Clinical decision support systems (CDSS) can support clinicians with the timely detection of patients' clinical deterioration, however, less than half of clinical decision support (CDS) systems implemented for clinical deterioration are used by clinicians. Poor design of CDS systems has emerged as a contributing factor.

The aim of this study was to 1) identify interface design features that have been used in CDS systems for real-time detection of clinical deterioration; (2) determine which interface design features are preferred by clinicians; and (3) examine other design features (external to the interface) which influence CDS acceptance.

Three databases (Medline, Scopus and CINAHL) were searched to identify relevant studies. All studies that met the eligibility criteria were included. A qualitative narrative synthesis was undertaken.

Of 24 eligible articles, 17 described CDS systems in the form of a dashboard and 7 described alerts. Of the 17 dashboards, graphs and tables were the most used interface design features to display vital signs. Colour was the most frequently used interface design feature to signal the presence of deterioration with half of colour-coded dashboards using a traffic light schema to classify patient risk level. Clinicians preferred dashboards that included both graphs and tables. Clinicians have expressed that they were disinclined to use CDS systems with manual recording of vital signs and high alert frequency.

This scoping review uncovered wide variability in design features across CDS systems for real-time detection of deterioration. Our research calls for better adherence to reporting checklists when reporting on interface designs, and the development of design guidelines to guide interface designs of CDS systems for detecting deterioration in real-time. Our scoping review may serve as a preliminary guide for the design of future CDS systems for detecting deterioration.

Clinical decision support systems (CDSSs) have the potential to assist health professionals in making informed and cost-effective clinical decisions while reducing medical errors. However, compared to physical health, CDSSs have been less investigated within the mental health context. In particular, despite mental health professionals being the primary users of mental health CDSSs, few studies have explored their experiences and/or views on these systems. Furthermore, we are not aware of any reviews specifically focusing on this topic. To address this gap, we conducted a scoping review to map the state of the art in studies examining CDSSs from the perspectives of mental health professionals.

In this review, following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guideline, we systematically searched the relevant literature in two databases, PubMed and PsycINFO.

We identified 23 articles describing 20 CDSSs Through the synthesis of qualitative findings, four key barriers and three facilitators to the adoption of CDSSs were identified. Although we did not synthesize quantitative findings due to the heterogeneity of the results and methodologies, we emphasize the issue of a lack of valid quantitative methods for evaluating CDSSs from the perspectives of mental health professionals.

To the best of our knowledge, this is the first review examining mental health professionals' experiences and views on CDSSs. We identified facilitators and barriers to adopting CDSSs and highlighted the need for standardizing research methods to evaluate CDSSs in the mental health space.

This review aims to explore the integration of artificial intelligence (AI) in laryngology, with specific focus on the barriers preventing translation from pilot studies into routine clinical practice and strategies for successful implementation.

Laryngology has seen an increasing number of pilot and proof-of-concept studies demonstrating AI's ability to enhance diagnostics, treatment planning, and patient outcomes. Despite these advancements, few tools have been successfully adopted in clinical settings. Effective implementation requires the application of established implementation science frameworks early in the design phase. Additional factors required for the successful integration of AI applications include addressing specific clinical needs, fostering diverse and interdisciplinary teams, and ensuring scalability without compromising model performance. Governance, epistemic, and ethical considerations must also be continuously incorporated throughout the project lifecycle to ensure the safe, responsible, and equitable use of AI technologies.

While AI hold significant promise for advancing laryngology, its implementation in clinical practice remains limited. Achieving meaningful integration will require a shift toward practical solutions that prioritize clinicians' and patients' needs, usability, sustainability, and alignment with clinical workflows.

Continuous advancements in diabetes technologies have improved self-management for people with type 1 diabetes. Continuous glucose monitoring and automated insulin delivery systems have enhanced the quality of life and glycemic outcomes while reducing severe hypoglycemia and diabetes ketoacidosis hospitalizations. Despite these benefits, racial inequities in the use of advanced diabetes technology (ADT) persist.

This study aims to develop and evaluate a best practice advisory (BPA) within the electronic medical record (EMR) to reduce racial and ethnic disparities in ADT use. We hypothesize that an EMR-based BPA designed to standardize the prescribing of ADTs will minimize racial and ethnic disparities in ADT adoption or progression in use among pediatric and adult people with type 1 diabetes.

The Best Practice Advisories to Reduce Inequities in Technology Use (BPA-TECH) study will use a nonrandomized matched pair intervention design. Phase 1 will use qualitative methods to develop and refine the BPA, including focus groups and surveys of health care providers and people with type 1 diabetes or their caregivers. Phase 2 will evaluate the effectiveness of the BPA through a controlled before-after study of people with type 1 diabetes seen at 7 T1D Exchange Quality Improvement Collaborative (T1DX-QI) centers, with control people with type 1 diabetes matched from nonintervention T1DX-QI centers. The baseline and postintervention periods will be the 12 months before and 12 months after deployment of the BPA at the intervention centers, respectively. Eligibility criteria include people with type 1 diabetes aged ≥2 years with an EMR diagnosis of T1D during the baseline period. The primary outcome is the progression in ADT use from the baseline to postintervention periods.

This 3-year study began in July 2024, with data collection from key stakeholders for phase 1 qualitative research beginning in August 2024. For phase 2, we estimate approximately 3000 eligible non-Hispanic Black and Hispanic people with type 1 diabetes at the intervention centers and 15,000 matched controls. Data on ADT use, glycated hemoglobin (HbA1c), severe hypoglycemic events, and diabetes ketoacidosis events will be collected via the T1DX-QI coordinating center. The study is powered to detect a between-group difference of 15% in the proportion of patients in the intervention and control groups in meeting the primary endpoint. We anticipate the completion of this study by May 2027.

The BPA-TECH study aims to leverage health IT to address racial and ethnic disparities in ADT use among people with type 1 diabetes. By standardizing the approach to ADT prescribing for people with type 1 diabetes, the BPA-TECH has the potential to promote equity in diabetes management and improve clinical outcomes. The outcomes of this study will inform future efforts to reduce health care disparities.

ClinicalTrials.gov NCT06931275; https://clinicaltrials.gov/search?term=NCT06931275.

DERR1-10.2196/71038.

While drugs have a limited role in the management of dental presentations, Australian dentists have high rates of inappropriate prescribing of antibiotics. There is also a lack of relevant drug resources for dentists in Australia. Our team developed Drugs4dent®, a medicines decision support tool, that provides dentists with relevant drug knowledge, assists with appropriate prescribing and provides safety checks to reduce prescribing errors. The aim of this pilot study was to improve Drugs4dent® with focus groups of dentists, and assess the acceptability, usability, and user experience, of Drugs4dent®.

Focus groups of ten dentists were established to inform the improvement of Drugs4dent®. Acceptability and usability testing of Drugs4dent® was then undertaken with a further ten dentists using interviews and a survey. The survey was based on the Framework for Acceptability and System Usability Scale. Inductive thematic analysis was undertaken using Nvivo for the focus groups and interviews, and descriptive statistics for analysis of survey results.

Dentists from the focus group and interviews found the content of Drugs4dent® acceptable and useful for dentistry, recognising that similar drug information is currently not available. The majority agreed that Drugs4dent® would improve their ability to prescribe according to guidance. Participants reported Drugs4dent® was intuitive, and information was easy to locate. Most dentists preferred Drugs4dent® integrated with their dental practice software. In the absence of this functionality, they preferred Drugs4dent® as a standalone resource, without needing to input patient data. Drugs4dent® was subsequently commercialised with MIMS Australia, to create the decision support tool: MIMS Drugs4dent®.

