Objectivizing issues in the diagnosis of complex rare diseases: lessons learned from testing existing diagnosis support systems on ciliopathies.

BACKGROUND: There are approximately 8,000 different rare diseases that affect roughly 400 million people worldwide. Many of them suffer from delayed diagnosis. Ciliopathies are rare monogenic disorders characterized by a significant phenotypic and genetic heterogeneity that raises an important challenge for clinical diagnosis. Diagnosis support systems (DSS) applied to electronic health record (EHR) data may help identify undiagnosed patients, which is of paramount importance to improve patients' care. Our objective was to evaluate three online-accessible rare disease DSSs using phenotypes derived from EHRs for the diagnosis of ciliopathies. METHODS: Two datasets of ciliopathy cases, either proven or suspected, and two datasets of controls were used to evaluate the DSSs. Patient phenotypes were automatically extracted from their EHRs and converted to Human Phenotype Ontology terms. We tested the ability of the DSSs to diagnose cases in contrast to controls based on Orphanet ontology. RESULTS: A total of 79 cases and 38 controls were selected. Performances of the DSSs on ciliopathy real world data (best DSS with area under the ROC curve = 0.72) were not as good as published performances on the test set used in the DSS development phase. None of these systems obtained results which could be described as "expert-level". Patients with multisystemic symptoms were generally easier to diagnose than patients with isolated symptoms. Diseases easily confused with ciliopathy generally affected multiple organs and had overlapping phenotypes. Four challenges need to be considered to improve the performances: to make the DSSs interoperable with EHR systems, to validate the performances in real-life settings, to deal with data quality, and to leverage methods and resources for rare and complex diseases. CONCLUSION: Our study provides insights into the complexities of diagnosing highly heterogenous rare diseases and offers lessons derived from evaluation existing DSSs in real-world settings. These insights are not only beneficial for ciliopathy diagnosis but also hold relevance for the enhancement of DSS for various complex rare disorders, by guiding the development of more clinically relevant rare disease DSSs, that could support early diagnosis and finally make more patients eligible for treatment.

Resident-Driven Clinical Decision Support Governance to Improve the Utility of Clinical Decision Support.

OBJECTIVES: This resident-driven quality improvement project aimed to better understand the known problem of a misaligned clinical decision support (CDS) strategy and improve CDS utilization. METHODS: An internal survey was sent to all internal medicine (IM) residents to identify the most bothersome CDS alerts. Survey results were supported by electronic health record (EHR) data of CDS firing rates and response rates which were collected for each of the three most bothersome CDS tools. Changes to firing criteria were created to increase utilization and to better align with the five rights of CDS. Findings and proposed changes were presented to our institution's CDS Governance Committee. Changes were approved and implemented. Postintervention firing rates were then collected for 1 week. RESULTS: Twenty nine residents participated in the CDS survey and identified sepsis alerts, lipid profile reminders, and telemetry renewals to be the most bothersome alerts. EHR data showed action rates for these CDS as low as 1%. We implemented changes to focus emergency department (ED)-based sepsis alerts to the right provider, better address the right information for lipid profile reminders, and select the right time in workflow for telemetry renewals to be most effective. With these changes we successfully eliminated ED-based sepsis CDS reminders for IM providers, saw a 97% reduction in firing rates for the lipid profile CDS, and noted a 55% reduction in firing rates for telemetry CDS. CONCLUSION: This project highlighted that alert improvements spearheaded by resident teams can be completed successfully using robust CDS governance strategies and can effectively optimize interruptive alerts.

Improving routine use of clinical pathway decision support through integration of an EHR with a clinical library resource designed to provide evidence-based guidance within oncology workflows.

BACKGROUND: The rapid evolution, complexity, and specialization of oncology treatment makes it challenging for physicians to provide care based on the latest and best evidence. We hypothesized that physicians would use evidence-based trusted care pathways if they were easy to use and integrated into clinical workflow at the point of care. METHODS: Within a large integrated care delivery system, we assembled clinical experts to define and update drug treatment pathways, encoded them as flowcharts in an online library integrated with the electronic medical record, communicated expectations that clinicians would use these pathways for every eligible patient, and combined data from multiple sources to understand usage over time. RESULTS: We were able to achieve > 75% utilization of eligible protocols ordered through these pathways within two years, with > 90% of individual oncologists having consulted the pathway at least once, despite no requirements or external incentives associated with pathway usage. Feedback from users contributed to improvements and updates to the guidance. CONCLUSIONS: By making our clinical decision support easily accessible and actionable, we find that we have made considerable progress toward our goal of having physicians consult the latest evidence in their treatment decisions.

