%0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 2 %P e25 %T Diabetes-Related Behavior Change Knowledge Transfer to Primary Care Practitioners and Patients: Implementation and Evaluation of a Digital Health Platform %+ Medical Informatics ProgramDepartment of Community Health and Epidemiology, Faculty of MedicineDalhousie UniversityCentre for Clinical Research5790 University AvenueHalifax, NS, B3H 1V7Canada1 902 448 24471 902 494 1597samina.abidi@dal.ca  %A Abidi,Samina %A Vallis,Michael %A Piccinini-Vallis,Helena %A Imran,Syed Ali %A Abidi,Syed Sibte Raza %K type 2 diabetes mellitus %K self-management %K health behavior %K knowledge management %K clinical decision support system %D 2018 %7 18.04.2018 %9 Original Paper %J JMIR Med Inform %G English %X Background: Behavioral science is now being integrated into diabetes self-management interventions. However, the challenge that presents itself is how to translate these knowledge resources during care so that primary care practitioners can use them to offer evidence-informed behavior change support and diabetes management recommendations to patients with diabetes. Objective: The aim of this study was to develop and evaluate a computerized decision support platform called “Diabetes Web-Centric Information and Support Environment” (DWISE) that assists primary care practitioners in applying standardized behavior change strategies and clinical practice guidelines–based recommendations to an individual patient and empower the patient with the skills and knowledge required to self-manage their diabetes through planned, personalized, and pervasive behavior change strategies. Methods: A health care knowledge management approach is used to implement DWISE so that it features the following functionalities: (1) assessment of primary care practitioners’ readiness to administer validated behavior change interventions to patients with diabetes; (2) educational support for primary care practitioners to help them offer behavior change interventions to patients; (3) access to evidence-based material, such as the Canadian Diabetes Association’s (CDA) clinical practice guidelines, to primary care practitioners; (4) development of personalized patient self-management programs to help patients with diabetes achieve healthy behaviors to meet CDA targets for managing type 2 diabetes; (5) educational support for patients to help them achieve behavior change; and (6) monitoring of the patients’ progress to assess their adherence to the behavior change program and motivating them to ensure compliance with their program. DWISE offers these functionalities through an interactive Web-based interface to primary care practitioners, whereas the patient’s self-management program and associated behavior interventions are delivered through a mobile patient diary via mobile phones and tablets. DWISE has been tested for its usability, functionality, usefulness, and acceptance through a series of qualitative studies. Results: For the primary care practitioner tool, most usability problems were associated with the navigation of the tool and the presentation, formatting, understandability, and suitability of the content. For the patient tool, most issues were related to the tool’s screen layout, design features, understandability of the content, clarity of the labels used, and navigation across the tool. Facilitators and barriers to DWISE use in a shared decision-making environment have also been identified. Conclusions: This work has provided a unique electronic health solution to translate complex health care knowledge in terms of easy-to-use, evidence-informed, point-of-care decision aids for primary care practitioners. Patients’ feedback is now being used to make necessary modification to DWISE. %M 29669705 %R 10.2196/medinform.9629 %U http://medinform.jmir.org/2018/2/e25/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 2 %P e24 %T Reasons For Physicians Not Adopting Clinical Decision Support Systems: Critical Analysis %+ Carolina Health Informatics ProgramUniversity of North Carolina at Chapel Hill428 Carrington HallChapel Hill, NC, 27514United States1 9198435413saif@unc.edu  %A Khairat,Saif %A Marc,David %A Crosby,William %A Al Sanousi,Ali %K decision support systems, clinical %K decision making, computer-assisted %K attitude to computers %D 2018 %7 18.04.2018 %9 Review %J JMIR Med Inform %G English %X Background: Clinical decision support systems (CDSSs) are an integral component of today’s health information technologies. They assist with interpretation, diagnosis, and treatment. A CDSS can be embedded throughout the patient safety continuum providing reminders, recommendations, and alerts to health care providers. Although CDSSs have been shown to reduce medical errors and improve patient outcomes, they have fallen short of their full potential. User acceptance has been identified as one of the potential reasons for this shortfall. Objective: The purpose of this paper was to conduct a critical review and task analysis of CDSS research and to develop a new framework for CDSS design in order to achieve user acceptance. Methods: A critical review of CDSS papers was conducted with a focus on user acceptance. To gain a greater understanding of the problems associated with CDSS acceptance, we conducted a task analysis to identify and describe the goals, user input, system output, knowledge requirements, and constraints from two different perspectives: the machine (ie, the CDSS engine) and the user (ie, the physician). Results: Favorability of CDSSs was based on user acceptance of clinical guidelines, reminders, alerts, and diagnostic suggestions. We propose two models: (1) the user acceptance and system adaptation design model, which includes optimizing CDSS design based on user needs/expectations, and (2) the input-process-output-engagemodel, which reveals to users the processes that govern CDSS outputs. Conclusions: This research demonstrates that the incorporation of the proposed models will improve user acceptance to support the beneficial effects of CDSSs adoption. Ultimately, if a user does not accept technology, this not only poses a threat to the use of the technology but can also pose a threat to the health and well-being of patients. %M 29669706 %R 10.2196/medinform.8912 %U http://medinform.jmir.org/2018/2/e24/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 2 %P e23 %T Agile Acceptance Test–Driven Development of Clinical Decision Support Advisories: Feasibility of Using Open Source Software %+ University of Texas Southwestern Medical Center5323 Harry Hines BoulevardDallas, TX, 75390United States1 214 648 1303mujeeb.basit@utsouthwestern.edu  %A Basit,Mujeeb A %A Baldwin,Krystal L %A Kannan,Vaishnavi %A Flahaven,Emily L %A Parks,Cassandra J %A Ott,Jason M %A Willett,Duwayne L %K clinical decision support systems %K electronic health records %K software validation %K software verification %K agile methods %K test driven development %D 2018 %7 13.04.2018 %9 Original Paper %J JMIR Med Inform %G English %X Background: Moving to electronic health records (EHRs) confers substantial benefits but risks unintended consequences. Modern EHRs consist of complex software code with extensive local configurability options, which can introduce defects. Defects in clinical decision support (CDS) tools are surprisingly common. Feasible approaches to prevent and detect defects in EHR configuration, including CDS tools, are needed. In complex software systems, use of test–driven development and automated regression testing promotes reliability. Test–driven development encourages modular, testable design and expanding regression test coverage. Automated regression test suites improve software quality, providing a “safety net” for future software modifications. Each automated acceptance test serves multiple purposes, as requirements (prior to build), acceptance testing (on completion of build), regression testing (once live), and “living” design documentation. Rapid-cycle development or “agile” methods are being successfully applied to CDS development. The agile practice of automated test–driven development is not widely adopted, perhaps because most EHR software code is vendor-developed. However, key CDS advisory configuration design decisions and rules stored in the EHR may prove amenable to automated testing as “executable requirements.” Objective: We aimed to establish feasibility of acceptance test–driven development of clinical decision support advisories in a commonly used EHR, using an open source automated acceptance testing framework (FitNesse). Methods: Acceptance tests were initially constructed as spreadsheet tables to facilitate clinical review. Each table specified one aspect of the CDS advisory’s expected behavior. Table contents were then imported into a test suite in FitNesse, which queried the EHR database to automate testing. Tests and corresponding CDS configuration were migrated together from the development environment to production, with tests becoming part of the production regression test suite. Results: We used test–driven development to construct a new CDS tool advising Emergency Department nurses to perform a swallowing assessment prior to administering oral medication to a patient with suspected stroke. Test tables specified desired behavior for (1) applicable clinical settings, (2) triggering action, (3) rule logic, (4) user interface, and (5) system actions in response to user input. Automated test suite results for the “executable requirements” are shown prior to building the CDS alert, during build, and after successful build. Conclusions: Automated acceptance test–driven development and continuous regression testing of CDS configuration in a commercial EHR proves feasible with open source software. Automated test–driven development offers one potential contribution to achieving high-reliability EHR configuration. Vetting acceptance tests with clinicians elicits their input on crucial configuration details early during initial CDS design and iteratively during rapid-cycle optimization. %M 29653922 %R 10.2196/medinform.9679 %U http://medinform.jmir.org/2018/2/e23/ %0 Journal Article %@ 1929-0748 %I JMIR Publications %V 7 %N 4 %P e105 %T Comparing a Mobile Decision Support System Versus the Use of Printed Materials for the Implementation of an Evidence-Based Recommendation: Protocol for a Qualitative Evaluation %+ Departamento de Epidemiología Clínica y BioestadísticaPontificia Universidad JaverianaCra 7 No 40-62 - Hospital Universitario San Ignacio, piso 2Bogotá,Colombia57 1 3208320 ext 2799jcamacho@javeriana.edu.co  %A Camacho,Jhon %A Medina Ch.,Ana María %A Landis-Lewis,Zach %A Douglas,Gerald %A Boyce,Richard %K practice guideline %K implementation science %K decision support systems %K mhealth %K technology acceptance %K computer-interpretable clinical guidelines %K Colombia %D 2018 %7 13.04.2018 %9 Protocol %J JMIR Res Protoc %G English %X Background: The distribution of printed materials is the most frequently used strategy to disseminate and implement clinical practice guidelines, although several studies have shown that the effectiveness of this approach is modest at best. Nevertheless, there is insufficient evidence to support the use of other strategies. Recent research has shown that the use of computerized decision support presents a promising approach to address some aspects of this problem. Objective: The aim of this study is to provide qualitative evidence on the potential effect of mobile decision support systems to facilitate the implementation of evidence-based recommendations included in clinical practice guidelines. Methods: We will conduct a qualitative study with two arms to compare the experience of primary care physicians while they try to implement an evidence-based recommendation in their clinical practice. In the first arm, we will provide participants with a printout of the guideline article containing the recommendation, while in the second arm, we will provide participants with a mobile app developed after formalizing the recommendation text into a clinical algorithm. Data will be collected using semistructured and open interviews to explore aspects of behavioral change and technology acceptance involved in the implementation process. The analysis will be comprised of two phases. During the first phase, we will conduct a template analysis to identify barriers and facilitators in each scenario. Then, during the second phase, we will contrast the findings from each arm to propose hypotheses about the potential impact of the system. Results: We have formalized the narrative in the recommendation into a clinical algorithm and have developed a mobile app. Data collection is expected to occur during 2018, with the first phase of analysis running in parallel. The second phase is scheduled to conclude in July 2019. Conclusions: Our study will further the understanding of the role of mobile decision support systems in the implementation of clinical practice guidelines. Furthermore, we will provide qualitative evidence to aid decisions made by low- and middle-income countries’ ministries of health about investments in these technologies. %M 29653921 %R 10.2196/resprot.9827 %U http://www.researchprotocols.org/2018/4/e105/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 1 %P e8 %T Automated Information Extraction on Treatment and Prognosis for Non–Small Cell Lung Cancer Radiotherapy Patients: Clinical Study %+ Penn MedicineDepartment of Radiation OncologyUniversity of Pennsylvania3400 Civic Center BlvdPhiladelphia, PA, 19104United States1 215 866 7087wei.zou@uphs.upenn.edu  %A Zheng,Shuai %A Jabbour,Salma K %A O'Reilly,Shannon E %A Lu,James J %A Dong,Lihua %A Ding,Lijuan %A Xiao,Ying %A Yue,Ning %A Wang,Fusheng %A Zou,Wei %K information extraction %K oncology %K chemoradiation treatment %K prognosis %K non–small cell lung %K information storage and retrieval %K natural language processing %D 2018 %7 01.02.2018 %9 Original Paper %J JMIR Med Inform %G English %X Background: In outcome studies of oncology patients undergoing radiation, researchers extract valuable information from medical records generated before, during, and after radiotherapy visits, such as survival data, toxicities, and complications. Clinical studies rely heavily on these data to correlate the treatment regimen with the prognosis to develop evidence-based radiation therapy paradigms. These data are available mainly in forms of narrative texts or table formats with heterogeneous vocabularies. Manual extraction of the related information from these data can be time consuming and labor intensive, which is not ideal for large studies. Objective: The objective of this study was to adapt the interactive information extraction platform Information and Data Extraction using Adaptive Learning (IDEAL-X) to extract treatment and prognosis data for patients with locally advanced or inoperable non–small cell lung cancer (NSCLC). Methods: We transformed patient treatment and prognosis documents into normalized structured forms using the IDEAL-X system for easy data navigation. The adaptive learning and user-customized controlled toxicity vocabularies were applied to extract categorized treatment and prognosis data, so as to generate structured output. Results: In total, we extracted data from 261 treatment and prognosis documents relating to 50 patients, with overall precision and recall more than 93% and 83%, respectively. For toxicity information extractions, which are important to study patient posttreatment side effects and quality of life, the precision and recall achieved 95.7% and 94.5% respectively. Conclusions: The IDEAL-X system is capable of extracting study data regarding NSCLC chemoradiation patients with significant accuracy and effectiveness, and therefore can be used in large-scale radiotherapy clinical data studies. %M 29391345 %R 10.2196/medinform.8662 %U http://medinform.jmir.org/2018/1/e8/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 20 %N 1 %P e22 %T Prediction of Incident Hypertension Within the Next Year: Prospective Study Using Statewide Electronic Health Records and Machine Learning %+ Department of SurgeryStanford UniversityS370, 300 Pasteur DriveStanford, CA, 94305United States1 6504279198bxling@stanford.edu  %A Ye,Chengyin %A Fu,Tianyun %A Hao,Shiying %A Zhang,Yan %A Wang,Oliver %A Jin,Bo %A Xia,Minjie %A Liu,Modi %A Zhou,Xin %A Wu,Qian %A Guo,Yanting %A Zhu,Chunqing %A Li,Yu-Ming %A Culver,Devore S %A Alfreds,Shaun T %A Stearns,Frank %A Sylvester,Karl G %A Widen,Eric %A McElhinney,Doff %A Ling,Xuefeng %K hypertension %K risk assessment %K electronic health records %K multiple chronic conditions %K mental disorders %K social determinants of health %D 2018 %7 30.01.2018 %9 Original Paper %J J Med Internet Res %G English %X Background: As a high-prevalence health condition, hypertension is clinically costly, difficult to manage, and often leads to severe and life-threatening diseases such as cardiovascular disease (CVD) and stroke. Objective: The aim of this study was to develop and validate prospectively a risk prediction model of incident essential hypertension within the following year. Methods: Data from individual patient electronic health records (EHRs) were extracted