Real-time Clinical Decision Support System Research Articles

Enhancing the accuracy and effectiveness of diagnosis of spontaneous bacterial peritonitis in cirrhotic patients: A machine learning approach utilizing clinical and laboratory data

PurposeSpontaneous bacterial peritonitis (SBP) is a bacterial infection of ascitic fluid that develops naturally, without being triggered by any surgical conditions or procedures, and is a common complication of cirrhosis. With a potential mortality rate of 40 ​%, accurate diagnosis and prompt initiation of appropriate antibiotic therapy are crucial for optimizing patient outcomes and preventing life-threatening complications. This study aimed to expand the use of computational models to improve the diagnostic accuracy of SBP in cirrhotic patients by incorporating a broader range of data, including clinical variables and laboratory values. Patients and methodsWe employed 5 machine learning classification methods - Decision Tree, Support Vector Machine, Naive Bayes, K-Nearest Neighbor, and Random Forest, utilizing a variety of demographic, clinical, and laboratory features and biomarkers. ResultsAscitic fluid markers, including white blood cell (WBC) count, lactate dehydrogenase (LDH), total protein, and polymorphonuclear cells (PMN), significantly differentiated between SBP and non-SBP patients. The Random Forest model demonstrated the highest overall accuracy at 86 ​%, while the Naive Bayes model achieved the highest sensitivity at 72 ​%. Utilizing 10 key features instead of the full feature set improved model performance, notably enhancing specificity and accuracy. ConclusionOur analysis highlights the potential of machine learning to enhance the accuracy of SBP diagnosis in cirrhotic patients. Integrating these models into clinical workflows could substantially improve patient outcomes. To achieve this, ongoing multidisciplinary research is crucial. Ensuring model interpretability, continuous monitoring, and rigorous validation will be essential for the successful implementation of real-time clinical decision support systems.

EHealth in pediatric respiratory allergy.

This review explores the relevance of eHealth technologies to address unmet needs in pediatric respiratory allergies, particularly allergic rhinitis (AR) and asthma. Given the increasing burden of these conditions, there is a pressing need for effective solutions to enhance disease surveillance, diagnosis, and management. Recent literature highlights the potential of eHealth tools to transform pediatric respiratory allergy care. The use of digital data for infodemiology, application of machine learning models to improve diagnostic sensitivity, smartphone apps with digital patient reported outcome measure (PROMs) and embedded sensors to monitor disease, healthcare professional dashboards with real-time data monitoring and clinical decision support systems (CDSS) are advances emerging to optimize pediatric respiratory allergy care. Integrating eHealth technologies into the pediatric respiratory allergy care pathway is a potential solution for current healthcare challenges to better meet the needs of children with AR and asthma. However, while the potential of eHealth is evident, its widespread implementation in real-world practice requires continued research, collaboration, and efforts to overcome existing barriers.

Artificial Intelligence-Based Models for Prediction of Mortality in ICU Patients: A Scoping Review.

Healthcare professionals may be able to anticipate more accurately a patient's timing of death and assess their possibility of recovery by implementing a real-time clinical decision support system. Using such a tool, the healthcare system can better understand a patient's condition and make more informed judgements about distributing limited resources. This scoping review aimed to analyze various death prediction AI (Artificial Intelligence) algorithms that have been used in ICU (Intensive Care Unit) patient populations. The search strategy of this study involved keyword combinations of outcome and patient setting such as mortality, survival, ICU, terminal care. These terms were used to perform database searches in MEDLINE, Embase, and PubMed up to July 2022. The variables, characteristics, and performance of the identified predictive models were summarized. The accuracy of the models was compared using their Area Under the Curve (AUC) values. Databases search yielded an initial pool of 8271 articles. A two-step screening process was then applied: first, titles and abstracts were reviewed for relevance, reducing the pool to 429 articles. Next, a full-text review was conducted, further narrowing down the selection to 400 key studies. Out of 400 studies on different tools or models for prediction of mortality in ICUs, 16 papers focused on AI-based models which were ultimately included in this study that have deployed different AI-based and machine learning models to make a prediction about negative patient outcome. The accuracy and performance of the different models varied depending on the patient populations and medical conditions. It was found that AI models compared with traditional tools like SAP3 or APACHE IV score were more accurate in death prediction, with some models achieving an AUC of up to 92.9%. The overall mortality rate ranged from 5% to more than 60% in different studies. We found that AI-based models exhibit varying performance across different patient populations. To enhance the accuracy of mortality prediction, we recommend customizing models for specific patient groups and medical contexts. By doing so, healthcare professionals may more effectively assess mortality risk and tailor treatments accordingly. Additionally, incorporating additional variables-such as genetic information-into new models can further improve their accuracy.