Drugs4dent® is the first dental medicines decision tool in Australia. The high acceptability and usability of the tool, and subsequent commercialisation indicates that MIMS Drugs4dent® has substantial promise for the future, and can transform access to relevant drug information for Australian dentists.

The opioid overdose epidemic continues to worsen in the United States, with opioid-related deaths increasing by 13 folds from 2000 to 2022 in Wisconsin. Naloxone, an opioid antagonist, can save lives by reversing opioid overdose in a matter of minutes. However, naloxone access and utilization remain suboptimal.

This study examined the development and implementation of best practice alerting (BPA) processes within community pharmacies. This study assessed to what extent the BPA processes (1a) prompted pharmacists to discuss naloxone with high-risk patients; (1b) increased the number of naloxone prescriptions dispensed; and (2) evaluated the facilitators and barriers to implementing the BPA processes.

A pilot study was conducted to develop and implement a BPA process in 3 rural community pharmacies in Wisconsin. The process involved staff identifying high-risk patients, initiating naloxone discussions, and offering naloxone prescriptions. Quantitative monthly data were recorded by pharmacies. Semi-structured interviews were conducted with one pharmacist from each pharmacy to assess the implementation process and outcomes. Descriptive statistics were used to analyze quantitative data. Interview transcripts were analyzed for key themes describing facilitators and barriers to the implementation process.

The naloxone alerting process resulted in a notable increase in naloxone discussions and naloxone prescriptions dispensed. Pharmacists reported that pharmacy staff buy-in and engagement, adequate staffing, developing meaningful partnerships, and using depersonalizing, destigmatizing, and normalizing approaches in discussing naloxone with patients were key facilitators. Key barriers included naloxone cost or co-payment and time constraints.

Implementing a BPA process in community pharmacies can notably increase naloxone prescribing for high-risk patients. Positive message framing, staffing, meaningful partnerships, and staff buy-in were key facilitators of implementation. Identified barriers were cost or co-payment and time constraints.

In radiology, technological progress has led to an enormous increase in data volumes. To effectively use these data during diagnostics or subsequent clinical evaluations, they have to be aggregated at a central location and be meaningfully retrievable in context. Radiology data warehouses undertake this task: they integrate diverse data sources, enable patient-specific and examination-specific evaluations, and thus offer numerous benefits in patient care, education, and clinical research.The international standard Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) is particularly suitable for the implementation of such a data warehouse. FHIR allows for easy and fast data access, supports modern web-based frontends, and offers high interoperability due to the integration of medical ontologies such as SNOMED-CT or RadLex. Furthermore, FHIR has a robust data security concept. Because of these properties, FHIR has been selected by the Medical Informatics Initiative (MII) as the data standard for the core data set and is intended to be promoted as an international standard in the European Health Data Space (EHDS).Implementing the FHIR standard in radiology data warehouses is therefore a logical and sensible step towards data-driven medicine. · A data warehouse is essential for data-driven medicine, clinical care, and research purposes.. · Data warehouses enable efficient integration of AI results and structured report templates.. · Fast Healthcare Interoperability Resources (FHIR) is a suitable standard for a data warehouse.. · FHIR provides an interoperable data standard, supported by proven web technologies.. · FHIR improves semantic consistency and facilitates secure data exchange.. · Arnold P, Pinto dos Santos D, Bamberg F et al. FHIR - Overdue Standard for Radiology Data Warehouses. Rofo 2025; 197: 518-524.

mobile Health (mHealth) refers to use of mobile wireless technologies for health. The potential for mHealth to enhance healthcare delivery is supported by near-universal availability of mobile phones and their expanding coverage in low- and middle-income countries. This systematic review analyses the available evidence on mHealth clinical decision-making tools in maternal and perinatal health, and whether they lead to improved maternal and perinatal health outcomes in Sub-Saharan Africa (SSA).

Eligibility criteria: Studies conducted in SSA describing mHealth tools piloted or used for clinical decision-making in maternal or perinatal healthcare. Exclusion criteria included mHealth tools used outside of maternal and perinatal healthcare, publications lacking sufficient detail (where information couldn't be obtained through contacting authors), articles where tools were used on a laptop or desktop computer, and articles not published in English. Data sources: PubMed, CINAHL, EMBASE, Global Health, and Web of Science were searched for relevant articles following a predetermined search strategy with no date restrictions. A limited grey literature search was conducted. Risk of bias: We assessed the quality of included studies using the Cochrane Risk of bias 2 tool, Newcastle- Ottawa scale and COREQ. This comprehensive approach ensured a rigorous evaluation of bias and validity in our systematic review. Data extraction and synthesis: Two independent reviewers screened articles and extracted data.

1119 records were screened, and 36 articles met the inclusion criteria. Fifteen mHealth tools were identified across 11 SSA countries.

mHealth tools for clinical decision-making in maternal and perinatal care were found to be feasible, usable, and acceptable. They demonstrated adequate user satisfaction, and some demonstrated improvement of pregnancy outcomes. However, technologies lack scalability, with only one scaled up nationally, and few tools interacted with existing health information systems or had plans for sustainability. This review will help establish best practice for developing and scaling up mHealth clinical decision-making tools, helping to improve maternal and perinatal healthcare in SSA.

There is a growing interest in understanding the effects of human-machine interaction on moral decision-making (Moral-DM) and sense of agency (SoA). Here, we investigated whether the "moral behavior" of an AI may affect both moral-DM and SoA in a military population, by using a task in which cadets played the role of drone operators on a battlefield. Participants had to decide whether or not to initiate an attack based on the presence of enemies and the risk of collateral damage. By combining three different types of trials (Moral vs. two No-Morals) in three blocks with three type of intelligent system support (No-AI support vs. Aggressive-AI vs. Conservative-AI), we showed that participants' decisions in the morally challenging situations were influenced by the inputs provided by the autonomous system. Furthermore, by measuring implicit and explicit agency, we found a significant increase in the SoA at the implicit level in the morally challenging situations, and a decrease in the explicit responsibility during the interaction with both AIs. These results suggest that the AI behavior influences human moral decision-making and alters the sense of agency and responsibility in ethical scenarios. These findings have implications for the design of AI-assisted decision-making processes in moral contexts.

Digital tracking on mobile devices, combined with clinical decision support systems and targeted client communication, can facilitate service delivery and potentially improve outcomes.

To assess the effects of using a mobile device to track service use when combined with clinical decision support (Tracking + CDSS), with targeted client communications (Tracking + TCC), or both (Tracking + CDSS + TCC).

Cochrane CENTRAL, MEDLINE, Embase, Ovid Population Information Online (POPLINE), K4Health and WHO Global Health Library (2000 to November 2022).

Randomised and non-randomised trials in community/primary care settings.

primary care providers and clients Interventions: 1. Tracking + CDSS 2. Tracking + TCC 3. Tracking + CDSS + TCC Comparators: usual care (without digital tracking) DATA COLLECTION AND ANALYSIS: Two authors independently screened trials, extracted data and assessed risk of bias using the RoB 1 tool. We used a random-effects model to meta-analyse data producing risk differences (RD), risk ratios (RR), or odds ratios (OR) for dichotomous outcomes and mean differences (MD) for continuous outcomes. Evidence certainty was assessed using GRADE.

