Search Articles

View query in Help articles search

Search Results (1 to 10 of 1005 Results)

Download search results: CSV END BibTex RIS


Evaluation of the Tu’Washindi Na PrEP Intervention to Reduce Gender-Based Violence and Increase Preexposure Prophylaxis Uptake and Adherence Among Kenyan Adolescent Girls and Young Women: Protocol for a Cluster Randomized Controlled Trial

Evaluation of the Tu’Washindi Na PrEP Intervention to Reduce Gender-Based Violence and Increase Preexposure Prophylaxis Uptake and Adherence Among Kenyan Adolescent Girls and Young Women: Protocol for a Cluster Randomized Controlled Trial

We have randomized 22 administrative wards in a 1:1 ratio and aim to enroll about 72 adolescent girls and young women from each (total N=about 1584) to receive either the Tu’Washindi intervention plus usual HIV prevention services, or usual HIV prevention services alone.

Sarah T Roberts, Alexandra M Minnis, Sue Napierala, Elizabeth T Montgomery, Lina Digolo, Mackenzie L Cottrell, Erica N Browne, Jacqueline Ndirangu, Joyce Boke, Kawango Agot

JMIR Res Protoc 2025;14:e55931

Ambulance Commanders’ Reluctance to Enter Road Tunnels in Simulated Incidents and the Effects of a Tunnel-Specific e-Learning Course on Decision-Making: Web-Based Randomized Controlled Trial

Ambulance Commanders’ Reluctance to Enter Road Tunnels in Simulated Incidents and the Effects of a Tunnel-Specific e-Learning Course on Decision-Making: Web-Based Randomized Controlled Trial

Participants (N=20) were enrolled and randomized to either a prerecorded lecture or the tunnel-specific e-learning course. Both groups completed follow-up simulations at 1 and 6 months, which included 15 decision points (each designed as a multiple-choice question). Outcomes included the decision to correctly enter the tunnel, the number of correct decisions, and time. Participants also self-evaluated their decision-making ability and sense of security in handling road tunnel incidents.

Johan Hylander, Lina Gyllencreutz, Michael Haney, Anton Westman

JMIR Form Res 2025;9:e58542

The AI Reviewer: Evaluating AI’s Role in Citation Screening for Streamlined Systematic Reviews

The AI Reviewer: Evaluating AI’s Role in Citation Screening for Streamlined Systematic Reviews

Hence, 121 citations (n=21, 17.4% included and n=100, 82.6% excluded) were tested against predefined eligibility criteria using Chat GPT 3.5 (version September 25, 2023), Chat GPT 4 (version September 25, 2023), Google Bard (version 1.15; released on September 2, 2023), Meta Llama 2 (70b parameters, version 2.1.1; released on October 10, 2023), and Claude AI 2 (version 1.3; released on July 11, 2023). We used descriptive statistics to evaluate sensitivity, specificity, and overall accuracy.

Jamie Ghossein, Brett N Hryciw, Tim Ramsay, Kwadwo Kyeremanteng

JMIR Form Res 2025;9:e58366

School-Partnered Collaborative Care (SPACE) for Pediatric Type 1 Diabetes: Development and Usability Study of a Virtual Intervention With Multisystem Community Partners

School-Partnered Collaborative Care (SPACE) for Pediatric Type 1 Diabetes: Development and Usability Study of a Virtual Intervention With Multisystem Community Partners

SPACEa design team community partner roles (n=17). a SPACE: school-partnered collaborative care. b Numbers add to more than 17 as partners could identify with more than one role. At the initial design meeting, participants generated 141 ideas for the SPACE redesign, of which 94 were unique. Partners assigned a numeric prioritization to ideas, which were then condensed to create a list of unique ideas (Multimedia Appendix 1).

Christine A March, Elissa Naame, Ingrid Libman, Chelsea N Proulx, Linda Siminerio, Elizabeth Miller, Aaron R Lyon

JMIR Diabetes 2025;10:e64096

Multilevel Intervention to Increase Patient Portal Use in Adults With Type 2 Diabetes Who Access Health Care at Community Health Centers: Single Arm, Pre-Post Pilot Study

Multilevel Intervention to Increase Patient Portal Use in Adults With Type 2 Diabetes Who Access Health Care at Community Health Centers: Single Arm, Pre-Post Pilot Study

The sample of completers (n=22) was recruited from 2 clinics, 68% (n=15) from 1 clinic, 32% (n=7) from the other clinic. Refer to Table 1 for demographic characteristics. The sample was predominately Latino or Hispanic (7/22, 77%) had a mean age of 56.32 (SD 10.93) years, 73% (16/22) were female, and 55% (12/22) were married or partnered. The majority reported low-income (86% Consolidated Standards of Reporting Trials (CONSORT) diagram. Participant demographics (n=22).

Robin Whittemore, Sangchoon Jeon, Samuel Akyirem, Helen N C Chen, Joanna Lipson, Maritza Minchala, Julie Wagner

JMIR Form Res 2025;9:e67293

Integration of a Patient-Centered mHealth Intervention (Support-Moms) Into Routine Antenatal Care to Improve Maternal Health Among Pregnant Women in Southwestern Uganda: Protocol for a Randomized Controlled Trial

Integration of a Patient-Centered mHealth Intervention (Support-Moms) Into Routine Antenatal Care to Improve Maternal Health Among Pregnant Women in Southwestern Uganda: Protocol for a Randomized Controlled Trial

In a randomized 3-arm pilot study (N=120) pregnant women who had not presented for ANC by their second trimester were equally randomized to receive (1) standard of care, which is the routine ANC information given at the maternity centers (control); (2) scheduled SMS audio messages from the final messaging prototype (scheduled messaging [SM]); and (3) SM, plus social supporter engagement through SMS (SS) [63].

Esther Cathyln Atukunda, Godfrey Rwambuka Mugyenyi, Jessica E Haberer, Mark J Siedner, Angella Musiimenta, Josephine N Najjuma, Celestino Obua, Lynn T Matthews

JMIR Res Protoc 2025;14:e67049

Enhancing Patient Outcome Prediction Through Deep Learning With Sequential Diagnosis Codes From Structured Electronic Health Record Data: Systematic Review

Enhancing Patient Outcome Prediction Through Deep Learning With Sequential Diagnosis Codes From Structured Electronic Health Record Data: Systematic Review

Among these 84 models, the most commonly applied DL technique for learning sequential diagnosis codes was RNNs and their derivatives (n=47, 56%), followed by transformers (n=22, 26%), which have been regularly applied in studies since their introduction in 2017 (Figure 2). Among the 38 studies that used embedding techniques to represent diagnostic data, the most frequently used embedding method was Word2 Vec (n=15, 39%), followed by GNNs (n=9, 24%) and transformers (n=3, 8%).

Tuankasfee Hama, Mohanad M Alsaleh, Freya Allery, Jung Won Choi, Christopher Tomlinson, Honghan Wu, Alvina Lai, Nikolas Pontikos, Johan H Thygesen

J Med Internet Res 2025;27:e57358