Innovations in Time Related Expression Recognition Using LSTM Networks

The proposed architecture leverages the strengths of both Convolutional Neural Network (CNN) and Bidirectional Long Short-Term (BLSTM) to create a robust model for temporal expression recognition in clinical texts. The CNN component effectively captures morphological and orthographic features at the character level, which enriches the semantic understanding of complex medical terminologies that are often abbreviated or have unique suffixes and prefixes. The BLSTM component excels in capturing long-range dependencies in text, which is crucial for understanding the context in which temporal expressions occur. By integrating these models with a CRF layer, the system not only predicts discrete labels but also ensures that the sequence of predicted labels is coherent and contextually appropriate, addressing the limitations of models that predict labels independently. The integration of pre-trained biomedical word vectors provides significant contextual grounding tailored to the medical domain, enhancing the model's ability to discern and interpret the nuances of medical language. This is crucial in clinical contexts where accurate interpretation of temporal phrases can be critical for patient management and treatment timelines. Further, experiments conducted on the dataset validate the effectiveness of the proposed model, demonstrating a notable improvement over traditional methods that rely heavily on hand-crafted features and rule-based approaches. Future work could explore the adaptability of this model to other subdomains of the medical field and its efficacy in processing multilingual texts, potentially increasing its applicability in global healthcare settings, with further refinement of the neural architecture and optimization of training strategies potentially yielding even better performance and faster processing times essential for real-time clinical decision support systems.

Improving McDiarmid Tree Performance for Predicting Heart Disease from Data Streams with Missing and Meaningless Values

Building machine learning algorithms for real-time clinical decision support systems has become a current research hotspot. The success of these algorithms depends on their ability to handle data stream characteristics. For example, we cite as characteristics the large amounts of data, the high speed and rate of incoming data, and the change in data nature and distribution over time. The Very Fast Decision Tree (VFDT) is a method for incrementally building decision trees. Since its proposition in the literature, it has become one of the most popular tools for data stream classification. This paper aims to optimize a new version of VFDT called McDiarmid Tree (MT) for the early prediction of heart diseases in real-time clinical decision support systems. The proposed method for improving MT performance consists of two main mechanisms: detecting the presence of missing and meaningless values in data attributes and handling the impact of this presence. The proposed MT has been compared with MT and VFDT. Simulation results show that the proposed MT attains significantly higher prediction accuracy with less time and model cost (RAM-Hours) than the other two algorithms.

An Accessible Clinical Decision Support System to Curtail Anesthetic Greenhouse Gases in a Large Health Network: Implementation Study.