We identified 18 eligible studies (11 randomised, seven non-randomised) conducted in Bangladesh, China, Ethiopia, India, Kenya, Palestine, Uganda, and the USA. All non-randomised studies had a high risk of bias. These results are from randomised studies. 'Probably/may/uncertain' indicates 'moderate/low/very low' certainty evidence. Tracking + CDSS Relating to antenatal/ postnatal care: Providers' adherence to recommendations May slightly increase home visits in the week following delivery (2 studies, 4531 participants; RD 0.10 [0.07, 0.14]) May slightly increase counselling for initiating complementary feeding (2 studies, 4397 participants; RD 0.12 [0.08, 0.15]) May slightly increase the mean number of home visits in the month following delivery (1 study, 3023 participants; MD 0.75 [0.47, 1.03]) Uncertain effect on home visits within 24 hours of delivery Clients' health behaviours May slightly increase skin-to-skin care (1 study, 1544 participants; RD 0.05 [0.00, 0.10]) May slightly increase early breastfeeding (2 studies, 4540 participants; RD 0.08 [0.05, 0.12]) Uncertain effects on applying nothing to the umbilical cord, taking ≥ 90 iron-folate tablets during pregnancy, exclusively breastfeeding for six months, delaying the newborn's bath at least two days and Kangaroo Mother Care. Clients' health status May reduce low birthweight babies (1 study, 3023 participants; RR 0.53 [0.38, 0.73]) May increase infants with pneumonia or fever seeking care (1 study, 3470 participants; RR 1.13 [1.03, 1.24]) Uncertain effects on stillbirths, neonatal and infant deaths, or testing positive for HIV during antenatal testing Tracking + TCC Clients' health status In stroke patients over 12 months: May slightly increase blood pressure (BP) medication adherence (1 study, 1226 participants; RR 1.10 [1.00, 1.21]) May reduce deaths (1 study, 1226 participants; RR 0.52 [0.28, 0.96]) May slightly reduce systolic BP (1 study, 1226 participants; MD -2.80 mmHg [-4.90, -0.70]) May slightly improve EQ-5D scores (1 study, 1226 participants; MD 0.04 [0.02, 0.06]) May reduce stroke hospitalisations (1 study, 1226 participants; RR 0.45 [0.32, 0.64]). Tracking + CDSS + TCC Providers' adherence to recommendations Probably increases guideline adherence for antenatal screening and management of anaemia (1 study, 10,502 participants; OR 1.88 [1.52, 2.32]), diabetes (1 study, 8669 participants; OR 1.45 [1.14, 1.84}), hypertension (1 study, 15,555 participants; OR 1.62 [1.29, 2.04]) and probably leads to lower adherence for abnormal foetal growth (1 study, 1165 participants; OR 0.59 [0.37, 0.95]). May slightly increase nevirapine prophylaxis in infants of HIV+ve mothers (1 study, 609 participants; OR 1.75 [0.73, 4.19]) Data quality In pregnant women (1 study, 6367 participants), tracking + CDSS + TCC: Probably slightly reduces missing data for haemoglobin (RR 0.77 [0.71, 0.84]) but slightly more for BP at delivery (RR 1.16 [1.08, 1.24]) May have little or no effect on missing data on gestational age (RR 0.96 [0.81, 1.14]) or birthweight (RR 0.90 [0.77, 1.04]) Clients' health behaviour May have little or no effect on being on anti-retroviral therapy at delivery (1 study, 438 participants; OR 1.41 [0.81, 2.45]) or exclusive breastfeeding for six months (1 study, 695 participants; OR 1.74 [0.95, 3.17]) in HIV+ve mothers Uncertain effects on physical activity in high cardiovascular-risk adults Clients' health status May reduce the number of deaths in patients with hypertension and diabetes (1 study, 3698 participants; OR 0.61 [0.35, 1.06]) May reduce new cardiovascular events in high-cardiovascular risk adults over 6-18 months (1 study, 8642 participants; OR 0.58 [0.42, 0.80}) May slightly decrease in antenatal women severe hypertension, but the confidence interval includes both a decrease and increase (1 study, 6367 participants; OR 0.61 [0.27, 1.37]) In women receiving antenatal care (1 study, 6367 participants), tracking + CDSS + TCC maymake little or no difference to adverse pregnancy outcomes (OR 0.99 [0.87, 1.12]), moderate or severe anaemia (OR 0.82 [0.51, 1.31]), or large-for-gestational-age babies (OR 1.06 [0.90, 1.25]). In adults with hypertension or diabetes (1 study, 3324 participants), tracking + CDSS + TCC maymake little or no difference to HbA1c (MD 0.08 [-0.27, 0.43]), total cholesterol (MD -2.50 [-7.10, 2.10]), 10-year cardiovascular risk (MD -0.40 [-2.30, 1.50]), tobacco use (MD-0.05 [-0.47, 0.37]), alcohol use (MD 0.70 [-3.70, 5.10]), or PHQ-9 (MD -1.60 [-4.40, 1.20]). Uncertain effects on maternal or infant mortality before the baby reaches 18 months in HIV-positive mothers, patients who achieve optimal BP, BP controlled at five years, diastolic or systolic BP, body mass index, fasting glucose and quality of life in adults with hypertension or diabetes Client service utilisation May have little or no effect on missed early infant diagnosis visits (1 study, 1183 participants; OR 0.92 [0.63, 1.35]). Uncertain effects on linkage to care Client satisfaction Probably increases slightly the number of adults with hypertension or diabetes reporting "slightly/much better" change in the quality of care (1 study, 3324 participants; RR 1.02 [1.00, 1.03]). No studies evaluated time between presentation and appropriate management, timeliness of receiving/accessing care, provider acceptability/satisfaction, resource use, or unintended consequences.

Digital tracking may improve primary care workers' ability to follow recommended antenatal and chronic disease practices, quality of patient records, patient health outcomes and service use. However, these interventions led to small or no outcome differences in most studies.

To assess the prevalence of recommended design elements in implemented electronic health record (EHR) interruptive alerts across pediatric care settings.

We conducted a 3-phase mixed-methods cross-sectional study. Phase 1 involved developing a codebook for alert content classification. Phase 2 identified the most frequently interruptive alerts at participating sites. Phase 3 applied the codebook to classify alerts. Inter-rater reliability (IRR) for the codebook and descriptive statistics for alert design contents were reported.

We classified alert content on design elements such as the rationale for the alert's appearance, the hazard of ignoring it, directive versus informational content, administrative purpose, and whether it aligned with one of the Institute of Medicine's (IOM) domains of healthcare quality. Most design elements achieved an IRR above 0.7, with the exceptions for identifying directive content outside of an alert (IRR 0.58) and whether an alert was for administrative purposes only (IRR 0.36). IRR was poor for all IOM domains except equity. Institutions varied widely in the number of unique alerts and their designs. 78% of alerts stated their purpose, over half were directive, and 13% were informational. Only 2%-20% of alerts explained the consequences of inaction.

This study raises important questions about the optimal balance of alert functions and desirable features of alert representation.

Our study provides the first multi-center analysis of EHR alert design elements in pediatric care settings, revealing substantial variation in content and design. These findings underline the need for future research to experimentally explore EHR alert design best practices to improve efficiency and effectiveness.

Significant quality gaps exist in the management of chronic liver diseases and cirrhosis. Clinical decision support systems-information-driven tools based in and launched from the electronic health record-are attractive and potentially scalable prospective interventions that could help standardize clinical care in hepatology. Yet, clinical decision support systems have had a mixed record in clinical medicine due to issues with interoperability and compatibility with clinical workflows. In this review, we discuss the conceptual origins of clinical decision support systems, existing applications in liver diseases, issues and challenges with implementation, and emerging strategies to improve their integration in hepatology care.

Digital health innovations represent unique opportunities to address maternal, newborn, and child health challenges in Sub-Saharan Africa. In 2021, the World Health Organization (WHO) launched the Digital Adaptation Kits (DAKs) for antenatal care (ANC) as part of its Standards-Based, Machine-Readable, Adaptive, Requirements-Based, and Testable (SMART) guidelines approach. DAKs are operational and software-neutral mechanisms that convert WHO guidelines into standardized formats that can be easily integrated into digital systems by various countries. This article outlines the methodology for updating and integrating WHO DAK content into the BornFyne-prenatal management system (PNMS) version 2.0.

