Sue & Bill Gross School of Nursing

Generative Artificial Intelligence is set to revolutionize healthcare delivery by transforming traditional patient care into a more personalized, efficient, and proactive process. Chatbots, serving as interactive conversational models, will probably drive this patient-centered transformation in healthcare. Through the provision of various services, including diagnosis, personalized lifestyle recommendations, dynamic scheduling of follow-ups, and mental health support, the objective is to substantially augment patient health outcomes, all the while mitigating the workload burden on healthcare providers. The life-critical nature of healthcare applications necessitates establishing a unified and comprehensive set of evaluation metrics for conversational models. Existing evaluation metrics proposed for various generic large language models (LLMs) demonstrate a lack of comprehension regarding medical and health concepts and their significance in promoting patients’ well-being. Moreover, these metrics neglect pivotal user-centered aspects, including trust-building, ethics, personalization, empathy, user comprehension, and emotional support. The purpose of this paper is to explore state-of-the-art LLM-based evaluation metrics that are specifically applicable to the assessment of interactive conversational models in healthcare. Subsequently, we present a comprehensive set of evaluation metrics designed to thoroughly assess the performance of healthcare chatbots from an end-user perspective. These metrics encompass an evaluation of language processing abilities, impact on real-world clinical tasks, and effectiveness in user-interactive conversations. Finally, we engage in a discussion concerning the challenges associated with defining and implementing these metrics, with particular emphasis on confounding factors such as the target audience, evaluation methods, and prompt techniques involved in the evaluation process.

Media exposure to graphic images of violence has proliferated in contemporary society, particularly with the advent of social media. Extensive exposure to media coverage immediately after the 9/11 attacks and the Boston Marathon bombings (BMB) was associated with more early traumatic stress symptoms; in fact, several hours of BMB-related daily media exposure was a stronger correlate of distress than being directly exposed to the bombings themselves. Researchers have replicated these findings across different traumatic events, extending this work to document that exposure to graphic images is independently and significantly associated with stress symptoms and poorer functioning. The media exposure-distress association also appears to be cyclical over time, with increased exposure predicting greater distress and greater distress predicting more media exposure following subsequent tragedies. The war in Israel and Gaza, which began on October 7, 2023, provides a current, real-time context to further explore these issues as journalists often share graphic images of death and destruction, making media-based graphic images once again ubiquitous and potentially challenging public well-being. For individuals sharing an identity with the victims or otherwise feeling emotionally connected to the Middle East, it may be difficult to avoid viewing these images. Through a review of research on the association between exposure to graphic images and public health, we discuss differing views on the societal implications of viewing such images and advocate for media literacy campaigns to educate the public to identify mis/disinformation and understand the risks of viewing and sharing graphic images with others.

Background

Methods

Results

Conclusions

The profound impact of food on health necessitates advanced nutrition-oriented food recommendation services. Conventional methods often lack the crucial elements of personalization, explainability, and interactivity. While Large Language Models (LLMs) bring interpretability and explainability, their standalone use falls short of achieving true personalization. In this paper, we introduce ChatDiet, a novel LLM-powered framework designed specifically for personalized nutrition-oriented food recommendation chatbots. ChatDiet integrates personal and population models, complemented by an orchestrator, to seamlessly retrieve and process pertinent information. The personal model leverages causal discovery and inference techniques to assess personalized nutritional effects for a specific user, whereas the population model provides generalized information on food nutritional content. The orchestrator retrieves, synergizes and delivers the output of both models to the LLM, providing tailored food recommendations designed to support targeted health outcomes. The result is a dynamic delivery of personalized and explainable food recommendations, tailored to individual user preferences. Our evaluation of ChatDiet includes a compelling case study, where we establish a causal personal model to estimate individual nutrition effects. Our assessments, including a food recommendation test showcasing a 92% effectiveness rate, coupled with illustrative dialogue examples, underscore ChatDiet's strengths in explainability, personalization, and interactivity.

The aim of this project was to increase willingness to receive the influenza vaccine to the optimal rate of ≥ 70%. Low acuity adult patients who visited an Emergency Department (ED) were assessed regarding their willingness to receive the influenza vaccine before and after an educational intervention that included a provider recommendation and an educational handout. A total of seventy-six patients (n = 76) were assessed. Patients willingness to receive the influenza vaccine rose from 29% pre-intervention to 72% post-intervention without disrupting the clinical flow in a busy ED. Similar vaccine educational strategies can be applied to influenza and other vaccines in EDs  to increase vaccination willingness in patients, including those who use the ED as a primary point of contact for healthcare, decreasing the burden of influenza illness in the community.

This dataset was collected from university students before, during, and after the COVID-19 lockdown in Southern California. Data collection happened continuously for the average of 7.8 months (SD=3.8, MIN=1.0, MAX=13.4) from a population of 21 students of which 12 have also completed an exit survey, and 7 started before the California COVID-19 lockdown order. This multimodal dataset included different means of data collection such as Samsung Galaxy Watch, Oura Ring, a Life-logger app named Personicle, a questionnaire mobile app named Personicle Questions, and periodical and personalised surveys. The dataset contains raw data from Photoplethysmogram (PPG), Inertial measurement unit (IMU), and pressure sensors in addition to processed data on heart rate, heart rate variability, sleep (bedtime, sleep stages, quality), and physical activity (step, active calories, type of activity). Ecological momentary assessments were collected from participants on daily and weekly bases containing their Positive and Negative Affect Schedule (PANAS) questionnaire and their emotional responses to COVID-19 and their health. Subjective data was also collected through monthly surveys containing standard mood and mental health surveys such as Beck Depression Inventory II (BDI-II), Brief Symptom Inventory (BSI), GAD-7, Inclusion of Other in the Self Scale (IOS-Partner), Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), Feasibility of Intervention Measure (FIM), Experiences in Close Relationships Scale Short Form (ECR-S), UCLA Three-Item Loneliness Scale (ULS), Multidimensional Scale of Perceived Social Support (MSPSS), Investment Model Scale (IMS), Conflict Management Scale (CMS), etc in addition to their response to important events and COVID-19. This dataset can be used to study emotions, mood, physical activity, and lifestyle of young adults through longitudinal subjective and objective measures. This dataset also contains valuable data regarding adjustment of lifestyle and emotions during the events of 2020 and 2021 including COVID-19 discovery and lockdown, Black Life Matter movement, 2020 US presidential elections, etc. On average, participants engaged in the EMA collection study at a rate of 86% (SD=10, MIN=65, MAX=99). Smartwatch usage saw an average participation rate of 51% (SD=20, MIN=16, MAX=88), while engagement with the Oura ring averaged at 85% (SD=12, MIN=60, MAX=99).

