JMIR Human Factors最新文献

Background: The use of artificial intelligence (AI) methods in palliative care research is increasing. Most AI palliative care research involves the use of routinely collected data from electronic health records; however, there are few data on the views of palliative care health care professionals on the role of AI in practice. Determining the opinions of palliative care health care professionals on the potential uses of AI in palliative care will be useful for policymakers and practitioners to determine and inform the meaningful use of AI in palliative care practice.

Objective: This study aimed to explore the views of palliative care health care professionals on the use of AI for the analysis of patient data in palliative care.

Methods: This was a phenomenological study using qualitative semistructured interviews with palliative care health care professionals with a minimum of 1 year of clinical experience in a hospice in the North West of England. Data were analyzed using inductive thematic analysis.

Results: We interviewed 6 palliative care professionals, including physicians, nurses, and occupational therapists. AI was viewed positively, although most participants had not used it in practice. None of the participants had received training in AI and stated that education in AI would be beneficial. Participants described the potential benefits of AI in palliative care, including the identification of people requiring palliative care interventions and the evaluation of patient experiences. Participants highlighted security and ethical concerns regarding AI related to data governance, efficacy, patient confidentiality, and consent issues.

Conclusions: This study highlights the importance of staff perceptions of AI in palliative care. Our findings support the role of AI in enhancing care, addressing educational needs, and tackling trust, ethics, and governance issues. This study lays the groundwork for guidelines on AI implementation, urging further research on the methodological, ethical, and practical aspects of AI in palliative care.

Background: The emergence of Artificial Intelligence (AI) is driving digital transformation and reshaping medical education in China. Numerous medical schools and institutions are actively implementing AI tools for case-based learning, literature analysis, and lecture support. This expanding application is accelerating the adoption of localized AI platforms, which are poised to become integral components in the coming years.

Objective: The primary aim of this study was to investigate the current use of AI tools among medical students, including usage frequency, commonly used platforms, and purposes of use. The second aim was to explore students' needs and expectations toward AI-powered medical education platforms by collecting and assessing student feedback and to identify practical requirements across disciplines and academic stages to inform more effective platform design.

Methods: Based on the Task-Technology Fit (TTF) model and five hypotheses, an anonymous online questionnaire was conducted to assess AI usage in learning, gather student feedback on AI-powered medical education platforms, and evaluate expected functionalities. The survey was conducted from March 1 to May 31, 2025, using a convenience sampling method to recruit medical students from various disciplines across Shanghai, China. The sample size was determined to be 422, accounting for a 10% rate of invalid responses. The questionnaire was developed and distributed online via Wenjuanxing and promoted through WeChat groups and in-person interviews. Data analysis was conducted employing IBM SPSS Statistics (v 27.0).

Results: A total of 428 valid questionnaires were collected. The average frequency of AI-assisted learning among medical students was 5.06±2.05 times per week. Over 90% of students (388/428) used more than two AI tools in their daily tasks. Students from different disciplines, educational stages, and academic systems demonstrated different usage patterns and expectations for AI-powered medical education platforms.

Conclusions: AI technology is widely accepted by medical students and is extensively applied across various aspects of medical education. Significant differences are observed in usage patterns across disciplines, educational stages, and academic systems. Understanding the actual needs of students is crucial for the construction of AI-powered medical education platforms.

Clinicaltrial:

Background: In blended care, digital mental health interventions (DMHIs) integrate with face-to-face psychotherapy provided in person or via telehealth. To incorporate DMHIs into routine care for depression and anxiety, it is important to understand the needs and expectations of mental health professionals for blended DMHIs.

Objective: The study objective was to partner with Australian mental health professionals in the design of a transdiagnostic, cognitive behavioral therapy-based blended model of care for adults experiencing depression and anxiety.

Methods: Participants were Australian health professionals who treat adults with depression and anxiety. The participatory design process included a web-based survey (N=258), one-on-one interviews (N=14), and a 2-part focus group (N=6). Quantitative and qualitative data were collected through the web-based survey. In-depth qualitative feedback from interviews and the 2-part focus group was subjected to reflexive thematic analysis.

Results: Mental health professionals found blended care with face-to-face therapy more acceptable than telehealth and blended care with telehealth, with standalone DMHIs being the least preferred option. The most common ways in which mental health professionals thought a DMHI could integrate with face-to-face psychotherapy included homework completion (129/178, 72.5%), skills practice to support in-session therapy (128/178, 71.9%), and psychoeducation (127/178, 71.3%). Mental health professionals expect the blended DMHI to be easy to use, flexible, protective of client data, and to include evidence-based content from several therapeutic modalities (eg, cognitive behavioral therapy and mindfulness). Other preferences included mental health professionals being able to prescribe specific program modules to their clients, track the treatment progress of clients, and receive alerts if their clients' symptoms worsened. In terms of implementation, mental health professionals were concerned about the time and effort needed to use blended care. They suggested that ongoing training and support would help mental health professionals implement blended care with their clients. Monitoring client risk and progress via a web-based dashboard and downloadable summaries was also important.

Conclusions: Designing DMHIs that support psychotherapy for adults with depression and anxiety has the potential to increase access to evidence-based treatment. Involving mental health professionals in DMHI design is expected to increase their acceptance of DMHIs and facilitate the integration of these digital products into routine care.