This study, which employs a participatory action research approach, is part of a larger research study for the BornFyne-PNMS project. A review of the ANC DAK operational document and data dictionaries was conducted to identify elements that were present in BornFyne-PNMS version 1.0. This was followed by a series of consultations and stakeholder meetings.

Five stakeholder meetings were held to engage stakeholders across Cameroon. Some of the registration elements, among other DAK aspects of ANC service provision, were identified in BornFyne version 1.0 but required reorganizing, remodeling, and reanalyzing to align with the International Classification of Diseases codes and DAK data content as part of the expansion for BornFyne version 2.0. Up to 40% of the DAK dictionary data content existed within the BornFyne-PNMS version 1.0, including additional DAK content adapted to update BornFyne-PNMS version 2.0. The digital health ecosystem in Cameroon is in an emerging phase with an increasing demand for digital health technologies, especially in the areas of reproductive, maternal, newborn, child, and adolescent health.

The digital health ecosystem in Cameroon is in an emerging phase with an increasing demand for digital health technologies, especially in the area of reproductive, maternal, newborn, child, and adolescent health. This article describes and documents the steps in operationalization of the ANC DAK content into the BornFyne-PNMS content, highlighting the DAK as an important tool for guiding and facilitating software engineers in developing and integrating recommended ANC guidelines into digital platforms to facilitate interoperability, going by the structure of the document, its workflow processes, and content mapping elements.

The return-to-sport (RTS) process is multifaceted and complex, as multiple variables may interact and influence the time to RTS. These variables include intrinsic factors related the player, such as anthropometrics and playing position, or extrinsic factors, such as competitive pressure. Providing an individualised estimation of time to return to play is often challenging, and clinical decision support tools are not common in sports medicine. This study uses epidemiological data to demonstrate a Bayesian Network (BN). We applied a BN that integrated clinical, non-clinical factors, and expert knowledge to classify time day to RTS and injury severity (minimal, mild, moderate and severe) for individual players. Retrospective injury data of 3374 player seasons and 6143 time-loss injuries from seven seasons of the professional German football league (Bundesliga, 2014/2015 through 2020/2021) were collected from public databases and media resources. A total of twelve variables from three categories (player's characteristics and anthropometrics, match information and injury information) were included. The response variables were 1) days to RTS (1-3, 4-7, 8-14, 15-28, 29-60, > 60, and 2) injury severity (minimal, mild, moderate, and severe). The sensitivity of the model for days to RTS was 0.24-0.97, while for severity categories it was 0.73-1.00. The user's accuracy of the model for days to RTS was 0.52-0.83, while for severity categories, it was 0.67-1.00. The BN can help to integrate different data types to model the probability of an outcome, such as days to return to sport. In our study, the BN may support coaches and players in 1) predicting days to RTS given an injury, 2) team planning via assessment of scenarios based on players' characteristics and injury risk, and 3) understanding the relationships between injury risk factors and RTS. This study demonstrates the how a Bayesian network may aid clinical decision making for RTS.

Accurately predicting the short-term in-hospital mortality risk for patients with stroke and TBI (Traumatic Brain Injury) is crucial for improving the quality of emergency medical care.

This study analyzed data from 2,125 emergency admission patients with stroke and traumatic brain injury at two Grade a hospitals in China from January 2021 to March 2024. LASSO regression was used for feature selection, and the predictive performance of logistic regression was compared with six machine learning algorithms. A 70:30 ratio was applied for cross-validation, and confidence intervals were calculated using the bootstrap method. Temporal validation was performed on the best-performing model. SHAP values were employed to assess variable importance.

The random forest algorithm excelled in predicting in-hospital 3-day mortality, achieving an AUC of 0.978 (95% CI: 0.966-0.986). Time series validation demonstrated the model's strong generalization capability, with an AUC of 0.975 (95% CI: 0.963-0.986). Key predictive factors in the final model included metabolic syndrome, NEWS2 score, Glasgow Coma Scale (GCS), whether surgery was performed, bowel movement status, potassium level (K), aspartate transaminase (AST) level, and temporal factors. SHAP value analysis further confirmed the significant contributions of these variables to the predictive outcomes. The random forest model developed in this study demonstrates good accuracy in predicting short-term in-hospital mortality rates for stroke and traumatic brain injury patients. The model integrates emergency scores, clinical signs, and key biochemical indicators, providing a comprehensive perspective for risk assessment. This approach, which incorporates emergency data, holds promise for assisting decision-making in clinical practice, thereby improving patient outcomes.

Clinical decision support systems (CDSS) frequently exhibit insufficient contextual adaptation, diminishing user engagement. To enhance the sensitivity of CDSS to contextual conditions, it is crucial first to develop a comprehensive understanding of the context factors influencing the clinical decision-making process. Therefore, this study aims to systematically identify and provide an extensive overview of contextual factors affecting clinical decision-making from the literature, enabling their consideration in the future implementation of CDSS.

A scoping review was conducted following the PRISMA-ScR guidelines to identify context factors in the clinical decision-making process. Searches were performed across nine databases: PubMed, APA PsycInfo, APA PsyArticles, PSYINDEX, CINAHL, Scopus, Embase, Web of Science, and LIVIVO. The search strategy focused on combined terms related to contextual factors and clinical decision-making. Included articles were original research articles written in English or German that involved empirical investigations related to clinical decision-making. The identified context factors were categorized using the card sorting method to ensure accurate classification.

The data synthesis included 84 publications, from which 946 context factors were extracted. These factors were assigned to six primary entities through card sorting: patient, physician, patient's family, institution, colleagues, and disease treatment. The majority of the identified context factors pertained to individual characteristics of the patient, such as health status and demographic attributes, as well as individual characteristics of the physician, including demographic data, skills, and knowledge.

This study provides a comprehensive overview of context factors in clinical decision-making previously investigated in the literature, highlighting the complexity and diversity of contextual influences on the decision-making process. By offering a detailed foundation of identified context factors, this study paves the way for future research to develop more effective, context-sensitive CDSS, enhancing personalized medicine and optimizing clinical outcomes with implications for practice and policy.

Background/Objectives: Clinical decision support systems (CDSSs) consisting of Computerized Physician Order Entry (CPOE) and oncology pathways serve as the foundation of high-quality cancer care. However, the resources needed to develop and maintain these systems have not been characterized for oncology enterprises. Methods: Executive leadership appointed a medical director and clinical pharmacist to develop and lead a Pathways and Protocols Program for the City of Hope (COH) enterprise. This involved developing a program charter and governance committee and a business case for resources to support CPOE in our Epic Beacon treatment orders. Missing CPOEs for oncology treatments were identified for treatments in COH's Elsevier ClinicalPath treatment pathways and for those few diseases not in the pathways for medical oncology and hematology. New FDA oncology drug approvals were used to estimate ongoing CPOE build needs. Time estimates for Beacon analysts to build Beacon protocols were developed from a prior CPOE catch-up project, from informal surveys of our clinical pharmacists and Beacon leads, and surveys of staff leads at two other large, multisite cancer programs using Epic. Informal surveys of oncology clinicians and pharmacists were carried out to understand the time they were using to build Beacon orders that were not in the COH system. This information was used to build a business case for additional project management and staffing to catch up on building 400 missing Beacon orders, to maintain Beacon orders as new therapies and regimens are needed, and to provide required regulatory oversight of Beacon orders. Given these standards had not been shared by others, this work was gathered into a manuscript to help others evaluate and support needed resources to manage oncology pathway programs and CPOE to improve efficiencies, safety, and quality of care for medical oncology and hematology programs. Results: A Pathways and Protocols program was developed with a governance committee, a program charter, and a charge for disease committees to prioritize, approve, and oversee the regulation of COH's Beacon treatment orders. CPOE resources to catch up and maintain COH's Beacon treatment orders were developed and shared with COH's executive leadership. Informal surveys were completed to benchmark Beacon resources with COH and two other Beacon enterprises as well as to estimate the time used by COH clinicians to build Beacon orders for orders not in the system. Conclusions: The resources for managing clinical oncology pathways and CPOE for an enterprise have not previously been published. Work components identified from our work at COH are shared so that other oncology leaders might have a starting framework to evaluate their own CDSS needs for oncology pathways and CPOE.