Integrating artificial intelligence techniques for advancements in colorectal cancer management: navigating past and predicting future direction.

Artificial Intelligence (AI) in the last few years has emerged as a valuable tool in managing colorectal cancer, revolutionizing its management at different stages. In early detection and diagnosis, AI leverages its prowess in imaging analysis, scrutinizing CT scans, MRI, and colonoscopy views to identify polyps and tumors. This ability enables timely and accurate diagnoses, initiating treatment at earlier stages. AI has helped in personalized treatment planning because of its ability to integrate diverse patient data, including tumor characteristics, medical history, and genetic information. Integrating AI into clinical decision support systems guarantees evidence-based treatment strategy suggestions in multidisciplinary clinical settings, thus improving patient outcomes. This narrative review explores the multifaceted role of AI, spanning early detection of colorectal cancer, personalized treatment planning, polyp detection, lymph node evaluation, cancer staging, robotic colorectal surgery, and training of colorectal surgeons.

Evaluation of an automated clinical decision system with deep learning dose prediction and NTCP model for prostate cancer proton therapy.

Objective.To evaluate the feasibility of using a deep learning dose prediction approach to identify patients who could benefit most from proton therapy based on the normal tissue complication probability (NTCP) model.Approach.Two 3D UNets were established to predict photon and proton doses. A dataset of 95 patients with localized prostate cancer was randomly partitioned into 55, 10, and 30 for training, validation, and testing, respectively. We selected NTCP models for late rectum bleeding and acute urinary urgency of grade 2 or higher to quantify the benefit of proton therapy. Propagated uncertainties of predicted ΔNTCPs resulting from the dose prediction errors were calculated. Patient selection accuracies for a single endpoint and a composite evaluation were assessed under different ΔNTCP thresholds.Main results.Our deep learning-based dose prediction technique can reduce the time spent on plan comparison from approximately 2 days to as little as 5 seconds. The expanded uncertainty of predicted ΔNTCPs for rectum and bladder endpoints propagated from the dose prediction error were 0.0042 and 0.0016, respectively, which is less than one-third of the acceptable tolerance. The averaged selection accuracies for rectum bleeding, urinary urgency, and composite evaluation were 90%, 93.5%, and 93.5%, respectively.Significance.Our study demonstrates that deep learning dose prediction and NTCP evaluation scheme could distinguish the NTCP differences between photon and proton treatment modalities. In addition, the dose prediction uncertainty does not significantly influence the decision accuracy of NTCP-based patient selection for proton therapy. Therefore, automated deep learning dose prediction and NTCP evaluation schemes can potentially be used to screen large patient populations and to avoid unnecessary delays in the start of prostate cancer radiotherapy in the future.

Do Shared Decision-Making and Patient Decision Aids Take Patient's Preferences Seriously?

Most clinical guidelines for the assessment and management of atrial fibrillation emphasize the importance of decision support provided by Patients Decision Aids, but they are to be used and evaluated only in the context of Shared Decision-Making. Detailed examination of 10 clinical decision support tools reveals that many do not engage with patient's preferences at all. Only two take them seriously in terms of their formation, elicitation and processing, aimed at identifying the optimal personalised decision for the patient. This failure is traced to a reluctance to accept the ontological nature of preferences, as instantiations of comparative magnitudes, and to set them in an analytical framework that facilitates their transparent integration with individualised evidence.

Stroke 5.0: A Technology Ecosystem to Support Acute Stroke Integrated Clinical Management.

Stroke remains a significant global health burden, with substantial costs and morbidity associated with its occurrence. To address this challenge, STROKE 5.0 proposes a comprehensive approach to stroke care management, integrating advanced digital technologies and clinical expertise. This paper presents the rationale, design, and potential impact of the STROKE 5.0 platform, which aims to optimize stroke care delivery from pre-hospital assessment through acute hospitalization. The platform facilitates early symptom recognition, efficient emergency response, and streamlined hospital management through intelligent decision support systems. By leveraging predictive analytics and personalized care pathways, STROKE 5.0 seeks to enhance clinical outcomes while providing a platform capable of optimizing the efficiency of service delivery. This innovative model represents a proactive shift towards evidence-based, patient-centered stroke care, with implications for healthcare quality improvement and resource allocation in the digital health domain.

