Clinical Decision Support Tool Research Articles

Deprescribing of antidepressants: development of indicators of high-risk and overprescribing using the RAND/UCLA Appropriateness Method

BackgroundAntidepressants are first-line medications for many psychiatric disorders. However, their widespread long-term use in some indications (e.g., mild depression and insomnia) is concerning. Particularly in older adults with comorbidities and polypharmacy, who are more susceptible to adverse drug reactions, the risks and benefits of treatment should be regularly reviewed. The aim of this consensus process was to identify explicit criteria of potentially inappropriate antidepressant use (indicators) in order to support primary care clinicians in identifying situations, where deprescribing of antidepressants should be considered.MethodsWe used the RAND/UCLA Appropriateness Method to identify the indicators of high-risk and overprescribing of antidepressants. We combined a structured literature review with a 3-round expert panel, with results discussed in moderated meetings in between rounds. Each of the 282 candidate indicators was scored on a 9-point Likert scale representing the necessity of a critical review of antidepressant continuation (1–3 = not necessary; 4–6 = uncertain; 7–9 = clearly necessary). Experts rated the indicators for the necessity of review, since decisions to deprescribe require considerations of patient risk/benefit balance and preferences. Indicators with a median necessity rating of ≥ 7 without disagreement after 3 rating rounds were accepted.ResultsThe expert panel comprised 2 general practitioners, 2 clinical pharmacologists, 1 gerontopsychiatrist, 2 psychiatrists, and 3 internists/geriatricians (total N = 10). After 3 assessment rounds, there was consensus for 37 indicators of high-risk and 25 indicators of overprescribing, where critical reviews were felt to be necessary. High-risk prescribing indicators included settings posing risks of drug-drug, drug-disease, and drug-age interactions or the occurrence of adverse drug reactions. Indicators with the highest ratings included those suggesting the possibility of cardiovascular risks (QTc prolongation), delirium, gastrointestinal bleeding, and liver injury in specific patient subgroups with additional risk factors. Overprescribing indicators target patients with long treatment durations for depression, anxiety, and insomnia as well as high doses for pain and insomnia.ConclusionsExplicit indicators of antidepressant high-risk and overprescribing may be used directly by patients and health care providers, and integrated within clinical decision support tools, in order to improve the overall risk/benefit balance of this commonly prescribed class of prescription drugs.

Understanding provider use of a new clinical decision support tool aimed at reducing excess telemetry in an academic health system: A retrospective study.

Cardiac monitoring has often been identified as an area of overutilization and remains a limited resource in many hospitals. With the aim of reducing telemetry overuse, we added clinical decision support to our health system's telemetry order with guidance on appropriate indications for monitoring. The new order requires selection of an appropriate clinical indication. In this study, we aimed to understand provider engagement with this tool by assessing concordance between selected indications within the order and the clinical presence of those conditions as documented within the patient chart. We randomly selected 100 telemetry orders from July to October 2022 across four different hospitals at NYU Langone Health. Two independent, blinded reviewers used a structured protocol to identify documentation of actual indications for telemetry in each selected chart. We calculated the rate of concordance between indications selected in the order and indications that were determined to be clinically present on chart review. There were 30,839 telemetry orders placed during the study timeframe. Overall concordance between the selection within the order and the actual indication was 48% (95% confidence interval [CI], 38.21%-57.79%). We observed especially low concordance rates for vague indications, such as 'Other', and for 'Confirmed Stroke', which was the only indication allowing for indefinite telemetry. The overall low concordance suggests a disconnect between the support tool and clinical practice. Providers are more likely to select an indication that reduces downstream work regardless of a patient's true clinical indication.

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

This study aims to assess previously developed Electronic Health Records System (EHRS) implementation models and identify successful models for decision support. 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. 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. 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.