from the Maine Health Information Exchange network. Retrospective (N=823,627, calendar year 2013) and prospective (N=680,810, calendar year 2014) cohorts were formed. A machine learning algorithm, XGBoost, was adopted in the process of feature selection and model building. It generated an ensemble of classification trees and assigned a final predictive risk score to each individual. Results: The 1-year incident hypertension risk model attained areas under the curve (AUCs) of 0.917 and 0.870 in the retrospective and prospective cohorts, respectively. Risk scores were calculated and stratified into five risk categories, with 4526 out of 381,544 patients (1.19%) in the lowest risk category (score 0-0.05) and 21,050 out of 41,329 patients (50.93%) in the highest risk category (score 0.4-1) receiving a diagnosis of incident hypertension in the following 1 year. Type 2 diabetes, lipid disorders, CVDs, mental illness, clinical utilization indicators, and socioeconomic determinants were recognized as driving or associated features of incident essential hypertension. The very high risk population mainly comprised elderly (age>50 years) individuals with multiple chronic conditions, especially those receiving medications for mental disorders. Disparities were also found in social determinants, including some community-level factors associated with higher risk and others that were protective against hypertension. Conclusions: With statewide EHR datasets, our study prospectively validated an accurate 1-year risk prediction model for incident essential hypertension. Our real-time predictive analytic model has been deployed in the state of Maine, providing implications in interventions for hypertension and related diseases and hopefully enhancing hypertension care. %M 29382633 %R 10.2196/jmir.9268 %U http://www.jmir.org/2018/1/e22/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 1 %P e3 %T Quality of Decision Support in Computerized Provider Order Entry: Systematic Literature Review %+ Division of PharmacyUniversity Hospitals of GenevaRue Gabrielle-Perret-Gentil 4Geneva, 1211Switzerland41 78653287141 223726255delphine.carli@chuv.ch  %A Carli,Delphine %A Fahrni,Guillaume %A Bonnabry,Pascal %A Lovis,Christian %K decision support systems, clinical %K medical order entry systems %K system, medication alert %K sensitivity %K specificity %K predictive value of tests %D 2018 %7 24.01.2018 %9 Review %J JMIR Med Inform %G English %X Background: Computerized decision support systems have raised a lot of hopes and expectations in the field of order entry. Although there are numerous studies reporting positive impacts, concerns are increasingly high about alert fatigue and effective impacts of these systems. One of the root causes of fatigue alert reported is the low clinical relevance of these alerts. Objective: The objective of this systematic review was to assess the reported positive predictive value (PPV), as a proxy to clinical relevance, of decision support systems in computerized provider order entry (CPOE). Methods: A systematic search of the scientific literature published between February 2009 and March 2015 on CPOE, clinical decision support systems, and the predictive value associated with alert fatigue was conducted using PubMed database. Inclusion criteria were as follows: English language, full text available (free or pay for access), assessed medication, direct or indirect level of predictive value, sensitivity, or specificity. When possible with the information provided, PPV was calculated or evaluated. Results: Additive queries on PubMed retrieved 928 candidate papers. Of these, 376 were eligible based on abstract. Finally, 26 studies qualified for a full-text review, and 17 provided enough information for the study objectives. An additional 4 papers were added from the references of the reviewed papers. The results demonstrate massive variations in PPVs ranging from 8% to 83% according to the object of the decision support, with most results between 20% and 40%. The best results were observed when patients’ characteristics, such as comorbidity or laboratory test results, were taken into account. There was also an important variation in sensitivity, ranging from 38% to 91%. Conclusions: There is increasing reporting of alerts override in CPOE decision support. Several causes are discussed in the literature, the most important one being the clinical relevance of alerts. In this paper, we tried to assess formally the clinical relevance of alerts, using a near-strong proxy, which is the PPV of alerts, or any way to express it such as the rate of true and false positive alerts. In doing this literature review, three inferences were drawn. First, very few papers report direct or enough indirect elements that support the use or the computation of PPV, which is a gold standard for all diagnostic tools in medicine and should be systematically reported for decision support. Second, the PPV varies a lot according to the typology of decision support, so that overall rates are not useful, but must be reported by the type of alert. Finally, in general, the PPVs are below or near 50%, which can be considered as very low. %M 29367187 %R 10.2196/medinform.7170 %U http://medinform.jmir.org/2018/1/e3/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 6 %N 1 %P e6 %T A Clinical Decision Support Engine Based on a National Medication Repository for the Detection of Potential Duplicate Medications: Design and Evaluation %+ Graduate Institute of Biomedical InformaticsCollege of Medical Science and TechnologyTaipei Medical University250 Wuxing StTaipei, 11030Taiwan886 266382736 ext 1509ctliu@tmu.edu.tw  %A Yang,Cheng-Yi %A Lo,Yu-Sheng %A Chen,Ray-Jade %A Liu,Chien-Tsai %K duplicate medication %K adverse drug reaction %K clinical decision support system %K PharmaCloud %D 2018 %7 19.01.2018 %9 Original Paper %J JMIR Med Inform %G English %X Background: A computerized physician order entry (CPOE) system combined with a clinical decision support system can reduce duplication of medications and thus adverse drug reactions. However, without infrastructure that supports patients’ integrated medication history across health care facilities nationwide, duplication of medication can still occur. In Taiwan, the National Health Insurance Administration has implemented a national medication repository and Web-based query system known as the PharmaCloud, which allows physicians to access their patients’ medication records prescribed by different health care facilities across Taiwan. Objective: This study aimed to develop a scalable, flexible, and thematic design-based clinical decision support (CDS) engine, which integrates a national medication repository to support CPOE systems in the detection of potential duplication of medication across health care facilities, as well as to analyze its impact on clinical encounters. Methods: A CDS engine was developed that can download patients’ up-to-date medication history from the PharmaCloud and support a CPOE system in the detection of potential duplicate medications. When prescribing a medication order using the CPOE system, a physician receives an alert if there is a potential duplicate medication. To investigate the impact of the CDS engine on clinical encounters in outpatient services, a clinical encounter log was created to collect information about time, prescribed drugs, and physicians’ responses to handling the alerts for each encounter. Results: The CDS engine was installed in a teaching affiliate hospital, and the clinical encounter log collected information for 3 months, during which a total of 178,300 prescriptions were prescribed in the outpatient departments. In all, 43,844/178,300 (24.59%) patients signed the PharmaCloud consent form allowing their physicians to access their medication history in the PharmaCloud. The rate of duplicate medication was 5.83% (1843/31,614) of prescriptions. When prescribing using the CDS engine, the median encounter time was 4.3 (IQR 2.3-7.3) min, longer than that without using the CDS engine (median 3.6, IQR 2.0-6.3 min). From the physicians’ responses, we found that 42.06% (1908/4536) of the potential duplicate medications were recognized by the physicians and the medication orders were canceled. Conclusions: The CDS engine could easily extend functions for detection of adverse drug reactions when more and more electronic health record systems are adopted. Moreover, the CDS engine can retrieve more updated and completed medication histories in the PharmaCloud, so it can have better performance for detection of duplicate medications. Although our CDS engine approach could enhance medication safety, it would make for a longer encounter time. This problem can be mitigated by careful evaluation of adopted solutions for implementation of the CDS engine. The successful key component of a CDS engine is the completeness of the patient’s medication history, thus further research to assess the factors in increasing the PharmaCloud consent rate is required. %M 29351893 %R 10.2196/medinform.9064 %U http://medinform.jmir.org/2018/1/e6/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 4 %P e45 %T Predicting Unplanned Transfers to the Intensive Care Unit: A Machine Learning Approach Leveraging Diverse Clinical Elements %+ The MITRE Corporation202 Burlington RdBedford, MA, 01730United States+1 781 271 7191+1 781 271 2352wellner@mitre.org  %A Wellner,Ben %A Grand,Joan %A Canzone,Elizabeth %A Coarr,Matt %A Brady,Patrick W %A Simmons,Jeffrey %A Kirkendall,Eric %A Dean,Nathan %A Kleinman,Monica %A Sylvester,Peter %K clinical deterioration %K machine learning %K data mining %K electronic health record %K patient acuity %K vital signs %K nursing assessment %K clinical laboratory techniques %D 2017 %7 22.11.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: Early warning scores aid in the detection of pediatric clinical deteriorations but include limited data inputs, rarely include data trends over time, and have limited validation. Objective: Machine learning methods that make use of large numbers of predictor variables are now commonplace. This work examines how different types of predictor variables derived from the electronic health record affect the performance of predicting unplanned transfers to the intensive care unit (ICU) at three large children’s hospitals. Methods: We trained separate models with data from three different institutions from 2011 through 2013 and evaluated models with 2014 data. Cases consisted of patients who transferred from the floor to the ICU and met one or more of 5 different priori defined criteria for suspected unplanned transfers. Controls were patients who were never transferred to the ICU. Predictor variables for the models were derived from vitals, labs, acuity scores, and nursing assessments. Classification models consisted of L1 and L2 regularized logistic regression and neural network models. We evaluated model performance over prediction horizons ranging from 1 to 16 hours. Results: Across the three institutions, the c-statistic values for our best models were 0.892 (95% CI 0.875-0.904), 0.902 (95% CI 0.880-0.923), and 0.899 (95% CI 0.879-0.919) for the task of identifying unplanned ICU transfer 6 hours before its occurrence and achieved 0.871 (95% CI 0.855-0.888), 0.872 (95% CI 0.850-0.895), and 0.850 (95% CI 0.825-0.875) for a prediction horizon of 16 hours. For our first model at 80% sensitivity, this resulted in a specificity of 80.5% (95% CI 77.4-83.7) and a positive predictive value of 5.2% (95% CI 4.5-6.2). Conclusions: Feature-rich models with many predictor variables allow for patient deterioration to be predicted accurately, even up to 16 hours in advance. %M 29167089 %R 10.2196/medinform.8680 %U http://medinform.jmir.org/2017/4/e45/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 4 %P e36 %T Computerized Decision Aids for Shared Decision Making in Serious Illness: Systematic Review %+ Health Data Science LabSchool of Public Health and Health SystemsUniversity of Waterloo200 University Ave WWaterloo, ON, N2L 3G1Canada1 5198884567 ext 31567joon.lee@uwaterloo.ca  %A Staszewska,Anna %A Zaki,Pearl %A Lee,Joon %K decision making %K decision aids %K evidence-based medicine %K user-computer interface %K chronic disease %D 2017 %7 06.10.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: Shared decision making (SDM) is important in achieving patient-centered care. SDM tools such as decision aids are intended to inform the patient. When used to assist in decision making between treatments, decision aids have been shown to reduce decisional conflict, increase ease of decision making, and increase modification of previous decisions. Objective: The purpose of this systematic review is to assess the impact of computerized decision aids on patient-centered outcomes related to SDM for seriously ill patients. Methods: PubMed and Scopus databases were searched to identify randomized controlled trials (RCTs) that assessed the impact of computerized decision aids on patient-centered outcomes and SDM in serious illness. Six RCTs were identified and data were extracted on study population, design, and results. Risk of bias was assessed by a modified Cochrane Risk of Bias Tool for Quality Assessment of Randomized Controlled Trials. Results: Six RCTs tested decision tools in varying serious illnesses. Three studies compared different computerized decision aids against each other and a control. All but one study demonstrated improvement in at least one patient-centered outcome. Computerized decision tools may reduce unnecessary treatment in patients with low disease severity in comparison with informational pamphlets. Additionally, electronic health record (EHR) portals may provide the opportunity to manage care from the home for individuals affected by illness. The quality of decision aids is of great importance. Furthermore, satisfaction with the use of tools is associated with increased patient satisfaction and reduced decisional conflict. Finally, patients may benefit from computerized decision tools without the need for increased physician involvement. Conclusions: Most computerized decision aids improved at least one patient-centered outcome. All RCTs identified were at a High Risk of Bias or Unclear Risk of Bias. Effort should be made to improve the quality of RCTs testing SDM aids in serious illness. %M 28986341 %R 10.2196/medinform.6405 %U https://medinform.jmir.org/2017/4/e36/ %0 Journal Article %@ 2291-5222 %I JMIR Publications %V 5 %N 9 %P e144 %T Tablet-Based Patient-Centered Decision Support for Minor Head Injury in the Emergency Department: Pilot Study %+ Department of Emergency MedicineYale School of Medicine464 Congress Ave, Suite 260New Haven, CT, 06519United States1 203 737 64541 203 785 4580edward.melnick@yale.edu  %A Singh,Navdeep %A Hess,Erik %A Guo,George %A Sharp,Adam %A Huang,Brian %A Breslin,Maggie %A Melnick,Edward %K clinical decision support %K decision aids %K head injury, minor %K medical informatics %K spiral computed tomography %K health services overuse %K patient-centered outcomes research %D 2017 %7 28.09.2017 %9 Original Paper %J JMIR Mhealth Uhealth %G English %X Background: The Concussion or Brain Bleed app is a clinician- and patient-facing electronic tool to guide decisions about head computed tomography (CT) use in patients presenting to the emergency department (ED) with minor head injury. This app integrates a patient decision aid and clinical decision support (using the Canadian CT Head Rule, CCHR) at the bedside on a tablet computer to promote conversations around individualized risk and patients’ specific concerns within the ED context. Objective: The objective of this study was to describe the use of the Concussion or Brain Bleed app in a high-volume ED and to establish preliminary efficacy estimates on patient experience, clinician experience, health care utilization, and patient safety. These data will guide the planning of a larger multicenter trial testing the effectiveness of the Concussion or Brain Bleed app. Methods: We conducted a prospective pilot study of adult (age 18-65 years) patients presenting to the ED after minor head injury who were identified by participating clinicians as low risk by the CCHR. The primary outcome was patient knowledge regarding the injury, risks, and CT use. Secondary outcomes included patient satisfaction, decisional conflict, trust in physician, clinician acceptability, system usability, Net Promoter scores, head CT rate, and patient safety at 7 days. Results: We enrolled 41 patients cared for by 29 different clinicians. Patient knowledge increased after the use of the app (questions correct out of 9: pre-encounter, 3.3 vs postencounter, 4.7; mean difference 1.4, 95% CI 0.8-2.0). Patients reported a mean of 11.7 (SD 13.5) on the Decisional Conflict Scale and 92.5 (SD 12.0) in the Trust in Physician Scale (both scales range from 0 to 100). Most patients were satisfied with the app’s clarity of information (35, 85%), helpfulness of information (36, 88%), and amount of information (36, 88%). In the 41 encounters, most clinicians thought the information was somewhat or extremely helpful to the patient (35, 85%), would want to use something similar for other decisions (27, 66%), and would recommend the app to other providers (28, 68%). Clinicians reported a mean system usability score of 85.1 (SD 15; scale from 0 to 100 with 85 in the “excellent” acceptability range). The total Net Promoter Score was 36.6 (on a scale from –100 to 100). A total of 7 (17%) patients received a head CT in the ED. No patients had a missed clinically important brain injury at 7 days. Conclusions: An app to help patients assess the utility of CT imaging after head injury in the ED increased patient knowledge. Nearly all clinicians reported the app to be helpful to patients. The high degree of patient satisfaction, clinician acceptability, and system usability support rigorous testing of the app in a larger multicenter trial. %M 28958987 %R 10.2196/mhealth.8732 %U http://mhealth.jmir.org/2017/9/e144/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 3 %P e24 %T Clinical Note Creation, Binning, and Artificial Intelligence %+ Harvard – MIT, Division of Health Science and TechnologyInstitute for Medical Engineering and ScienceMassachusetts Institute of Technology77 Massachusetts Ave, MIT E25-505Cambridge, MA, MA 02139United States1 617 253 79371 617 258 07859lceli@mit.edu  %A Deliberato,Rodrigo Octávio %A Celi,Leo Anthony %A Stone,David J %K electronic health records %K artificial Intelligence %K clinical informatics %D 2017 %7 03.08.2017 %9 Viewpoint %J JMIR Med Inform %G English %X The creation of medical notes in software applications poses an intrinsic problem in workflow as the technology inherently intervenes in the processes of collecting and assembling information, as well as the production of a data-driven note that meets both individual and healthcare system requirements. In addition, the note writing applications in currently available electronic health records (EHRs) do not function to support decision making to any substantial degree. We suggest that artificial intelligence (AI) could be utilized to facilitate the workflows of the data collection and assembly processes, as well as to support the development of personalized, yet data-driven assessments and plans. %M 28778845 %R 10.2196/medinform.7627 %U http://medinform.jmir.org/2017/3/e24/ %0 Journal Article %@ 2292-9495 %I JMIR Publications %V 4 %N 3 %P e17 %T A Technological Innovation to Reduce Prescribing Errors Based on Implementation Intentions: The Acceptability and Feasibility of MyPrescribe %+ Manchester Centre for Health PsychologyDivision of Psychology and Mental Health, School of Health Sciences, Faculty of Biology, Medicine and HealthUniversity of Manchester, Manchester Academic Health Science CentreCoupland 1 BuildingOxford RoadManchester, M13 9PLUnited Kingdom44 161 275 2589chris.keyworth@manchester.ac.uk  %A Keyworth,Chris %A Hart,Jo %A Thoong,Hong %A Ferguson,Jane %A Tully,Mary %K drug prescribing %K behavior and behavior mechanisms %K clinical competence %K qualitative research %K mobile applications %K pharmacists %K patient safety %K telemedicine %D 2017 %7 01.08.2017 %9 Original Paper %J JMIR Hum Factors %G English %X Background: Although prescribing of medication in hospitals is rarely an error-free process, prescribers receive little feedback on their mistakes and ways to change future practices. Audit and feedback interventions may be an effective approach to modifying the clinical practice of health professionals, but these may pose logistical challenges when used in hospitals. Moreover, such interventions are often labor intensive. Consequently, there is a need to develop effective and innovative interventions to overcome these challenges and to improve the delivery of feedback on prescribing. Implementation intentions, which have been shown to be effective in changing behavior, link critical situations with an appropriate response; however, these have rarely been used in the context of improving prescribing practices. Objective: Semistructured qualitative interviews were conducted to evaluate the acceptability and feasibility of providing feedback on prescribing errors via MyPrescribe, a mobile-compatible website informed by implementation intentions. Methods: Data relating to 200 prescribing errors made by 52 junior doctors were collected by 11 hospital pharmacists. These errors were populated into MyPrescribe, where prescribers were able to construct their own personalized action plans. Qualitative interviews with a subsample of 15 junior doctors were used to explore issues regarding feasibility and acceptability of MyPrescribe and their experiences of using implementation intentions to construct prescribing action plans. Framework analysis was used to identify prominent themes, with findings mapped to the behavioral components of the COM-B model (capability, opportunity, motivation, and behavior) to inform the development of future interventions. Results: MyPrescribe was perceived to be effective in providing opportunities for critical reflection on prescribing errors and to complement existing training (such as junior doctors’ e-portfolio). The participants were able to provide examples of how they would use “If-Then” plans for patient management. Technology, as opposed to other methods of learning (eg, traditional “paper based” learning), was seen as a positive advancement for continued learning. Conclusions: MyPrescribe was perceived as an acceptable and feasible learning tool for changing prescribing practices, with participants suggesting that it would make an important addition to medical prescribers’ training in reflective practice. MyPrescribe is a novel theory-based technological innovation that provides the platform for doctors to create personalized implementation intentions. Applying the COM-B model allows for a more detailed understanding of the perceived mechanisms behind prescribing practices and the ways in which interventions aimed at changing professional practice can be implemented. %M 28765104 %R 10.2196/humanfactors.7153 %U http://humanfactors.jmir.org/2017/3/e17/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 3 %P e18 %T Issues Associated With the Use of Semantic Web Technology in Knowledge Acquisition for Clinical Decision Support Systems: Systematic Review of the Literature %+ Department of Computer ScienceAuckland University of TechnologyPrivate Bag 92006Auckland,New Zealand64 9921999964 99219944szolhava@aut.ac.nz  %A Zolhavarieh,Seyedjamal %A Parry,David %A Bai,Quan %K semantic web technology %K clinical decision support system %K systematic review %K medical informatics %K knowledge %K Internet %D 2017 %7 05.07.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: Knowledge-based clinical decision support system (KB-CDSS) can be used to help practitioners make diagnostic decisions. KB-CDSS may use clinical knowledge obtained from a wide variety of sources to make decisions. However, knowledge acquisition is one of the well-known bottlenecks in KB-CDSSs, partly because of the enormous growth in health-related knowledge available and the difficulty in assessing the quality of this knowledge as well as identifying the “best” knowledge to use. This bottleneck not only means that lower-quality knowledge is being used, but also that KB-CDSSs are difficult to develop for areas where expert knowledge may be limited or unavailable. Recent methods have been developed by utilizing Semantic Web (SW) technologies in order to automatically discover relevant knowledge from knowledge sources. Objective: The two main objectives of this study were to (1) identify and categorize knowledge acquisition issues that have been addressed through using SW technologies and (2) highlight the role of SW for acquiring knowledge used in the KB-CDSS. Methods: We conducted a systematic review of the recent work related to knowledge acquisition MeM for clinical decision support systems published in scientific journals. In this regard, we used the keyword search technique to extract relevant papers. Results: The retrieved papers were categorized based on two main issues: (1) format and data heterogeneity and (2) lack of semantic analysis. Most existing approaches will be discussed under these categories. A total of 27 papers were reviewed in this study. Conclusions: The potential for using SW technology in KB-CDSS has only been considered to a minor extent so far despite its promise. This review identifies some questions and issues regarding use of SW technology for extracting relevant knowledge for a KB-CDSS. %M 28679487 %R 10.2196/medinform.6169 %U http://medinform.jmir.org/2017/3/e18/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 2 %P e17 %T Validation of an Improved Computer-Assisted Technique for Mining Free-Text Electronic Medical Records %+ School of Veterinary Medicine and ScienceUniversity of NottinghamSutton Bonington CampusCollege RoadLoughborough, LE12 5RDUnited Kingdom44 115 951 679644 1159516415marco.duz@nottingham.ac.uk  %A Duz,Marco %A Marshall,John F %A Parkin,Tim %K text mining %K data mining %K electronic medical record %K validation studies %D 2017 %7 29.06.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: The use of electronic medical records (EMRs) offers opportunity for clinical epidemiological research. With large EMR databases, automated analysis processes are necessary but require thorough validation before they can be routinely used. Objective: The aim of this study was to validate a computer-assisted technique using commercially available content analysis software (SimStat-WordStat v.6 (SS/WS), Provalis Research) for mining free-text EMRs. Methods: The dataset used for the validation process included life-long EMRs from 335 patients (17,563 rows of data), selected at random from a larger dataset (141,543 patients, ~2.6 million rows of data) and obtained from 10 equine veterinary practices in the United Kingdom. The ability of the computer-assisted technique to detect rows of data (cases) of colic, renal failure, right dorsal colitis, and non-steroidal anti-inflammatory drug (NSAID) use in the population was compared with manual classification. The first step of the computer-assisted analysis process was the definition of inclusion dictionaries to identify cases, including terms identifying a condition of interest. Words in inclusion dictionaries were selected from the list of all words in the dataset obtained in SS/WS. The second step consisted of defining an exclusion dictionary, including combinations of words to remove cases erroneously classified by the inclusion dictionary alone. The third step was the definition of a reinclusion dictionary to reinclude cases that had been erroneously classified by the exclusion dictionary. Finally, cases obtained by the exclusion dictionary were removed from cases obtained by the inclusion dictionary, and cases from the reinclusion dictionary were subsequently reincluded using Rv3.0.2 (R Foundation for Statistical Computing, Vienna, Austria). Manual analysis was performed as a separate process by a single experienced clinician reading through the dataset once and classifying each row of data based on the interpretation of the free-text notes. Validation was performed by comparison of the computer-assisted method with manual analysis, which was used as the gold standard. Sensitivity, specificity, negative predictive values (NPVs), positive predictive values (PPVs), and F values of the computer-assisted process were calculated by comparing them with the manual classification. Results: Lowest sensitivity, specificity, PPVs, NPVs, and F values were 99.82% (1128/1130), 99.88% (16410/16429), 94.6% (223/239), 100.00% (16410/16412), and 99.0% (100×2×0.983×0.998/[0.983+0.998]), respectively. The computer-assisted process required few seconds to run, although an estimated 30 h were required for dictionary creation. Manual classification required approximately 80 man-hours. Conclusions: The critical step in this work is the creation of accurate and inclusive dictionaries to ensure that no potential cases are missed. It is significantly easier to remove false positive terms from a SS/WS selected subset of a large database than search that original database for potential false negatives. The benefits of using this method are proportional to the size of the dataset to be analyzed. %M 28663163 %R 10.2196/medinform.7123 %U http://medinform.jmir.org/2017/2/e17/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 6 %P e221 %T The Effectiveness of Information Technology-Supported Shared Care for Patients With Chronic Disease: A Systematic Review %+ The Netherlands Cancer InstituteDivision of Psychosocial Research and EpidemiologyPlesmanlaan 121Amsterdam, 1066CXNetherlands31 88 005 7531 51 223 22w.v.harten@nki.nl  %A Kooij,Laura %A Groen,Wim G %A van Harten,Wim H %K review %K integrated healthcare systems %K health information systems %K chronic disease %D 2017 %7 22.06.2017 %9 Original Paper %J J Med Internet Res %G English %X Background: In patients with chronic disease, many health care professionals are involved during treatment and follow-up. This leads to fragmentation that in turn may lead to suboptimal care. Shared care is a means to improve the integration of care delivered by various providers, specifically primary care physicians (PCPs) and specialty care professionals, for patients with chronic disease. The use of information technology (IT) in this field seems promising. Objective: Our aim was to systematically review the literature regarding the effectiveness of IT-supported shared care interventions in chronic disease in terms of provider or professional, process, health or clinical and financial outcomes. Additionally, our aim was to provide an inventory of the IT applications' characteristics that support such interventions. Methods: PubMed, Scopus, and EMBASE were searched from 2006 to 2015 to identify relevant studies using search terms related to shared care, chronic disease, and IT. Eligible studies were in the English language, and the randomized controlled trials (RCTs), controlled trials, or single group pre-post studies used reported on the effects of IT-supported shared care in patients with chronic disease and cancer. The interventions had to involve providers from both primary and specialty health care. Intervention and IT characteristics and effectiveness—in terms of provider or professional (proximal), process (intermediate), health or clinical and financial (distal) outcomes—were extracted. Risk of bias of (cluster) RCTs was assessed using the Cochrane tool. Results: The initial search yielded 4167 results. Thirteen publications were used, including 11 (cluster) RCTs, a controlled trial, and a pre-post feasibility study. Four main categories of IT applications were identified: (1) electronic decision support tools, (2) electronic platform with a call-center, (3) electronic health records, and (4) electronic communication applications. Positive effects were found for decision support-based interventions on financial and health outcomes, such as physical activity. Electronic health record use improved PCP visits and reduced rehospitalization. Electronic platform use resulted in fewer readmissions and better clinical outcomes—for example, in terms of body mass index (BMI) and dyspnea. The use of electronic communication applications using text-based information transfer between professionals had a positive effect on the number of PCPs contacting hospitals, PCPs’ satisfaction, and confidence. Conclusions: IT-supported shared care can improve proximal outcomes, such as confidence and satisfaction of PCPs, especially in using electronic communication applications. Positive effects on intermediate and distal outcomes were also reported but were mixed. Surprisingly, few studies were found that substantiated these anticipated benefits. Studies showed a large heterogeneity in the included populations, outcome measures, and IT applications used. Therefore, a firm conclusion cannot be drawn. As IT applications are developed and implemented rapidly, evidence is needed to test the specific added value of IT in shared care interventions. This is expected to require innovative research methods. %M 28642218 %R 10.2196/jmir.7405 %U http://www.jmir.org/2017/6/e221/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 2 %P e15 %T Applying STOPP Guidelines in Primary Care Through Electronic Medical Record Decision Support: Randomized Control Trial Highlighting the Importance of Data Quality %+ LEAD LabDepartment of Family Practice, Island Medical ProgramUniversity of British ColumbiaMedical Science Building University of Victoria PO Box 1700 STN CSCVictoria, BC, V8W 2Y2Canada1 25021677091 2504725505morgan@leadlab.ca  %A Price,Morgan %A Davies,Iryna %A Rusk,Raymond %A Lesperance,Mary %A Weber,Jens %K electronic medical records %K clinical decision support %K randomized control trial %K electronic prescribing %K data quality %D 2017 %7 15.06.