., cigarette use and cannabis use) among the adult population. However, their relationship remains poorly understood. This study estimated the prevalence of economic instability indicators (i.e., employment instability, received food stamps, food insecurity, housing insecurity, and threatened utility shutoffs), exclusive cigarette use (TU), exclusive cannabis use (CU), and cigarette-cannabis co-use (TCU), and investigated the associations between economic instability indicators and these distinct substance use patterns among adults aged ≥18 years. Methods: This study used cross-sectional population-based data from the 2023 Behavioral Risk Factor Surveillance System (BRFSS) Marijuana Use and Social Determinants and Health Equity modules, administered in 10 US states. The 2023 BRFSS achieved a response rate of 44.7%. The sample included (unweighted n = 50,481; weighted N = 24,896,969) adults ≥18 years. The outcome variables were TU, CU, and TCU. Indicators of economic instability included employment instability, receipt of food stamps, food insecurity, housing insecurity, and threats of utility shutoffs. Weighted descriptive statistics and multivariable multinomial logistic regression analyses were conducted. Results: Overall, 48.6% of the participants were aged ≤50 years, 52.6% were women, and 64.4% self-identified as non-Hispanic White. The prevalences of TU, CU, and TCU were 8.1%, 11.1%, and 3.6%, respectively. Participants experiencing employment instability, housing insecurity, or threats of utility shutoffs were more likely to engage in CU compared to TU or TCU ( P ≤ 0.01). All economic instability indicators were significantly associated with TCU, with the strongest association observed for those who received food stamps (aRRR = 2.27, 95% CI: 1.75–2.94), which was also significantly associated with TU (aRRR = 1.55, 95% CI: 1.24–1.93). Threats of utility shutoffs (aRRR = 1.37, 95% CI: 1.10–1.70) and housing insecurity (aRRR = 1.27, 95% CI: 1.04–1.55) were significantly associated with CU. Conclusions: Economic instability indicators are significantly associated with cigarette and cannabis use, as well as their co-use patterns among US adults residing in the 10 participating states. Effective, targeted public health interventions to mitigate substance use risk in economically precarious populations are needed. Healthcare providers should routinely screen for economic instability to identify individuals at high risk for substance use and provide them with social services and behavioral health interventions.

HIV-related stigma diminishes efforts to contain the HIV epidemic, as it compromises testing and access to quality services. Various HIV prevention and control measures have been put in place, including antiretroviral medications. However, HIV-related stigma still compromises human rights for people with HIV. We used the social ecological model to examine HIV-related stigma in a cross-sectional and multinational sample of 1,598 people with HIV. We found moderate HIV-related stigma, mainly related to concerns about HIV status disclosure and negative attitudes of the public. Overall HIV-related stigma and its dimensions were associated with factors at the individual, community, and structural levels of the social ecological model, with patterns of association depending on a given dimension of HIV stigma and contextual factors. The findings suggest that efforts to address HIV-related stigma need a comprehensive approach that considers not only people with HIV but also their immediate and distal social ecological environment.

The purpose of this study was to understand the impact of the coronavirus disease 2019 (COVID-19) pandemic and mitigation efforts on health and social outcomes for people with HIV at the individual, social, and structural levels of the Social Ecological Model. The International Nursing Network for HIV collected data for a cross-sectional survey of people with HIV in Botswana, Canada, Colombia, Hong Kong, Kenya, Nigeria, South Africa, Thailand, and the United States from August 2021 through June 2023. Among 1,400 participants, 47.5% experienced decreased quality of life, 40.9% experienced increased anxiety, 33.0% had reduced connection with friends, and 38.8% had reduced access to resources. Participants’ reported impacts of COVID-19 varied by socioeconomic factors. Among these people with HIV, changes in quality of life, anxiety, social connectedness, and access to resources due to the COVID-19 pandemic were significantly associated with individual, social, and structural level factors using the Social Ecological Framework.