Abstract PO2-02-10: Predicting the feasibility of breast conserving surgery using pre-treatment standard of care DCE-MRI: a novel clinical decision support tool for breast cancer surgical planning

Abstract Background: The decision-making process regarding breast cancer treatment options, specifically breast conserving surgery (BCS) versus mastectomy, is a complex and sensitive matter that takes into account the patient's preferences and their eligibility for BCS. A range of anatomical factors, such as tumor size, breast size, and proximity of the tumor to the skin, can influence the surgeon's recommendation regarding BCS or mastectomy. While mastectomy may be the only viable option for patients with large tumors, those with smaller tumors often have the opportunity to collaboratively decide with their surgeon on the most appropriate approach to ensure oncologic success and minimize potential cosmetic deformity. To facilitate this decision-making process, we developed a tool capable of providing informed recommendations on BCS or mastectomy based on standard of care imaging patient data. By integrating these recommendation capabilities, we aim to enhance the decision-making process and improve patient outcomes in breast cancer treatment. Methods: To assist surgeons and patients in determining a patient’s eligibility for breast conserving surgery (BCS), we developed the BCS Feasibility Score. The BCS Feasibility Score is generated from a binary classifier machine learning model and is representative of the probability that a given patient could receive BCS. The score is based solely on the output from TumorSight Viz, a software platform that uses only pre-treatment T1-weighted DCE MRI and applies a suite of deep learning algorithms to segment the patient’s disease and the surrounding breast tissues. From the multi-tissue segmentation generated using TumorSight Viz, we extracted 25 spatial morphology features to use in the model development process. The training and testing sets were a subsample of a publicly available dataset (Saha, et al, 2018). We selected patients that received either BCS or mastectomy and had a viable T1-weighted DCE MRI. Bi-lateral cases were excluded. The total training/testing set consisted of n = 749 cases, 363 of which received mastectomy and 386 who received BCS. Before model training occurred, we split 20% of the sample to use as the test set. We then trained a random forest classifier across a range of pre-set hyper-parameters, using a total of 25 features as inputs, and using 5x cross-validation in the training set. We then assessed model performance in the testing set only. Results: In the test set, 76 cases received BCS versus 74 that received mastectomy. We successfully predicted BCS in 56 out of 76 cases (74%), and mastectomy in 46 out of 74 cases (62%). Overall, we observed an AUC = 0.76 and an F1 score = 0.66, indicating moderate to strong model performance. The most important features in the model, as measured by SHAP values, included the axis aligned tumor longest dimension (mm), the closest distance between the tumor and the nipple (mm), the volume of the tumor convex hull (mL; dilated uni-centrically), and the ratio of tumor convex hull volume to breast volume. We tested the model’s performance in a fully independent holdout validation set of 579 cases. Within this validation set, 335 patients received BCS versus 244 received mastectomy. We successfully predicted BCS in 254 out of 335 cases (76%), and mastectomy in 149 out of 244 cases (61%). Overall, we observed an AUC = 0.75 and an F1 score = 0.63. Validation set performance mirrored results observed in the test set, indicating strong generalizability. In the validation set, comparing patients who underwent BCS to those who underwent mastectomy, we observed significant differences across all four of the most predictive features. Conclusion: TumorSight Viz and the BCS Feasibility Score help empower both patients and clinicians with information and tools to facilitate surgical planning decisions. Citation Format: John Pfeiffer, Bradley Feiger, Anuja Antony, Joseph Peterson. Predicting the feasibility of breast conserving surgery using pre-treatment standard of care DCE-MRI: a novel clinical decision support tool for breast cancer surgical planning [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-02-10.

Improving Antibiotic Overuse in Primary Care: A Multimodal Quality Improvement Project.

Antibiotic overuse has increased over time related to provider knowledge gaps about best practices, provider perception of patient expectations on receiving an antibiotic, possible pressure to see patients in a timely fashion, and concerns about decreased patient satisfaction when an antibiotic is not prescribed. The Centers for Disease Control and Prevention estimates that up to 30% of antibiotics are inappropriately prescribed in the outpatient setting. This quality improvement project consisted of a multimodal approach to decrease inappropriate antibiotic prescribing for viral upper respiratory infections (URIs) by using provider education, passive patient education, and clinical decision support tools based on Centers for Disease Control and Prevention recommendations and the Be Antibiotic Aware tool. Following implementation, there was an 11% decrease in viral URI antibiotic prescribing, from a rate of 29.33% to 18.33% following the multimodal implementation. The use of evidence-based education and treatment guidelines was found to decrease inappropriate antibiotic prescribing for patients diagnosed with viral URIs.