Appropriateness of intended antibiotic prescribing using clinical case vignettes in primary care, and related factors.

BACKGROUND: Factors associated with the appropriateness of antibiotic prescribing in primary care have been poorly explored. In particular, the impact of computerised decision-support systems (CDSS) remains unknown. OBJECTIVES: We aim at investigating the uptake of CDSS and its association with physician characteristics and professional activity. METHODS: Since May 2022, users of a CDSS for antibiotic prescribing in primary care in France have been invited, when registering, to complete three case vignettes assessing clinical situations frequently encountered in general practice and identified as at risk of antibiotic misuse. Appropriateness of antibiotic prescribing was defined as the rate of answers in line with the current guidelines, computed by individuals and by specific questions. Physician's characteristics associated with individual appropriate antibiotic prescribing (< 50%, 50-75% and > 75% appropriateness) were identified by multivariate ordinal logistic regression. RESULTS: In June 2023, 60,067 physicians had registered on the CDSS. Among the 13,851 physicians who answered all case vignettes, the median individual appropriateness level of antibiotic prescribing was 77.8% [Interquartile range, 66.7%-88.9%], and was < 50% for 1,353 physicians (10%). In the multivariate analysis, physicians' characteristics associated with appropriateness were prior use of the CDSS (OR = 1.71, 95% CI 1.56-1.87), being a general practitioner vs. other specialist (OR = 1.34, 95% CI 1.20-1.49), working in primary care (OR = 1.14, 95% CI 1.02-1.27), mentoring students (OR = 1.12, 95% CI 1.04-1.21) age (OR = 0.69 per 10 years increase, 95% CI 0.67-0.71). CONCLUSION: Individual appropriateness for antibiotic prescribing was high among CDSS users, with a higher rate in young general practitioners, previously using the system. CDSS could improve antibiotic prescribing in primary care.

Individual appropriateness for antibiotic prescribing is high among CDSS users.CDSS use could passively improve antibiotic prescribing in primary care.Factors associated with appropriateness for antibiotic prescribing for primary care diseases are: prior use of CDSS, general practice speciality vs. other specialities, younger age and mentoring of students.

User-Centered Framework for Implementation of Technology (UFIT): Development of an Integrated Framework for Designing Clinical Decision Support Tools Packaged With Tailored Implementation Strategies.

BACKGROUND: Electronic health record-based clinical decision support (CDS) tools can facilitate the adoption of evidence into practice. Yet, the impact of CDS beyond single-site implementation is often limited by dissemination and implementation barriers related to site- and user-specific variation in workflows and behaviors. The translation of evidence-based CDS from initial development to implementation in heterogeneous environments requires a framework that assures careful balancing of fidelity to core functional elements with adaptations to ensure compatibility with new contexts. OBJECTIVE: This study aims to develop and apply a framework to guide tailoring and implementing CDS across diverse clinical settings. METHODS: In preparation for a multisite trial implementing CDS for pediatric overweight or obesity in primary care, we developed the User-Centered Framework for Implementation of Technology (UFIT), a framework that integrates principles from user-centered design (UCD), human factors/ergonomics theories, and implementation science to guide both CDS adaptation and tailoring of related implementation strategies. Our transdisciplinary study team conducted semistructured interviews with pediatric primary care clinicians and a diverse group of stakeholders from 3 health systems in the northeastern, midwestern, and southeastern United States to inform and apply the framework for our formative evaluation. RESULTS: We conducted 41 qualitative interviews with primary care clinicians (n=21) and other stakeholders (n=20). Our workflow analysis found 3 primary ways in which clinicians interact with the electronic health record during primary care well-child visits identifying opportunities for decision support. Additionally, we identified differences in practice patterns across contexts necessitating a multiprong design approach to support a variety of workflows, user needs, preferences, and implementation strategies. CONCLUSIONS: UFIT integrates theories and guidance from UCD, human factors/ergonomics, and implementation science to promote fit with local contexts for optimal outcomes. The components of UFIT were used to guide the development of Improving Pediatric Obesity Practice Using Prompts, an integrated package comprising CDS for obesity or overweight treatment with tailored implementation strategies. TRIAL REGISTRATION: ClinicalTrials.gov NCT05627011; https://clinicaltrials.gov/study/NCT05627011.

Perceptions of primary care patients on the use of electronic clinical decision support tools to facilitate health care: A systematic review.

