Clinical Decision Support System Tool Research Articles

Supporting clinical decision making in the emergency department for paediatric patients using machine learning: A scoping review protocol.

Machine learning as a clinical decision support system tool has the potential to assist clinicians who must make complex and accurate medical decisions in fast paced environments such as the emergency department. This paper presents a protocol for a scoping review, with the objective of summarising the existing research on machine learning clinical decision support system tools in the emergency department, focusing on models that can be used for paediatric patients, where a knowledge gap exists. The methodology used will follow the scoping study framework of Arksey and O'Malley, along with other guidelines. Machine learning clinical decision support system tools for any outcome and population (paediatric/adult/mixed) for use in the emergency department will be included. Articles such as grey literature, letters, pre-prints, editorials, scoping/literature/narrative reviews, non-English full text papers, protocols, surveys, abstract or full text not available and models based on synthesised data will be excluded. Articles from the last five years will be included. Four databases will be searched: Medline (EBSCO), CINAHL (EBSCO), EMBASE and Cochrane Central. Independent reviewers will perform the screening in two sequential stages (stage 1: clinician expertise and stage 2: computer science expertise), disagreements will be resolved by discussion. Data relevant to the research question will be collected. Quantitative analysis will be performed to generate the results. The study results will summarise the existing research on machine learning clinical decision support tools in the emergency department, focusing on models that can be used for paediatric patients. This holds the promise to identify opportunities to both incorporate models in clinical practice and to develop future models by utilising reviewers from diverse backgrounds and relevant expertise.

Human-Centered Design of a Clinical Decision Support for Anemia Screening in Children with Inflammatory Bowel Disease.

Inflammatory bowel disease (IBD) commonly leads to iron deficiency anemia (IDA). Rates of screening and treatment of IDA are often low. A clinical decision support system (CDSS) embedded in an electronic health record could improve adherence to evidence-based care. Rates of CDSS adoption are often low due to poor usability and fit with work processes. One solution is to use human-centered design (HCD), which designs CDSS based on identified user needs and context of use and evaluates prototypes for usefulness and usability. this study aimed to use HCD to design a CDSS tool called the IBD Anemia Diagnosis Tool, IADx. Interviews with IBD practitioners informed creation of a process map of anemia care that was used by an interdisciplinary team that used HCD principles to create a prototype CDSS. The prototype was iteratively tested with "Think Aloud" usability evaluation with clinicians as well as semi-structured interviews, a survey, and observations. Feedback was coded and informed redesign. Process mapping showed that IADx should function at in-person encounters and asynchronous laboratory review. Clinicians desired full automation of clinical information acquisition such as laboratory trends and analysis such as calculation of iron deficit, less automation of clinical decision selection such as laboratory ordering, and no automation of action implementation such as signing medication orders. Providers preferred an interruptive alert over a noninterruptive reminder. Providers preferred an interruptive alert, perhaps due to the low likelihood of noticing a noninterruptive advisory. High levels of desire for automation of information acquisition and analysis with less automation of decision selection and action may be generalizable to other CDSSs designed for chronic disease management. This underlines the ways in which CDSSs have the potential to augment rather than replace provider cognitive work.

Screening and routine diagnosis of mental disorders among migrants in primary care: A cross-sectional study

