Ming-Yuan Chih

A rapidly expanding array of Artificial Intelligence (AI) tools, with continually evolving features and functionalities, offers unprecedented opportunities to streamline literature reviews, expediting the screening, extraction, and synthesis phases. We present preliminary findings of evaluating various AI tools’ strengths and limitations.

Postural Orthostatic Tachycardia Syndrome (POTS) patients experience debilitating symptoms and struggle to obtain a timely diagnosis. Mobile health (mHealth) and wearable technology may provide health information to empower patients when communicating with providers and in turn shorten the time to diagnosis. We conducted three online co-design focus groups with POTS patients to determine elements of a desirable mHealth intervention. Themes emerged that helped us develop a smartphone- and wearable-based system, using a risk communication model as a message design framework. A pilot randomized trial is ongoing to test the POTS App. Initial results showed the app reaches high usability.

Consumer involvement in the co-design of diabetes self-management smartphone apps is vital. This scoping review explored how consumers are involved in the co-design processes and methods and approaches guiding this research. Our review was guided by Arksey and O'Malley's five-stage framework, PRISMA-ScR guidelines, and Witteman and colleagues' 11-item user-centered design (UCD-11) framework. We searched literature across five databases and examined types of consumer involvement in co-design and frequency of methods and approaches (i.e., co-design approaches, behavioral theories, and other frameworks), synthesizing findings in SPSS and Excel. Of the 14,206 initial items, 283 articles were included. Most studies were conducted in Asia (33.2%) and focused on type 2 diabetes (43.1%). All articles addressed at least one UCD principle, and prototype evaluation (UCD-3) was the most frequent (82.3%); 85.2% addressed iterative responsiveness (factor 2). Most articles (66.8%) did not report a particular method or approach; 20.5% used design-related approaches, with user-centered design being the most common (7.4%). Few articles (3.9%) utilized social cognitive theory. Overall, co-design activities were isolated by phase. Consumers were primarily involved in evaluating prototypes and had limited engagement in the early stages. Iterative responsiveness factor activities were underreported or limited in scope. The use of approaches, theories, and frameworks was inconsistent. Consumer involvement in the co-design of diabetes self-management apps is often limited to later phases, with minimal engagement during the critical preprototype phase. To enhance the relevance, effectiveness, and adoption of diabetes self-management apps, app designers should improve the reporting of co-design activities and engage consumers across all co-design phases. This project did not include human or animal subjects. ^* Co-first author. ^** Co-senior author. Received: 11 December 2024 Accepted: 21 March 2025 Article published online: 23 July 2025 © 2025. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Background: Despite the increased accessibility and availability of technology in recent years, equality and access to health-related technology remain limited to some demographics. In particular, patients who are older or from rural communities represent a large segment of people who are currently underusing mobile health (mHealth) solutions. System usability continues to hinder mHealth adoption among users with nontraditional digital literacy. Objective: This study aims to investigate if state-of-the-art mobile app interfaces from open-source libraries provide sufficient usability for rural patients with cancer, with minimal design changes and forgoing the co-design process. Methods: We developed Assuage (Network Reconnaissance Lab) as a research platform for any mHealth study. We conducted a pilot study using Assuage to assess the usability of 4 mobile user interfaces (UIs) based on open-source libraries from Apple’s ResearchKit and CareKit. These UIs varied in complexity for reporting distress symptoms. Patients with cancer were recruited at the Markey Cancer Center, and all research procedures were conducted in person. Participants completed the distress assessment using a randomly selected UI in Assuage with little to no assistance. Data were collected on participant age, location, mobile app use, and familiarity with mHealth apps. Participants rated usability with the System Usability Scale (SUS), and usability issues were documented and compared. A one-way ANOVA was used to compare the effect of the UIs on the SUS scores. Results: We recruited 30 current or postsurgery patients with cancer for this pilot study. Most participants were aged >50 years (24/30, 80%), from rural areas (25/30, 83%), had up to a high school education (19/30, 63%), and were unfamiliar with mHealth apps (21/30, 70%). General mobile app use was split, with 43% (14/30) of the patients not regularly using mobile apps. The mean SUS score across the UIs was 75.8 (SD 22.2), with UI 3 and UI 4 achieving an SUS score ≥80, meeting the industry standard for good usability of 80. Critical usability issues were related to data input and navigation with touch devices, such as scale-format questions, vertical scrolling, and traversing multiple screens. Conclusions: The findings from this study show that most patients with cancer (20/30, 67%) who participated in this study rated the different interfaces of Assuage as above-average usability (SUS score >68). This suggests that Apple’s ResearchKit and CareKit libraries can provide usable UIs for older and rural users with minimal interface alterations. When resources are limited, the design stage can be simplified by omitting the co-design process while preserving suitable usability for users with nontraditional technical proficiency. Usability comparable to industry standards can be achieved by considering heuristics for interface and electronic survey design, specifically how to segment and navigate surveys, present important interface elements, and signal gestural interactions.