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: Potentially Inappropriate Prescriptions (PIPs) are a common cause of morbidity, particularly in the elderly. Objective: We sought to understand how the Screening Tool of Older People’s Prescriptions (STOPP) prescribing criteria, implemented in a routinely used primary care Electronic Medical Record (EMR), could impact PIP rates in community (non-academic) primary care practices. Methods: We conducted a mixed-method, pragmatic, cluster, randomized control trial in research naïve primary care practices. Phase 1: In the randomized controlled trial, 40 fully automated STOPP rules were implemented as EMR alerts during a 16-week intervention period. The control group did not receive the 40 STOPP rules (but received other alerts). Participants were recruited through the OSCAR EMR user group mailing list and in person at user group meetings. Results were assessed by querying EMR data PIPs. EMR data quality probes were included. Phase 2: physicians were invited to participate in 1-hour semi-structured interviews to discuss the results. Results: In the EMR, 40 STOPP rules were successfully implemented. Phase 1: A total of 28 physicians from 8 practices were recruited (16 in intervention and 12 in control groups). The calculated PIP rate was 2.6% (138/5308) (control) and 4.11% (768/18,668) (intervention) at baseline. No change in PIPs was observed through the intervention (P=.80). Data quality probes generally showed low use of problem list and medication list. Phase 2: A total of 5 physicians participated. All the participants felt that they were aware of the alerts but commented on workflow and presentation challenges. Conclusions: The calculated PIP rate was markedly less than the expected rate found in literature (2.6% and 4.0% vs 20% in literature). Data quality probes highlighted issues related to completeness of data in areas of the EMR used for PIP reporting and by the decision support such as problem and medication lists. Users also highlighted areas for better integration of STOPP guidelines with prescribing workflows. Many of the STOPP criteria can be implemented in EMRs using simple logic. However, data quality in EMRs continues to be a challenge and was a limiting step in the effectiveness of the decision support in this study. This is important as decision makers continue to fund implementation and adoption of EMRs with the expectation of the use of advanced tools (such as decision support) without ongoing review of data quality and improvement. Trial Registration: Clinicaltrials.gov NCT02130895; https://clinicaltrials.gov/ct2/show/NCT02130895 (Archived by WebCite at http://www.webcitation.org/6qyFigSYT) %M 28619704 %R 10.2196/medinform.6226 %U http://medinform.jmir.org/2017/2/e15/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 5 %P e162 %T Enhancing Comparative Effectiveness Research With Automated Pediatric Pneumonia Detection in a Multi-Institutional Clinical Repository: A PHIS+ Pilot Study %+ Medical University of South Carolina135 Cannon St, 4th FloorCharleston, SC,United States1 843 792 00151 843 792 5587meystre@musc.edu  %A Meystre,Stephane %A Gouripeddi,Ramkiran %A Tieder,Joel %A Simmons,Jeffrey %A Srivastava,Rajendu %A Shah,Samir %K natural language processing %K pneumonia, bacterial %K medical informatics %K comparative effectiveness research %D 2017 %7 15.05.2017 %9 Original Paper %J J Med Internet Res %G English %X Background: Community-acquired pneumonia is a leading cause of pediatric morbidity. Administrative data are often used to conduct comparative effectiveness research (CER) with sufficient sample sizes to enhance detection of important outcomes. However, such studies are prone to misclassification errors because of the variable accuracy of discharge diagnosis codes. Objective: The aim of this study was to develop an automated, scalable, and accurate method to determine the presence or absence of pneumonia in children using chest imaging reports. Methods: The multi-institutional PHIS+ clinical repository was developed to support pediatric CER by expanding an administrative database of children’s hospitals with detailed clinical data. To develop a scalable approach to find patients with bacterial pneumonia more accurately, we developed a Natural Language Processing (NLP) application to extract relevant information from chest diagnostic imaging reports. Domain experts established a reference standard by manually annotating 282 reports to train and then test the NLP application. Findings of pleural effusion, pulmonary infiltrate, and pneumonia were automatically extracted from the reports and then used to automatically classify whether a report was consistent with bacterial pneumonia. Results: Compared with the annotated diagnostic imaging reports reference standard, the most accurate implementation of machine learning algorithms in our NLP application allowed extracting relevant findings with a sensitivity of .939 and a positive predictive value of .925. It allowed classifying reports with a sensitivity of .71, a positive predictive value of .86, and a specificity of .962. When compared with each of the domain experts manually annotating these reports, the NLP application allowed for significantly higher sensitivity (.71 vs .527) and similar positive predictive value and specificity . Conclusions: NLP-based pneumonia information extraction of pediatric diagnostic imaging reports performed better than domain experts in this pilot study. NLP is an efficient method to extract information from a large collection of imaging reports to facilitate CER. %M 28506958 %R 10.2196/jmir.6887 %U http://www.jmir.org/2017/5/e162/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 4 %P e120 %T Use of Machine Learning Classifiers and Sensor Data to Detect Neurological Deficit in Stroke Patients %+ Department of NeurologyYonsei University College of Medicine50 Yonsei-ro, Seodaemoon-guSeoul, 03722Republic Of Korea82 2 2228 161782 2 393 0705hsnam@yuhs.ac  %A Park,Eunjeong %A Chang,Hyuk-Jae %A Nam,Hyo Suk %K medical informatics %K machine learning %K motor %K neurological examination %K stroke %D 2017 %7 18.04.2017 %9 Original Paper %J J Med Internet Res %G English %X Background: The pronator drift test (PDT), a neurological examination, is widely used in clinics to measure motor weakness of stroke patients. Objective: The aim of this study was to develop a PDT tool with machine learning classifiers to detect stroke symptoms based on quantification of proximal arm weakness using inertial sensors and signal processing. Methods: We extracted features of drift and pronation from accelerometer signals of wearable devices on the inner wrists of 16 stroke patients and 10 healthy controls. Signal processing and feature selection approach were applied to discriminate PDT features used to classify stroke patients. A series of machine learning techniques, namely support vector machine (SVM), radial basis function network (RBFN), and random forest (RF), were implemented to discriminate stroke patients from controls with leave-one-out cross-validation. Results: Signal processing by the PDT tool extracted a total of 12 PDT features from sensors. Feature selection abstracted the major attributes from the 12 PDT features to elucidate the dominant characteristics of proximal weakness of stroke patients using machine learning classification. Our proposed PDT classifiers had an area under the receiver operating characteristic curve (AUC) of .806 (SVM), .769 (RBFN), and .900 (RF) without feature selection, and feature selection improves the AUCs to .913 (SVM), .956 (RBFN), and .975 (RF), representing an average performance enhancement of 15.3%. Conclusions: Sensors and machine learning methods can reliably detect stroke signs and quantify proximal arm weakness. Our proposed solution will facilitate pervasive monitoring of stroke patients. %M 28420599 %R 10.2196/jmir.7092 %U http://www.jmir.org/2017/4/e120/ %0 Journal Article %@ 2291-5222 %I JMIR Publications %V 5 %N 3 %P e38 %T mHealth for Clinical Decision-Making in Sub-Saharan Africa: A Scoping Review %+ Athena Institute for Research on Innovation and Communication in Health and Life SciencesVrije Universiteit AmsterdamDe Boelelaan 1105WN Building, Room S-544Amsterdam, 1081HVNetherlands31 20598314331 205987031i.o.adepoju@vu.nl  %A Adepoju,Ibukun-Oluwa Omolade %A Albersen,Bregje Joanna Antonia %A De Brouwere,Vincent %A van Roosmalen,Jos %A Zweekhorst,Marjolein %K mHealth %K decision support systems, clinical %K sub-Saharan Africa %K clinical decision-making %D 2017 %7 23.03.2017 %9 Original Paper %J JMIR Mhealth Uhealth %G English %X Background: In a bid to deliver quality health services in resource-poor settings, mobile health (mHealth) is increasingly being adopted. The role of mHealth in facilitating evidence-based clinical decision-making through data collection, decision algorithms, and evidence-based guidelines, for example, is established in resource-rich settings. However, the extent to which mobile clinical decision support systems (mCDSS) have been adopted specifically in resource-poor settings such as Africa and the lessons learned about their use in such settings are yet to be established. Objective: The aim of this study was to synthesize evidence on the use of mHealth for point-of-care decision support and improved quality of care by health care workers in Africa. Methods: A scoping review of 4 peer-reviewed and 1 grey literature databases was conducted. No date limits were applied, but only articles in English language were selected. Using pre-established criteria, 2 reviewers screened articles and extracted data. Articles were analyzed using Microsoft Excel and MAXQDA. Results: We retained 22 articles representing 11 different studies in 7 sub-Saharan African countries. Interventions were mainly in the domain of maternal health and ranged from simple text messaging (short message service, SMS) to complex multicomponent interventions. Although health workers are generally supportive of mCDSS and perceive them as useful, concerns about increased workload and altered workflow hinder sustainability. Facilitators and barriers to use of mCDSS include technical and infrastructural support, ownership, health system challenges, and training. Conclusions: The use of mCDSS in sub-Saharan Africa is an indication of progress in mHealth, although their effect on quality of service delivery is yet to be fully explored. Lessons learned are useful for informing future research, policy, and practice for technologically supported health care delivery, especially in resource-poor settings. %M 28336504 %R 10.2196/mhealth.7185 %U http://mhealth.jmir.org/2017/3/e38/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 3 %P e54 %T A Learning Health Care System Using Computer-Aided Diagnosis %+ IBM TJ Watson Research Center1101 Kitchawan RdRoute 134Yorktown Heights, NY, 10598United States1 914 945 25901 914 945 2590acahan@us.ibm.com  %A Cahan,Amos %A Cimino,James J %K diagnostic errors %K diagnosis, computer-assisted %K decision support systems, clinical %K pattern recognition, automated %K knowledge bases %K knowledge management %K diagnosis support systems %K crowdsourcing %K structured knowledge representation %D 2017 %7 08.03.2017 %9 Viewpoint %J J Med Internet Res %G English %X Physicians intuitively apply pattern recognition when evaluating a patient. Rational diagnosis making requires that clinical patterns be put in the context of disease prior probability, yet physicians often exhibit flawed probabilistic reasoning. Difficulties in making a diagnosis are reflected in the high rates of deadly and costly diagnostic errors. Introduced 6 decades ago, computerized diagnosis support systems are still not widely used by internists. These systems cannot efficiently recognize patterns and are unable to consider the base rate of potential diagnoses. We review the limitations of current computer-aided diagnosis support systems. We then portray future diagnosis support systems and provide a conceptual framework for their development. We argue for capturing physician knowledge using a novel knowledge representation model of the clinical picture. This model (based on structured patient presentation patterns) holds not only symptoms and signs but also their temporal and semantic interrelations. We call for the collection of crowdsourced, automatically deidentified, structured patient patterns as means to support distributed knowledge accumulation and maintenance. In this approach, each structured patient pattern adds to a self-growing and -maintaining knowledge base, sharing the experience of physicians worldwide. Besides supporting diagnosis by relating the symptoms and signs with the final diagnosis recorded, the collective pattern map can also provide disease base-rate estimates and real-time surveillance for early detection of outbreaks. We explain how health care in resource-limited settings can benefit from using this approach and how it can be applied to provide feedback-rich medical education for both students and practitioners. %M 28274905 %R 10.2196/jmir.6663 %U http://www.jmir.org/2017/3/e54/ %0 Journal Article %@ 2291-5222 %I JMIR Publications %V 5 %N 3 %P e29 %T A Mobile Clinical Decision Support Tool for Pediatric Cardiovascular Risk-Reduction Clinical Practice Guidelines: Development and Description %+ Digital Health and Clinical InformaticsRTI International3040 Cornwallis RdResearch Triangle Park, NC, 27709United States1 919 316 37261 919 541 6621rfurberg@rti.org  %A Furberg,Robert D %A Williams,Pamela %A Bagwell,Jacqueline %A LaBresh,Kenneth %K pediatrics %K cardiovascular risk reduction %K mHealth %K clinical decision support %K clinical practice guidelines %D 2017 %7 07.03.2017 %9 Original Paper %J JMIR Mhealth Uhealth %G English %X Background: Widespread application of research findings to improve patient outcomes remains inadequate, and failure to routinely translate research findings into daily clinical practice is a major barrier for the implementation of any evidence-based guideline. Strategies to increase guideline uptake in primary care pediatric practices and to facilitate adherence to recommendations are required. Objective: Our objective was to operationalize the US National Heart, Lung, and Blood Institute’s Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents into a mobile clinical decision support (CDS) system for healthcare providers, and to describe the process development and outcomes. Methods: To overcome the difficulty of translating clinical practice guidelines into a computable form that can be used by a CDS system, we used a multilayer framework to convert the evidence synthesis into executable knowledge. We used an iterative process of design, testing, and revision through each step in the translation of the guidelines for use in a CDS tool to support the development of 4 validated modules: an integrated risk assessment; a blood pressure calculator; a body mass index calculator; and a lipid management instrument. Results: The iterative revision process identified several opportunities to improve the CDS tool. Operationalizing the integrated guideline identified numerous areas in which the guideline was vague or incorrect and required more explicit operationalization. Iterative revisions led to workable solutions to problems and understanding of the limitations of the tool. Conclusions: The process and experiences described provide a model for other mobile CDS systems that translate written clinical practice guidelines into actionable, real-time clinical recommendations. %M 28270384 %R 10.2196/mhealth.6291 %U http://mhealth.jmir.org/2017/3/e29/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 3 %P e69 %T Exacerbations in Chronic Obstructive Pulmonary Disease: Identification and Prediction Using a Digital Health System %+ Nuffield Department of Clinical NeurosciencesUniversity of OxfordLevel 6, West Wing, John Radcliffe HospitalOxford,United Kingdom44 1865 23476444 1865 273010syed.shah@ndcn.ox.ac.uk  %A Shah,Syed Ahmar %A Velardo,Carmelo %A Farmer,Andrew %A Tarassenko,Lionel %K COPD %K disease exacerbation %K mobile health %K self-management %K pulse oximetry %K respiratory rate %K clinical prediction rule %K algorithms %D 2017 %7 07.03.2017 %9 Original Paper %J J Med Internet Res %G English %X Background: Chronic obstructive pulmonary disease (COPD) is a progressive, chronic respiratory disease with a significant socioeconomic burden. Exacerbations, the sudden and sustained worsening of symptoms, can lead to hospitalization and reduce quality of life. Major limitations of previous telemonitoring interventions for COPD include low compliance, lack of consensus on what constitutes an exacerbation, limited numbers of patients, and short monitoring periods. We developed a telemonitoring system based on a digital health platform that was used to collect data from the 1-year EDGE (Self Management and Support Programme) COPD clinical trial aiming at daily monitoring in a heterogeneous group of patients with moderate to severe COPD. Objective: The objectives of the study were as follows: first, to develop a systematic and reproducible approach to exacerbation identification and to track the progression of patient condition during remote monitoring; and second, to develop a robust algorithm able to predict COPD exacerbation, based on vital signs acquired from a pulse oximeter. Methods: We used data from 110 patients, with a combined monitoring period of more than 35,000 days. We propose a finite-state machine–based approach for modeling COPD exacerbation to gain a deeper insight into COPD patient condition during home monitoring