BACKGROUND The integration of artificial intelligence in health care presents a significant opportunity to revolutionize patient care. In the United States, an estimated 129 million people have at least 1 chronic illness, with 42% having 2 or more. Despite being largely preventable, the prevalence of chronic illness is expected to rise and impose significant economic burdens and financial toxicity on health care consumers. OBJECTIVE We leveraged an interdisciplinary team encompassing nursing, public health, and computer science to optimize health through prevention education for cardiovascular and metabolic comorbidities in persons living with HIV. In this tutorial, we describe the iterative, data-based development and evaluation of an intersectionality-informed large language model designed to support patient teaching in this population. METHODS First, we curated data by scraping publicly available, authoritative, evidence-based sources to capture a comprehensive dataset, supplemented by publicly available HIV forum content. Second, we benchmarked candidate large language models and generated a fine-tuning dataset using GPT-4 through multiturn question and answer conversations, using standardized metrics to assess baseline model performance. Third, we iteratively refined the selected model via low-rank adaptation and reinforcement learning, integrating quantitative metrics with qualitative expert evaluations. RESULTS Pre-existing large language models (LLMs) demonstrated poor n-gram agreement, dissonance from model answers (accuracy 4.16, readability 4.63, and professionalism 4.58), and difficult readability (Kincaid 8.54 and Jargon 4.44). After prompt adjustments and fine-tuning, preliminary results demonstrate the potential of a customized Llama-based LLM to provide personalized, culturally salient patient education. CONCLUSIONS We present a data-based, step-by-step tutorial for interdisciplinary development of CARDIO, a specialized LLM, for cardiovascular health education in HIV care. Through comprehensive data curation and scraping, systematic benchmarking, and a dual-stage fine-tuning pipeline, CARDIO’s performance improved markedly (accuracy 5.0, readability 4.98, professionalism 4.98, Kincaid 7.17, and Jargon 2.92). Although patient pilot testing remains forthcoming, our results demonstrate that targeted data curation, rigorous benchmarking, and iterative fine-tuning have provided a robust evaluation of the model’s potential. By building an LLM tailored to cardiovascular health promotion and patient education, this work lays the foundation for innovative artificial intelligence–driven strategies to manage comorbid conditions in people living with HIV.

The integration of Artificial Intelligence in healthcare presents as a significant opportunity to revolutionize patient care. In the United States, an estimated 129 million people have at least one chronic illness, with 42% having two or more. Despite being largely preventable, the prevalence of chronic illness is expected to rise and impose significant economic burdens and financial toxicity on healthcare consumers. We leveraged an interdisciplinary team encompassing nursing, public health, and computer science to optimize health through prevention education for cardiovascular and metabolic comorbidities in persons living with HIV. In this tutorial, we describe the iterative, data-based development and evaluation of an intersectionality-informed large language model designed to support patient teaching in this population. First, we curated data by scraping publicly available, authoritative, evidence-based sources to capture a comprehensive dataset, supplemented by publicly available HIV forum content. Second, we benchmarked candidate large language models and generated a fine-tuning dataset using GPT-4 through multi-turn question-answer conversations, employing standardized metrics to assess baseline model performance. Third, we iteratively refined the selected model via Low-Rank Adaptation and reinforcement learning, integrating quantitative metrics with qualitative expert evaluations. Pre-existing LLM models demonstrated poor n-gram agreement, dissonance from model answers (Accuracy 4.16, Readability 4.63, Professionalism 4.58), and difficult readability (Kincaid 8.54, Jargon 4.44). After prompt adjustments and fine-tuning, preliminary results demonstrate the potential of a customized LLaMA-based LLM to provide personalized, culturally salient patient education. We present a data-based, step-by-step tutorial for interdisciplinary development of CARDIO, a specialized LLM, for cardiovascular health education in HIV care. Through comprehensive data curation and scraping, systematic benchmarking, and a dual-stage fine-tuning pipeline, CARDIO's performance improved markedly (Accuracy 5.0, Readability 4.98, Professionalism 4.98, Kincaid 7.17, Jargon 2.92). Although patient pilot testing remains forthcoming, our results demonstrate that targeted data curation, rigorous benchmarking, and iterative fine-tuning have provided a robust evaluation of the model's potential. By building an LLM tailored to cardiovascular health promotion and patient education, this work lays the foundation for innovative AI-driven strategies to manage comorbid conditions in people living with HIV.

Introduction: Globalization has increased the importance of multicultural research to address health disparities and improve healthcare outcomes for underrepresented communities. The International Nursing Network for HIV Research (The Network) serves as a platform for researchers to collaborate on cross-cultural and cross-national HIV studies. This article discusses the Network’s approach to overcoming barriers in multicultural and multinational research in a qualitative context. Methods: The network created a protocol to guide decision-making throughout the translation process of qualitative data collected from participants in their native languages. The protocol includes aspects of why, when, what, who, how, where, and by what means the translation is completed. Results: The protocol has allowed researchers to enhance the validity, reliability, and cultural sensitivity of translation process, ensuring the clarity and impact of their research findings. Discussion: Rigorous translation practices promote cross-cultural understanding and respect for participants’ perspectives, fostering global collaborations and knowledge exchange.

As the COVID-19 pandemic spread across the world, immunocompromised individuals such as people with HIV (PWH) may have faced a disproportionate impact on their health and HIV outcomes, both from COVID-19 and from the strategies enacted to contain it. Based on the SPIRIT guidelines, we describe the protocol for an international multisite observational study being conducted by The International Nursing Network for HIV Research, with the Coordinating Center based at the University of California, San Francisco (UCSF) School of Nursing. Site Principal Investigators implement a standardized protocol to recruit PWH to complete the study online or in-person. Questions address demographics; HIV continuum of care indicators; mental and social health; COVID-19 and vaccination knowledge, attitudes, behaviors, and fears; and overall outcomes. Results of this study will contribute to knowledge that can inform responses to future public health crises to minimize their impacts on vulnerable populations such as PWH.