Inhaled anesthetics in the operating room are potent greenhouse gases and are a key contributor to carbon emissions from health care facilities. Real-time clinical decision support (CDS) systems lower anesthetic gas waste by prompting anesthesia professionals to reduce fresh gas flow (FGF) when a set threshold is exceeded. However, previous CDS systems have relied on proprietary or highly customized anesthesia information management systems, significantly reducing other institutions' accessibility to the technology and thus limiting overall environmental benefit. In 2018, a CDS system that lowers anesthetic gas waste using methods that can be easily adopted by other institutions was developed at the University of California San Francisco (UCSF). This study aims to facilitate wider uptake of our CDS system and further reduce gas waste by describing the implementation of the FGF CDS toolkit at UCSF and the subsequent implementation at other medical campuses within the University of California Health network. We developed a noninterruptive active CDS system to alert anesthesia professionals when FGF rates exceeded 0.7 L per minute for common volatile anesthetics. The implementation process at UCSF was documented and assembled into an informational toolkit to aid in the integration of the CDS system at other health care institutions. Before implementation, presentation-based education initiatives were used to disseminate information regarding the safety of low FGF use and its relationship to environmental sustainability. Our FGF CDS toolkit consisted of 4 main components for implementation: sustainability-focused education of anesthesia professionals, hardware integration of the CDS technology, software build of the CDS system, and data reporting of measured outcomes. The FGF CDS system was successfully deployed at 5 University of California Health network campuses. Four of the institutions are independent from the institution that created the CDS system. The CDS system was deployed at each facility using the FGF CDS toolkit, which describes the main components of the technology and implementation. Each campus made modifications to the CDS tool to best suit their institution, emphasizing the versatility and adoptability of the technology and implementation framework. It has previously been shown that the FGF CDS system reduces anesthetic gas waste, leading to environmental and fiscal benefits. Here, we demonstrate that the CDS system can be transferred to other medical facilities using our toolkit for implementation, making the technology and associated benefits globally accessible to advance mitigation of health care-related emissions.

A time-aware attention model for prediction of acute kidney injury after pediatric cardiac surgery.

Acute kidney injury (AKI) is a common complication after pediatric cardiac surgery, and the early detection of AKI may allow for timely preventive or therapeutic measures. However, current AKI prediction researches pay less attention to time information among time-series clinical data and model building strategies that meet complex clinical application scenario. This study aims to develop and validate a model for predicting postoperative AKI that operates sequentially over individual time-series clinical data. A retrospective cohort of 3386 pediatric patients extracted from PIC database was used for training, calibrating, and testing purposes. A time-aware deep learning model was developed and evaluated from 3 clinical perspectives that use different data collection windows and prediction windows to answer different AKI prediction questions encountered in clinical practice. We compared our model with existing state-of-the-art models from 3 clinical perspectives using the area under the receiver operating characteristic curve (ROC AUC) and the area under the precision-recall curve (PR AUC). Our proposed model significantly outperformed the existing state-of-the-art models with an improved average performance for any AKI prediction from the 3 evaluation perspectives. This model predicted 91% of all AKI episodes using data collected at 24h after surgery, resulting in a ROC AUC of 0.908 and a PR AUC of 0.898. On average, our model predicted 83% of all AKI episodes that occurred within the different time windows in the 3 evaluation perspectives. The calibration performance of the proposed model was substantially higher than the existing state-of-the-art models. This study showed that a deep learning model can accurately predict postoperative AKI using perioperative time-series data. It has the potential to be integrated into real-time clinical decision support systems to support postoperative care planning.

Prediction of postoperative cardiac events in multiple surgical cohorts using a multimodal and integrative decision support system

Postoperative patients are at risk of life-threatening complications such as hemodynamic decompensation or arrhythmia. Automated detection of patients with such risks via a real-time clinical decision support system may provide opportunities for early and timely interventions that can significantly improve patient outcomes. We utilize multimodal features derived from digital signal processing techniques and tensor formation, as well as the electronic health record (EHR), to create machine learning models that predict the occurrence of several life-threatening complications up to 4 hours prior to the event. In order to ensure that our models are generalizable across different surgical cohorts, we trained the models on a cardiac surgery cohort and tested them on vascular and non-cardiac acute surgery cohorts. The best performing models achieved an area under the receiver operating characteristic curve (AUROC) of 0.94 on training and 0.94 and 0.82, respectively, on testing for the 0.5-hour interval. The AUROCs only slightly dropped to 0.93, 0.92, and 0.77, respectively, for the 4-hour interval. This study serves as a proof-of-concept that EHR data and physiologic waveform data can be combined to enable the early detection of postoperative deterioration events.