OBJECTIVE: Electronic clinical decision support tools (eCDSTs) are interventions designed to facilitate clinical decision-making using targeted medical knowledge and patient information. While eCDSTs have been demonstrated to improve quality of care, there is a paucity of research relating to the acceptability of eCDSTs in primary care from the patients' perspective. This study aims to summarize current evidence relating to primary care patients' perceptions and experiences on the use of eCDSTs by their clinician to provide care. METHODS: Four databases (Medline, Embase, CINAHL and Cochrane Library) were searched for qualitative and quantitative studies with outcomes relating to patients' perceptions of the use of clinician-facing or shared-eCDSTs. Data extraction and critical appraisal using the Johanna Briggs Institute Critical Appraisal checklists were carried out independently by reviewers. Qualitative and quantitative outcomes were synthesized independently. We used Richardson et al. 'Patient Evaluation of Artificial Intelligence (AI) in Healthcare' framework for qualitative analysis. FINDINGS: 20 papers were included for synthesis. eCDSTs were generally well-regarded by patients. The key facilitators for use were promoting informed decision-making, prompting discussions, aiding clinical decision-making, and enabling information sharing. Key barriers for use were lack of holistic care, 'medicalized' language, and confidentiality concerns. CONCLUSION: Our study identified important aspects to consider in the development of future eCDSTs. Patients were generally positive regarding the use of eCDSTs; however, patient's perspectives should be included from the conception of new eCDSTs to ensure recommendations align with the needs of patients and clinicians. PRACTICE IMPLICATIONS: The study results contribute to ensuring the acceptability of eCDSTs for patients and their unique needs. Encouragement is given for future development to adopt and build upon these findings. Additional research focusing on patients' perceptions of using eCDSTs for specific health conditions is deemed necessary.

Machine Learning for Clinical Decision Support of Acute Streptococcal Pharyngitis: A Pilot Study.

BACKGROUND: Group A Streptococcus (GAS) is the predominant bacterial pathogen of pharyngitis in children. However, distinguishing GAS from viral pharyngitis is sometimes difficult. Unnecessary antibiotic use contributes to unwanted side effects, such as allergic reactions and diarrhea. It also may increase antibiotic resistance. OBJECTIVES: To evaluate the effect of a machine learning algorithm on the clinical evaluation of bacterial pharyngitis in children. METHODS: We assessed 54 children aged 2-17 years who presented to a primary healthcare clinic with a sore throat and fever over 38°C from 1 November 2021 to 30 April 2022. All children were tested with a streptococcal rapid antigen detection test (RADT). If negative, a throat culture was performed. Children with a positive RADT or throat culture were considered GAS-positive and treated antibiotically for 10 days, as per guidelines. Children with negative RADT tests throat cultures were considered positive for viral pharyngitis. The children were allocated into two groups: Group A streptococcal pharyngitis (GAS-P) (n=36) and viral pharyngitis (n=18). All patients underwent a McIsaac score evaluation. A linear support vector machine algorithm was used for classification. RESULTS: The machine learning algorithm resulted in a positive predictive value of 80.6 % (27 of 36) for GAS-P infection. The false discovery rates for GAS-P infection were 19.4 % (7 of 36). CONCLUSIONS: Applying the machine-learning strategy resulted in a high positive predictive value for the detection of streptococcal pharyngitis and can contribute as a medical decision aid in the diagnosis and treatment of GAS-P.

Revolutionizing decision support: a systematic literature review of contextual implementation models for electronic health records systems.

PURPOSE: This study aims to assess previously developed Electronic Health Records System (EHRS) implementation models and identify successful models for decision support. DESIGN/METHODOLOGY/APPROACH: A systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The data sources used were Scopus, PubMed and Google Scholar. The review identified peer-reviewed papers published in the English Language from January 2010 to April 2023, targeting well-defined implementation of EHRS with decision-support capabilities in healthcare. To comprehensively address the research question, we ensured that all potential sources of evidence were considered, and quantitative and qualitative studies reporting primary data and systematic review studies that directly addressed the research question were included in the review. By including these studies in our analysis, we aimed to provide a more thorough and reliable evaluation of the available evidence. FINDINGS: The findings suggest that the success of EHRS implementation is determined by organizational and human factors rather than technical factors alone. Successful implementation is dependent on a suitable implementation framework and management of EHRS. The review identified the capabilities of Clinical Decision Support (CDS) tools as essential in the effectiveness of EHRS in supporting decision-making. ORIGINALITY/VALUE: This study contributes to the existing literature on EHRS implementation models and identifies successful models for decision support. The findings can inform future implementations and guide decision-making in healthcare facilities.