Background: Adolescents need comprehensive education and resources to promote their sexual and reproductive health (SRH) and make informed decisions about their SRH. However, many adolescents fail to secure an opportunity to learn accurate and reliable SRH information, as they face sociocultural barriers, fear of prejudice, and societal stigma. Information available on digital platforms may not always be evidence-based and can further spread misinformation. Digital knowledge translation interventions that provide evidence-based SRH information in North America are limited.

Objective: This study aimed to co-design a digital knowledge translation intervention, reflective of adolescents' voices, to provide evidence-based, accessible, and accurate SRH resources.

Methods: We conducted a large multisite project across 3 cities in Canada. Using a community-based participatory approach and principles of human-centered design, we established adolescent advisory groups (AAGs) to actively engage them in the design, development, and implementation of the intervention.

Results: A total of 26 participants were recruited from Edmonton, Vancouver, and Toronto to be part of the advisory groups. AAG members participated in design-thinking sessions to brainstorm ideas for website design, identify informational needs for content development, provide iterative feedback on the design of the intervention, and suggest strategies to improve engagement and interaction. With their input, 6 priority areas were identified to develop SRH resources (ie, puberty, menstruation, sexually transmitted infections, healthy relationships, sexual assault, and contraception). Their feedback informed the design's language, visual appeal, and engagement factors. To promote meaningful engagement of AAGs at each step, we used strategies such as gamification activities, group discussions, and flexible scheduling, resulting in high retention and ownership of the process among AAG members.

Conclusions: Co-designing with adolescents strengthened the intervention's cultural relevance, youth-friendliness, and credibility. Our process emphasizes the significance of involving adolescents in co-designing SRH interventions, which results in more meaningful, long-term, and youth-friendly solutions.

Background: Remote patient monitoring (RPM) has the potential to reduce in-clinic visits and promote proactive and preventive care for patients with chronic diseases in primary care. However, a decentralized approach to RPM in a primary health care (PHC) setting has not met stakeholders' expectations regarding scalability. This study introduces a centralized virtual ward (CVW)-led RPM, utilizing a multidisciplinary team approach to monitor patients with chronic diseases by clinicians who do not belong to the patients' PHC center.

Objective: This study aimed to gain a better understanding of patients' perceptions of CVW-led RPM for managing chronic diseases in a PHC setting.

Methods: In-depth interviews were conducted with 22 patients with chronic diseases enrolled at a PHC center in Stockholm, Sweden. The RPM project ran between October 2018 and April 2019 and included a total of 395 patients. Interviews followed a semistructured interview guide and were analyzed using qualitative content analysis.

Results: Primary care patients with chronic diseases expressed that their contact with the CVW felt impersonal but at the same time secure and accessible. They noted a lack of coordination and communication between the clinicians of the CVW and their PHC providers. Captured data resulted in 1 overarching theme "Sense of security and accessibility, but impersonal and uncoordinated" based on 5 categories: sense of security, care and self-care, accessibility, quality of care, and communication.

Conclusions: Our findings suggest that by addressing patients' needs for new organizational routines for patient-caregiver communication, RPM via centralized virtual wards can better realize the potential of this technology.

Background: Digital health applications (DiGA) have been integrated into Germany's health care system since 2019, offering certified medical devices for various health conditions. This study focuses on deprexis and Selfapy, the first 2 permanently approved DiGA for depressive disorders in Germany, to evaluate their usability for people ≥60 years. The study's significance is underscored by the underrepresentation of older people in previous DiGA studies, accompanied by an emergent risk of inequalities in distribution for this vulnerable population.

Objective: This study assessed the usability of DiGA deprexis and Selfapy for adults aged ≥60 years with mild to moderate depression. The more user-friendly option will be chosen for the DiGA4Aged project's upcoming randomized controlled trial.

Methods: The prospective observational study uses the People at the Centre of Mobile Application Development (PACMAD) usability model in a mixed methods design. The study's multistage data collection encompasses sociodemographic data and quantitative questionnaires about health literacy (European Health Literacy Survey Questionnaire [HLS-EU-Q16]), electronic health literacy (revised German eHealth Literacy Scale [GR-eHEALS]), media affinity, depressive symptoms (9-item Patient Health Questionnaire [PHQ-9]), and perceived usability (System Usability Scale [SUS]), as well as a qualitative think-aloud and semistructured interview. Participants were equally allocated to use either deprexis or Selfapy. Recruitment of 18 participants was conducted at 3 hospital departments (ie, psychiatry, psychosomatics, and geriatrics) in spring 2024. Participants were eligible if they were aged ≥60 years, were diagnosed with mild or moderate depressive disorder, owned a digital device, and gave written consent to participate.

Results: Quantitative analysis revealed age, gender, depressive severity, and health literacy parity between both groups. Selfapy users displayed marginally lower technical proficiency and lower usability scores. Qualitative data showed lower usability among participants in the Selfapy group due to design-related errors and higher cognitive load. Despite visual, psychomotor, and cognitive challenges, participants endorsed both DiGA for older users, stressing the importance of assistance and practicing the usage.

Conclusions: Reported difficulties in usability may help to improve future DiGA development for older people, especially as the willingness to use DiGA exists.