Process Mining the Performance of a Real-Time Healthcare 4.0 Systems Using Conditional Survival Models

As the world moves into the exciting age of Healthcare 4.0, it is essential that patients and clinicians have confidence and reassurance that the real-time clinical decision support systems being used throughout their care guarantee robustness and optimal quality of care. However, current systems involving autonomic behaviour and those with no prior clinical feedback, have generally to date had little focus on demonstrating robustness in the use of data and final output, thus generating a lack of confidence. This paper wishes to address this challenge by introducing a new process mining approach based on a statistically robust methodology that relies on the utilisation of conditional survival models for the purpose of evaluating the performance of Healthcare 4.0 systems and the quality of the care provided. Its effectiveness is demonstrated by analysing the performance of a clinical decision support system operating in an intensive care setting with the goal to monitor ventilated patients in real-time and to notify clinicians if the patient is predicted at risk of receiving injurious mechanical ventilation. Additionally, we will also demonstrate how the same metrics can be used for evaluating the patient quality of care. The proposed methodology can be used to analyse the performance of any Healthcare 4.0 system and the quality of care provided to the patient.

A Real-Time Clinical Decision Support System, for Mild Cognitive Impairment Detection, Based on a Hybrid Neural Architecture.

Clinical procedure for mild cognitive impairment (MCI) is mainly based on clinical records and short cognitive tests. However, low suspicion and difficulties in understanding test cut-offs make diagnostic accuracy being low, particularly in primary care. Artificial neural networks (ANNs) are suitable to design computed aided diagnostic systems because of their features of generating relationships between variables and their learning capability. The main aim pursued in that work is to explore the ability of a hybrid ANN-based system in order to provide a tool to assist in the clinical decision-making that facilitates a reliable MCI estimate. The model is designed to work with variables usually available in primary care, including Minimental Status Examination (MMSE), Functional Assessment Questionnaire (FAQ), Geriatric Depression Scale (GDS), age, and years of education. It will be useful in any clinical setting. Other important goal of our study is to compare the diagnostic rendering of ANN-based system and clinical physicians. A sample of 128 MCI subjects and 203 controls was selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The ANN-based system found the optimal variable combination, being AUC, sensitivity, specificity, and clinical utility index (CUI) calculated. The ANN results were compared with those from medical experts which include two family physicians, a neurologist, and a geriatrician. The optimal ANN model reached an AUC of 95.2%, with a sensitivity of 90.0% and a specificity of 84.78% and was based on MMSE, FAQ, and age inputs. As a whole, physician performance achieved a sensitivity of 46.66% and a specificity of 91.3%. CUIs were also better for the ANN model. The proposed ANN system reaches excellent diagnostic accuracy although it is based only on common clinical tests. These results suggest that the system is especially suitable for primary care implementation, aiding physicians work with cognitive impairment suspicions.

Combining structured and unstructured data in EMRs to create clinically-defined EMR-derived cohorts