BACKGROUND Despite the increased accessibility and availability of technology in recent years, equality and access to health-related technology remain limited to some demographics. In particular, patients who are older or from rural communities represent a large segment of people who are currently underusing mobile health (mHealth) solutions. System usability continues to hinder mHealth adoption among users with nontraditional digital literacy. OBJECTIVE This study aims to investigate if state-of-the-art mobile app interfaces from open-source libraries provide sufficient usability for rural patients with cancer, with minimal design changes and forgoing the co-design process. METHODS We developed Assuage (Network Reconnaissance Lab) as a research platform for any mHealth study. We conducted a pilot study using Assuage to assess the usability of 4 mobile user interfaces (UIs) based on open-source libraries from Apple’s ResearchKit and CareKit. These UIs varied in complexity for reporting distress symptoms. Patients with cancer were recruited at the Markey Cancer Center, and all research procedures were conducted in person. Participants completed the distress assessment using a randomly selected UI in Assuage with little to no assistance. Data were collected on participant age, location, mobile app use, and familiarity with mHealth apps. Participants rated usability with the System Usability Scale (SUS), and usability issues were documented and compared. A one-way ANOVA was used to compare the effect of the UIs on the SUS scores. RESULTS We recruited 30 current or postsurgery patients with cancer for this pilot study. Most participants were aged >50 years (24/30, 80%), from rural areas (25/30, 83%), had up to a high school education (19/30, 63%), and were unfamiliar with mHealth apps (21/30, 70%). General mobile app use was split, with 43% (14/30) of the patients not regularly using mobile apps. The mean SUS score across the UIs was 75.8 (SD 22.2), with UI 3 and UI 4 achieving an SUS score ≥80, meeting the industry standard for good usability of 80. Critical usability issues were related to data input and navigation with touch devices, such as scale-format questions, vertical scrolling, and traversing multiple screens. CONCLUSIONS The findings from this study show that most patients with cancer (20/30, 67%) who participated in this study rated the different interfaces of Assuage as above-average usability (SUS score >68). This suggests that Apple’s ResearchKit and CareKit libraries can provide usable UIs for older and rural users with minimal interface alterations. When resources are limited, the design stage can be simplified by omitting the co-design process while preserving suitable usability for users with nontraditional technical proficiency. Usability comparable to industry standards can be achieved by considering heuristics for interface and electronic survey design, specifically how to segment and navigate surveys, present important interface elements, and signal gestural interactions.