BackgroundMigrants in host countries are at risk for the development of mental health conditions. The two aims of the study were to describe routine diagnoses of mental disorders among migrant patients at primary healthcare level and the associated risk factors, and to test the utility of an innovative migrant mental health assessment by evaluating whether the health professionals followed the recommendations proposed by the clinical decision support system (CDSS) tool. MethodsA cross-sectional study was carried out in eight primary care centres (PCCs) in four non-randomly selected health regions of Catalonia, Spain from March to December 2018. Routine health data and mental health diagnoses based on the International Classification of Diseases (10th edition), including mental, behavioural and neuro developmental disorders (F01-F99), symptoms and signs involving emotional state (R45), and sleep disorders (G47), were extracted from the electronic health records. The proportion of mental health conditions was estimated and logistic regression models were used to assess any possible association with mental health disorders. The utility of the mental health assessment was assessed with the proportion of questionnaires performed by health professionals for migrants fulfilling the mental health screening criterion (country of origin with an active conflict in 2017) and the diagnoses given to the screened patients. ResultsOf 14,130 migrants that visited any of the PCCs during the study period, 7,358 (52.1 %) were women with a median age of 38.0 years-old. There were 520/14,130 (3.7 %) migrant patients diagnosed with a mental disorder, being more frequent among women (342/7,358; 4.7 %, p-value < 0.001), migrants from Latin-America (177/3,483; 5.1 %, p < 0.001) and those who recently arrived in Spain (170/3,672; 4.6 %, p < 0.001). A lower proportion of mental disorders were reported in migrants coming from conflicted countries in 2017 (116/3,669, 3.2 %, p = 0.053).Out of the 547 mental health diagnoses reported in 520 patients, 69/14,130 (0.5 %) were mood disorders, 346/14,130 (2.5 %) anxiety disorders and 127/14,130 (0.9 %) sleeping disorders. Mood disorders were more common in migrants from Eastern Europe (25/2,971; 0.8 %, p < 0.001) and anxiety disorders in migrants from Latin-America (126/3,483; 3.6 %, p < 0.001), while both type of disorders were more often reported in women (p < 0.001).In the adjusted model, women (aOR: 1.5, [95 % CI 1.2–1.8, p < 0.001]), migrants with more than one visit to the health center during the study period (aOR: 4.4, [95 %CI 2.8–6.8, p < 0.001]) and who presented an infectious disease (aOR: 2.1, [95 %CI 1.5–3.1, p < 0.001]) had higher odds of having a mental disorder.Lastly, out of the 1,840 migrants coming from a conflicted country in 2017 who were attended in centres where the CDSS tool was implemented, 29 (1.6 %) had a mental health assessment performed and the tool correctly identified one individual. ConclusionsMental health is a condition that may be overlooked in migrants at primary healthcare. Interventions at this level of care must be reinforced and adapted to the needs and circumstances of migrants to ensure equity in health services.

Development and implementation of a clinical decision support system tool for the evaluation of suspected monkeypox infection.

Monkeypox virus was historically rare outside of West and Central Africa until the current 2022 global outbreak, which has required clinicians to be alert to identify individuals with possible monkeypox, institute isolation, and take appropriate next steps in evaluation and management. Clinical decision support systems (CDSS), which have been shown to improve adherence to clinical guidelines, can support frontline clinicians in applying the most current evaluation and management guidance in the setting of an emerging infectious disease outbreak when those guidelines are evolving over time. Here, we describe the rapid development and implementation of a CDSS tool embedded in the electronic health record to guide frontline clinicians in the diagnostic evaluation of monkeypox infection and triage patients with potential monkeypox infection to individualized infectious disease physician review. We also present data on the initial performance of this tool in a large integrated healthcare system.

Physicians' Perceptions and Expectations of an Artificial Intelligence-Based Clinical Decision Support System in Cancer Care in an Underserved Setting

Abstract Objectives Artificial intelligence (AI) tools are being increasingly incorporated into health care. However, few studies have evaluated users' expectations of such tools, prior to implementation, specifically in an underserved setting. Methods We conducted a qualitative research study employing semistructured interviews of physicians at The Instituto do Câncer do Ceará, Fortaleza, Brazil. The interview guide focused on anticipated, perceived benefits and challenges of using an AI-based clinical decision support system tool, Watson for Oncology. We recruited physician oncologists, working full or part-time, without prior experience with any AI-based tool. The interviews were taped and transcribed in Portuguese and then translated into English. Thematic analysis using the constant comparative approach was performed. Results Eleven oncologists participated in the study. The following overarching themes and subthemes emerged from the analysis of interview transcripts: theme-1, “general context” including (1) current setting, workload, and patient population and (2) existing challenges in cancer treatment, and theme-2, “perceptions around the potential use of an AI-based tool,” including (1) perceived benefits and (2) perceived challenges. Physicians expected that the implementation of an AI-based tool would result in easy access to the latest clinical recommendations, facilitate standardized cancer care, and allow it to be delivered with greater confidence and efficiency. Participants had several concerns such as availability of innovative treatments in resource-poor settings, treatment acceptance, trust, physician autonomy, and workflow disruptions. Conclusion This study provides physicians' anticipated perspectives, both benefits and challenges, about the use of an AI-based tool in cancer treatment in a resource-limited setting.