to take account of the time course of symptoms. A robust algorithm based on short-period trend analysis and logistic regression using vital signs derived from a pulse oximeter is also developed to predict exacerbations. Results: On the basis of 27,260 sessions recorded during the clinical trial (average usage of 5.3 times per week for 12 months), there were 361 exacerbation events. There was considerable variation in the length of exacerbation events, with a mean length of 8.8 days. The mean value of oxygen saturation was lower, and both the pulse rate and respiratory rate were higher before an impending exacerbation episode, compared with stable periods. On the basis of the classifier developed in this work, prediction of COPD exacerbation episodes with 60%-80% sensitivity will result in 68%-36% specificity. Conclusions: All 3 vital signs acquired from a pulse oximeter (pulse rate, oxygen saturation, and respiratory rate) are predictive of COPD exacerbation events, with oxygen saturation being the most predictive, followed by respiratory rate and pulse rate. Combination of these vital signs with a robust algorithm based on machine learning leads to further improvement in positive predictive accuracy. Trial Registration: International Standard Randomized Controlled Trial Number (ISRCTN): 40367841; http://www.isrctn.com/ISRCTN40367841 (Archived by WebCite at http://www.webcitation.org/6olpMWNpc) %M 28270380 %R 10.2196/jmir.7207 %U http://www.jmir.org/2017/3/e69/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 1 %P e7 %T Patient Similarity in Prediction Models Based on Health Data: A Scoping Review %+ Health Data Science LabSchool of Public Health and Health SystemsUniversity of Waterloo200 University Avenue WestLyle Hallman NorthWaterloo, ON, N2L 3G1Canada1 519 888 4567 ext 315671 519 746 6776joon.lee@uwaterloo.ca  %A Sharafoddini,Anis %A Dubin,Joel A %A Lee,Joon %K patient similarity %K predictive modeling %K health data %K medical records %K electronic health records %K personalized medicine %K data-driven prediction %K review %D 2017 %7 03.03.2017 %9 Review %J JMIR Med Inform %G English %X Background: Physicians and health policy makers are required to make predictions during their decision making in various medical problems. Many advances have been made in predictive modeling toward outcome prediction, but these innovations target an average patient and are insufficiently adjustable for individual patients. One developing idea in this field is individualized predictive analytics based on patient similarity. The goal of this approach is to identify patients who are similar to an index patient and derive insights from the records of similar patients to provide personalized predictions.. Objective: The aim is to summarize and review published studies describing computer-based approaches for predicting patients’ future health status based on health data and patient similarity, identify gaps, and provide a starting point for related future research. Methods: The method involved (1) conducting the review by performing automated searches in Scopus, PubMed, and ISI Web of Science, selecting relevant studies by first screening titles and abstracts then analyzing full-texts, and (2) documenting by extracting publication details and information on context, predictors, missing data, modeling algorithm, outcome, and evaluation methods into a matrix table, synthesizing data, and reporting results. Results: After duplicate removal, 1339 articles were screened in abstracts and titles and 67 were selected for full-text review. In total, 22 articles met the inclusion criteria. Within included articles, hospitals were the main source of data (n=10). Cardiovascular disease (n=7) and diabetes (n=4) were the dominant patient diseases. Most studies (n=18) used neighborhood-based approaches in devising prediction models. Two studies showed that patient similarity-based modeling outperformed population-based predictive methods. Conclusions: Interest in patient similarity-based predictive modeling for diagnosis and prognosis has been growing. In addition to raw/coded health data, wavelet transform and term frequency-inverse document frequency methods were employed to extract predictors. Selecting predictors with potential to highlight special cases and defining new patient similarity metrics were among the gaps identified in the existing literature that provide starting points for future work. Patient status prediction models based on patient similarity and health data offer exciting potential for personalizing and ultimately improving health care, leading to better patient outcomes. %M 28258046 %R 10.2196/medinform.6730 %U http://medinform.jmir.org/2017/1/e7/ %0 Journal Article %@ 1438-8871 %I JMIR Publications %V 19 %N 1 %P e25 %T Ecological Assessment of Clinicians’ Antipsychotic Prescription Habits in Psychiatric Inpatients: A Novel Web- and Mobile Phone–Based Prototype for a Dynamic Clinical Decision Support System %+ Department of PsychiatryBrest Medical University Hospital at BrestUrgences Psychiatrques CHRU BrestBrest, 29200France33 66820417833 298015218sofian.berrouiguet@gmail.com  %A Berrouiguet,Sofian %A Barrigón,Maria Luisa %A Brandt,Sara A %A Nitzburg,George C %A Ovejero,Santiago %A Alvarez-Garcia,Raquel %A Carballo,Juan %A Walter,Michel %A Billot,Romain %A Lenca,Philippe %A Delgado-Gomez,David %A Ropars,Juliette %A de la Calle Gonzalez,Ivan %A Courtet,Philippe %A Baca-García,Enrique %K clinical decision-making %K antipsychotic agents %K software %K mobile applications %K off-label use %K prescriptions %D 2017 %7 26.01.2017 %9 Original Paper %J J Med Internet Res %G English %X Background: Electronic prescribing devices with clinical decision support systems (CDSSs) hold the potential to significantly improve pharmacological treatment management. Objective: The aim of our study was to develop a novel Web- and mobile phone–based application to provide a dynamic CDSS by monitoring and analyzing practitioners’ antipsychotic prescription habits and simultaneously linking these data to inpatients’ symptom changes. Methods: We recruited 353 psychiatric inpatients whose symptom levels and prescribed medications were inputted into the MEmind application. We standardized all medications in the MEmind database using the Anatomical Therapeutic Chemical (ATC) classification system and the defined daily dose (DDD). For each patient, MEmind calculated an average for the daily dose prescribed for antipsychotics (using the N05A ATC code), prescribed daily dose (PDD), and the PDD to DDD ratio. Results: MEmind results found that antipsychotics were used by 61.5% (217/353) of inpatients, with the largest proportion being patients with schizophrenia spectrum disorders (33.4%, 118/353). Of the 217 patients, 137 (63.2%, 137/217) were administered pharmacological monotherapy and 80 (36.8%, 80/217) were administered polytherapy. Antipsychotics were used mostly in schizophrenia spectrum and related psychotic disorders, but they were also prescribed in other nonpsychotic diagnoses. Notably, we observed polypharmacy going against current antipsychotics guidelines. Conclusions: MEmind data indicated that antipsychotic polypharmacy and off-label use in inpatient units is commonly practiced. MEmind holds the potential to create a dynamic CDSS that provides real-time tracking of prescription practices and symptom change. Such feedback can help practitioners determine a maximally therapeutic drug treatment while avoiding unproductive overprescription and off-label use. %M 28126703 %R 10.2196/jmir.5954 %U http://www.jmir.org/2017/1/e25/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 5 %N 1 %P e3 %T Patient-Specific Predictive Modeling Using Random Forests: An Observational Study for the Critically Ill %+ Health Data Science LabSchool of Public Health and Health SystemsUniversity of Waterloo200 University Ave WWaterloo, ON, N2L 3G1Canada1 519 888 4567 ext 315671 519 746 6776joon.lee@uwaterloo.ca  %A Lee,Joon %K forecasting %K critical care %K predictive analytics %K patient similarity %K random forest %D 2017 %7 17.01.2017 %9 Original Paper %J JMIR Med Inform %G English %X Background: With a large-scale electronic health record repository, it is feasible to build a customized patient outcome prediction model specifically for a given patient. This approach involves identifying past patients who are similar to the present patient and using their data to train a personalized predictive model. Our previous work investigated a cosine-similarity patient similarity metric (PSM) for such patient-specific predictive modeling. Objective: The objective of the study is to investigate the random forest (RF) proximity measure as a PSM in the context of personalized mortality prediction for intensive care unit (ICU) patients. Methods: A total of 17,152 ICU admissions were extracted from the Multiparameter Intelligent Monitoring in Intensive Care II database. A number of predictor variables were extracted from the first 24 hours in the ICU. Outcome to be predicted was 30-day mortality. A patient-specific predictive model was trained for each ICU admission using an RF PSM inspired by the RF proximity measure. Death counting, logistic regression, decision tree, and RF models were studied with a hard threshold applied to RF PSM values to only include the M most similar patients in model training, where M was varied. In addition, case-specific random forests (CSRFs), which uses RF proximity for weighted bootstrapping, were trained. Results: Compared to our previous study that investigated a cosine similarity PSM, the RF PSM resulted in superior or comparable predictive performance. RF and CSRF exhibited the best performances (in terms of mean area under the receiver operating characteristic curve [95% confidence interval], RF: 0.839 [0.835-0.844]; CSRF: 0.832 [0.821-0.843]). RF and CSRF did not benefit from personalization via the use of the RF PSM, while the other models did. Conclusions: The RF PSM led to good mortality prediction performance for several predictive models, although it failed to induce improved performance in RF and CSRF. The distinction between predictor and similarity variables is an important issue arising from the present study. RFs present a promising method for patient-specific outcome prediction. %M 28096065 %R 10.2196/medinform.6690 %U http://medinform.jmir.org/2017/1/e3/ %0 Journal Article %@ 2369-3762 %I JMIR Publications %V 2 %N 2 %P e16 %T Information and Communication Technologies for the Dissemination of Clinical Practice Guidelines to Health Professionals: A Systematic Review %+ School of Rehabilitation SciencesFaculty of Health SciencesUniversity of Ottawa451 Smyth RoadOttawa, ON, K1H 8M5Canada1 61356258001 6135625428gdean053@uottawa.ca  %A De Angelis,Gino %A Davies,Barbara %A King,Judy %A McEwan,Jessica %A Cavallo,Sabrina %A Loew,Laurianne %A Wells,George A %A Brosseau,Lucie %K health information technologies %K electronic mail %K email %K Web 2.0 %K practice guidelines %K health professions %K information dissemination %D 2016 %7 30.11.2016 %9 Review %J JMIR Med Educ %G English %X Background: The transfer of research knowledge into clinical practice can be a continuous challenge for researchers. Information and communication technologies, such as websites and email, have emerged as popular tools for the dissemination of evidence to health professionals. Objective: The objective of this systematic review was to identify research on health professionals’ perceived usability and practice behavior change of information and communication technologies for the dissemination of clinical practice guidelines. Methods: We used a systematic approach to retrieve and extract data about relevant studies. We identified 2248 citations, of which 21 studies met criteria for inclusion; 20 studies were randomized controlled trials, and 1 was a controlled clinical trial. The following information and communication technologies were evaluated: websites (5 studies), computer software (3 studies), Web-based workshops (2 studies), computerized decision support systems (2 studies), electronic educational game (1 study), email (2 studies), and multifaceted interventions that consisted of at least one information and communication technology component (6 studies). Results: Website studies demonstrated significant improvements in perceived usefulness and perceived ease of use, but not for knowledge, reducing barriers, and intention to use clinical practice guidelines. Computer software studies demonstrated significant improvements in perceived usefulness, but not for knowledge and skills. Web-based workshop and email studies demonstrated significant improvements in knowledge, perceived usefulness, and skills. An electronic educational game intervention demonstrated a significant improvement from baseline in knowledge after 12 and 24 weeks. Computerized decision support system studies demonstrated variable findings for improvement in skills. Multifaceted interventions demonstrated significant improvements in beliefs about capabilities, perceived usefulness, and intention to use clinical practice guidelines, but variable findings for improvements in skills. Most multifaceted studies demonstrated significant improvements in knowledge. Conclusions: The findings suggest that health professionals’ perceived usability and practice behavior change vary by type of information and communication technology. Heterogeneity and the paucity of properly conducted studies did not allow for a clear comparison between studies and a conclusion on the effectiveness of information and communication technologies as a knowledge translation strategy for the dissemination of clinical practice guidelines. %M 27903488 %R 10.2196/mededu.6288 %U http://mededu.jmir.org/2016/2/e16/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 4 %N 4 %P e36 %T Increasing Complexity in Rule-Based Clinical Decision Support: The Symptom Assessment and Management Intervention %+ School of MedicineDepartment of Community & Family MedicineDuke University Medical CenterDuke UniversityBox 3886Durham, NC, 22710United States1 919 438 23461 919 681 7085David.Lobach@klesishealthcare.com  %A Lobach,David F %A Johns,Ellis B %A Halpenny,Barbara %A Saunders,Toni-Ann %A Brzozowski,Jane %A Del Fiol,Guilherme %A Berry,Donna L %A Braun,Ilana M %A Finn,Kathleen %A Wolfe,Joanne %A Abrahm,Janet L %A Cooley,Mary E %K rule-based clinical decision support %K clinical algorithms %K Web services %K software as a service %K symptom management %K patient-reported outcomes %K lung cancer %D 2016 %7 08.11.2016 %9 Original Paper %J JMIR Med Inform %G English %X Background: Management of uncontrolled symptoms is an important component of quality cancer care. Clinical guidelines are available for optimal symptom management, but are not often integrated into the front lines of care. The use of clinical decision support (CDS) at the point-of-care is an innovative way to incorporate guideline-based symptom management into routine cancer care. Objective: The objective of this study was to develop and evaluate a rule-based CDS system to enable management of multiple symptoms in lung cancer patients at the point-of-care. Methods: This study was conducted in three phases involving a formative evaluation, a system evaluation, and a contextual evaluation of clinical use. In Phase 1, we conducted iterative usability testing of user interface prototypes with patients and health care providers (HCPs) in two thoracic oncology clinics. In Phase 2, we programmed complex algorithms derived from clinical practice guidelines into a rules engine that used Web services to communicate with the end-user application. Unit testing of algorithms was conducted using a stack-traversal tree-spanning methodology to identify all possible permutations of pathways through each algorithm, to validate accuracy. In Phase 3, we evaluated clinical use of the system among patients and HCPs in the two clinics via observations, structured interviews, and questionnaires. Results: In Phase 1, 13 patients and 5 HCPs engaged in two rounds of formative testing, and suggested improvements leading to revisions until overall usability scores met a priori benchmarks. In Phase 2, symptom management algorithms contained between 29 and 1425 decision nodes, resulting in 19 to 3194 unique pathways per algorithm. Unit testing required 240 person-hours, and integration testing required 40 person-hours. In Phase 3, both patients and HCPs found the system usable and acceptable, and offered suggestions for improvements. Conclusions: A rule-based CDS system for complex symptom management was systematically developed and tested. The complexity of the algorithms required extensive development and innovative testing. The Web service-based approach allowed remote access to CDS knowledge, and could enable scaling and sharing of this knowledge to accelerate availability, and reduce duplication of effort. Patients and HCPs found the system to be usable and useful. %M 27826132 %R 10.2196/medinform.5728 %U http://medinform.jmir.org/2016/4/e36/ %0 Journal Article %@ 2369-2960 %I JMIR Publications %V 2 %N 2 %P e157 %T IBM Watson Analytics: Automating Visualization, Descriptive, and Predictive Statistics %+ College of