B-34 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Contrast Induced Acute Kidney Injury (CI-AKI) in Patients Undergoing Percutaneous Coronary Intervention (PCI)

B-34 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Contrast Induced Acute Kidney Injury (CI-AKI) in Patients Undergoing Percutaneous Coronary Intervention (PCI)

OR12-6 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Post-Procedure Bleeding Events in Patients Undergoing Percutaneous Coronary Intervention (PCI)

OR12-6 | Effects of Artificial Intelligence Clinical Decision Support Tools (Terumo Health Outcomes, ePRISM) on Post-Procedure Bleeding Events in Patients Undergoing Percutaneous Coronary Intervention (PCI)

Evaluation of a Clinical Decision Support Tool to Guide Adoption of the American Heart Association Telemetry Monitoring Practice Standards.

The objectives of this study were to (1) evaluate telemetry use pre- and postimplementation of clinical decision support tools to support American Heart Association practice standards for telemetry monitoring and (2) understand the factors that may contribute to variation of telemetry monitoring in practice. First, we captured overall variability in telemetry use pre- and postimplementation of the clinical decision support intervention. We then conducted semistructured interviews with telemetry-ordering providers to identify key barriers and facilitators to adoption. During the study period, 399 physicians met criteria for inclusion and were divided into excessive and nonexcessive orderers. Distribution of telemetry use was bimodal. Among nonexcessive users, 24.4% of patient days were with telemetry compared with 51.6% among excessive users. On average, both excessive (6.1% reduction) and nonexcessive users (2.8% reduction) decreased telemetry use postimplementation, and these reductions were sustained over a 16-month period. Sixteen interviews were conducted. Physicians believed that the tool was successful because it caused them to more closely consider if telemetry was indicated for each patient. Physicians also voiced frustration with interruptions to their workflow, and some noted that they commonly use telemetry outside of practice standards to monitor patients who were acutely but not critically ill. Embedding telemetry practice standards into the electronic health record in the form of clinical decision support is effective at reducing excess telemetry use. Although the intervention was well received, there are persistent barriers, such as preexisting views on telemetry and existing workflow habits, that may inhibit higher adoption of standards.

Infection Control Through Clinical Pipelines Built with Arden Syntax MLM Building Blocks.

Healthcare-associated infections (HAIs) may have grave consequences for patients. In the case of sepsis, the 30-day mortality rate is about 25%. HAIs cost EU member states an estimated 7 billion Euros annually. Clinical decision support tools may be useful for infection monitoring, early warning, and alerts. MONI, a tool for monitoring nosocomial infections, is used at University Hospital Vienna, but needs to be clinically and technically revised and updated. A new, completely configurable pipeline-based system for defining and processing HAI definitions was developed and validated. A network of data access points, clinical rules, and explanatory output is arranged as an inference network, a clinical pipeline as it is called, and processed in a stepwise manner. Arden-Syntax-based medical logic modules were used to implement the respective rules. The system was validated by creating a pipeline for the ECDC PN5 pneumonia rule. It was tested on a set of patient data from intensive care medicine. The results were compared with previously obtained MONI output as a suitable reference, yielding a sensitivity of 93.8% and a specificity of 99.8%. Clinical pipelines show promise as an open and configurable approach to graphically-based, human-readable, machine-executable HAI definitions.

Academic Detailing as a Health Information Technology Implementation Method: Supporting the Design and Implementation of an Emergency Department-Based Clinical Decision Support Tool to Prevent Future Falls.