Systematic review of antibiotic stewardship interventions for urinary tract infection management in the ambulatory setting

Background: Urinary tract infections (UTIs) are common indications for antibiotics in ambulatory setting, and inappropriate use is prevalent. Fluoroquinolones account for 40% of antibiotics prescribed for uncomplicated UTIs, despite clinical guidance against their use as firstline agents. We conducted a systematic review to determine which antibiotic stewardship intervention(s) are effective in improving antibiotic prescribing for UTIs in the ambulatory setting. Methods: Following PRISMA guidelines, English-language literature from 1995 to September 21, 2021, was searched for articles about antimicrobial stewardship, UTI, and ambulatory setting from PubMed, Embase, and Central. Additional articles were identified from authors’ collections and references of pertinent articles. Studies were included if the authors implemented intervention targeting adults 18 years and older in outpatient setting (excluding emergency departments). Interventions were categorized into Guideline Development and Dissemination (GDD), Audit and Feedback, Clinical Decision Support System (CDSS), and Multimodal Interventions. Results: The literature search identified 1,899 papers; 14 papers were included in this review; and 4 additional papers were identified from other sources. The main interventions were GDD in 6 studies, audit and feedback in 3 studies, CDSS in 4 studies, and multimodal interventions in 5 stidues. These studies had heterogeneity of the practice settings and interventions. Moreover, 11 studies targeted primary care, 2 studies targeted urgent care, 1 study targeted both primary and urgent care, 2 studies were conducted in spinal cord injury clinics, and 2 studies were conducted in hospital-wide outpatient sites. Outcomes included (1) statistically significant increase in guideline-concordant antibiotic prescribing in 12 studies (range, 4.6%–246%); (2) statistically significant decrease in fluoroquinolone prescriptions (range, 9.1%–86.3%) in 7 of 9 studies focusing on fluoroquinolones; (3) significant decreases in drug resistance in urine pathogens in 2 studies that evaluated this. Provider education, in conjunction with passive CDSS tools, such as integrating order sets for UTI prescriptions with prefilled instructions into electronic medical records appeared most beneficial. Several studies have investigated negative impact and have found no increase in retreatment rates or worse outcomes. Conclusions: Our systematic literature review identified a limited number of studies with a variety of interventions that improved antibiotic use for UTIs in the ambulatory care setting. Provider education, in conjunction with CDSS tools, can be less time-consuming than audit and feedback and can target a large number of providers and practices. Future studies need to address sustainability over longer periods and should target specialty clinic populations because they have high burden of patients with multidrug-resistant UTI organisms.Funding: NoneDisclosures: None

Clostridioides difficile: A diagnostic intervention

Background:Clostridioides difficile infection (CDI) is a leading cause of healthcare-associated infection and is associated with increased morbidity and mortality. Multiple strategies have been implemented to optimize the diagnostic accuracy of CDI testing algorithms. However, overdiagnosis of C. difficile colonization remains a challenge especially in the era of highly sensitive Nucleic acid amplification testing (NAAT). We implemented 2 interventions to reduce the rates of inappropriate C. difficile orders and tests. Methods: We performed a quasi-experimental retrospective study to examine the rates of all inpatient C. difficile test orders and results relative to 2 interventions between January 2018 and February 2021. We defined 3 periods: preintervention, after the first intervention, and after the second intervention. The first intervention, implemented May 2019, was a clinical decision support system (CDSS) tool to guide clinicians to order testing only if CDI criteria were met. The second intervention, implemented July 2020, was the requirement of mandatory antimicrobial team approval of PCR reflex testing for indeterminate toxin or antigen test results. This intervention included a discussion between clinicians and members of stewardship team prior to approval. To evaluate the impact of interventions on ordering appropriateness, chart review was conducted on a random subsample of 100 orders from each period. Hospital-onset CDI (HO-CDI) was calculated using CDC criteria. Results: In total, 3,004 C. difficile test orders were placed during the study period. The rates of reportable HO-CDI were significantly reduced by 57.1% (P = .003). We detected a significant reduction in the number of tests ordered over time from 11.6 to 7.51 per 1,000 patient days. (p Conclusions: CDSS tools target patients at high pretest probability of CDI. The restriction of PCR-reflex testing to the antimicrobial stewardship team is a novel effective measure to minimize the misdiagnosis of CDI. The incorporation of multiple strategies is necessary to improve the diagnostic accuracy of C. difficile testing.Funding: NoneDisclosures: None