HealthDepartment of Health Sciences and AdministrationUniversity of West Florida11000 University ParkwayPensacola, FL, 32514United States1 85038452351 850 474 2173rhoyt@uwf.edu  %A Hoyt,Robert Eugene %A Snider,Dallas %A Thompson,Carla %A Mantravadi,Sarita %K data analysis %K data mining %K machine learning %K statistical data analysis %K natural language processing %D 2016 %7 11.10.2016 %9 Original Paper %J JMIR Public Health Surveill %G English %X Background: We live in an era of explosive data generation that will continue to grow and involve all industries. One of the results of this explosion is the need for newer and more efficient data analytics procedures. Traditionally, data analytics required a substantial background in statistics and computer science. In 2015, International Business Machines Corporation (IBM) released the IBM Watson Analytics (IBMWA) software that delivered advanced statistical procedures based on the Statistical Package for the Social Sciences (SPSS). The latest entry of Watson Analytics into the field of analytical software products provides users with enhanced functions that are not available in many existing programs. For example, Watson Analytics automatically analyzes datasets, examines data quality, and determines the optimal statistical approach. Users can request exploratory, predictive, and visual analytics. Using natural language processing (NLP), users are able to submit additional questions for analyses in a quick response format. This analytical package is available free to academic institutions (faculty and students) that plan to use the tools for noncommercial purposes. Objective: To report the features of IBMWA and discuss how this software subjectively and objectively compares to other data mining programs. Methods: The salient features of the IBMWA program were examined and compared with other common analytical platforms, using validated health datasets. Results: Using a validated dataset, IBMWA delivered similar predictions compared with several commercial and open source data mining software applications. The visual analytics generated by IBMWA were similar to results from programs such as Microsoft Excel and Tableau Software. In addition, assistance with data preprocessing and data exploration was an inherent component of the IBMWA application. Sensitivity and specificity were not included in the IBMWA predictive analytics results, nor were odds ratios, confidence intervals, or a confusion matrix. Conclusions: IBMWA is a new alternative for data analytics software that automates descriptive, predictive, and visual analytics. This program is very user-friendly but requires data preprocessing, statistical conceptual understanding, and domain expertise. %M 27729304 %R 10.2196/publichealth.5810 %U http://publichealth.jmir.org/2016/2/e157/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 4 %N 3 %P e28 %T Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data: A Machine Learning Approach %+ Dascena, Inc1135 Martin Luther King DriveHayward, CA, 94541United States1 (872) 228 53321 (872) 228 5332jana@dascena.com  %A Desautels,Thomas %A Calvert,Jacob %A Hoffman,Jana %A Jay,Melissa %A Kerem,Yaniv %A Shieh,Lisa %A Shimabukuro,David %A Chettipally,Uli %A Feldman,Mitchell D %A Barton,Chris %A Wales,David J %A Das,Ritankar %K sepsis %K machine learning %K clinical decision support systems %K electronic health records %K medical informatics %D 2016 %7 30.09.2016 %9 Original Paper %J JMIR Med Inform %G English %X Background: Sepsis is one of the leading causes of mortality in hospitalized patients. Despite this fact, a reliable means of predicting sepsis onset remains elusive. Early and accurate sepsis onset predictions could allow more aggressive and targeted therapy while maintaining antimicrobial stewardship. Existing detection methods suffer from low performance and often require time-consuming laboratory test results. Objective: To study and validate a sepsis prediction method, InSight, for the new Sepsis-3 definitions in retrospective data, make predictions using a minimal set of variables from within the electronic health record data, compare the performance of this approach with existing scoring systems, and investigate the effects of data sparsity on InSight performance. Methods: We apply InSight, a machine learning classification system that uses multivariable combinations of easily obtained patient data (vitals, peripheral capillary oxygen saturation, Glasgow Coma Score, and age), to predict sepsis using the retrospective Multiparameter Intelligent Monitoring in Intensive Care (MIMIC)-III dataset, restricted to intensive care unit (ICU) patients aged 15 years or more. Following the Sepsis-3 definitions of the sepsis syndrome, we compare the classification performance of InSight versus quick sequential organ failure assessment (qSOFA), modified early warning score (MEWS), systemic inflammatory response syndrome (SIRS), simplified acute physiology score (SAPS) II, and sequential organ failure assessment (SOFA) to determine whether or not patients will become septic at a fixed period of time before onset. We also test the robustness of the InSight system to random deletion of individual input observations. Results: In a test dataset with 11.3% sepsis prevalence, InSight produced superior classification performance compared with the alternative scores as measured by area under the receiver operating characteristic curves (AUROC) and area under precision-recall curves (APR). In detection of sepsis onset, InSight attains AUROC = 0.880 (SD 0.006) at onset time and APR = 0.595 (SD 0.016), both of which are superior to the performance attained by SIRS (AUROC: 0.609; APR: 0.160), qSOFA (AUROC: 0.772; APR: 0.277), and MEWS (AUROC: 0.803; APR: 0.327) computed concurrently, as well as SAPS II (AUROC: 0.700; APR: 0.225) and SOFA (AUROC: 0.725; APR: 0.284) computed at admission (P<.001 for all comparisons). Similar results are observed for 1-4 hours preceding sepsis onset. In experiments where approximately 60% of input data are deleted at random, InSight attains an AUROC of 0.781 (SD 0.013) and APR of 0.401 (SD 0.015) at sepsis onset time. Even with 60% of data missing, InSight remains superior to the corresponding SIRS scores (AUROC and APR, P<.001), qSOFA scores (P=.0095; P<.001) and superior to SOFA and SAPS II computed at admission (AUROC and APR, P<.001), where all of these comparison scores (except InSight) are computed without data deletion. Conclusions: Despite using little more than vitals, InSight is an effective tool for predicting sepsis onset and performs well even with randomly missing data. %M 27694098 %R 10.2196/medinform.5909 %U http://medinform.jmir.org/2016/3/e28/ %0 Journal Article %@ 2368-7959 %I JMIR Publications Inc. %V 3 %N 3 %P e19 %T Predicting Risk of Suicide Attempt Using History of Physical Illnesses From Electronic Medical Records %+ Centre for Pattern Recognition and Data AnalyticsDeakin UniversityWaurn PondsGeelong,Australia61 35227307961 352273079karmakar@deakin.edu.au  %A Karmakar,Chandan %A Luo,Wei %A Tran,Truyen %A Berk,Michael %A Venkatesh,Svetha %K suicide risk %K electronic medical record %K history of physical illnesses %K ICD-10 codes %K suicide risk prediction model %D 2016 %7 11.07.2016 %9 Original Paper %J JMIR Mental Health %G English %X Background: Although physical illnesses, routinely documented in electronic medical records (EMR), have been found to be a contributing factor to suicides, no automated systems use this information to predict suicide risk. Objective: The aim of this study is to quantify the impact of physical illnesses on suicide risk, and develop a predictive model that captures this relationship using EMR data. Methods: We used history of physical illnesses (except chapter V: Mental and behavioral disorders) from EMR data over different time-periods to build a lookup table that contains the probability of suicide risk for each chapter of the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) codes. The lookup table was then used to predict the probability of suicide risk for any new assessment. Based on the different lengths of history of physical illnesses, we developed six different models to predict suicide risk. We tested the performance of developed models to predict 90-day risk using historical data over differing time-periods ranging from 3 to 48 months. A total of 16,858 assessments from 7399 mental health patients with at least one risk assessment was used for the validation of the developed model. The performance was measured using area under the receiver operating characteristic curve (AUC). Results: The best predictive results were derived (AUC=0.71) using combined data across all time-periods, which significantly outperformed the clinical baseline derived from routine risk assessment (AUC=0.56). The proposed approach thus shows potential to be incorporated in the broader risk assessment processes used by clinicians. Conclusions: This study provides a novel approach to exploit the history of physical illnesses extracted from EMR (ICD-10 codes without chapter V-mental and behavioral disorders) to predict suicide risk, and this model outperforms existing clinical assessments of suicide risk. %M 27400764 %R 10.2196/mental.5475 %U http://mental.jmir.org/2016/3/e19/ %0 Journal Article %@ 2291-9694 %I JMIR Publications %V 4 %N 3 %P e23 %T Evaluation of an Expert System for the Generation of Speech and Language Therapy Plans %+ Grupo de Investigación en Inteligencia Artificial y Tecnologías de AsistenciaUniversidad Politécnica SalesianaCalle Vieja, 12-30Elia LiutCuenca, 010102Ecuador593 7862213 ext 1278593 7862213vrobles@ups.edu.ec  %A Robles-Bykbaev,Vladimir %A López-Nores,Martín %A García-Duque,Jorge %A Pazos-Arias,José J %A Arévalo-Lucero,Daysi %K speech-language pathology %K rehabilitation of speech and language disorders %K decision support systems, clinical %K expert systems %D 2016 %7 01.07.2016 %9 Original Paper %J JMIR Med Inform %G English %X Background: Speech and language pathologists (SLPs) deal with a wide spectrum of disorders, arising from many different conditions, that affect voice, speech, language, and swallowing capabilities in different ways. Therefore, the outcomes of Speech and Language Therapy (SLT) are highly dependent on the accurate, consistent, and complete design of personalized therapy plans. However, SLPs often have very limited time to work with their patients and to browse the large (and growing) catalogue of activities and specific exercises that can be put into therapy plans. As a consequence, many plans are suboptimal and fail to address the specific needs of each patient. Objective: We aimed to evaluate an expert system that automatically generates plans for speech and language therapy, containing semiannual activities in the five areas of hearing, oral structure and function, linguistic formulation, expressive language and articulation, and receptive language. The goal was to assess whether the expert system speeds up the SLPs’ work and leads to more accurate, consistent, and complete therapy plans for their patients. Methods: We examined the evaluation results of the SPELTA expert system in supporting the decision making of 4 SLPs treating children in three special education institutions in Ecuador. The expert system was first trained with data from 117 cases, including medical data; diagnosis for voice, speech, language and swallowing capabilities; and therapy plans created manually by the SLPs. It was then used to automatically generate new therapy plans for 13 new patients. The SLPs were finally asked to evaluate the accuracy, consistency, and completeness of those plans. A four-fold cross-validation experiment was also run on the original corpus of 117 cases in order to assess the significance of the results. Results: The evaluation showed that 87% of the outputs provided by the SPELTA expert system were considered valid therapy plans for the different areas. The SLPs rated the overall accuracy, consistency, and completeness of the proposed activities with 4.65, 4.6, and 4.6 points (to a maximum of 5), respectively. The ratings for the subplans generated for the areas of hearing, oral structure and function, and linguistic formulation were nearly perfect, whereas the subplans for expressive language and articulation and for receptive language failed to deal properly with some of the subject cases. Overall, the SLPs indicated that over 90% of the subplans generated automatically were “better than” or “as good as” what the SLPs would have created manually if given the average time they can devote to the task. The cross-validation experiment yielded very similar results. Conclusions: The results show that the SPELTA expert system provides valuable input for SLPs to design proper therapy plans for their patients, in a shorter time and considering a larger set of activities than proceeding manually. The algorithms worked well even in the presence of a sparse corpus, and the evidence suggests that the system will become more reliable as it is trained with more subjects. %M 27370070 %R 10.2196/medinform.5660 %U http://medinform.jmir.org/2016/3/e23/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 4 %N 2 %P e16 %T Putting Meaning into Meaningful Use: A Roadmap to Successful Integration of Evidence at the Point of Care %+ Hofstra North Shore LII School of Medicine4th Floor300 Community DriveManhasset, NY, 11030United States1 516 562 43101 516 562 2526tmcginn@nshs.edu  %A McGinn,Thomas %K clinical decision support tools %K framework %K implementation %D 2016 %7 19.05.2016 %9 Viewpoint %J JMIR Med Inform %G English %X Pressures to contain health care costs, personalize patient care, use big data, and to enhance health care quality have highlighted the need for integration of evidence at the point of care. The application of evidence-based medicine (EBM) has great promise in the era of electronic health records (EHRs) and health technology. The most successful integration of evidence into EHRs has been complex decision tools that trigger at a critical point of the clinical visit and include patient specific recommendations. The objective of this viewpoint paper is to investigate why the incorporation of complex CDS tools into the EMR is equally complex and continues to challenge health service researchers and implementation scientists. Poor adoption and sustainability of EBM guidelines and CDS tools at the point of care have persisted and continue to document low rates of usage. The barriers cited by physicians include efficiency, perception of usefulness, information content, user interface, and over-triggering. Building on the traditional EHR implementation frameworks, we review keys strategies for successful CDSs: (1) the quality of the evidence, (2) the potential to reduce unnecessary care, (3) ease of integrating evidence at the point of care, (4) the evidence’s consistency with clinician perceptions and preferences, (5) incorporating bundled sets or automated documentation, and (6) shared decision making tools. As EHRs become commonplace and insurers demand higher quality and evidence-based care, better methods for integrating evidence into everyday care are warranted. We have outlined basic criteria that should be considered before attempting to integrate evidenced-based decision support tools into the EHR. %M 27199223 %R 10.2196/medinform.4553 %U http://medinform.jmir.org/2016/2/e16/ %0 Journal Article %@ 1929-0748 %I JMIR Publications Inc. %V 5 %N 1 %P e41 %T Predicting Appropriate Admission of Bronchiolitis Patients in the Emergency Department: Rationale and Methods %+ School of MedicineDepartment of Biomedical InformaticsUniversity of Utah421 Wakara Way, Suite 140Salt Lake City, UT, 84108United States1 801 213 35651 801 581 4297gangluo@cs.wisc.edu  %A Luo,Gang %A Stone,Bryan L %A Johnson,Michael D %A Nkoy,Flory L %K Decision support techniques %K forecasting %K computer simulation %K machine learning %D 2016 %7 07.03.2016 %9 Protocol %J JMIR Res Protoc %G English %X Background: In young children, bronchiolitis is the most common illness resulting in hospitalization. For children less than age 2, bronchiolitis incurs an annual total inpatient cost of $1.73 billion. Each year in the United States, 287,000 emergency department (ED) visits occur because of bronchiolitis, with a hospital admission rate of 32%-40%. Due to a lack of evidence and objective criteria for managing bronchiolitis, ED disposition decisions (hospital admission or discharge to home) are often made subjectively, resulting in significant practice variation. Studies reviewing admission need suggest that up to 29% of admissions from the ED are unnecessary. About 6% of ED discharges for bronchiolitis result in ED returns with admission. These inappropriate dispositions waste limited health care resources, increase patient and parental distress, expose patients to iatrogenic risks, and worsen outcomes. Existing clinical guidelines for bronchiolitis offer limited improvement in patient outcomes. Methodological shortcomings include that the guidelines provide no specific thresholds for ED decisions to admit or to discharge, have an insufficient level of detail, and do not account for differences in patient and illness