Background: The extent of human interaction needed to achieve effective and cost-effective use of mobile health (mHealth) apps for individuals with mild to moderate alcohol use disorder (AUD) remains largely unexamined. This study seeks to understand how varying levels of human interaction affect the ways in which an mHealth intervention for the prevention and treatment of AUDs works or does not work, for whom, and under what circumstances. Objective: The primary aim is to detect the effectiveness of an mHealth intervention by assessing differences in self-reported risky drinking patterns and quality of life between participants in three study groups (self-monitored, peer-supported, and clinically integrated). The cost-effectiveness of each approach will also be assessed. Methods: This hybrid type 1 study is an unblinded patient-level randomized clinical trial testing the effects of using an evidence-based mHealth system on participants’ drinking patterns and quality of life. There are two groups of participants for this study: individuals receiving the intervention and health care professionals practicing in the broader health care environment. The intervention is a smartphone app that encourages users to reduce their alcohol consumption within the context of integrative medicine using techniques to build healthy habits. The primary outcomes for quantitative analysis will be participant data on their risky drinking days and quality of life as well as app use from weekly and quarterly surveys. Cost measures include intervention and implementation costs. The cost per participant will be determined for each study arm, with intervention and implementation costs separated within each group. There will also be a qualitative assessment of health care professionals’ engagement with the app as well as their thoughts on participant experience with the app. Results: This protocol was approved by the Health Sciences Minimal Risk Institutional Review Board on November 18, 2019, with subsequent annual reviews. Recruitment began on March 6, 2020, but was suspended on March 13, 2020, due to the COVID-19 pandemic restrictions. Limited recruitment resumed on July 6, 2020. Trial status as of November 17, 2021, is as follows: 357 participants were enrolled in the study for a planned enrollment of 546 participants. Conclusions: The new knowledge gained from this study could have wide and lasting benefits related to the integration of mHealth systems for individuals with mild to moderate AUDs. The results of this study will guide policy makers and providers toward cost-effective ways to incorporate technology in health care and community settings.

Background: Smartphone-based apps are increasingly used to prevent relapse among those with substance use disorders (SUDs). These systems collect a wealth of data from participants, including the content of messages exchanged in peer-to-peer support forums. How individuals self-disclose and exchange social support in these forums may provide insight into their recovery course, but a manual review of a large corpus of text by human coders is inefficient. Objective: The study sought to evaluate the feasibility of applying supervised machine learning (ML) to perform large-scale content analysis of an online peer-to-peer discussion forum. Machine-coded data were also used to understand how communication styles relate to writers’ substance use and well-being outcomes. Methods: Data were collected from a smartphone app that connects patients with SUDs to online peer support via a discussion forum. Overall, 268 adult patients with SUD diagnoses were recruited from 3 federally qualified health centers in the United States beginning in 2014. Two waves of survey data were collected to measure demographic characteristics and study outcomes: at baseline (before accessing the app) and after 6 months of using the app. Messages were downloaded from the peer-to-peer forum and subjected to manual content analysis. These data were used to train supervised ML algorithms using features extracted from the Linguistic Inquiry and Word Count (LIWC) system to automatically identify the types of expression relevant to peer-to-peer support. Regression analyses examined how each expression type was associated with recovery outcomes. Results: Our manual content analysis identified 7 expression types relevant to the recovery process (emotional support, informational support, negative affect, change talk, insightful disclosure, gratitude, and universality disclosure). Over 6 months of app use, 86.2% (231/268) of participants posted on the app’s support forum. Of these participants, 93.5% (216/231) posted at least 1 message in the content categories of interest, generating 10,503 messages. Supervised ML algorithms were trained on the hand-coded data, achieving F1-scores ranging from 0.57 to 0.85. Regression analyses revealed that a greater proportion of the messages giving emotional support to peers was related to reduced substance use. For self-disclosure, a greater proportion of the messages expressing universality was related to improved quality of life, whereas a greater proportion of the negative affect expressions was negatively related to quality of life and mood. Conclusions: This study highlights a method of natural language processing with potential to provide real-time insights into peer-to-peer communication dynamics. First, we found that our ML approach allowed for large-scale content coding while retaining moderate-to-high levels of accuracy. Second, individuals’ expression styles were associated with recovery outcomes. The expression types of emotional support, universality disclosure, and negative affect were significantly related to recovery outcomes, and attending to these dynamics may be important for appropriate intervention.