BackgroundThere have been few studies describing how production EMR systems can be systematically queried to identify clinically-defined populations and limited studies utilising free-text in this process. The aim of this study is to provide a generalisable methodology for constructing clinically-defined EMR-derived patient cohorts using structured and unstructured data in EMRs.MethodsPatients with possible acute coronary syndrome (ACS) were used as an exemplar. Cardiologists defined clinical criteria for patients presenting with possible ACS. These were mapped to data tables within the production EMR system creating seven inclusion criteria comprised of structured data fields (orders and investigations, procedures, scanned electrocardiogram (ECG) images, and diagnostic codes) and unstructured clinical documentation. Data were extracted from two local health districts (LHD) in Sydney, Australia. Outcome measures included examination of the relative contribution of individual inclusion criteria to the identification of eligible encounters, comparisons between inclusion criterion and evaluation of consistency of data extracts across years and LHDs.ResultsAmong 802,742 encounters in a 5 year dataset (1/1/13–30/12/17), the presence of an ECG image (54.8% of encounters) and symptoms and keywords in clinical documentation (41.4–64.0%) were used most often to identify presentations of possible ACS. Orders and investigations (27.3%) and procedures (1.4%), were less often present for identified presentations. Relevant ICD-10/SNOMED CT codes were present for 3.7% of identified encounters. Similar trends were seen when the two LHDs were examined separately, and across years.ConclusionsClinically-defined EMR-derived cohorts combining structured and unstructured data during cohort identification is a necessary prerequisite for critical validation work required for development of real-time clinical decision support and learning health systems.

Transitioning Clinical Practice Guidelines into the Electronic Health Record through Clinical Pathways.

Clinical practice guidelines in oncology provide an evidence-based roadmap for most cancer care delivery but often lack directions for specific patient factors and disease conditions. Clinical pathways serve as a real-time clinical decision support system to translate guidelines to clinical practice. Pathways allow for the creation of a standardized, multidimensional roadmap for the continuum of care that can support clinical decision-making, maintain optimal outcomes, and limit unnecessary variation in cancer care. Here we describe the process to develop and implement clinical pathways in the electronic health record. This process includes building the appropriate foundation for a clinical pathways team with supports in the institutional ecosystem, creating visual representations of care paths, formalizing the pathway approval process, and translating clinical pathways into an electronic health record-integrated clinical decision support tool.

P1559PERFORMANCE EVALUATION OF REAL-TIME MULTI-BED MONITORING SYSTEM FOR BLOOD LEAKAGE IN DIALYSIS

Abstract Background and Aims In HD treatment, venous needle dislodgement (VND) is a major concern in healthcare systems worldwide. Although VND is not an extremely common occurrence, it could be life-threatening or even cause death. The report of the Taiwan Patient Safety Reporting System listed the dislodgement of needles, catheters, and similar devices as third among thirteen common incident types that threaten patient safety. Hospitals are jointly liable for the blood leakage and have to provide emergency treatment such as emergency hemostasis, blood transfusion, Erythropoietin (EPO) injection and hospitalization. To a worse extent, hospitals, physicians and paramedics have to face criminal charges and the claims raised by the patients’ families. This study employed digital technology to develop a real-time multi-bed monitoring system to detect fluid leakage and control the state of multi-bed blood leakage simultaneously, allowing paramedics to respond to and solve issues swiftly. The research evaluated the IoT system for the leakage detector system with multi-bed monitoring interface by clinical trials. Method Our research used the Acusense IoT platform, which is a multi-bed monitoring system that includes leakage detection device, IoT gateway, and cloud analysis system to monitor multiple HD patients in real-time. Any single IoT gateway can be utilized to pair with eight blood leakage detection devices and collects data with multi-point Bluetooth 4.2, and the information gathered by IoT gateway will be uploaded to database via WiFi. The selecting criteria for participants, who has low risk of blood leakage and the participants who did not have blood leakage within two months, were all excluded from our analysis. Before a clinical trial, 20 nurses were trained to use the leakage detection system for two times. And then we have set this platform on different areas for 1 month. For the results of blood leakage detection system in the clinical settings, we calculated accuracy, recall, precision, and averaged F score, which are the most common metrics for evaluating the performance of binary classification. Results In this study, a total of 50 subjects with an average age and standard deviation (SD) of 62.2 and 9.1 were recruited from the Tainan Municipal An-Nan Hospital (TMANH) in Taiwan. After deploying Acusense IoT platform in clinical workflow, there were a total of 682 dialysis sessions and 15 events of venous needle dislodgement or bleeding occurred of which 13 were detected by the device. The performance metrics reach 99.7% accuracy, 100% recall, 99.7% precision, and 99.9% Averaged F-measure. There are no adverse reactions to patients. It indicates that the proposed blood leakage detection system showed favorable performance, and was proven to be capable of accurately detecting blood leakage. Conclusion The blood leakage detector combined with information system realizes the multi-bed monitoring system and provides support for real-time clinical decision support system. When venous needle dislodgement happens, the monitoring system will timely warn nurse to increase the relationship between doctors and patients. It may result in the prevention of VND, and decreased hospitalization to reduce the side effects of hemodialysis.