Clinical decision support (CDS) tools that incorporate machine learning-derived content have the potential to transform clinical care by augmenting clinicians' expertise. To realize this potential, such tools must be designed to fit the dynamic work systems of the clinicians who use them. We propose the use of academic detailing-personal visits to clinicians by an expert in a specific health IT tool-as a method for both ensuring the correct understanding of that tool and its evidence base and identifying factors influencing the tool's implementation. This study aimed to assess academic detailing as a method for simultaneously ensuring the correct understanding of an emergency department-based CDS tool to prevent future falls and identifying factors impacting clinicians' use of the tool through an analysis of the resultant qualitative data. Previously, our team designed a CDS tool to identify patients aged 65 years and older who are at the highest risk of future falls and prompt an interruptive alert to clinicians, suggesting the patient be referred to a mobility and falls clinic for an evidence-based preventative intervention. We conducted 10-minute academic detailing interviews (n=16) with resident emergency medicine physicians and advanced practice providers who had encountered our CDS tool in practice. We conducted an inductive, team-based content analysis to identify factors that influenced clinicians' use of the CDS tool. The following categories of factors that impacted clinicians' use of the CDS were identified: (1) aspects of the CDS tool's design (2) clinicians' understanding (or misunderstanding) of the CDS or referral process, (3) the busy nature of the emergency department environment, (4) clinicians' perceptions of the patient and their associated fall risk, and (5) the opacity of the referral process. Additionally, clinician education was done to address any misconceptions about the CDS tool or referral process, for example, demonstrating how simple it is to place a referral via the CDS and clarifying which clinic the referral goes to. Our study demonstrates the use of academic detailing for supporting the implementation of health information technologies, allowing us to identify factors that impacted clinicians' use of the CDS while concurrently educating clinicians to ensure the correct understanding of the CDS tool and intervention. Thus, academic detailing can inform both real-time adjustments of a tool's implementation, for example, refinement of the language used to introduce the tool, and larger scale redesign of the CDS tool to better fit the dynamic work environment of clinicians.

Prediction of heart failure and all-cause mortality using cardiac ultrasomics in patients with breast cancer.

Breast cancer chemotherapy/immunotherapy can be associated with treatment-limiting cardiotoxicity. Radiomics techniques applied to ultrasound, known as ultrasomics, can be used in cardio-oncology to leverage echocardiography for added prognostic value. To utilize ultrasomics features collected prior to antineoplastic therapy to enhance prediction of mortality and heart failure (HF) in patients with breast cancer. Patients were retrospectively recruited in a study at the West Virginia University Cancer Institute. The final inclusion criteria were met by a total of 134 patients identified for the study. Patients were imaged using echocardiography in the parasternal long axis prior to receiving chemotherapy. All-cause mortality and HF, developed during treatment, were the primary outcomes. 269 features were assessed, grouped into four major classes: demographics (n = 21), heart function (n = 7), antineoplastic medication (n = 17), and ultrasomics (n = 224). Data was split into an internal training (60%, n = 81) and testing (40%, n = 53) set. Ultrasomics features augmented classification of mortality (area under the curve (AUC) 0.89 vs. 0.65, P = 0.003), when compared to demographic variables. When developing a risk prediction score for each feature category, ultrasomics features were significantly associated with both mortality (P = 0.031, log-rank test) and HF (P = 0.002, log-rank test). Further, only ultrasomics features provided significant improvement over demographic variables when predicting mortality (C-Index: 0.78 vs. 0.65, P = 0.044) and HF (C-Index: 0.77 vs. 0.60, P = 0.017), respectively. With further investigation, a clinical decision support tool could be developed utilizing routinely obtained patient data alongside ultrasomics variables to augment treatment regimens.

Application of clinical decision support tools for predicting outcomes with vedolizumab therapy in patients with inflammatory bowel disease: A KASID multicentre study.