Antimicrobial resistances pose significant challenges in health care systems. Clinical decision support systems (CDSSs) represent a potential strategy for promoting a more targeted and guideline-based use of antibiotics. The integration of artificial intelligence (AI) into these systems has the potential to support physicians in selecting the most effective drug therapy for a given patient.

This study aimed to analyze the feasibility of an AI-based CDSS pilot version for antibiotic therapy in sepsis patients and identify facilitating and inhibiting conditions for its implementation in intensive care medicine.

The evaluation was conducted in 2 steps, using a qualitative methodology. Initially, expert interviews were conducted, in which intensive care physicians were asked to assess the AI-based recommendations for antibiotic therapy in terms of plausibility, layout, and design. Subsequently, focus group interviews were conducted to examine the technology acceptance of the AI-based CDSS. The interviews were anonymized and evaluated using content analysis.

In terms of the feasibility, barriers included variability in previous antibiotic administration practices, which affected the predictive ability of AI recommendations, and the increased effort required to justify deviations from these recommendations. Physicians' confidence in accepting or rejecting recommendations depended on their level of professional experience. The ability to re-evaluate CDSS recommendations and an intuitive, user-friendly system design were identified as factors that enhanced acceptance and usability. Overall, barriers included low levels of digitization in clinical practice, limited availability of cross-sectoral data, and negative previous experiences with CDSSs. Conversely, facilitators to CDSS implementation were potential time savings, physicians' openness to adopting new technologies, and positive previous experiences.

Early integration of users is beneficial for both the identification of relevant context factors and the further development of an effective CDSS. Overall, the potential of AI-based CDSSs is offset by inhibiting contextual conditions that impede its acceptance and implementation. The advancement of AI-based CDSSs and the mitigation of these inhibiting conditions are crucial for the realization of its full potential.

Acute kidney injury (AKI) is common in hospitalized children which could rapidly progress into chronic kidney disease if not timely diagnosed. Prognostic prediction models for AKI were established to identify AKI early and improve children's prognosis.

To appraise prognostic prediction models for pediatric AKI.

Four English and four Chinese databases were systematically searched from January 1, 2010, to June 6, 2022. Articles describing prognostic prediction models for pediatric AKI were included. The data extraction was based on the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. The risk of bias (ROB) was assessed according to the Prediction model Risk of Bias Assessment Tool guideline. The quantitative synthesis of the models was not performed due to the lack of methods regarding the meta-analysis of prediction models.

Eight studies with 16 models were included. There were significant deficiencies in reporting and all models were considered at high ROB. The area under the receiver operating characteristic curve to predict AKI ranged from 0.69 to 0.95. However, only about one-third of models have completed internal or external validation. The calibration was provided only in four models. Three models allowed easy bedside calculation or electronic automation, and two models were evaluated for their impacts on clinical practice.

Besides the modeling algorithm, the challenges for developing prediction models for pediatric AKI reflected by the reporting deficiencies included ways of handling baseline serum creatinine and age-dependent blood biochemical indexes. Moreover, few prediction models for pediatric AKI were performed for external validation, let alone the transformation in clinical practice. Further investigation should focus on the combination of prediction models and electronic automatic alerts.

 Adolescents are at high risk for sexually transmitted infections (STIs) and frequently present to emergency departments (EDs) for care. Screening for STIs using confidential patient-reported outcomes represents an ideal use of electronic screening methodology.

 The objectives of this study were to implement a patient-facing, confidential electronic survey to assess adolescent risk for STIs and consent for testing with integrated provider-facing electronic clinical decision support (CDS) across six geographically dispersed pediatric EDs and evaluate implementation based on survey and CDS usage metrics.

 A pilot site provided code for the electronic survey, data query, and CDS templates to six EDs. Institutions identified necessary information technology (IT) personnel, completed the local build, and made modifications to suit individual site workflow variations with all sites successfully deploying the electronic survey with electronic health record (EHR)-embedded CDS.

 Out of 79,780 eligible adolescents, 6,165 adolescents completed the confidential health survey between April 12, 2021 and September 25, 2022. The CDS was triggered indicating the patient was at risk or consented to STI testing across all six sites 2,058 times. The average percentage of time the CDS was acknowledged by a provider was 81.6% (range 45.7-97.6%). The median number of providers who acknowledged each instance of the CDS was 2.0. STI testing was ordered from the CDS on average 47.3% of the time. CDS acknowledged selection of "other" and "[testing] already ordered" was the most frequent indication STI testing was not ordered from the CDS.

 Successful deployment of patient-facing screeners with integrated electronic CDS across multiple healthcare institutions is feasible. A combination of different types of IT and informatics expertise with local knowledge of clinical workflows is essential to success.

Healthcare avoidance or delays for wounds and related skin- and soft-tissue infections are often attributed to negative interactions with medical providers. An infrastructural violence framework posits that healthcare infrastructure serves as a material channel for structural violence, maintaining inequities in healthcare experiences and outcomes. Infrastructural violence ensues when infrastructure is designed for some members or groups within a society while perpetuating violence among others. This study draws on the concept to understand how healthcare infrastructure creates and perpetuates inequities within the healthcare system for people who inject drugs for their wound care-related needs. Between January and September 2023, semi-structured interviews were conducted with 12 medical providers in Maryland. An abductive thematic analysis approach was followed. Healthcare infrastructure mediated the relationship between structural factors, such as policies on prescribing privileges of medications for opioid use disorder and subsequent individual health experiences. Medical providers also described how their access to training, protocols, and other resources was insufficient to meet the needs of people who inject drugs presenting to healthcare settings for wound care. A new conceptual grounding provides recommendations on extending beyond medical provider behavior change interventions in healthcare settings to create supportive infrastructure, which includes readily available and accessible policies, protocols, and resources to care for this patient population.

Low back pain (LBP) is a highly prevalent problem causing substantial personal and societal burden. Although there are specific types of LBP, each with evidence-based treatment recommendations, most patients receive a nonspecific diagnosis that does not facilitate evidence-based and individualized care.

We designed, developed, and initially tested the usability of a LBP diagnosis and treatment decision support tool based on the available evidence for use by clinicians who treat LBP, with an initial focus on chiropractic care.

Our 3-step user-centered design approach consisted of identifying clinical requirements through the analysis of evidence reviews, iteratively identifying task-based user requirements and developing a working web-based prototype, and evaluating usability through scenario-based interviews and the System Usability Scale.