Introducing the DM-P approach for analysing the performances of real-time clinical decision support systems

This paper introduces the DM-P approach for assessing the performance of the alerts produced from a clinical decisions support system when there is no information to indicate a genuine observation and one that is not. The DM-P approach allows the representation of the p-charts, normally used in a statistical process control chart setting, as a defects map which is ordered according to the occurrence of similar behaviour as determined using agglomerative hierarchical clustering. The effectiveness of this new approach is demonstrated using real life data from the VILIAlert clinical decisions support system deployed in real-time in an intensive care unit to monitor patients on the ventilator and used to trigger alerts to clinicians when a patient is at risk of receiving injurious mechanical ventilation. The VILIAlert data was successfully used as a case study for the DM-P approach proving both the usefulness of the technique and its ability to identify the genuine cases of patients at risk. Such an approach also has scale up quality being designed in such a way to be able to deal with large volumes of data and can be extended to include any form of statistical process control chart not just p-charts.

Classification of the Cardiotocogram Data for Anticipation of Fetal Risks using Bagging Ensemble Classifier

Abstract Cardiotocography (CTG) is utilized for monitoring fetal status during antepartum and intrapartum periods to predict the condition of the fetal wellbeing, broadly in pregnant women having potential difficulties to designate the risk of a fetal acidosis. These predictions are assessed in a real-time clinical decision support system and gives valuable information which can be utilized for additional information about the fetal state. The improvements in modern obstetric practice permitted numerous reliable and robust machine learning approaches to be employed in classifying fetal heart rate signals. The role of machine learning algorithms in identifying illnesses is becoming crucial. The purpose of this study is to evaluate the classification performances of ensemble machine learning algorithms on the antepartum CTG data. Hence, this paper is focused on the Bagging ensemble machine learning algorithm to classify fetal heart rate signals as normal or abnormal. The accuracy, F-measure and ROC area is utilized as performance metrics to assess the success of the classifiers. Experimental results have revealed that the Bagging ensemble classifier produced satisfactory results, and Bagging with Random Forest achieved better results with an accuracy of 99.02%.

Gaussian process emulation to accelerate parameter estimation in a mechanical model of the left ventricle: a critical step towards clinical end-user relevance

In recent years, we have witnessed substantial advances in the mathematical modelling of the biomechanical processes underlying the dynamics of the cardiac soft-tissue. Gao et al. (Gao et al. 2017 J. R. Soc. Interface 14, 20170203 ( doi:10.1098/rsif.2017.0203 )) demonstrated that the parameters underlying the biomechanical model have diagnostic value for prognosticating the risk of myocardial infarction. However, the computational costs of parameter estimation are prohibitive when the goal lies in building real-time clinical decision support systems. This is due to the need to repeatedly solve the mathematical equations numerically using finite-element discretization during an iterative optimization routine. The present article presents a method for accelerating the inference of the constitutive parameters by using statistical emulation with Gaussian processes. We demonstrate how the computational costs can be reduced by about three orders of magnitude, with hardly any loss in accuracy, and we assess various alternative techniques in a comparative evaluation study based on simulated data obtained by solving the left ventricular model with the finite-element method, and real magnetic resonance images data for a human volunteer.

