Understanding Patients' Negative Experiences with Telehealth: A Content Analysis of Survey Data.

Understanding differences in how demographic groups experience telehealth may be relevant in addressing potential disparities in telehealth usage. We seek to identify and examine themes most pertinent to patients' negative telehealth experiences by age and race in order to inform interventions to improve patients' future telehealth experiences. We performed a content analysis of Press Ganey patient experience surveys from adult patients at 17 primary care sites of a large, public healthcare system with visits from April 30, 2020 to August 27, 2021. We used sentiment analysis to identify negative comments. We coded for content themes and analyzed their frequency, stratifying by age and race. We analyzed 745 negative comments. Most frequent themes differed by demographic categories, but overall, the most commonly applied codes were "Contacting the Clinic" (n = 97), "Connectivity" (n = 84), and "Webside Manner" (n = 79). The top three codes accounted for >40% of the negative codes in each race category and >35% of the negative codes in each age category. While there were common negative experiences among groups, patients of different demographics highlighted different aspects of their telehealth experiences for potential improvement.

Performance of Artificial Intelligence Chatbots on Glaucoma Questions Adapted From Patient Brochures.

Introduction With the potential for artificial intelligence (AI) chatbots to serve as the primary source of glaucoma information to patients, it is essential to characterize the information that chatbots provide such that providers can tailor discussions, anticipate patient concerns, and identify misleading information. Therefore, the purpose of this study was to evaluate glaucoma information from AI chatbots, including ChatGPT-4, Bard, and Bing, by analyzing response accuracy, comprehensiveness, readability, word count, and character count in comparison to each other and glaucoma-related American Academy of Ophthalmology (AAO) patient materials. Methods Section headers from AAO glaucoma-related patient education brochures were adapted into question form and asked five times to each AI chatbot (ChatGPT-4, Bard, and Bing). Two sets of responses from each chatbot were used to evaluate the accuracy of AI chatbot responses and AAO brochure information, and the comprehensiveness of AI chatbot responses compared to the AAO brochure information, scored 1-5 by three independent glaucoma-trained ophthalmologists. Readability (assessed with Flesch-Kincaid Grade Level (FKGL), corresponding to the United States school grade levels), word count, and character count were determined for all chatbot responses and AAO brochure sections. Results Accuracy scores for AAO, ChatGPT, Bing, and Bard were 4.84, 4.26, 4.53, and 3.53, respectively. On direct comparison, AAO was more accurate than ChatGPT (p=0.002), and Bard was the least accurate (Bard versus AAO, p<0.001; Bard versus ChatGPT, p<0.002; Bard versus Bing, p=0.001). ChatGPT had the most comprehensive responses (ChatGPT versus Bing, p<0.001; ChatGPT versus Bard p=0.008), with comprehensiveness scores for ChatGPT, Bing, and Bard at 3.32, 2.16, and 2.79, respectively. AAO information and Bard responses were at the most accessible readability levels (AAO versus ChatGPT, AAO versus Bing, Bard versus ChatGPT, Bard versus Bing, all p<0.0001), with readability levels for AAO, ChatGPT, Bing, and Bard at 8.11, 13.01, 11.73, and 7.90, respectively. Bing responses had the lowest word and character count. Conclusion AI chatbot responses varied in accuracy, comprehensiveness, and readability. With accuracy scores and comprehensiveness below that of AAO brochures and elevated readability levels, AI chatbots require improvements to be a more useful supplementary source of glaucoma information for patients. Physicians must be aware of these limitations such that patients are asked about existing knowledge and questions and are then provided with clarifying and comprehensive information.

Patient Experiences With Telehealth During Versus After a System-Wide Telehealth Mandate During the COVID-19 Pandemic.

This study examines whether patients' telehealth experiences differed during a health system mandate for telehealth encounters due to the COVID-19 pandemic versus after the mandate was relaxed. Patient experience surveys from telehealth visits across 17 adult (age 18+) primary care sites at a large, urban public health system were analyzed during two periods: when a mandate was active (March 1, 2020-June 30, 2020) and when the mandate was relaxed and any appointment modality was available (July 1, 2020-November 30, 2021). Primary outcomes were odds ratios (ORs) comparing top-box percentages of survey responses at multiple levels: individual questions, four domains, and all questions together as a composite. Key findings: Patients had higher odds of selecting top-box answers in the elective telehealth period for the Care Provider (1.09 [95% confidence interval 1.03, 1.16]) and General Assessment (1.13 [1.02, 1.24]) domains and the survey composite (1.08 [1.04, 1.13]), but there was no difference for individual questions.Women reported more positive experiences during the elective telehealth period in the Access (1.22 [1.01, 1.47]), Care Provider (1.32 [1.17, 1.50]), and Telemedicine Technology (1.24 [1.04, 1.50]) domains.Our findings suggest that patients had better telehealth experiences when mandates were relaxed.

Colorectal Cancer Screening Decisions From Student-Led Telemedicine Initiative During COVID-19.

Background and Objectives: With the onset of the COVID-19 pandemic, there has been a significant decrease in colorectal cancer (CRC) screening leading to delayed diagnoses and increased cancer deaths. To mitigate these burgeoning gaps in care, we developed a medical student-led service learning project aimed at improving rates of colorectal cancer screening at the Farrell Health Center (FHC), a primary care practice within the Ambulatory Care Network (ACN) at New York-Presbyterian Hospital. Methods: A cohort of 973 FHC patients aged 50-75 years were identified as possibly overdue for screening. Student volunteers reviewed patient charts to confirm screening eligibility and then contacted patients to offer colonoscopy or stool DNA test. Following the patient outreach intervention, medical student volunteers completed a questionnaire to assess the educational value for the service-learning experience. Results: Fifty-three percent of identified patients were due for CRC screening; 67.8% of eligible patients were reached by volunteers. Among the patients reached, 47.0% were referred for CRC screening. No statistical significance was observed between likelihood of CRC screening acceptance and patient age or sex; 87% of medical student volunteers felt that the service-learning project was a valuable educational experience. Conclusion: The student-led patient telehealth outreach program is an effective model for identifying and referring patients overdue for CRC screening and an enriching educational experience for preclinical medical students. The structure provides a valuable framework to address gaps in health care maintenance.

Efficient Sampling of Protein Loop Regions Using Conformational Hashing Complemented with Random Coordinate Descent.

De novo construction of loop regions is an important problem in computational structural biology. Compared to regions with well-defined secondary structure, loops tend to exhibit significant conformational heterogeneity. As a result, their structures are often ambiguous when determined using experimental data obtained by crystallography, cryo-EM, or NMR. Although structurally diverse models could provide a more relevant representation of proteins in their native states, obtaining large numbers of biophysically realistic and physiologically relevant loop conformations is a resource-consuming task. To address this need, we developed a novel loop construction algorithm, Hash/RCD, that combines knowledge-based conformational hashing with random coordinate descent (RCD). This hybrid approach achieved a closure rate of 100% on a benchmark set of 195 loops in 29 proteins that range from 3 to 31 residues. More importantly, the use of templates allows Hash/RCD to maintain the accuracy of state-of-the-art coordinate descent methods while reducing sampling time from over 400 to 141 ms. These results highlight how the integration of coordinate descent with knowledge-based sampling overcomes barriers inherent to either approach in isolation. This method may facilitate the identification of native-like loop conformations using experimental data or full-atom scoring functions by allowing rapid sampling of large numbers of loops. In this manuscript, we investigate and discuss the advantages, bottlenecks, and limitations of combining conformational hashing with RCD. By providing a detailed technical description of the Hash/RCD algorithm, we hope to facilitate its implementation by other researchers.