The 5 participating users had an average of 18.5 years of practicing chiropractic medicine. Clinical requirements included 44 patient interview and examination items. Of these, 13 interview items were enabled for all patients and 13 were enabled conditional on other input items. One examination item was enabled for all patients and 16 were enabled conditional on other items. One item was a synthesis of interview and examination items. These items provided evidence of 12 possible working diagnoses of which 3 were macrodiagnoses and 9 were microdiagnoses. Each diagnosis had relevant treatment recommendations and corresponding patient educational materials. User requirements focused on tasks related to inputting data, and reviewing and selecting working diagnoses, treatments, and patient education. User input led to key refinements in the design, such as organizing the input questions by microdiagnosis, adding a patient summary screen that persists during data input and when reviewing output, adding more information buttons and graphics to input questions, and providing traceability by highlighting the input items used by the clinical logic to suggest a working diagnosis. Users believed that it would be important to have the tool accessible from within an electronic health record for adoption within their workflows. The System Usability Scale score for the prototype was 84.75 (range: 67.5-95), considered as the top 10th percentile. Users believed that the tool was easy to use although it would require training and practice on the clinical content to use it effectively. With such training and practice, users believed that it would improve care and shed light on the "black hole" of LBP diagnosis and treatment.

Our systematic process of defining clinical requirements and eliciting user requirements to inform a clinician-facing decision support tool produced a prototype application that was viewed positively and with enthusiasm by clinical users. With further planned development, this tool has the potential to guide clinical evaluation, inform more specific diagnosis, and encourage patient education and individualized treatment planning for patients with LBP through the application of evidence at the point of care.

Use of biologic and targeted synthetic disease-modifying antirheumatic drugs (b/tsDMARDs) in patients with preexisting tuberculosis (TB), hepatitis B virus (HBV), or hepatitis C virus (HCV) infection can have serious consequences. Although various society guidelines recommend routine screening for these infections before initiating certain b/tsDMARDs, adherence to these recommendations varies widely. This quality improvement initiative evaluated local compliance with screening and assessed whether an automated computerized decision support system in the form of a best practice advisory (BPA) in the electronic health record could improve patient screening.

Established patients with autoimmune rheumatic disease (ARD) aged 18 years or older with at least one visit to our rheumatology practice between October 1, 2017, and March 3, 2022, were included. When prescribing a new b/tsDMARD, clinicians were alerted via a BPA that showed the most recent results for TB, HBV, and HCV. Screening proportions for TB, HBV, and HCV before BPA initiation were compared with those of eligible patients after the BPA implementation.

A total of 711 patients pre-BPA and 257 patients post-BPA implementation were included in the study. The BPA implementation was associated with statistically significant improvement in screening for TB from 66% to 82% (P ≤ 0.001), HCV from 60% to 79% (P ≤ 0.001), hepatitis B core antibody 32% to 51% (P ≤ 0.001), and hepatitis B surface antigen from 51% to 70% (P ≤ 0.001).

Implementation of a BPA can improve infectious disease screening for patients with ARD who are started on b/tsDMARDs and has potential to improve patient safety.

Radiology referral quality impacts patient care, yet factors influencing quality are poorly understood. This study assessed the quality of computed tomography (CT) referrals, identified associated characteristics, and evaluated the ESR-iGuide clinical decision support tool's ability to optimize referrals. A retrospective review analyzed 300 consecutive CT referrals from an acute care hospital. Referral quality was evaluated on a 5-point scale by three expert reviewers (inter-rater reliability κ = 0.763-0.97). The ESR-iGuide tool provided appropriateness scores and estimated radiation exposure levels for the actual referred exams and recommended exams. Scores were compared between actual and recommended exams. Associations between ESR-iGuide scores and referral characteristics, including the specialty of the ordering physician (surgical vs. non-surgical), were explored. Of the referrals, 67.1% were rated as appropriate. The most common exams were head and abdomen/pelvis CTs. The ESR-iGuide deemed 70% of the actual referrals "usually appropriate" and found that the recommended exams had lower estimated radiation exposure compared to the actual exams. Logistic regression analysis showed that non-surgical physicians were more likely to order inappropriate exams compared to surgical physicians. Over one-third of the referrals showed suboptimal quality in the unstructured system. The ESR-iGuide clinical decision support tool identified opportunities to optimize appropriateness and reduce radiation exposure. Implementation of such a tool warrants consideration to improve communication and maximize patient care quality.

The objective of this research is to assess the impact of clinical decision support (CDS) tools on the practices of Indian physicians.

Descriptive statistics and frequency distributions are used to assess the data.

Through a primary survey, it was found that about 69% of the physicians frequently use clinical decision tools in their practice. The author found that the clinical decision tools affect 1-5 decisions every week (for about 54% of the sample). Nonetheless, a great many (31%) stated that they do not use the tools frequently; therefore, none of their decisions are affected by the technology on a usual basis. There is a slight improvement in diagnosis post the use of the app. Although 46% of doctors stated that they have made zero errors in decision making post the use of the application, 54% admitted making errors in 1-5 decisions per week. This shows that the tool has not been able to address all the needs of the doctors. A great many agreed that the tool helped in reducing diagnostic tests. Although a majority of doctors stated that they order fewer than five diagnostic tests post the use of the application, a great many doctors agreed that they order >10 tests after using the application. This could be due to less faith in the technology or could be an attribute of a small sample. The author intended to assess whether clinical decision tools are cost-effective. The author found that not all decision tools are cost-effective. The variation could be due to differences in comprehensiveness of information, product features, and area of practice.

This study exhibits that there is less faith in the technology and the application is favored by younger doctors. By and large, doctors agreed that the tool provides quicker diagnosis and is user-friendly.

While there are numerous tools available to inform if and when to use total knee arthroplasty (TKA), very few tools exist to help guide the recovery period after surgery.

We piloted a decision support tool that promotes a "people-like-me" (PLM) approach to rehabilitation after total knee arthroplasty (TKA). The PLM approach encourages person-centered care by "using historical outcomes data from similar (past) patients as a template of what to expect for a new patient". In this study, we evaluated how successfully the PLM tool was implemented and examined contextual factors that may have influenced its implementation.

Two outpatient physical therapy clinics (Clinics A and B) piloted the PLM tool from September 2020 - December 2022. We gathered data related to its implementation from multiple sources including the electronic health record (EHR), the tool itself, and surveys and interviews with patients and clinicians. We used an explanatory sequential mixed methods design to analyze the data overall and separately by each clinic.

Overall, the clinics met most pre-specified implementation targets, but did not use the tool as frequently as intended. Both clinics identified time, technology, and scheduling barriers to using the tool, but Clinic A scored higher in nearly every implementation outcome. Clinic A's success may have been related to its clinicians' higher level of experience, more positive attitudes towards the tool, and more active approach to implementation compared to Clinic B.

The clinics met most of our implementation targets, but Clinic A experienced more success than Clinic B. Future efforts to implement this PLM tool should (1) engage clinicians as active participants in the implementation process, (2) explore whether incorporating treatment recommendations into the PLM tool and/or using alternative training strategies can enhance its ability to alter clinician behavior, (3) integrate the tool within the EHR to complement existing workflows and mitigate implementation barriers, and (4) include randomized controlled trials that evaluate the tool's effectiveness and scalability across diverse clinical settings.

In prehospital emergency care, providers face significant challenges in making informed decisions due to factors such as limited cognitive support, high-stress environments, and lack of experience with certain patient conditions. Effective Clinical Decision Support Systems (CDSS) have great potential to alleviate these challenges. However, such systems have not yet been widely adopted in real-world practice and have been found to cause workflow disruptions and usability issues. Therefore, it is critical to investigate how to design CDSS that meet the needs of prehospital providers while accounting for the unique characteristics of prehospital workflows.

We conducted semi-structured interviews with 20 prehospital providers recruited from four Emergency Medical Services (EMS) agencies in an urban area in the northeastern U.S. The interviews focused on the decision-making challenges faced by prehospital providers, their technological needs for decision support, and key considerations for the design and implementation of a CDSS that can seamlessly integrate into prehospital care workflows. The data were analyzed using content analysis to identify common themes.