Clinical decision support system recommendations: how often do radiologists and clinicians accept them?

To assess the acceptance and reliability of clinical decision support system (CDSS) imaging referral scores (ESR iGuide). A pilot study was conducted in a tertiary hospital. Four different experts were invited to rate 40 simulated clinical cases on a 5-level scale, for the level of agreement with the ESR iGuide's recommended procedures. In cases of disagreement, physicians were asked to indicate the reason. Descriptive measures were calculated for the level of agreement. We also explored the degree of agreement between four different specialists, and examined the cases in which clinicians disagreed with ESR iGuide best practice recommendations. The mean rating of the four experts for the 40 clinical simulated cases was 4.17 ± 0.65, median 4.25 (on a scale of 1-5). All four raters totally agreed with the system recommendation in 75% of cases. No significant relationship was found between the degree of agreement and the number of indications and the patient's age or gender. In an optimistic scenario, using a binary agree/disagree variable, the Overall Percentage Agreement for the rating of the 40 simulated cases between the four experts was 77.28%. There were a total of 20 disagreements out of 160 cases with the ESR iGuide, of which 7 were among the two radiologists. CDSS can be an effective tool for guiding the selection of appropriate imaging examinations, thus cutting costs due to unnecessary imaging scans. Since this is a pilot study, further research on a larger scale, preferably at national level, is required. • The average of the mean rating of the four experts was 4.17 ±0.65, median 4.25, on a scale of 1-5 where 5 represents total agreement with the CDSS tool. • In an optimistic scenario, using a binary agree/disagree variable, the Overall Percentage Agreement between the four experts was 77.28%. • Radiologists had fewer disagreements with the recommendations of the CDSS tool than other physicians, indicating a better fit of the support system to radiology experts' perspective.

Role of a Digital Clinical Decision-Support System in General Practitioners' Management of COPD in Norway.

BackgroundThe study investigated if a web-based clinical decision–support system (CDSS) tool would improve general practitioners’ (GPs) accuracy of diagnosis and classification of patients with chronic obstructive pulmonary disease (COPD), and whether nonpharmacological and pharmacological treatment would be better aligned with the COPD guidelines.MethodsGPs were randomized to either a single use of the CDSS or continuing standard of care. The clinical recommendations of the CDSS were based on the GOLD guidelines and provided suggestions for treatment and management of COPD. Data were collected digitally from GPs and patients in both groups using a tablet computer. A follow-up questionnaire was sent to the GPs 1 year after the conclusion of the study.ResultsA total of 25 GPs (31% women, mean age 41 years) participated, 12 randomized to using the CDSS tool and 13 followed standard of care when assessing their next five to ten COPD patients. In sum, 149 patients with presumed COPD were included (88 CDSS group, 61 standard-of-care group). In the CDSS group, no COPD misdiagnoses occurred, 98% received vaccine recommendations, and all smokers (n=39) received smoking-cessation advice. The standard-of-care group had 23% misdiagnosis (P<0.001), only 67% received vaccine recommendations (P<0.001), and 87% smoking-cessation advice (P=0.022. All told, 31% of patients did not receive medication as recommended according to guidelines, with no significant differences between the groups. GPs rated the CDSS as very useful. Mean usage time was 3 minutes, 26 seconds. A majority (13 of 19, 68%) of the GPs continued using the CDSS after the conclusion of the study. CAT score identified twice as many patients as having more symptoms than the mMRC, indicating the added value of the multi-item questionnaire.ConclusionUse of the CDSS was associated with preventing misdiagnosis of COPD and improved adherence to recommended nonpharmacological measures, but a single use did not improve pharmacological treatment considerations.