characteristics including co-morbidities. Predictive models are frequently used to complement clinical guidelines, reduce practice variation, and improve clinicians’ decision making. Used in real time, predictive models can present objective criteria supported by historical data for an individualized disease management plan and guide admission decisions. However, existing predictive models for ED patients with bronchiolitis have limitations, including low accuracy and the assumption that the actual ED disposition decision was appropriate. To date, no operational definition of appropriate admission exists. No model has been built based on appropriate admissions, which include both actual admissions that were necessary and actual ED discharges that were unsafe. Objective: The goal of this study is to develop a predictive model to guide appropriate hospital admission for ED patients with bronchiolitis. Methods: This study will: (1) develop an operational definition of appropriate hospital admission for ED patients with bronchiolitis, (2) develop and test the accuracy of a new model to predict appropriate hospital admission for an ED patient with bronchiolitis, and (3) conduct simulations to estimate the impact of using the model on bronchiolitis outcomes. Results: We are currently extracting administrative and clinical data from the enterprise data warehouse of an integrated health care system. Our goal is to finish this study by the end of 2019. Conclusions: This study will produce a new predictive model that can be operationalized to guide and improve disposition decisions for ED patients with bronchiolitis. Broad use of the model would reduce iatrogenic risk, patient and parental distress, health care use, and costs and improve outcomes for bronchiolitis patients. %M 26952700 %R 10.2196/resprot.5155 %U http://www.researchprotocols.org/2016/1/e41/ %0 Journal Article %@ 1929-0748 %I JMIR Publications Inc. %V 5 %N 1 %P e24 %T Development of a Decision Aid for Cardiopulmonary Resuscitation Involving Intensive Care Unit Patients' and Health Professionals' Participation Using User-Centered Design and a Wiki Platform for Rapid Prototyping: A Research Protocol %+ Centre de recherche du Centre hospitalier affilié universitaire Hôtel-Dieu de Lévis143 rue WolfeLévis, QC, G6V 2Y1Canada1 41890330581 4189033058arianeplaisance@gmail.com  %A Plaisance,Ariane %A Witteman,Holly O %A Heyland,Daren Keith %A Ebell,Mark H %A Dupuis,Audrey %A Lavoie-Bérard,Carole-Anne %A Légaré,France %A Archambault,Patrick Michel %K cardiopulmonary resuscitation %K end-of-life planning %K goals of care discussions %K intensive care medicine %K medical informatics %K shared decision making %K user-centered design %K wikis %D 2016 %7 11.02.2016 %9 Protocol %J JMIR Res Protoc %G English %X Background: Cardiopulmonary resuscitation (CPR) is an intervention used in cases of cardiac arrest to revive patients whose heart has stopped. Because cardiac arrest can have potentially devastating outcomes such as severe neurological deficits even if CPR is performed, patients must be involved in determining in advance if they want CPR in the case of an unexpected arrest. Shared decision making (SDM) facilitates discussions about goals of care regarding CPR in intensive care units (ICUs). Patient decision aids (DAs) are proven to support the implementation of SDM. Many patient DAs about CPR exist, but they are not universally implemented in ICUs in part due to lack of context and cultural adaptation. Adaptation to local context is an important phase of implementing any type of knowledge tool such as patient DAs. User-centered design supported by a wiki platform to perform rapid prototyping has previously been successful in creating knowledge tools adapted to the needs of patients and health professionals (eg, asthma action plans). This project aims to explore how user-centered design and a wiki platform can support the adaptation of an existing DA for CPR to the local context. Objective: The primary objective is to use an existing DA about CPR to create a wiki-based DA that is adapted to the context of a single ICU and tailorable to individual patient’s risk factors while employing user-centered design. The secondary objective is to document the use of a wiki platform for the adaptation of patient DAs. Methods: This study will be conducted in a mixed surgical and medical ICU at Hôtel-Dieu de Lévis, Quebec, Canada. We plan to involve all 5 intensivists and recruit at least 20 alert and oriented patients admitted to the ICU and their family members if available. In the first phase of this study, we will observe 3 weeks of daily interactions between patients, families, intensivists, and other allied health professionals. We will specifically observe 5 dyads of attending intensivists and alert and oriented patients discussing goals of care concerning CPR to understand how a patient DA could support this decision. We will also conduct individual interviews with the 5 intensivists to identify their needs concerning the implementation of a DA. In the second phase of the study, we will build a first prototype based on the needs identified in Phase I. We will start by translating an existing DA entitled “Cardiopulmonary resuscitation: a decision aid for patients and their families.” We will then adapt this tool to the needs we identified in Phase I and archive this first prototype in a wiki. Building on the wiki’s programming architecture, we intend to integrate the Good Outcome Following Attempted Resuscitation risk calculator into our DA to determine personal risks and benefits of CPR for each patient. We will then present the first prototype to 5 new patient-intensivist dyads. Feedback about content and visual presentation will be collected from the intensivists through short interviews while longer interviews will be conducted with patients and their family members to inform the visual design and content of the next prototype. After each rapid prototyping cycle, 2 researchers will perform qualitative content analysis of data collected through interviews and direct observations. We will attempt to solve all content and visual design issues identified before moving to the next round of prototyping. In all, we will conduct 3 prototyping cycles with a total of 15 patient-intensivist dyads. Results: We expect to develop a multimedia wiki-based DA to support goals of care discussions about CPR adapted to the local needs of patients, their family members, and intensivists and tailorable to individual patient risk factors. The final version of the DA as well as the development process will be housed in an open-access wiki and free to be adapted and used in other contexts. Conclusions: This study will shed new light on the development of DAs adapted to local context and tailorable to individual patient risk factors employing user-centered design and a wiki to support rapid prototyping of content and visual design issues. %M 26869137 %R 10.2196/resprot.5107 %U http://www.researchprotocols.org/2016/1/e24/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 4 %P e36 %T Real-Time and Retrospective Health-Analytics-as-a-Service: A Novel Framework %+ IBMCanada Research and Development CenterIBM Canada3600 Steeles Avenue EastMarkham, Toronto, ON, L3R 1H5Canada1 905 721 8668 ext 36971 416 567 8167hamzeh.k.h@ieee.org  %A Khazaei,Hamzeh %A McGregor,Carolyn %A Eklund,J Mikael %A El-Khatib,Khalil %K premature babies %K physiological data %K decision support system %K analytics-as-a-service %K cloud computing %K big data, health informatics %K real-time analytics %K retrospective analysis %K performance modeling %D 2015 %7 18.11.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: Analytics-as-a-service (AaaS) is one of the latest provisions emerging from the cloud services family. Utilizing this paradigm of computing in health informatics will benefit patients, care providers, and governments significantly. This work is a novel approach to realize health analytics as services in critical care units in particular. Objective: To design, implement, evaluate, and deploy an extendable big-data compatible framework for health-analytics-as-a-service that offers both real-time and retrospective analysis. Methods: We present a novel framework that can realize health data analytics-as-a-service. The framework is flexible and configurable for different scenarios by utilizing the latest technologies and best practices for data acquisition, transformation, storage, analytics, knowledge extraction, and visualization. We have instantiated the proposed method, through the Artemis project, that is, a customization of the framework for live monitoring and retrospective research on premature babies and ill term infants in neonatal intensive care units (NICUs). Results: We demonstrated the proposed framework in this paper for monitoring NICUs and refer to it as the Artemis-In-Cloud (Artemis-IC) project. A pilot of Artemis has been deployed in the SickKids hospital NICU. By infusing the output of this pilot set up to an analytical model, we predict important performance measures for the final deployment of Artemis-IC. This process can be carried out for other hospitals following the same steps with minimal effort. SickKids’ NICU has 36 beds and can classify the patients generally into 5 different types including surgical and premature babies. The arrival rate is estimated as 4.5 patients per day, and the average length of stay was calculated as 16 days. Mean number of medical monitoring algorithms per patient is 9, which renders 311 live algorithms for the whole NICU running on the framework. The memory and computation power required for Artemis-IC to handle the SickKids NICU will be 32 GB and 16 CPU cores, respectively. The required amount of storage was estimated as 8.6 TB per year. There will always be 34.9 patients in SickKids NICU on average. Currently, 46% of patients cannot get admitted to SickKids NICU due to lack of resources. By increasing the capacity to 90 beds, all patients can be accommodated. For such a provisioning, Artemis-IC will need 16 TB of storage per year, 55 GB of memory, and 28 CPU cores. Conclusions: Our contributions in this work relate to a cloud architecture for the analysis of physiological data for clinical decisions support for tertiary care use. We demonstrate how to size the equipment needed in the cloud for that architecture based on a very realistic assessment of the patient characteristics and the associated clinical decision support algorithms that would be required to run for those patients. We show the principle of how this could be performed and furthermore that it can be replicated for any critical care setting within a tertiary institution. %M 26582268 %R 10.2196/medinform.4640 %U http://medinform.jmir.org/2015/4/e36/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 4 %P e34 %T Disrupting Electronic Health Records Systems: The Next Generation %+ Beth Israel Deaconess Medical CenterDivision of Pulmonary, Critical Care, and Sleep Medicine330 Brookline AvenueBoston, MA, 02215United States1 617 667 58641 617 667 1604lceli@bidmc.harvard.edu  %A Celi,Leo Anthony %A Marshall,Jeffrey David %A Lai,Yuan %A Stone,David J %K clinical decision making %K clinical decision support %K electronic health records %K electronic notes %D 2015 %7 23.10.2015 %9 Viewpoint %J JMIR Med Inform %G English %X The health care system suffers from both inefficient and ineffective use of data. Data are suboptimally displayed to users, undernetworked, underutilized, and wasted. Errors, inefficiencies, and increased costs occur on the basis of unavailable data in a system that does not coordinate the exchange of information, or adequately support its use. Clinicians’ schedules are stretched to the limit and yet the system in which they work exerts little effort to streamline and support carefully engineered care processes. Information for decision-making is difficult to access in the context of hurried real-time workflows. This paper explores and addresses these issues to formulate an improved design for clinical workflow, information exchange, and decision making based on the use of electronic health records. %M 26500106 %R 10.2196/medinform.4192 %U http://medinform.jmir.org/2015/4/e34/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 4 %P e33 %T Technology for Large-Scale Translation of Clinical Practice Guidelines: A Pilot Study of the Performance of a Hybrid Human and Computer-Assisted Approach %+ EBMPracticeNetKapucijnenvoer 33 Blok JLeuven, 3000Belgium32 1633269732 16332697stijn.vandevelde@med.kuleuven.be  %A Van de Velde,Stijn %A Macken,Lieve %A Vanneste,Koen %A Goossens,Martine %A Vanschoenbeek,Jan %A Aertgeerts,Bert %A Vanopstal,Klaar %A Vander Stichele,Robert %A Buysschaert,Joost %K practice guidelines as topic %K translations %K technology %K education, medical, continuing %K evidence-based practice
 %D 2015 %7 09.10.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: The construction of EBMPracticeNet, a national electronic point-of-care information platform in Belgium, began in 2011 to optimize quality of care by promoting evidence-based decision making. The project involved, among other tasks, the translation of 940 EBM Guidelines of Duodecim Medical Publications from English into Dutch and French. Considering the scale of the translation process, it was decided to make use of computer-aided translation performed by certificated translators with limited expertise in medical translation. Our consortium used a hybrid approach, involving a human translator supported by a translation memory (using SDL Trados Studio), terminology recognition (using SDL MultiTerm terminology databases) from medical terminology databases, and support from online machine translation. This resulted in a validated translation memory, which is now in use for the translation of new and updated guidelines. Objective: The objective of this experiment was to evaluate the performance of the hybrid human and computer-assisted approach in comparison with translation unsupported by translation memory and terminology recognition. A comparison was also made with the translation efficiency of an expert medical translator. Methods: We conducted a pilot study in which two sets of 30 new and 30 updated guidelines were randomized to one of three groups. Comparable guidelines were translated (1) by certificated junior translators without medical specialization using the hybrid method, (2) by an experienced medical translator without this support, and (3) by the same junior translators without the support of the validated translation memory. A medical proofreader who was blinded for the translation procedure, evaluated the translated guidelines for acceptability and adequacy. Translation speed was measured by recording translation and post-editing time. The human translation edit rate was calculated as a metric to evaluate the quality of the translation. A further evaluation was made of translation acceptability and adequacy. Results: The average number of words per guideline was 1195 and the mean total translation time was 100.2 minutes/1000 words. No meaningful differences were found in the translation speed for new guidelines. The translation of updated guidelines was 59 minutes/1000 words faster (95% CI 2-115; P=.044) in the computer-aided group. Revisions due to terminology accounted for one third of the overall revisions by the medical proofreader. Conclusions: Use of the hybrid human and computer-aided translation by a non-expert translator makes the translation of updates of clinical practice guidelines faster and cheaper because of the benefits of translation memory. For the translation of new guidelines, there was no apparent benefit in comparison with the efficiency of translation unsupported by translation memory (whether by an expert or non-expert translator). %M 26453372 %R 10.2196/medinform.4450 %U http://medinform.jmir.org/2015/4/e33/ %0 Journal Article %@ 1438-8871 %I JMIR Publications Inc. %V 17 %N 9 %P e215 %T How Consumers and Physicians View New Medical Technology: Comparative Survey %+ Scripps Translational Science InstituteScripps HealthThe Scripps Research Institute3344 North Torrey Pines CtSuite 300La Jolla, CA, 92037United States1 858 554 57081 858 546 9272etopol@scripps.edu  %A Boeldt,Debra L %A Wineinger,Nathan E %A Waalen,Jill %A Gollamudi,Shreya %A Grossberg,Adam %A Steinhubl,Steven R %A McCollister-Slipp,Anna %A Rogers,Marc A %A Silvers,Carey %A Topol,Eric J %K digital revolution %K healthcare %K medical technology %K physician and consumer attitudes %K electronic health record %K mobile health %D 2015 %7 14.09.2015 %9 Original Paper %J J Med Internet Res %G English %X Background: As a result of the digital revolution coming to medicine, a number of new tools are becoming available and are starting to be introduced in clinical practice. Objective: We aim to assess health care professional and consumer attitudes toward new medical technology including smartphones, genetic testing, privacy, and patient-accessible electronic health records. Methods: We performed a survey with 1406 health care providers and 1102 consumer responders. Results: Consumers who completed the survey were more likely to prefer new