We aimed to validate clinical decision support tools (CDSTs) to predict real-life effectiveness of vedolizumab (VDZ) in patients with inflammatory bowel disease. We retrospectively enrolled patients with Crohn's disease (CD) or ulcerative colitis (UC) treated with VDZ at 10 tertiary referral centres in Korea between January 2017 and November 2021. We assessed clinical remission (CREM) and response (CRES), corticosteroid-free clinical remission (CSF-CREM) and response (CSF-CRES), biochemical response based on C-reactive protein (BioRES[CRP]) and faecal calprotectin (BioRES[FC]), endoscopic healing (EH), and the need to optimise or switch drugs based on CDST-defined response groups. Additionally, the area under the receiver operating characteristics curve (AUC) for the CDSTs was calculated. We included 143 patients with CD and 219 with UC. We observed incremental trends on CSF-CRES at week 14 (W14) (ptrend = 0.004) and decreasing trends for the need to optimise or switch drugs (ptrend = 0.016) in CD from the low to high probability groups. Except for CSF-CREM at W54, we noticed incremental trends for all clinical responses at W14, W26 and W54 (ptrend <0.001) in UC. W26 and W54 BioRES[CRP] and W14 EH also showed increasing trends (ptrend <0.05) in UC. With increasing probabilities of response, drug optimisation or switching was less frequently required in UC (ptrend = 0.013). With 26 points cut-off, CDSTs effectively identified W14 CSF-CRES, W26 BioRES[CRP], BioRES[FC] and W54 BioRES[CRP] in UC, all with AUCs >0.600, whereas CDSTs showed poor accuracy in CD. CDSTs for VDZ had acceptable accuracy in predicting effectiveness outcomes including clinical and biochemical outcomes in UC. However, their utility in CD was limited.

Development and validation of machine learning-based models for predicting healthcare-associated bacterial/fungal infections among COVID-19 inpatients: a retrospective cohort study

BackgroundCOVID-19 and bacterial/fungal coinfections have posed significant challenges to human health. However, there is a lack of good tools for predicting coinfection risk to aid clinical work.ObjectiveWe aimed to investigate the risk factors for bacterial/fungal coinfection among COVID-19 patients and to develop machine learning models to estimate the risk of coinfection.MethodsIn this retrospective cohort study, we enrolled adult inpatients confirmed with COVID-19 in a tertiary hospital between January 1 and July 31, 2023, in China and collected baseline information at admission. All the data were randomly divided into a training set and a testing set at a ratio of 7:3. We developed the generalized linear and random forest models for coinfections in the training set and assessed the performance of the models in the testing set. Decision curve analysis was performed to evaluate the clinical applicability.ResultsA total of 1244 patients were included in the training cohort with 62 healthcare-associated bacterial/fungal infections, while 534 were included in the testing cohort with 22 infections. We found that patients with comorbidities (diabetes, neurological disease) were at greater risk for coinfections than were those without comorbidities (OR = 2.78, 95%CI = 1.61–4.86; OR = 1.93, 95%CI = 1.11–3.35). An indwelling central venous catheter or urinary catheter was also associated with an increased risk (OR = 2.53, 95%CI = 1.39–4.64; OR = 2.28, 95%CI = 1.24–4.27) of coinfections. Patients with PCT > 0.5 ng/ml were 2.03 times (95%CI = 1.41–3.82) more likely to be infected. Interestingly, the risk of coinfection was also greater in patients with an IL-6 concentration < 10 pg/ml (OR = 1.69, 95%CI = 0.97–2.94). Patients with low baseline creatinine levels had a decreased risk of bacterial/fungal coinfections(OR = 0.40, 95%CI = 0.22–0.71). The generalized linear and random forest models demonstrated favorable receiver operating characteristic curves (ROC = 0.87, 95%CI = 0.80–0.94; ROC = 0.88, 95%CI = 0.82–0.93) with high accuracy, sensitivity and specificity of 0.86vs0.75, 0.82vs0.86, 0.87vs0.74, respectively. The corresponding calibration evaluation P statistics were 0.883 and 0.769.ConclusionsOur machine learning models achieved strong predictive ability and may be effective clinical decision-support tools for identifying COVID-19 patients at risk for bacterial/fungal coinfection and guiding antibiotic administration. The levels of cytokines, such as IL-6, may affect the status of bacterial/fungal coinfection.

System-Wide Thromboprophylaxis Interventions for Hospitalized Patients at Risk of Venous Thromboembolism: Focus on Cross-Platform Clinical Decision Support.

