Patterns and predictors of viral suppression: A Big Data approach

抽象/总结

Abstract/Summary Sustained viral suppression, an indicator of long-term treatment success and mortality reduction, is one of four strategic areas of the “Ending the HIV Epidemic (EtHE): A Plan for America” federal campaign launched in 2019. Underrepresented populations, such as racial or ethnic minority populations, sexual and gender minority groups, and socioeconomically disadvantaged populations are usually disproportionately affected by HIV and subsequently experience a more striking virological failure. The COVID-19 pandemic is affecting People living with HIV (PLWH) in unique ways. It reveals the more apparent systemic inequities of HIV care due to the exacerbated preexisting structural disparities among underrepresented populations and consequently puts the already vulnerable populations at increased risk of worse HIV outcomes, including viral suppression. The parent grant (R01 AI164947) funded in 2021 aims to examine the longitudinal dynamic pattern of viral suppression, develop optimal predictive models of various viral suppression indicators, and translate the models to service-ready tools for clinical use using the South Carolina (SC) statewide HIV electronic health record (EHR) data. However, the SC statewide HIV database, a real-world data, cannot capture an adequate sample of underrepresented populations due to their historically limited access to specialty care and academic medical centers that serve as the primary sources for EHR data. The All of Us Research Program, a national historic effort supported by the NIH, aims to recruit a broad diverse group of the US population with more than 50% of the participants from racial and ethnic minority groups and more than 80% from populations historically underrepresented in biomedical research. The All of Us Research Program is harmonizing data from multiple sources on an ongoing basis and currently it has recruited ~4800 PLWH with a series of self- reported survey data (e.g., Lifestyle, Healthcare Access, COVID-19 Participant Experience) and relevant longitudinal EHR data (laboratory and medication). Given the limitations of the parent grant (R01 AI164947), this administrative supplement expands the parent grant to target a broadly defined underrepresented HIV population and develop a personalized viral suppression prediction model using machine learning techniques by incorporating multilevel factors (e.g., COVID-19 interruption, psychological wellbeing, healthcare utilization, and social environmental factors) using All of Us big data resources. The availability of comprehensive phenotypic data and the Researcher Workbench in All of Us platform fully assures the transparency and reproducibility of the proposed project and thus increases the generalizability of research findings. The proposed personalized viral suppression prediction can provide data driven evidence on tailored HIV treatment strategies to different underrepresented populations particularly in the face of the unexpected interruptions like the COVID-19 pandemic, and eventually serve towards the goal of ending the HIV epidemic in the US.