Achieving large-scale clinician adoption of AI-enabled decision support.

Computerised decision support (CDS) tools enabled by artificial intelligence (AI) seek to enhance accuracy and efficiency of clinician decision-making at the point of care. Statistical models developed using machine learning (ML) underpin most current tools. However, despite thousands of models and hundreds of regulator-approved tools internationally, large-scale uptake into routine clinical practice has proved elusive. While underdeveloped system readiness and investment in AI/ML within Australia and perhaps other countries are impediments, clinician ambivalence towards adopting these tools at scale could be a major inhibitor. We propose a set of principles and several strategic enablers for obtaining broad clinician acceptance of AI/ML-enabled CDS tools.

The Effect of Artificial Intelligence on Patient-Physician Trust: Cross-Sectional Vignette Study.

BACKGROUND: Clinical decision support systems (CDSSs) based on routine care data, using artificial intelligence (AI), are increasingly being developed. Previous studies focused largely on the technical aspects of using AI, but the acceptability of these technologies by patients remains unclear. OBJECTIVE: We aimed to investigate whether patient-physician trust is affected when medical decision-making is supported by a CDSS. METHODS: We conducted a vignette study among the patient panel (N=860) of the University Medical Center Utrecht, the Netherlands. Patients were randomly assigned into 4 groups-either the intervention or control groups of the high-risk or low-risk cases. In both the high-risk and low-risk case groups, a physician made a treatment decision with (intervention groups) or without (control groups) the support of a CDSS. Using a questionnaire with a 7-point Likert scale, with 1 indicating "strongly disagree" and 7 indicating "strongly agree," we collected data on patient-physician trust in 3 dimensions: competence, integrity, and benevolence. We assessed differences in patient-physician trust between the control and intervention groups per case using Mann-Whitney U tests and potential effect modification by the participant's sex, age, education level, general trust in health care, and general trust in technology using multivariate analyses of (co)variance. RESULTS: In total, 398 patients participated. In the high-risk case, median perceived competence and integrity were lower in the intervention group compared to the control group but not statistically significant (5.8 vs 5.6; P=.16 and 6.3 vs 6.0; P=.06, respectively). However, the effect of a CDSS application on the perceived competence of the physician depended on the participant's sex (P=.03). Although no between-group differences were found in men, in women, the perception of the physician's competence and integrity was significantly lower in the intervention compared to the control group (P=.009 and P=.01, respectively). In the low-risk case, no differences in trust between the groups were found. However, increased trust in technology positively influenced the perceived benevolence and integrity in the low-risk case (P=.009 and P=.04, respectively). CONCLUSIONS: We found that, in general, patient-physician trust was high. However, our findings indicate a potentially negative effect of AI applications on the patient-physician relationship, especially among women and in high-risk situations. Trust in technology, in general, might increase the likelihood of embracing the use of CDSSs by treating professionals.

A computational clinical decision-supporting system to suggest effective anti-epileptic drugs for pediatric epilepsy patients based on deep learning models using patient's medical history.

BACKGROUND: Epilepsy, a chronic brain disorder characterized by abnormal brain activity that causes seizures and other symptoms, is typically treated using anti-epileptic drugs (AEDs) as the first-line therapy. However, due to the variations in their modes of action, identification of effective AEDs often relies on ad hoc trials, which is particularly challenging for pediatric patients. Thus, there is significant value in computational methods capable of assisting in the selection of AEDs, aiming to minimize unnecessary medication and improve treatment efficacy. RESULTS: In this study, we collected 7,507 medical records from 1,000 pediatric epilepsy patients and developed a computational clinical decision-supporting system for AED selection. This system leverages three multi-channel convolutional neural network (CNN) models tailored to three specific AEDs (vigabatrin, prednisolone, and clobazam). Each CNN model predicts whether a respective AED is effective on a given patient or not. The CNN models showed AUROCs of 0.90, 0.80, and 0.92 in 10-fold cross-validation, respectively. Evaluation on a hold-out test dataset further revealed positive predictive values (PPVs) of 0.92, 0.97, and 0.91 for the three respective CNN models, representing that suggested AEDs by our models would be effective in controlling epilepsy with a high accuracy and thereby reducing unnecessary medications for pediatric patients. CONCLUSION: Our CNN models in the system demonstrated high PPVs for the three AEDs, which signifies the potential of our approach to support the clinical decision-making by assisting doctors in recommending effective AEDs within the three AEDs for patients based on their medical history. This would result in a reduction in the number of unnecessary ad hoc attempts to find an effective AED for pediatric epilepsy patients.