Use of quick sequential organ failure assessment score-based sepsis clinical decision support system may be helpful to predict sepsis development

Objectives: A sepsis clinical decision support system (CDSS) can facilitate quicker sepsis detection and treatment and consequently improve outcomes. We developed a qSOFA-based sepsis CDSS and evaluated its impact on compliance with a 3-hour resuscitation bundle for patients with sepsis. Methods: This before-and-after study included consecutive adult patients with suspected infection and qSOFA scores of ≥ 2 at their emergency department (ED) presentation of a tertiary care hospital. Sepsis was defined according to the Sepsis-3 criteria. We evaluated the 3-hour resuscitation bundle compliance rate for control patients from July through August 2016, for patients using the qSOFA-based sepsis CDSS from September through December 2016, and the impact of the system using multivariable logistic regression analysis. Results: Of 306 patients with suspected infection and positive qSOFA scores at presentation, 265 patients (86.6%) developed sepsis (including 71 patients with septic shock). The 3-hour resuscitation bundle compliance rate did not differ significantly between the patients before and after the routine implementation of the qSOFA-based sepsis CDSS (63.7% vs. 52.6%; P = 0.071). Multivariate analysis showed that age (AOR [adjusted odds ratio], 1.033; P = 0.002) and body temperature (AOR, 1.677; P &lt; 0.001) were associated with bundle compliance. Conclusions: Among patients with a positive qSOFA score at presentation, sepsis developed in 86.6%, which means the qSOFA-based sepsis CDSS may be helpful; however, it was not associated with improved bundle compliance. Future quality improvement studies with multifactorial, hospital-wide approaches using sepsis CDSS tools are warranted.

Mobile Clinical Decision Support System for the Management of Diabetic Patients With Kidney Complications in UK Primary Care Settings: Mixed Methods Feasibility Study.

BackgroundAttempts to utilize eHealth in diabetes mellitus (DM) management have shown promising outcomes, mostly targeted at patients; however, few solutions have been designed for health care providers.ObjectiveThe purpose of this study was to conduct a feasibility project developing and evaluating a mobile clinical decision support system (CDSS) tool exclusively for health care providers to manage chronic kidney disease (CKD) in patients with DM.MethodsThe design process was based on the 3 key stages of the user-centered design framework. First, an exploratory qualitative study collected the experiences and views of DM specialist nurses regarding the use of mobile apps in clinical practice. Second, a CDSS tool was developed for the management of patients with DM and CKD. Finally, a randomized controlled trial examined the acceptability and impact of the tool.ResultsWe interviewed 15 DM specialist nurses. DM specialist nurses were not currently using eHealth solutions in their clinical practice, while most nurses were not even aware of existing medical apps. However, they appreciated the potential benefits that apps may bring to their clinical practice. Taking into consideration the needs and preferences of end users, a new mobile CDSS app, “Diabetes & CKD,” was developed based on guidelines. We recruited 39 junior foundation year 1 doctors (44% male) to evaluate the app. Of them, 44% (17/39) were allocated to the intervention group, and 56% (22/39) were allocated to the control group. There was no significant difference in scores (maximum score=13) assessing the management decisions between the app and paper-based version of the app’s algorithm (intervention group: mean 7.24 points, SD 2.46 points; control group: mean 7.39, SD 2.56; t37=–0.19, P=.85). However, 82% (14/17) of the participants were satisfied with using the app.ConclusionsThe findings will guide the design of future CDSS apps for the management of DM, aiming to help health care providers with a personalized approach depending on patients’ comorbidities, specifically CKD, in accordance with guidelines.

Driving quality improvement: How clinical decision support can facilitate compliance with evidence-based pathways.

