Application of Artificial Intelligence in Community-Based Primary Health Care: Systematic Scoping Review and Critical Appraisal.

BACKGROUND: Research on the integration of artificial intelligence (AI) into community-based primary health care (CBPHC) has highlighted several advantages and disadvantages in practice regarding, for example, facilitating diagnosis and disease management, as well as doubts concerning the unintended harmful effects of this integration. However, there is a lack of evidence about a comprehensive knowledge synthesis that could shed light on AI systems tested or implemented in CBPHC. OBJECTIVE: We intended to identify and evaluate published studies that have tested or implemented AI in CBPHC settings. METHODS: We conducted a systematic scoping review informed by an earlier study and the Joanna Briggs Institute (JBI) scoping review framework and reported the findings according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analysis-Scoping Reviews) reporting guidelines. An information specialist performed a comprehensive search from the date of inception until February 2020, in seven bibliographic databases: Cochrane Library, MEDLINE, EMBASE, Web of Science, Cumulative Index to Nursing and Allied Health Literature (CINAHL), ScienceDirect, and IEEE Xplore. The selected studies considered all populations who provide and receive care in CBPHC settings, AI interventions that had been implemented, tested, or both, and assessed outcomes related to patients, health care providers, or CBPHC systems. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Two authors independently screened the titles and abstracts of the identified records, read the selected full texts, and extracted data from the included studies using a validated extraction form. Disagreements were resolved by consensus, and if this was not possible, the opinion of a third reviewer was sought. A third reviewer also validated all the extracted data. RESULTS: We retrieved 22,113 documents. After the removal of duplicates, 16,870 documents were screened, and 90 peer-reviewed publications met our inclusion criteria. Machine learning (ML) (41/90, 45%), natural language processing (NLP) (24/90, 27%), and expert systems (17/90, 19%) were the most commonly studied AI interventions. These were primarily implemented for diagnosis, detection, or surveillance purposes. Neural networks (ie, convolutional neural networks and abductive networks) demonstrated the highest accuracy, considering the given database for the given clinical task. The risk of bias in diagnosis or prognosis studies was the lowest in the participant category (4/49, 4%) and the highest in the outcome category (22/49, 45%). CONCLUSIONS: We observed variabilities in reporting the participants, types of AI methods, analyses, and outcomes, and highlighted the large gap in the effective development and implementation of AI in CBPHC. Further studies are needed to efficiently guide the development and implementation of AI interventions in CBPHC settings.

Evidence Supporting the Management of Medical Conditions During Long-Duration Spaceflight: Protocol for a Scoping Review.

BACKGROUND: Future long-duration space exploration missions, such as traveling to Mars, will create an increase in communication time delays and disruptions and remove the viability of emergency returns to Earth for timely medical treatment. Thus, higher levels of medical autonomy are necessary. Crew selection is proposed as the first line of defense to minimize medical risk for future missions; however, the second proposed line of defense is medical preparedness and crew member autonomy. In an effort to develop a decision support system, the Canadian Space Agency mandated a team of scientists from Thales Research and Technology Canada (Québec, QC) and Université Laval (Québec, QC) to create an evidence-based medical condition database linking mission-critical human conditions with key causal factors, diagnostic and treatment information, and probable outcomes. OBJECTIVE: To complement this database, we are currently conducting a scoping review to better understand the depth and breadth of evidence about managing medical conditions in space. METHODS: This scoping review will adhere to quality standards for scoping reviews, employing Levac, Colquhoun, and O'Brien's 6-stage methodology; the reported results will follow the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) extension for scoping reviews. In stage 1, we identified the research question in collaboration with the Canadian Space Agency (CSA), the main knowledge user. We prioritized 10 medical conditions: (1) acute coronary syndrome, (2) atrial fibrillation, (3) eye penetration, (4) herniated disk, (5) nephrolithiasis, (6) pulmonary embolism, (7) retinal detachment, (8) sepsis, (9) stroke, and (10) spaceflight associated neuro-ocular syndrome. In stage 2, with the help of an information specialist from Cochrane Canada Francophone, papers were identified through searches of the following databases: ARC, Embase, IeeeXplore, Medline Ovid, PsychINFO, and Web of Science. In stage 3, studies will be selected and assessed using a 3-step process and emerging, refined exclusion criteria. In stage 4, the data will be charted in a table based on parameters required by the CSA and developed using Google spreadsheets for shared access. In stage 5, evidence-based descriptive summaries will be produced for each condition, as well as descriptive analyses of collected data. Finally, in stage 6, the findings will be shared with the CSA to guide the completion of this project. RESULTS: This study was planned in December 2018. Stage 1 has been completed. The initial database search strategy with all target conditions combined identified a total of 10,403 citations to review through title and abstract screening and after duplicate removal. We plan to complete stages 2-6 by the beginning of 2021. CONCLUSIONS: This scoping review will map the literature on the management of 10 priority medical conditions in space. It will also enable us to identify knowledge gaps that must be addressed in future research, ensuring successful and medically safe future missions as humankind embarks upon new frontiers of space exploration. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/24323.