Thromboprophylaxis of hospitalized patients at risk of venous thromboembolism (VTE) presents challenges owing to patient heterogeneity and lack of adoption of evidence-based methods. Intuitive practices for thromboprophylaxis have resulted in many patients being inappropriately prophylaxed. We conducted a narrative review summarizing system-wide thromboprophylaxis interventions in hospitalized patients. Multiple interventions for thromboprophylaxis have been tested, including multifaceted approaches such as national VTE prevention programs with audits, pre-printed order entry, passive alerts (either human or electronic), and more recently, the use of active clinical decision support (CDS) tools incorporated into electronic health records (EHRs). Multifaceted health-system and order entry interventions have shown mixed results in their ability to increase appropriate thromboprophylaxis and reduce VTE unless mandated through a national VTE prevention program, though the latter approach is potentially costly and effort- and time-dependent. Studies utilizing passive human or electronic alerts have also shown mixed results in increasing appropriate thromboprophylaxis and reducing VTE. Recently, a universal cloud-based and EHR-agnostic CDS VTE tool incorporating a validated VTE risk score revealed high adoption and effectiveness in increasing appropriate thromboprophylaxis and reducing major thromboembolism. Active CDS tools hold promise in improving appropriate thromboprophylaxis, especially with further refinement and widespread implementation within various EHRs and clinical workflows.

Assessment of drug utilization and potentially inappropriate medications in hemodialysis patients with end-stage renal dysfunction: A study in a tertiary care hospital in Bahrain.

Patients undergoing dialysis pose therapeutic challenges in terms of polypharmacy, administration of potentially inappropriate drugs, and drugs with the potential risk of toxicity. This study evaluated the use of drugs, potentially inappropriate medicines (PIM), drugs with risk of Torsades de Pointes (TdP), and the complexity of the prescribed regimen using the medication regimen complexity index scale in patients undergoing hemodialysis. A retrospective cohort study was carried out amongst patients receiving hemodialysis. Drugs were classified into one of four classes: (i) drugs used in managing renal complications, (ii) cardiovascular drugs, (iii) anti-diabetic drugs, (iv) drugs for symptomatic management, and (v) others. Drugs were considered as PIM according to the Can-SOLVE CKD working group from a network of Canadian nephrology health professionals. The study adhered to the CredibleMeds classification of drugs with known, possible, and conditional risk of TdP and the complexity of prescribed medicines was evaluated based on the pre-validated medication regimen complexity index scale based on form/route, frequency of dosing, and requirement of special instructions. Sixty-three participants were included in the study (49 males and 14 females) with the median (range) age of 45 (21-66) years. Cardiovascular drugs followed by drugs used for managing renal complications were the most common classes administered. Notably, 12 (19.1%) patients received one of the non-steroidal anti-inflammatory drugs, 21 (33.3%) received a proton pump inhibitor, three (4.8%) received pregabalin, two (3.2%) received opioid drugs, and one (1.6%) was administered celecoxib. Atorvastatin, furosemide, omeprazole, and allopurinol were the most common PIM drugs administered to the study participants followed by others. Drugs used for symptomatic management had significantly more PIM compared to other classes (p <0.0001). Six (9.5%) patients received drugs with known TdP risk, one with possible TdP risk, and 61 with conditional risk. Median (range) medical regimen complexity index score was 26.5 (2-62.5). A huge burden of drug therapy was observed in the hemodialysis patients in terms of higher proportions of PIM, complex medical regimen, and prescription of drugs with risk of TdP. Implementation of clinical decision support tools enhancing rational prescription and identification of drugs with TdP risk, introducing antimicrobial stewardship, and stepwise deprescription of the drugs with the least benefit-risk ratio are warranted.

Understanding Physician's Perspectives on AI in Health Care: Protocol for a Sequential Multiple Assignment Randomized Vignette Study.