Decision support systems for antibiotic prescription in hospitals: a survey with hospital managers on factors for implementation.

BACKGROUND: Inappropriate antimicrobial use, such as antibiotic intake in viral infections, incorrect dosing and incorrect dosing cycles, has been shown to be an important determinant of the emergence of antimicrobial resistance. Artificial intelligence-based decision support systems represent a potential solution for improving antimicrobial prescribing and containing antimicrobial resistance by supporting clinical decision-making thus optimizing antibiotic use and improving patient outcomes. OBJECTIVE: The aim of this research was to examine implementation factors of artificial intelligence-based decision support systems for antibiotic prescription in hospitals from the perspective of the hospital managers, who have decision-making authority for the organization. METHODS: An online survey was conducted between December 2022 and May 2023 with managers of German hospitals on factors for decision support system implementation. Survey responses were analyzed from 118 respondents through descriptive statistics. RESULTS: Survey participants reported openness towards the use of artificial intelligence-based decision support systems for antibiotic prescription in hospitals but little self-perceived knowledge in this field. Artificial intelligence-based decision support systems appear to be a promising opportunity to improve quality of care and increase treatment safety. Along with the Human-Organization-Technology-fit model attitudes were presented. In particular, user-friendliness of the system and compatibility with existing technical structures are considered to be important for implementation. The uptake of decision support systems also depends on the ability of an organization to create a facilitating environment that helps to address the lack of user knowledge as well as trust in and skepticism towards these systems. This includes the training of user groups and support of the management level. Besides, it has been assessed to be important that potential users are open towards change and perceive an added value of the use of artificial intelligence-based decision support systems. CONCLUSION: The survey has revealed the perspective of hospital managers on different factors that may help to address implementation challenges for artificial intelligence-based decision support systems in antibiotic prescribing. By combining factors of user perceptions about the systems´ perceived benefits with external factors of system design requirements and contextual conditions, the findings highlight the need for a holistic implementation framework of artificial intelligence-based decision support systems.

Decision discovery using clinical decision support system decision log data for supporting the nurse decision-making process.

BACKGROUND: Decision-making in healthcare is increasingly complex; notably in hospital environments where the information density is high, e.g., emergency departments, oncology departments, and psychiatry departments. This study aims to discover decisions from logged data to improve the decision-making process. METHODS: The Design Science Research Methodology (DSRM) was chosen to design an artifact (algorithm) for the discovery and visualization of decisions. The DSRM's different activities are explained, from the definition of the problem to the evaluation of the artifact. During the design and development activities, the algorithm itself is created. During the demonstration and evaluation activities, the algorithm was tested with an authentic synthetic dataset. RESULTS: The results show the design and simulation of an algorithm for the discovery and visualization of decisions. A fuzzy classifier algorithm was adapted for (1) discovering decisions from a decision log and (2) visualizing the decisions using the Decision Model and Notation standard. CONCLUSIONS: In this paper, we show that decisions can be discovered from a decision log and visualized for the improvement of the decision-making process of healthcare professionals or to support the periodic evaluation of protocols and guidelines.

Contextual design requirements for decision-support tools involved in weaning patients from mechanical ventilation in intensive care units.

Weaning patients from ventilation in intensive care units (ICU) is a complex task. There is a growing desire to build decision-support tools to help clinicians during this process, especially those employing Artificial Intelligence (AI). However, tools built for this purpose should fit within and ideally improve the current work environment, to ensure they can successfully integrate into clinical practice. To do so, it is important to identify areas where decision-support tools may aid clinicians, and associated design requirements for such tools. This study analysed the work context surrounding the weaning process from mechanical ventilation in ICU environments, via cognitive task and work domain analyses. In doing so, both what cognitive processes clinicians perform during weaning, and the constraints and affordances of the work environment itself, were described. This study found a number of weaning process tasks where decision-support tools may prove beneficial, and from these a set of contextual design requirements were created. This work benefits researchers interested in creating human-centred decision-support tools for mechanical ventilation that are sensitive to the wider work system.