Learning Integrated Health System to Mobilize Context-Adapted Knowledge With a Wiki Platform to Improve the Transitions of Frail Seniors From Hospitals and Emergency Departments to the Community (LEARNING WISDOM): Protocol for a Mixed-Methods Implementation Study.

BACKGROUND: Elderly patients discharged from hospital experience fragmented care, repeated and lengthy emergency department (ED) visits, relapse into their earlier condition, and rapid cognitive and functional decline. The Acute Care for Elders (ACE) program at Mount Sinai Hospital in Toronto, Canada uses innovative strategies, such as transition coaches, to improve the care transition experiences of frail elderly patients. The ACE program reduced the lengths of hospital stay and readmission for elderly patients, increased patient satisfaction, and saved the health care system over Can $4.2 million (US $2.6 million) in 2014. In 2016, a context-adapted ACE program was implemented at one hospital in the Centre intégré de santé et de services sociaux de Chaudière-Appalaches (CISSS-CA) with a focus on improving transitions between hospitals and the community. The quality improvement project used an intervention strategy based on iterative user-centered design prototyping and a "Wiki-suite" (free web-based database containing evidence-based knowledge tools) to engage multiple stakeholders. OBJECTIVE: The objectives of this study are to (1) implement a context-adapted CISSS-CA ACE program in four hospitals in the CISSS-CA and measure its impact on patient-, caregiver-, clinical-, and hospital-level outcomes; (2) identify underlying mechanisms by which our context-adapted CISSS-CA ACE program improves care transitions for the elderly; and (3) identify underlying mechanisms by which the Wiki-suite contributes to context-adaptation and local uptake of knowledge tools. METHODS: Objective 1 will involve staggered implementation of the context-adapted CISSS-CA ACE program across the four CISSS-CA sites and interrupted time series to measure the impact on hospital-, patient-, and caregiver-level outcomes. Objectives 2 and 3 will involve a parallel mixed-methods process evaluation study to understand the mechanisms by which our context-adapted CISSS-CA ACE program improves care transitions for the elderly and by which our Wiki-suite contributes to adaptation, implementation, and scaling up of geriatric knowledge tools. RESULTS: Data collection started in January 2019. As of January 2020, we enrolled 1635 patients and 529 caregivers from the four participating hospitals. Data collection is projected to be completed in January 2022. Data analysis has not yet begun. Results are expected to be published in 2022. Expected results will be presented to different key internal stakeholders to better support the effort and resources deployed in the transition of seniors. Through key interventions focused on seniors, we are expecting to increase patient satisfaction and quality of care and reduce readmission and ED revisit. CONCLUSIONS: This study will provide evidence on effective knowledge translation strategies to adapt best practices to the local context in the transition of care for elderly people. The knowledge generated through this project will support future scale-up of the ACE program and our wiki methodology in other settings in Canada. TRIAL REGISTRATION: ClinicalTrials.gov NCT04093245; https://clinicaltrials.gov/ct2/show/NCT04093245. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/17363.