2045 Background: Cancer care is changing rapidly with more detailed understanding of disease and more numerous therapeutic choices. As treatment choice is more complex, mechanisms to improve compliance with evidence based treatment can improve the quality of cancer care. Methods: A retrospective cohort study was conducted from January 2014-May 2016 evaluating the impact of a clinical decision support system (CDSS) on compliance with evidence based pathways (EBP) across 9 statewide community based oncology practices. These EBP are developed with physician input on efficacy toxicity and value and incorporated in to a CDSS that is used within the Electronic Health Record (EHR) at point of care to alter the choice architecture a clinician sees when prescribing therapy. A multi-level logistic regression model was used to adjust for group effects on physician or practice behavior. SAS 9.4 software was used and GLIMMIX was applied. Individual physician benchmark compliance was evaluated using McNemar's test. Results: Regimen compliance with EBP was measured pre- and post- implementation of the CDSS tool across a large network encompassing 9 statewide practices and 633 physicians who prescribed over 30,000 individual patient treatment regimens over a 6 month period. The CDSS that is incorporated within the EHR significantly improved compliance with EBP across the entire cohort of practices, and in individual practices (see Table). Individual oncologists reached a target of 75% compliance more often (58% vs 72%) after implementation of the tool (p &lt; 0.001). Conclusions: CDSS is a tool that improves compliance with EBP that is effective at improving targets of compliance broadly, at the practice, and at the individual clinician level. Clinical informatics solutions that influence physician behavior can be inclusive of physicians in design, iterative in process, and nudge as opposed to force clinician behavior to drive quality improvement. These clinical informatics solutions grow in importance as the complexity of cancer care continues to increase and we seek to improve upon the quality and value of care delivery. [Table: see text]

A clinical decision support system improves antibiotic therapy for upper urinary tract infection in a randomized single-blinded study

BackgroundDue to increasing bacterial resistance rates choosing a correct empiric antibiotic therapy is getting more and more complex. Often medical doctors use information tools to make the right treatment choice.MethodsOne hundred sixty six participants (77 medical doctors and 89 medical students) were asked to provide a diagnosis and antibiotic therapy in a simple fictive paper case of upper urinary tract infection (UTI) in a randomized single-blinded study. Participants were randomized to one of four information tools they were allowed to use in the study or control: 1. free internet access, 2. pharmaceutical pocket guide, 3. pocket guide antibiotic therapy, 4. clinical decision support system (CDSS), and control (no information tool). The CDSS was designed for the study. The adherence to the national German UTI guideline was evaluated.ResultsOnly 27.1% (n = 45/166) provided a correct diagnosis of upper UTI and 19.4% (n = 32/166) an antibiotic treatment recommended by national German treatment guidelines indicating their need for information tools. This result was not significantly different between medical doctors and medical students, residents and medical specialists or level of working experience. Using CDSS improved results significantly compared to conventional tools (diagnosis 57.1%; treatment recommendation 40.5%; p < 0,01). Processing time was not different between the use of CDSS and conventional information tools. CDSS users based their decision making on their assigned information tool more than users of conventional tools (73.8% vs. 48.0%; p < 0.01). Using CDSS improved the confidence of participants in their recommendation significantly compared to conventional tools (p < 0.01).ConclusionsOur study suggests that medical professionals require information tools in diagnosing and treating a simple case of upper UTI correctly. CDSS appears to be superior to conventional tools as an information source.

Case Not Closed: Prescription Errors 12 Years after Computerized Physician Order Entry Implementation

To assess the prolonged impact of computerized physician order entry (CPOE) on medication prescription errors in pediatric intensive care patients. This observational study was conducted at a pediatric intensive care unit in which a CPOE (Metavision, iMDsoft, Israel) with a limited clinical decision support system was implemented between 2004 and 2007. Since then, no changes were made to the systems. We analyzed 2500 electronic prescriptions (1250 prescriptions from 2015 and 1250 prescriptions from 2016). Prescription errors were identified by a pediatric intensive care physician and classified as potential adverse drug events, medication prescription errors, or rule violations. Their prevalence was compared with the rate in 2007, reported in a previous study from the same unit. A randomly selected 10% of the prescriptions were also analyzed by the pediatric intensive care unit pharmacist, and the level of agreement was determined. The rate of prescription errors increased from 1.4% in 2007 to 3.2% in 2015 (P = .03). Following revision of the clinical decision support system tools, prescription errors decreased to 1% in 2016 (P < .0001). The potential adverse drug event rate dropped from 2% in 2015 to 0.7% in 2016 (P = .006), and the medication prescription error rate, from 1% to 0.2% (P = .01). The agreement between the 2 reviewers was excellent (k = 0.96). The rate of prescription errors may increase with time from implementation of a CPOE. Repeated surveillance of prescription errors is highly advised to plan strategies to reduce them. This approach should be considered in quality improvement of computerized information systems in general.