Our qualitative study identified several challenges in prehospital decision-making, including limited access to diagnostic tools, insufficient experience with certain critical patient conditions, and a lack of cognitive support. Participants highlighted several desired features to make CDSS more effective in the dynamic, hands-busy, and cognitively demanding prehospital context, such as automatic prompts for possible patient conditions and treatment options, alerts for critical patient safety events, AI-powered medication identification, and easy retrieval of protocols using hands-free methods (e.g., voice commands). Key considerations for successful CDSS adoption included balancing the frequency and urgency of alerts to reduce alarm fatigue and workflow disruptions, facilitating real-time data collection and documentation to enable decision generation, and ensuring trust and accountability while preventing over-reliance when using CDSS.

This study provides empirical insights into the challenges and user needs in prehospital decision-making and offers practical and system design implications for addressing these issues.

Electronic decision-support systems (EDSSs) aim to improve the quality of antenatal care (ANC) through adherence to evidence-based guidelines. We assessed the potential of the mHealth integrated model of hypertension, diabetes, and ANC EDSS and the World Health Organization EDSS to improve the quality of ANC in primary-level health care facilities in Nepal.

From December 2021 to January 2023, we conducted a mixed-methods evaluation in 19 primary-level ANC facilities in Bagmati Province, Nepal. Implementation was from March 2022 to August 2022. We conducted a health facility survey, ANC clinical observations, longitudinal case studies and validation workshop, in-depth interviews, monitoring visits, research team debriefing meetings, health care provider attitude survey, and stakeholder engagement and feedback meetings. Results were integrated using concurrent triangulation to develop explanations about the EDSS implementation process and the effects observed.

We identified 9 themes on implementation challenges that hindered the EDSS from generating the desired improvements to ANC quality. Facility readiness and provider confidence in using the EDSS were mixed. It was not always used or used as intended, and the approach to ANC provision did not change. EDSS inflexibility did not reflect how staff made decisions about pregnant women's needs or ensure that tests were done at the right time. There was mixed evidence that ANC staff believed that the EDSS benefited their work. The EDSS did not become fully integrated into existing health systems. Engagement of essential stakeholders fell short.

Different understandings of and inconsistent use of the EDSS highlighted the need for increased training and support periods, greater stakeholder engagement, and further integration into existing health systems. Our novel approach to integrating findings from multiple substudies offers uniquely valuable insights into the many factors needed for the successful implementation of an EDSS to improve the quality of ANC in Nepal.

Epilepsy care generates multiple sources of high-dimensional data, including clinical, imaging, electroencephalographic, genomic, and neuropsychological information, that are collected routinely to establish the diagnosis and guide management. Thanks to high-performance computing, sophisticated graphics processing units, and advanced analytics, we are now on the cusp of being able to use these data to significantly improve individualized care for people with epilepsy. Despite this, many clinicians, health care providers, and people with epilepsy are apprehensive about implementing Big Data and accompanying technologies such as artificial intelligence (AI). Practical, ethical, privacy, and climate issues represent real and enduring concerns that have yet to be completely resolved. Similarly, Big Data and AI-related biases have the potential to exacerbate local and global disparities. These are highly germane concerns to the field of epilepsy, given its high burden in developing nations and areas of socioeconomic deprivation. This educational paper from the International League Against Epilepsy's (ILAE) Big Data Commission aims to help clinicians caring for people with epilepsy become familiar with how Big Data is collected and processed, how they are applied to studies using AI, and outline the immense potential positive impact Big Data can have on diagnosis and management.

This review aims to synthesise the literature on the efficacy, evolution, and challenges of implementing Clincian Decision Support Systems (CDSS) in the realm of mental health, addiction, and concurrent disorders.

Following PRISMA guidelines, a systematic review and meta-analysis were performed. Searches conducted in databases such as MEDLINE, Embase, CINAHL, PsycINFO, and Web of Science through 25 May 2023, yielded 27,344 records. After necessary exclusions, 69 records were allocated for detailed synthesis. In the examination of patient outcomes with a focus on metrics such as therapeutic efficacy, patient satisfaction, and treatment acceptance, meta-analytic techniques were employed to synthesise data from randomised controlled trials.

A total of 69 studies were included, revealing a shift from knowledge-based models pre-2017 to a rise in data-driven models post-2017. The majority of models were found to be in Stage 2 or 4 of maturity. The meta-analysis showed an effect size of -0.11 for addiction-related outcomes and a stronger effect size of -0.50 for patient satisfaction and acceptance of CDSS.

The results indicate a shift from knowledge-based to data-driven CDSS approaches, aligned with advances in machine learning and big data. Although the immediate impact on addiction outcomes is modest, higher patient satisfaction suggests promise for wider CDSS use. Identified challenges include alert fatigue and opaque AI models.

CDSS shows promise in mental health and addiction treatment but requires a nuanced approach for effective and ethical implementation. The results emphasise the need for continued research to ensure optimised and equitable use in healthcare settings.

Clinical decision support systems (CDSSs) are designed to assist in health care delivery by supporting medical practice with clinical knowledge, patient information, and other relevant types of health information. CDSSs are integral parts of health care technologies assisting in disease management, including diagnosis, treatment, and monitoring. While electronic medical records (EMRs) serve as data repositories, CDSSs are used to assist clinicians in providing personalized, context-specific recommendations derived by comparing individual patient data to evidence-based guidelines.

This targeted literature review (TLR) aimed to identify characteristics and features of both stand-alone and EMR-integrated CDSSs that influence their outcomes and benefits based on published scientific literature.

A TLR was conducted using the Embase, MEDLINE, and Cochrane databases to identify data on CDSSs published in a 10-year frame (2012-2022). Studies on computerized, guideline-based CDSSs used by health care practitioners with a focus on chronic disease areas and reporting outcomes for CDSS utilization were eligible for inclusion.

A total of 49 publications were included in the TLR. Studies predominantly reported on EMR-integrated CDSSs (ie, connected to an EMR database; n=32, 65%). The implementation of CDSSs varied globally, with substantial utilization in the United States and within the domain of cardio-renal-metabolic diseases. CDSSs were found to positively impact "quality assurance" (n=35, 69%) and provide "clinical benefits" (n=20, 41%), compared to usual care. Among CDSS features, treatment guidance and flagging were consistently reported as the most frequent elements for enhancing health care, followed by risk level estimation, diagnosis, education, and data export. The effectiveness of a CDSS was evaluated most frequently in primary care settings (n=34, 69%) across cardio-renal-metabolic disease areas (n=32, 65%), especially in diabetes (n=13, 26%). Studies reported CDSSs to be commonly used by a mixed group (n=27, 55%) of users including physicians, specialists, nurses or nurse practitioners, and allied health care professionals.

Overall, both EMR-integrated and stand-alone CDSSs showed positive results, suggesting their benefits to health care providers and potential for successful adoption. Flagging and treatment recommendation features were commonly used in CDSSs to improve patient care; other features such as risk level estimation, diagnosis, education, and data export were tailored to specific requirements and collectively contributed to the effectiveness of health care delivery. While this TLR demonstrated that both stand-alone and EMR-integrated CDSSs were successful in achieving clinical outcomes, the heterogeneity of included studies reflects the evolving nature of this research area, underscoring the need for further longitudinal studies to elucidate aspects that may impact their adoption in real-world scenarios.

To evaluate acceptability and effectiveness of midwives as trainers for NoviGuide, a neonatal clinical decision support system (CDSS).

A 20-months, mixed-methods open cohort study.

Nurse-midwives at four rural health facilities in eastern Uganda.