BACKGROUND: The existing literature has limited detail on theory-driven interventions, particularly in pain studies. We adapted Banduras self-efficacy framework toward a theory-driven, non-pharmacological intervention using auricular point acupressure (APA) and evaluated participants perceptions of this intervention on their pain self-management. APA is a non-invasive modality based on auricular acupuncture principles. METHODS: We mapped our study intervention components according to Banduras key sources of self-efficacy (performance accomplishments, vicarious experience, verbal persuasion, and emotional arousal) to facilitate the self-management of pain. Through a qualitative study design, we conducted virtual interviews at one and three months after a 4-week APA intervention among 23 participants using purposive sampling to describe their experiences in managing their pain based on our theory-driven APA intervention. RESULTS: Using thematic analyses, we found four themes: the enhanced self-management of pain, improved pain outcomes, the feasibility of technology, and the sustainability of APA. CONCLUSIONS: Describing how interventions are mapped according to the elements of theoretical frameworks can help to guide intervention development, advance science and knowledge development, and promote the implementation of interventions. As such, using Banduras self-efficacy theory as a foundation for the APA intervention, APA was found to be feasible and sustainable, improving self-management, pain intensity, and pain-related outcomes. Participants provided recommendations for the further improvement of this theory-driven intervention.

AIM: This paper describes the trajectory during 1 year of four patient-reported outcomes (PROs), namely, sleep, depressive symptoms, health-related quality of life (HrQoL), and well-being, in patients with heart failure (HF), their relationship and the patient characteristics associated with changes in these PROs. METHODS AND RESULTS: Data analyses of PROs from 603 patients (mean age 67 years; 29% female, 60% NYHA II) enrolled in the HF-Wii study. On short term, between baseline and 3 months, 16% of the patients experienced continuing poor sleep, 11% had sustained depressive symptoms, 13% had consistent poor HrQoL, and 13% consistent poor well-being. Across the entire 1-year period only 21% of the patients had good PRO scores at all timepoints (baseline, 3, 6, and 12 months). All others had at least one low score in any of the PROs at some timepoint during the study. Over the 12 months, 17% had consistently poor sleep, 17% had sustained symptoms of depression, 15% consistently rated a poor HrQoL, and 13% poor well-being. Different patient characteristics per PRO were associated with a poor outcomes across the 12 months. Age, education, New York Heart Association, and length of disease were related to two PRO domains and submaximal exercise capacity (6 min test), co-morbidity, and poor physical activity to one. CONCLUSION: In total, 79% of the patients with HF encountered problems related to sleep, depressive symptoms, HrQoL, and well-being at least once during a 1-year period. This underscores the need for continuous monitoring and follow-up of patients with HF and the need for dynamic adjustments in treatment and care regularly throughout the HF trajectory.

Background

Purpose

Methods

Results

Conclusions

AIMS: To describe sources of health information and health-seeking behaviours of adults (aged ≥18) living in medically underserved communities in the Philippines. DESIGN: This is a secondary, quantitative analysis from a cross-sectional parent study. Participants completed a 10-item, self-report survey on their sources of health information, healthcare providers sought for health and wellness and health-seeking behaviours when ill. Responses were evaluated across two age groups (<60 vs. ≥60 years) and genders using generalized linear mixed models. RESULTS: Surveys were completed by 1202 participants in rural settings (64.6% female, mean age 49.5 ± 17.6). Friends and/or family were their key source of health information (59.6%), followed by traditional media (37%) and healthcare professionals (12.2%). For health promotion, participants went to healthcare professionals (60.9%), informal healthcare providers (17.2%) or others (7.2%). When ill, they visited a healthcare professional 69.1% of the time, self-medicated (43.9%), prayed (39.5%) or sought treatment from a rural health clinic (31.5%). We also found differences in health-seeking behaviours based on age and gender. CONCLUSIONS: Our findings highlight the need to organize programs that explicitly deliver accurate health information and adequate care for wellness and illness. Study findings emphasize the importance of integrating family, friends, media and healthcare professionals, including public health nurses, to deliver evidence-based health information, health promotion and sufficient treatment to medically underserved Filipinos. IMPLICATIONS: New knowledge provides valuable information to healthcare providers, including public health nurses, in addressing health disparities among medically underserved Filipinos. IMPACT: This study addresses the current knowledge gap in a medically vulnerable population. Healthcare professionals are not the primary sources of health information. Approximately one-third of participants do not seek them for health promotion or treatment even when ill, exacerbating health inequities. More work is necessary to support initiatives in low- and middle-income countries such as the Philippines to reduce health disparities. REPORTING METHOD: We adhered to the reporting guidelines of STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) for cross-sectional studies. PATIENT OR PUBLIC CONTRIBUTION: There was no patient or public contribution as our study design and methodology do not make this necessary.