Avoiding alert fatigue in pulmonary embolism decision support: a new method to examine ‘trigger rates’

A clinical decision support system (CDSS) is integrated into the electronic health record (EHR) and allows physicians to easily use a clinical decision support (CDS) tool. However, often CDSSs are...

Optimizing Clinical Decision Support in the Electronic Health Record. Clinical Characteristics Associated with the Use of a Decision Tool for Disposition of ED Patients with Pulmonary Embolism.

Adoption of clinical decision support (CDS) tools by clinicians is often limited by workflow barriers. We sought to assess characteristics associated with clinician use of an electronic health record-embedded clinical decision support system (CDSS). In a prospective study on emergency department (ED) activation of a CDSS tool across 14 hospitals between 9/1/14 to 4/30/15, the CDSS was deployed at 10 active sites with an on-site champion, education sessions, iterative feedback, and up to 3 gift cards/clinician as an incentive. The tool was also deployed at 4 passive sites that received only an introductory educational session. Activation of the CDSS - which calculated the Pulmonary Embolism Severity Index (PESI) score and provided guidance - and associated clinical data were collected prospectively. We used multivariable logistic regression with random effects at provider/facility levels to assess the association between activation of the CDSS tool and characteristics at: 1) patient level (PESI score), 2) provider level (demographics and clinical load at time of activation opportunity), and 3) facility level (active vs. passive site, facility ED volume, and ED acuity at time of activation opportunity). Out of 662 eligible patient encounters, the CDSS was activated in 55%: active sites: 68% (346/512); passive sites 13% (20/150). In bivariate analysis, active sites had an increase in activation rates based on the number of prior gift cards the physician had received (96% if 3 prior cards versus 60% if 0, p<0.0001). At passive sites, physicians < age 40 had higher rates of activation (p=0.03). In multivariable analysis, active site status, low ED volume at the time of diagnosis and PESI scores I or II (compared to III or higher) were associated with higher likelihood of CDSS activation. Performing on-site tool promotion significantly increased odds of CDSS activation. Optimizing CDSS adoption requires active education.

Streamlining patient consultations for sleep disorders with a knowledge-based CDSS

ObjectivesThis paper examines the workflow of sleep physicians during a patient consultation and how an innovative clinical decision support system (CDSS) provides efficiency and effectiveness gains. MethodsThe CDSS tools consisted of two input applications for patient data, with a knowledge based decision support system developed participatively with physicians and an international panel. An argument tree approach was used to produce diagnostic explanations and an evidence-based report for the physician using medically correct and shared terminology. A usability evaluation using a qualitative approach was carried out to ensure that the CDSS met the physicians׳ information needs, as well as the wider needs of a Sleep Investigation Unit. ResultsThe physicians found the CDSS both useful and usable with clear applications in triage and diagnostic decision-making, and in patient education. ConclusionThe CDSS both reduces the time and number of visits needed for consultations, and helps focus consultation on better individual patient care through informed explanation of diagnostic and treatment decisions.