As the availability and performance of artificial intelligence (AI)-based clinical decision support (CDS) systems improve, physicians and other care providers poised to be on the front lines will be increasingly tasked with using these tools in patient care and incorporating their outputs into clinical decision-making processes. Vignette studies provide a means to explore emerging hypotheses regarding how context-specific factors, such as clinical risk, the amount of information provided about the AI, and the AI result, may impact physician acceptance and use of AI-based CDS tools. To best anticipate how such factors influence the decision-making of frontline physicians in clinical scenarios involving AI decision-support tools, hypothesis-driven research is needed that enables scenario testing before the implementation and deployment of these tools. This study's objectives are to (1) design an original, web-based vignette-based survey that features hypothetical scenarios based on emerging or real-world applications of AI-based CDS systems that will vary systematically by features related to clinical risk, the amount of information provided about the AI, and the AI result; and (2) test and determine causal effects of specific factors on the judgments and perceptions salient to physicians' clinical decision-making. US-based physicians with specialties in family or internal medicine will be recruited through email and mail (target n=420). Through a web-based survey, participants will be randomized to a 3-part "sequential multiple assignment randomization trial (SMART) vignette" detailing a hypothetical clinical scenario involving an AI decision support tool. The SMART vignette design is similar to the SMART design but adapted to a survey design. Each respondent will be randomly assigned to 1 of the possible vignette variations of the factors we are testing at each stage, which include the level of clinical risk, the amount of information provided about the AI, and the certainty of the AI output. Respondents will be given questions regarding their hypothetical decision-making in response to the hypothetical scenarios. The study is currently in progress and data collection is anticipated to be completed in 2024. The web-based vignette study will provide information on how contextual factors such as clinical risk, the amount of information provided about an AI tool, and the AI result influence physicians' reactions to hypothetical scenarios that are based on emerging applications of AI in frontline health care settings. Our newly proposed "SMART vignette" design offers several benefits not afforded by the extensively used traditional vignette design, due to the 2 aforementioned features. These advantages are (1) increased validity of analyses targeted at understanding the impact of a factor on the decision outcome, given previous outcomes and other contextual factors; and (2) balanced sample sizes across groups. This study will generate a better understanding of physician decision-making within this context. DERR1-10.2196/54787.

Implementation and Evaluation of Educational Workshops to Increase Primary Care Provider and Medical Student Comfort in Psychiatric Care.

Mental health treatment is often initiated in primary care settings, but many primary care providers (PCPs), residents, and medical students report discomfort in managing psychiatric conditions. This study evaluated the effect of an educational workshop that featured an evidence-based psychopharmacology clinical decision support tool (CDST) on trainee confidence and willingness to treat psychiatric conditions. Participants completed pre- and post-workshop surveys. Nine months after the workshop, a subset of trainees participated in a focus group. Of the participants, 62.5% of the obstetrics-gynecology (OB-GYN) resident physicians (10/16) and 100% of the medical students (18/18) completed both pre- and post-surveys. Following the workshop, OB-GYN resident physicians reported significantly improved confidence in treating psychiatric disorders (p < 0.001), sense of having psychiatric support tools (p < 0.001), and knowledge of treating psychiatric disorders (p = 0.021). Medical students reported significantly improved confidence in treating psychiatric disorders (p < 0.001), willingness to devise treatment plans for psychiatric disorders (p = 0.024), sense of having psychiatric support tools (p < 0.001), knowledge of treating psychiatric disorders (p < 0.001), and comfort in presenting a psychiatric treatment plan to an attending (p = 0.003). Most focus group participants (93.75%; 15/16) reported that they continued to use the CDST, and it increased their confidence in formulating psychiatric treatment plans. These findings suggest that educational workshops that introduce high-quality psychopharmacology CDSTs may be an effective method for improving provider comfort in treating psychiatric disorders.

The Effect of an Oxytocin Decision Support Checklist on Oxytocin Use and Maternal and Neonatal Outcomes: A Retrospective Cohort Study.