Practical impact of a decision support for goal-directed fluid therapy on protocol adherence: a clinical implementation study in patients undergoing major abdominal surgery.

The purpose of this study was to assess the effects of using a real time clinical decision-support system, "Assisted Fluid Management" (AFM), to guide goal-directed fluid therapy (GDFT) during major abdominal surgery. We compared a group of patients managed using the AFM system with a historical cohort of patients (control group) who had been managed using a manual GDFT strategy. Adherence to the protocol was defined as the relative intraoperative time spent with a stroke volume variation (SVV) < 13%. We hypothesised that patients in the AFM group would have more time during surgery with a SVV < 13% compared to the control group. All patients had a radial arterial line connected to a pulse contour analysis monitor and received a 2ml/kg/h maintenance crystalloid infusion. Additional 250ml crystalloid boluses were administered whenever measured SVV ≥ 13% in the control group, and when the software suggested a fluid bolus in the AFM group. We compared 46 AFM-guided patients to 38 controls. Patients in the AFM group spent significantly more time during surgery with a SVV < 13% compared to the control group (median 92% [82, 96] vs. 76% [54, 86]; P < 0.0005), and received less fluid overall (1775ml [1225, 2425] vs. 2350ml [1825, 3250]; P = 0.010). The incidence of postoperative complications was comparable in the two groups. Implementation of a decision support system for GDFT guidance resulted in a significantly longer period during surgery with a SVV < 13% with a reduced total amount of fluid administered. Trial registration: Clinical Trials.gov (NCT03141411).

Development and Feasibility of a Real-Time Clinical Decision Support System for Traumatic Brain Injury Anesthesia Care.

Real-time clinical decision support (CDS) integrated with anesthesia information management systems (AIMS) can generate point of care reminders to improve quality of care. To develop, implement and evaluate a real-time clinical decision support system for anesthetic management of pediatric traumatic brain injury (TBI) patients undergoing urgent neurosurgery. We iteratively developed a CDS system for pediatric TBI patients undergoing urgent neurosurgery. The system automatically detects eligible cases and evidence-based key performance indicators (KPIs). Unwanted clinical events trigger and display real-time messages on the AIMS computer screen. Main outcomes were feasibility of detecting eligible cases and KPIs, and user acceptance. The CDS system was triggered in 22 out of 28 (79%) patients. The sensitivity of detecting continuously sampled KPIs reached 93.8%. For intermittently sampled KPIs, sensitivity and specificity reached 90.9% and 100%, respectively. 88% of providers reported that CDS helped with TBI anesthesia care. CDS implementation is feasible and acceptable with a high rate of case capture and appropriate generation of alert and guidance messages for TBI anesthesia care.

Influencing Anesthesia Provider Behavior Using Anesthesia Information Management System Data for Near Real-Time Alerts and Post Hoc Reports.

In this review article, we address issues related to using data from anesthesia information management systems (AIMS) to deliver near real-time alerts via AIMS workstation popups and/or alphanumeric pagers and post hoc reports via e-mail. We focus on reports and alerts for influencing the behavior of anesthesia providers (i.e., anesthesiologists, anesthesia residents, and nurse anesthetists). Multiple studies have shown that anesthesia clinical decision support (CDS) improves adherence to protocols and increases financial performance through facilitation of billing, regulatory, and compliance documentation; however, improved clinical outcomes have not been demonstrated. We inform developers and users of feedback systems about the multitude of concerns to consider during development and implementation of CDS to increase its effectiveness and to mitigate its potentially disruptive aspects. We discuss the timing and modalities used to deliver messages, implications of outlier-only versus individualized feedback, the need to consider possible unintended consequences of such feedback, regulations, sustainability, and portability among systems. We discuss statistical issues related to the appropriate evaluation of CDS efficacy. We provide a systematic review of the published literature (indexed in PubMed) of anesthesia CDS and offer 2 case studies of CDS interventions using AIMS data from our own institution illustrating the salient points. Because of the considerable expense and complexity of maintaining near real-time CDS systems, as compared with providing individual reports via e-mail after the fact, we suggest that if the same goal can be accomplished via delayed reporting versus immediate feedback, the former approach is preferable. Nevertheless, some processes require near real-time alerts to produce the desired improvement. Post hoc e-mail reporting from enterprise-wide electronic health record systems is straightforward and can be accomplished using system-independent pathways (e.g., via built-in e-mail support provided by the relational database management system). However, for some of these enterprise-wide systems, near real-time data access, necessary for CDS that generates concurrent alerts, has been challenging to implement.

