Background The highest rates of new HIV infections are observed in African Americans and Hispanics/Latinos (ethnic minority) adolescents and young adults (youth). HIV-infected ethnic minority youth are less likely to initiate and maintain adherence to antiretroviral treatment (ART) and medical care, as compared with their adult counterparts. Objective The objective of this research protocol was to describe our proposed methods for testing a peer-led mobile health cognitive behavioral intervention, delivered via remote videoconferencing and smartphones with HIV-infected ethnic minority youth, Adherence Connection for Counseling, Education, and Support (ACCESS). Our secondary aim was to obtain initial estimates of the biobehavioral impact of ACCESS on HIV virologic outcomes and self-reported ART adherence, beliefs and knowledge about ART treatment, adherence self-efficacy, and health care utilization (retention in care). Methods An exploratory, sequential mixed-methods study design will be used with conceptual determinants of adherence behavior informed by the information-motivation-behavioral skills model. HIV-infected ethnic minority youth aged 16 to 29 years with a detectable HIV serum viral load of more than 200 copies/ml (N=25) will be recruited. Qualitative pretesting will be conducted, including semistructured, in-depth, individual interviews with a convenience sample meeting the study inclusion criteria. Preliminary analysis of qualitative data will be used to inform and tailor the ACCESS intervention. Testing and implementation will include a one-group pre-posttest pilot, delivered by a trained successful peer health coach who lives with HIV and is well-engaged in HIV care and taking ART. A total of 5 peer-led remote videoconferencing sessions will be delivered using study-funded smartphones and targeting adherence information (HIV knowledge), motivation (beliefs and perceptions), and behavioral skills (self-efficacy). Participant satisfaction will be assessed with poststudy focus groups and quantitative survey methodology. Bivariate analyses will be computed to compare pre- and postintervention changes in HIV biomarkers, self-reported ART adherence, beliefs and knowledge about ART, adherence self-efficacy, and retention in care. Results As of December 2018, we are in the data analysis phase of this pilot and anticipate completion with dissemination of final study findings by spring/summer 2019. The major outcomes will include intervention feasibility, acceptability, and preliminary evidence of impact on serum HIV RNA quantitative viral load (primary adherence outcome variable). Self-reported ART adherence and retention in care will be assessed as secondary outcomes. Findings from the qualitative pretesting will contribute to an improved understanding of adherence behavior. Conclusions Should the ACCESS intervention prove feasible and acceptable, th...

Background Accurately identifying patients with hypoglycemia is key to preventing adverse events and mortality. Natural language processing (NLP), a form of artificial intelligence, uses computational algorithms to extract information from text data. NLP is a scalable, efficient, and quick method to extract hypoglycemia-related information when using electronic health record data sources from a large population. Objective The objective of this systematic review was to synthesize the literature on the application of NLP to extract hypoglycemia from electronic health record clinical notes. Methods Literature searches were conducted electronically in PubMed, Web of Science Core Collection, CINAHL (EBSCO), PsycINFO (Ovid), IEEE Xplore, Google Scholar, and ACL Anthology. Keywords included hypoglycemia, low blood glucose, NLP, and machine learning. Inclusion criteria included studies that applied NLP to identify hypoglycemia, reported the outcomes related to hypoglycemia, and were published in English as full papers. Results This review (n=8 studies) revealed heterogeneity of the reported results related to hypoglycemia. Of the 8 included studies, 4 (50%) reported that the prevalence rate of any level of hypoglycemia was 3.4% to 46.2%. The use of NLP to analyze clinical notes improved the capture of undocumented or missed hypoglycemic events using International Classification of Diseases, Ninth Revision (ICD-9), and International Classification of Diseases, Tenth Revision (ICD-10), and laboratory testing. The combination of NLP and ICD-9 or ICD-10 codes significantly increased the identification of hypoglycemic events compared with individual methods; for example, the prevalence rates of hypoglycemia were 12.4% for International Classification of Diseases codes, 25.1% for an NLP algorithm, and 32.2% for combined algorithms. All the reviewed studies applied rule-based NLP algorithms to identify hypoglycemia. Conclusions The findings provided evidence that the application of NLP to analyze clinical notes improved the capture of hypoglycemic events, particularly when combined with the ICD-9 or ICD-10 codes and laboratory testing.

Background: Improving the recruitment and retention of underrepresented groups in all research areas is essential for health equity. However, achieving and retaining diverse samples is challenging. Barriers to recruitment and retention of diverse participants include socioeconomic and cultural factors and practical challenges (e.g., time and travel commitments).Objectives: The purpose of this article is to describe the successful recruitment and retention strategies used by two related studies within a P20 center funded by the National Institute of Nursing Research focused on precision health research in diverse populations with multiple chronic conditions, including metabolic syndrome.Methods: To address the complexity, biodiversity, and effect of metabolic syndrome and multiple chronic conditions, we developed culturally appropriate, multipronged recruitment and retention strategies for a pilot intervention study and a longitudinal observational pilot study within our P20 center. The following are the underlying principles that guided the recruitment and retention strategies: (a) flexibility, (b) active listening and bidirectional conversations, and (c) innovative problem solving.Results: The intervention study (Pilot 1) enrolled 49 participants. The longitudinal observational study (Pilot 2) enrolled 45 participants. Women and racial/ethnic minorities were significantly represented in both. In Pilot 1, most of the participants completed the intervention and all phases of data collection. In Pilot 2, most participants completed all phases of data collection and chose to provide biorepository specimens.Discussion: We developed a recruitment and retention plan building on standard strategies for a general medical population. Our real-world experiences informed the adaption of these strategies to facilitate the participation of individuals who often do not participate in research-specifically, women and racial/ethnic populations. Our experience across two pilot studies suggests that recruiting diverse populations should build flexibility in the research plan at the outset.