To assess the association between use of an oxytocin decision support checklist with oxytocin usage and clinical outcomes. We conducted a retrospective cohort study of patients with singleton gestations at 370/7 weeks or greater who received oxytocin during labor from October 2012 to February 2017 at an integrated community health care system during three exposure periods: (1) pre-checklist; (2) after paper checklist implementation; and (3) after checklist integration into the electronic medical record (EMR). The checklist was a clinical decision support tool to standardize the dosing and management of oxytocin. Thus, our primary outcomes included oxytocin infusion rates and cumulative dose. Secondary outcomes included maternal and neonatal outcomes. We controlled for maternal risk factors with multivariable regression analysis and stratified by mode of delivery. A total of 34,269 deliveries were included. Unadjusted analyses showed that compared with pre-checklist, deliveries during the paper and EMR-integrated periods had a lower cumulative dose (4,670 ± 6,174 vs. 4,318 ± 5,719 and 4,286 ± 5,579 mU, p < 0.001 for both), lower maximal infusion rate (9.9 ± 6.8 vs. 8.7 ± 5.8 and 8.4 ± 5.6 mU/min, p < 0.001 for both), and longer duration of oxytocin use (576 ± 442 vs. 609 ± 476 and 627 ± 488 minutes, p < 0.001 and p = 0.01, respectively). The unadjusted rates of cesarean, 5-minute Apgar <7, mechanical ventilation, and neonatal hospital length of stay were similar between periods. The adjusted mean difference in time from admission to delivery was longer during the EMR-integrated period compared with pre-checklist (3.0 [95% confidence interval: 2.7-3.3] hours, p < 0.001). Oxytocin checklist use was associated with decreased oxytocin use patterns at the expense of longer labor times. Findings were more pronounced with EMR integration. · An oxytocin decision support checklist is associated with reduced amounts of oxytocin used.. · However, checklists were associated with longer duration of oxytocin use and of labor.. · Results were more pronounced in the EMR-integrated checklist compared with paper checklist..

Machine learning in cardiac surgery: a narrative review.

Machine learning (ML) is increasingly being utilized to provide data driven solutions to challenges in medicine. Within the field of cardiac surgery, ML methods have been employed as risk stratification tools to predict a variety of operative outcomes. However, the clinical utility of ML in this domain is unclear. The aim of this review is to provide an overview of ML in cardiac surgery, particularly with regards to its utility in predictive analytics and implications for use in clinical decision support. We performed a narrative review of relevant articles indexed in PubMed since 2000 using the MeSH terms "Machine Learning", "Supervised Machine Learning", "Deep Learning", or "Artificial Intelligence" and "Cardiovascular Surgery" or "Thoracic Surgery". ML methods have been widely used to generate pre-operative risk profiles, consistently resulting in the accurate prediction of clinical outcomes in cardiac surgery. However, improvement in predictive performance over traditional risk metrics has proven modest and current applications in the clinical setting remain limited. Studies utilizing high volume, multidimensional data such as that derived from electronic health record (EHR) data appear to best demonstrate the advantages of ML methods. Models trained on post cardiac surgery intensive care unit data demonstrate excellent predictive performance and may provide greater clinical utility if incorporated as clinical decision support tools. Further development of ML models and their integration into EHR's may result in dynamic clinical decision support strategies capable of informing clinical care and improving outcomes in cardiac surgery.

Genetically guided precision medicine clinical decision support tools: a systematic review.

Patient care using genetics presents complex challenges. Clinical decision support (CDS) tools are a potential solution because they provide patient-specific risk assessments and/or recommendations at the point of care. This systematic review evaluated the literature on CDS systems which have been implemented to support genetically guided precision medicine (GPM). A comprehensive search was conducted in MEDLINE and Embase, encompassing January 1, 2011-March 14, 2023. The review included primary English peer-reviewed research articles studying humans, focused on the use of computers to guide clinical decision-making and delivering genetically guided, patient-specific assessments, and/or recommendations to healthcare providers and/or patients. The search yielded 3832 unique articles. After screening, 41 articles were identified that met the inclusion criteria. Alerts and reminders were the most common form of CDS used. About 27 systems were integrated with the electronic health record; 2 of those used standards-based approaches for genomic data transfer. Three studies used a framework to analyze the implementation strategy. Findings include limited use of standards-based approaches for genomic data transfer, system evaluations that do not employ formal frameworks, and inconsistencies in the methodologies used to assess genetic CDS systems and their impact on patient outcomes. We recommend that future research on CDS system implementation for genetically GPM should focus on implementing more CDS systems, utilization of standards-based approaches, user-centered design, exploration of alternative forms of CDS interventions, and use of formal frameworks to systematically evaluate genetic CDS systems and their effects on patient care.