Optimal and early detection of acute kidney injury requires effective clinical decision support systems.

We would like to commend Porter et al. for their development of a fully automated real-time clinical decision support system using internationally recognized criteria for acute kidney injury (AKI) [1]. They reported the incidence of AKI Stages 1–3, in-hospital mortality, length of stay and distribution of alerts by speciality. During the 2-year study, their system generated 59 921 alerts that was associated with 15 550 unique patients, and detected an overall incidence of AKI of 10.7%. The major findings of this study include a distribution of highest AKI stage reached 7.2, 2.2 and 1.3% for Stage 1, 2 and 3, respectively. Older adults experienced a majority of the AKI episodes, with the median age being 74 years old, which was the same across all stages of AKI. Overall mortality for all stages of AKI was 18.5% (2873 patients) compared with general in-hospital mortality for non-AKI patients of 2.2%. The median length of stay was 9 days for all stages of AKI. Finally, there was substantial variability in the distribution of AKI by clinical service-line. In the article, the authors decided that a high degree of sensitivity was desirable for their clinical decision support system, given the size of their hospital and the fact that many of the acute specialties did not have access to a renal consult service. Moreover, they preferred a low threshold for alerting AKI, giving the clinician the choice of whether to follow up on the alert, rather than risk an episode being missed. Finally, they chose the lowest serum creatinine in the previous 7–365 days prior to admission, which also increased sensitivity. Although the authors selected this as design priority, that is to increase sensitivity over specificity, this could lead to a significant amount of false-positive alerts. A high false-positive alert rate can result in alert burden or fatigue, which manifests itself in physicians not responding to important alerts because they receive too many [2]. Future work intent on reducing alert fatigue should focus on assessing the clinical performance of the alerts as measured by sensitivity, specificity, likelihood ratios and positive predictive values [3, 4]. The performance of these alerts can likely be improved by the inclusion of patientspecific risk factors, which may differ by AKI stage and age [5]. The alerts that were presented to the clinician in real-time as a statement accompanying the qualifying serum creatine result. These non-interruptive alerts included the AKI stage and referred the clinician to hospital guidelines on the intranet. It is important to note that the authors did not assess the impact of the alerts on process or outcome measures as others have done previously [6–8] including an assessment of how the alerts (the majority of which were Stage 1) might affect progression of AKI to more advanced stages. However, for those who may wish to formally assess the impact of a clinical decision support system to detect and manage AKI, certain concepts should be followed. First, non-interruptive or passive alerts may be useful in some situations, but interruptive alerts have been shown to be more effective in having physicians respond to AKI-alerts [9]. Second, although counterintuitive, alerts should not be presented in electronic charting or order entry. Rather for maximum effectiveness, they should be delivered to clinicians outside of an electronic medical record or computerized provider order entry system such as to a pager, email or direct phone communication [10]. Third, effort should be made to enhance the presentation and personalization of the alerts, as well as what content should be included in them [11]. Lastly, in an effort to reduce alert fatigue, streamlining alerts to those that are actionable (i.e., require intervention) may be less sensitive but more specific with higher impact on patient care. We applaud the efforts of Porter et al. to advance the use of a clinical decision support system to improve patient care by enhancing detection of AKI. Coupling an alert with existing IN F O C U S