Time-dependent Estimates of Recurrence and Survival in Colon Cancer: Clinical Decision Support System Tool Development for Adjuvant Therapy and Oncological Outcome Assessment

Unanswered questions remain in determining which high-risk node-negative colon cancer (CC) cohorts benefit from adjuvant therapy and how it may differ in an equal access population. Machine-learned Bayesian Belief Networks (ml-BBNs) accurately estimate outcomes in CC, providing clinicians with Clinical Decision Support System (CDSS) tools to facilitate treatment planning. We evaluated ml-BBNs ability to estimate survival and recurrence in CC. We performed a retrospective analysis of registry data of patients with CC to train-test-crossvalidate ml-BBNs using the Department of Defense Automated Central Tumor Registry (January 1993 to December 2004). Cases with events or follow-up that passed quality control were stratified into 1-, 2-, 3-, and 5-year survival cohorts. ml-BBNs were trained using machine-learning algorithms and k-fold crossvalidation and receiver operating characteristic curve analysis used for validation. BBNs were comprised of 5301 patients and areas under the curve ranged from 0.85 to 0.90. Positive predictive values for recurrence and mortality ranged from 78 to 84 per cent and negative predictive values from 74 to 90 per cent by survival cohort. In the 12-month model alone, 1,132,462,080 unique rule sets allow physicians to predict individual recurrence/mortality estimates. Patients with Stage II (N0M0) CC benefit from chemotherapy at different rates. At one year, all patients older than 73 years of age with T2-4 tumors and abnormal carcinoembryonic antigen levels benefited, whereas at five years, all had relative reduction in mortality with the largest benefit amongst elderly, highest T-stage patients. ml-BBN can readily predict which high-risk patients benefit from adjuvant therapy. CDSS tools yield individualized, clinically relevant estimates of outcomes to assist clinicians in treatment planning.

Time-dependent estimates of recurrence and survival in colon cancer: clinical decision support system tool development for adjuvant therapy and oncological outcome assessment.

e14500 Background: Unanswered questions remain regarding treatment efficacy in colon cancer (CC), especially those determining high-risk node-negative cohorts that may benefit from adjuvant therapy. We sought to evaluate the use of machine learning and classification modeling to estimate survival and recurrence in CC. Methods: We used the Department of Defense Automated Central Tumor Registry (ACTUR) to identify primary CC patients treated between January 1993 and December 2004. Cases with events or follow-up that passed quality control were stratified into one-, two-, three-, and five-year survival cohorts. ml-BBNs were trained using machine-learning algorithms and k-fold cross-validation, and receiver operating characteristic (ROC) curve analysis used for validation. Results: There were 5,301 cases stratified into cohorts. Survival cohort Areas-Under-the-Curve (AUCs) ranged from 0.85–0.90, positive-predictive-values (PPVs) for recurrence and mortality ranged from 78-84% and negative-predictive-values (NPVs) from 74-90%. Cross-validation showed that the ml-BBNs produce robust individual estimates of recurrence (p&lt;0.001) and mortality (p&lt;0.001) based on readily available clinical-pathological information in the context of adjuvant chemotherapy. Conclusions: Tumor registry data and machine-learned Bayesian Belief Networks produce robust classifiers. These Clinical Decision Support System tools yield clinically relevant estimates of outcomes that may assist clinicians in treatment planning.

Impact of health information technology on detection of potential adverse drug events at the ordering stage

The impact of implementing commercially available health care information technologies at hospitals in a large health system on the identification of potential adverse drug events (ADEs) at the medication ordering stage was studied. All hospitals in the health system had implemented a clinical decision-support system (CDSS) consisting of a centralized clinical data repository, interfaces for reports, a results reviewer, and a package of ADE alert rules. Additional technology including computerized provider order entry (CPOE), an advanced CDSS, and evidence-based order sets was implemented in nine hospitals. ADE alerts at these hospitals were compared with alerts at nine hospitals without the advanced technology. A linear mixed-effects model was used in determining the mean response profile of six dependent variables over 28 total months for each experimental group. Overall, hospitals with CPOE and an advanced CDSS captured significantly more ADE alerts for pharmacist review; an average of 336 additional potential ADEs per month per hospital were reviewed. Pharmacists identified some 94% of the alerts as false positives. Alerts identified as potentially true positives were reviewed with physicians, and order changes were recommended. The number of true-positive alerts per 1000 admissions increased. The implementation of CPOE and advanced CDSS tools significantly increased the number of potential ADE alerts for pharmacist review and the number of true-positive ADE alerts identified per 1000 admissions.