We developed a midwife-led trainer programme and instructed two midwives as NoviGuide Trainers in three 3-hour-long sessions. Trainers trained all nurse-midwives at each site in single 3-hour-long sessions. Using the Kirkpatrick model, we evaluated acceptability at level 1 for participant's reaction and level 3 for participant's attitudes towards the programme. We evaluated effectiveness at level 2 for newly learnt skills, and level 3 for participant's uptake of NoviGuide and perception of newborn care practices. We used surveys and focus groups at baseline, 3 months and 6 months and viewed usage data from September 2020 through May 2022.

All 49 participants were female, 23 (46.9%) owned smartphones, 12 (24.5%) accessed the internet daily and 17 (34.7%) were present by study end following staff changes. All participants perceived the use of midwives as NoviGuide Trainers to be an acceptable approach to introduce NoviGuide (mean 5.9 out of 6, SD 0.37). Participants reported gaining new skills and confidence to use NoviGuide; some, in turn, trained others. Participants reported improvement in newborn care. Uptake of NoviGuide was high. Of 49 trained participants, 48 (98%) used NoviGuide. A total of 4045 assessments of newborns were made. Of these, 13.8% (558/4045) were preterm, 17.5% (709/4045) weighed under 2.5 kg and 21.1% (855/4045) had a temperature <36.5°C.

This midwife-led programme was acceptable and led to self-reported improvement in newborn care and high uptake of NoviGuide among nurse-midwives. Task shifting CDSS expert roles to midwives could facilitate large-scale implementation. However, resources like internet coverage, reliable electricity and mobile devices should be considered in low-resource settings.

In prehospital emergency care, providers face significant challenges in making informed decisions due to factors such as limited cognitive support, high-stress environments, and lack of experience with certain patient conditions. Effective Clinical Decision Support Systems (CDSS) have great potential to alleviate these challenges. However, such systems have not yet been widely adopted in real-world practice and have found to cause workflow disruptions and usability issues. Therefore, it is critical to investigate how to design CDSS that meet the needs of prehospital providers while accounting for the unique characteristics of prehospital workflows.

We conducted semi-structured interviews with 20 prehospital providers recruited from four emergency medical services (EMS) agencies in an urban area in the northeastern U.S. The interviews focused on the decision-making challenges faced by prehospital providers, their technological needs for decision support, and key considerations for the design and implementation of a CDSS that can seamlessly integrate into prehospital care workflows. The data were analyzed using content analysis to identify common themes.

Our qualitative study identified several challenges in prehospital decision-making, including limited access to diagnostic tools, insufficient experience with certain critical patient conditions, and a lack of cognitive support. Participants highlighted several desired features to make CDSS more effective in the dynamic, hands-busy, and cognitively demanding prehospital context, such as automatic prompts for possible patient conditions and treatment options, alerts for critical patient safety events, AI-powered medication identification, and easy retrieval of protocols and guidelines using voice commands. Key considerations for successful CDSS adoption included prioritizing alerts to reduce alert fatigue and workflow disruptions, facilitating real-time data collection and documentation to enable decision generation, and ensuring trust and accountability while preventing over-reliance when using CDSS.

This study provides empirical insights into the challenges prehospital providers face and offers design recommendations for developing CDSS solutions that align with prehospital workflows.

More than 150 million people in India need mental health care but few have access to affordable care, especially in rural areas.

To determine whether a multifaceted intervention involving a digital health care model along with a community-based antistigma campaign leads to reduced depression risk and lower mental health-related stigma among adults residing in rural India.

This parallel, cluster randomized, usual care-controlled trial was conducted from September 2020 to December 2021 with blinded follow-up assessments at 3, 6, and 12 months at 44 rural primary health centers across 3 districts in Haryana and Andhra Pradesh states in India. Adults aged 18 years and older at high risk of depression or self-harm defined by either a Patient Health Questionnaire-9 item (PHQ-9) score of 10 or greater, a Generalized Anxiety Disorder-7 item (GAD-7) score of 10 or greater, or a score of 2 or greater on the self-harm/suicide risk question on the PHQ-9. A second cohort of adults not at high risk were selected randomly from the remaining screened population. Data were cleaned and analyzed from April 2022 to February 2023.

The 12-month intervention included a community-based antistigma campaign involving all participants and a digital mental health intervention involving only participants at high risk. Primary health care workers were trained to identify and manage participants at high risk using the Mental Health Gap Action Programme guidelines from the World Health Organization.

The 2 coprimary outcomes assessed at 12 months were mean PHQ-9 scores in the high-risk cohort and mean behavior scores in the combined high-risk and non-high-risk cohorts using the Mental Health Knowledge, Attitude, and Behavior scale.

Altogether, 9928 participants were recruited (3365 at high risk and 6563 not at high risk; 5638 [57%] female and 4290 [43%] male; mean [SD] age, 43 [16] years) with 9057 (91.2%) followed up at 12 months. Mean PHQ-9 scores at 12 months for the high-risk cohort were lower in the intervention vs control groups (2.77 vs 4.48; mean difference, -1.71; 95% CI, -2.53 to -0.89; P < .001). The remission rate in the high-risk cohort (PHQ-9 and GAD-7 scores <5 and no risk of self-harm) was higher in the intervention vs control group (74.7% vs 50.6%; odds ratio [OR], 2.88; 95% CI, 1.53 to 5.42; P = .001). Across both cohorts, there was no difference in 12-month behavior scores in the intervention vs control group (17.39 vs 17.74; mean difference, -0.35; 95% CI, -1.11 to 0.41; P = .36).

A multifaceted intervention was effective in reducing depression risk but did not improve intended help-seeking behaviors for mental illness.

Clinical Trial Registry India: CTRI/2018/08/015355.

There has been a substantial increase in the development of artificial intelligence (AI) tools for clinical decision support. Historically, these were mostly knowledge-based systems, but recent advances include non-knowledge-based systems using some form of machine learning. The ability of health care professionals to trust technology and understand how it benefits patients or improves care delivery is known to be important for their adoption of that technology. For non-knowledge-based AI tools for clinical decision support, these issues are poorly understood.

The aim of this study is to qualitatively synthesize evidence on the experiences of health care professionals in routinely using non-knowledge-based AI tools to support their clinical decision-making.

In June 2023, we searched 4 electronic databases, MEDLINE, Embase, CINAHL, and Web of Science, with no language or date limit. We also contacted relevant experts and searched reference lists of the included studies. We included studies of any design that reported the experiences of health care professionals using non-knowledge-based systems for clinical decision support in their work settings. We completed double independent quality assessment for all included studies using the Mixed Methods Appraisal Tool. We used a theoretically informed thematic approach to synthesize the findings.

After screening 7552 titles and 182 full-text articles, we included 25 studies conducted in 9 different countries. Most of the included studies were qualitative (n=13), and the remaining were quantitative (n=9) and mixed methods (n=3). Overall, we identified 7 themes: health care professionals' understanding of AI applications, level of trust and confidence in AI tools, judging the value added by AI, data availability and limitations of AI, time and competing priorities, concern about governance, and collaboration to facilitate the implementation and use of AI. The most frequently occurring are the first 3 themes. For example, many studies reported that health care professionals were concerned about not understanding the AI outputs or the rationale behind them. There were issues with confidence in the accuracy of the AI applications and their recommendations. Some health care professionals believed that AI provided added value and improved decision-making, and some reported that it only served as a confirmation of their clinical judgment, while others did not find it useful at all.

Our review identified several important issues documented in various studies on health care professionals' use of AI tools in real-world health care settings. Opinions of health care professionals regarding the added value of AI tools for supporting clinical decision-making varied widely, and many professionals had concerns about their understanding of and trust in this technology. The findings of this review emphasize the need for concerted efforts to optimize the integration of AI tools in real-world health care settings.

PROSPERO CRD42022336359; https://tinyurl.com/2yunvkmb.