technologies for a medical diagnosis (437/1102, 39.66%) compared with providers (194/1406, 13.80%; P<.001), with more providers (393/1406, 27.95%) than consumers (175/1102, 15.88%) reporting feeling uneasy about using technology for a diagnosis. Both providers and consumers supported genetic testing for various purposes, with providers (1234/1406, 87.77%) being significantly more likely than consumers (806/1102, 73.14%) to support genetic testing when planning to have a baby (P<.001). Similarly, 91.68% (1289/1406) of providers and 81.22% (895/1102) of consumers supported diagnosing problems in a fetus (P<.001). Among providers, 90.33% (1270/1406) were concerned that patients would experience anxiety after accessing health records, and 81.95% (1149/1406) felt it would lead to requests for unnecessary medical evaluations, but only 34.30% (378/1102; P<.001) and 24.59% (271/1102; P<.001) of consumers expressed the same concerns, respectively. Physicians (137/827, 16.6%) reported less concern about the use of technology for diagnosis compared to medical students (21/235, 8.9%; P=.03) and also more frequently felt that patients owned their medical record (323/827, 39.1%; and 30/235, 12.8%, respectively; P<.001). Conclusions: Consumers and health professionals differ significantly and broadly in their views of emerging medical technology, with more enthusiasm and support expressed by consumers. %M 26369254 %R 10.2196/jmir.4456 %U http://www.jmir.org/2015/9/e215/ %0 Journal Article %@ 2292-9495 %I Gunther Eysenbach %V 2 %N 2 %P e14 %T Usability Testing of a Complex Clinical Decision Support Tool in the Emergency Department: Lessons Learned %+ Hofstra North Shore-LIJ School of MedicineDepartment of Medicine4th Floor300 Community DriveManhasset, NY, 11030United States1 267 979 79401 516 562 2526apress@nshs.edu  %A Press,Anne %A McCullagh,Lauren %A Khan,Sundas %A Schachter,Andy %A Pardo,Salvatore %A McGinn,Thomas %K clinical decision support %K emergency department %K usability testing %K clinical prediction rules %K Wells criteria %K pulmonary embolism %D 2015 %7 10.09.2015 %9 Original Paper %J JMIR Human Factors %G English %X Background: As the electronic health record (EHR) becomes the preferred documentation tool across medical practices, health care organizations are pushing for clinical decision support systems (CDSS) to help bring clinical decision support (CDS) tools to the forefront of patient-physician interactions. A CDSS is integrated into the EHR and allows physicians to easily utilize CDS tools. However, often CDSS are integrated into the EHR without an initial phase of usability testing, resulting in poor adoption rates. Usability testing is important because it evaluates a CDSS by testing it on actual users. This paper outlines the usability phase of a study, which will test the impact of integration of the Wells CDSS for pulmonary embolism (PE) diagnosis into a large urban emergency department, where workflow is often chaotic and high stakes decisions are frequently made. We hypothesize that conducting usability testing prior to integration of the Wells score into an emergency room EHR will result in increased adoption rates by physicians. Objective: The objective of the study was to conduct usability testing for the integration of the Wells clinical prediction rule into a tertiary care center’s emergency department EHR. Methods: We conducted usability testing of a CDS tool in the emergency department EHR. The CDS tool consisted of the Wells rule for PE in the form of a calculator and was triggered off computed tomography (CT) orders or patients’ chief complaint. The study was conducted at a tertiary hospital in Queens, New York. There were seven residents that were recruited and participated in two phases of usability testing. The usability testing employed a “think aloud” method and “near-live” clinical simulation, where care providers interacted with standardized patients enacting a clinical scenario. Both phases were audiotaped, video-taped, and had screen-capture software activated for onscreen recordings. Results: Phase I: Data from the “think-aloud” phase of the study showed an overall positive outlook on the Wells tool in assessing a patient for a PE diagnosis. Subjects described the tool as “well-organized” and “better than clinical judgment”. Changes were made to improve tool placement into the EHR to make it optimal for decision-making, auto-populating boxes, and minimizing click fatigue. Phase II: After incorporating the changes noted in Phase 1, the participants noted tool improvements. There was less toggling between screens, they had all the clinical information required to complete the tool, and were able to complete the patient visit efficiently. However, an optimal location for triggering the tool remained controversial. Conclusions: This study successfully combined “think-aloud” protocol analysis with “near-live” clinical simulations in a usability evaluation of a CDS tool that will be implemented into the emergency room environment. Both methods proved useful in the assessment of the CDS tool and allowed us to refine tool usability and workflow. %M 27025540 %R 10.2196/humanfactors.4537 %U http://humanfactors.jmir.org/2015/2/e14/ %0 Journal Article %@ 2291-5222 %I JMIR Publications Inc. %V 3 %N 3 %P e80 %T Use of Mobile Clinical Decision Support Software by Junior Doctors at a UK Teaching Hospital: Identification and Evaluation of Barriers to Engagement %+ University of LeicesterDepartment of Medical & Social Care Education107 Princess Road EastLeicester, LE1 7LAUnited Kingdom44 116 252 366844 116 252 3668rp299@le.ac.uk  %A Patel,Rakesh %A Green,William %A Shahzad,Muhammad Waseem %A Larkin,Chris %K clinical decision support systems %K health care technology %K human-centered computing %K medical education %K patient safety %K ubiquitous and mobile computing %D 2015 %7 13.08.2015 %9 Original Paper %J JMIR mHealth uHealth %G English %X Background: Clinical decision support (CDS) tools improve clinical diagnostic decision making and patient safety. The availability of CDS to health care professionals has grown in line with the increased prevalence of apps and smart mobile devices. Despite these benefits, patients may have safety concerns about the use of mobile devices around medical equipment. Objective: This research explored the engagement of junior doctors (JDs) with CDS and the perceptions of patients about their use. There were three objectives for this research: (1) to measure the actual usage of CDS tools on mobile devices (mCDS) by JDs, (2) to explore the perceptions of JDs about the drivers and barriers to using mCDS, and (3) to explore the perceptions of patients about the use of mCDS. Methods: This study used a mixed-methods approach to study the engagement of JDs with CDS accessed through mobile devices. Usage data were collected on the number of interactions by JDs with mCDS. The perceived drivers and barriers for JDs to using CDS were then explored by interviews. Finally, these findings were contrasted with the perception of patients about the use of mCDS by JDs. Results: Nine of the 16 JDs made a total of 142 recorded interactions with the mCDS over a 4-month period. Only 27 of the 114 interactions (24%) that could be categorized as on-shift or off-shift occurred on-shift. Eight individual, institutional, and cultural barriers to engagement emerged from interviews with the user group. In contrast to reported cautions and concerns about the impact of clinicians’ use of mobile phone on patient health and safety, patients had positive perceptions about the use of mCDS. Conclusions: Patients reported positive perceptions toward mCDS. The usage of mCDS to support clinical decision making was considered to be positive as part of everyday clinical practice. The degree of engagement was found to be limited due to a number of individual, institutional, and cultural barriers. The majority of mCDS engagement occurred outside of the workplace. Further research is required to verify these findings and assess their implications for future policy and practice. %M 26272411 %R 10.2196/mhealth.4388 %U http://mhealth.jmir.org/2015/3/e80/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 2 %P e23 %T A Web-Based Tool for Patient Triage in Emergency Department Settings: Validation Using the Emergency Severity Index %+ Duke Clinical Research InstituteDuke University School of MedicineDCRI 7th Floor2400 Pratt StDurham, NC, 27705United States1 407 782 22661 407 782 2266pierre.elias@duke.edu  %A Elias,Pierre %A Damle,Ash %A Casale,Michael %A Branson,Kim %A Peterson,Nick %A Churi,Chaitanya %A Komatireddy,Ravi %A Feramisco,Jamison %K triage %K emergency severity index %K differential diagnosis %K clinical decision support %D 2015 %7 10.6.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: We evaluated the concordance between triage scores generated by a novel Internet clinical decision support tool, Clinical GPS (cGPS) (Lumiata Inc, San Mateo, CA), and the Emergency Severity Index (ESI), a well-established and clinically validated patient severity scale in use today. Although the ESI and cGPS use different underlying algorithms to calculate patient severity, both utilize a five-point integer scale with level 1 representing the highest severity. Objective: The objective of this study was to compare cGPS results with an established gold standard in emergency triage. Methods: We conducted a blinded trial comparing triage scores from the ESI: A Triage Tool for Emergency Department Care, Version 4, Implementation Handbook to those generated by cGPS from the text of 73 sample case vignettes. A weighted, quadratic kappa statistic was used to assess agreement between cGPS derived severity scores and those published in the ESI handbook for all 73 cases. Weighted kappa concordance was defined a priori as almost perfect (kappa > 0.8), substantial (0.6 < kappa < 0.8), moderate (0.4 < kappa < 0.6), fair (0.2 < kappa< 0.4), or slight (kappa < 0.2). Results: Of the 73 case vignettes, the cGPS severity score matched the ESI handbook score in 95% of cases (69/73 cases), in addition, the weighted, quadratic kappa statistic showed almost perfect agreement (kappa = 0.93, 95% CI 0.854-0.996). In the subanalysis of 41 case vignettes assigned ESI scores of level 1 or 2, the cGPS and ESI severity scores matched in 95% of cases (39/41 cases). Conclusions: These results indicate that the cGPS is a reliable indicator of triage severity, based on its comparison to a standardized index, the ESI. Future studies are needed to determine whether the cGPS can accurately assess the triage of patients in real clinical environments. %R 10.2196/medinform.3508 %U http://medinform.jmir.org/2015/2/e23/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 2 %P e21 %T A Telesurveillance System With Automatic Electrocardiogram Interpretation Based on Support Vector Machine and Rule-Based Processing %+ National Taiwan UniversityGraduate Institute of Communication EngineeringNo. 1, Section 4, Roosevelt RoadTaipei, 10617Taiwan886 233669652886 233663662jjding@ntu.edu.tw  %A Ho,Te-Wei %A Huang,Chen-Wei %A Lin,Ching-Miao %A Lai,Feipei %A Ding,Jian-Jiun %A Ho,Yi-Lwun %A Hung,Chi-Sheng %K telehealth care %K telesurveillance system %K electrocardiogram %K ECG classification %K support vector machine %D 2015 %7 07.05.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: Telehealth care is a global trend affecting clinical practice around the world. To mitigate the workload of health professionals and provide ubiquitous health care, a comprehensive surveillance system with value-added services based on information technologies must be established. Objective: We conducted this study to describe our proposed telesurveillance system designed for monitoring and classifying electrocardiogram (ECG) signals and to evaluate the performance of ECG classification. Methods: We established a telesurveillance system with an automatic ECG interpretation mechanism. The system included: (1) automatic ECG signal transmission via telecommunication, (2) ECG signal processing, including noise elimination, peak estimation, and feature extraction, (3) automatic ECG interpretation based on the support vector machine (SVM) classifier and rule-based processing, and (4) display of ECG signals and their analyzed results. We analyzed 213,420 ECG signals that were diagnosed by cardiologists as the gold standard to verify the classification performance. Results: In the clinical ECG database from the Telehealth Center of the National Taiwan University Hospital (NTUH), the experimental results showed that the ECG classifier yielded a specificity value of 96.66% for normal rhythm detection, a sensitivity value of 98.50% for disease recognition, and an accuracy value of 81.17% for noise detection. For the detection performance of specific diseases, the recognition model mainly generated sensitivity values of 92.70% for atrial fibrillation, 89.10% for pacemaker rhythm, 88.60% for atrial premature contraction, 72.98% for T-wave inversion, 62.21% for atrial flutter, and 62.57% for first-degree atrioventricular block. Conclusions: Through connected telehealth care devices, the telesurveillance system, and the automatic ECG interpretation system, this mechanism was intentionally designed for continuous decision-making support and is reliable enough to reduce the need for face-to-face diagnosis. With this value-added service, the system could widely assist physicians and other health professionals with decision making in clinical practice. The system will be very helpful for the patient who suffers from cardiac disease, but for whom it is inconvenient to go to the hospital very often. %M 25953306 %R 10.2196/medinform.4397 %U http://medinform.jmir.org/2015/2/e21/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 2 %P e17 %T Prioritization of Free-Text Clinical Documents: A Novel Use of a Bayesian Classifier %+ Massachusetts Institute of Technology77 Mass AveCambridge, MA, 02139United States1 4128057254ambshun@alum.mit.edu  %A Singh,Mark %A Murthy,Akansh %A Singh,Shridhar %K clinical reports %K prioritization %K Bayesian classifier %K radiology %K natural language processing %D 2015 %7 10.04.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: The amount of incoming data into physicians’ offices is increasing, thereby making it difficult to process information efficiently and accurately to maximize positive patient outcomes. Current manual processes of screening for individual terms within long free-text documents are tedious and error-prone. This paper explores the use of statistical methods and computer systems to assist clinical data management. Objective: The objective of this study was to verify and validate the use of a naive Bayesian classifier as a means of properly prioritizing important clinical data, specifically that of free-text radiology reports. Methods: There were one hundred reports that were first used to train the algorithm based on physicians’ categorization of clinical reports as high-priority or low-priority. Then, the algorithm was used to evaluate 354 reports. Additional beautification procedures such as section extraction, text preprocessing, and negation detection were performed. Results: The algorithm evaluated the 354 reports with discrimination between high-priority and low-priority reports, resulting in a bimodal probability distribution. In all scenarios tested, the false negative rates were below 1.1% and the recall rates ranged from 95.65% to 98.91%. In the case of 50% prior probability and 80% threshold probability, the accuracy of this Bayesian classifier was 93.50%, with a positive predictive value (precision) of 80.54%. It also showed a sensitivity (recall) of 98.91% and a F-measure of 88.78%. Conclusions: The results showed that the algorithm could be trained to detect abnormal radiology results by accurately screening clinical reports. Such a technique can potentially be used to enable automatic flagging of critical results. In addition to accuracy, the algorithm was able to minimize false negatives, which is important for clinical applications. We conclude that a Bayesian statistical classifier, by flagging reports with abnormal findings, can assist a physician in reviewing radiology reports more efficiently. This higher level of prioritization allows physicians to address important radiologic findings in a timelier manner and may also aid in minimizing errors of omission. %M 25863643 %R 10.2196/medinform.3793 %U http://medinform.jmir.org/2015/2/e17/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 1 %P e12 %T CIMIDx: Prototype for a Cloud-Based System to Support Intelligent Medical Image Diagnosis With Efficiency %+ Department of Computer Science and EngineeringCollege of EngineeringAnna UniversityGuindy CampusSardar Patel Road, GuindyChennai, Tamilnadu, 600025India91 950003789591 4422351956adhityaresearch@hotmail.com  %A Bhavani,Selvaraj Rani %A Senthilkumar,Jagatheesan %A Chilambuchelvan,Arul Gnanaprakasam %A Manjula,Dhanabalachandran %A Krishnamoorthy,Ramasamy %A Kannan,Arputharaj %K association rules %K cloud computing %K breast cancer %K pre-processing %K segmentation %K feature extraction %K intelligent system %K UDDI %K SOAP %K Web-based intervention %K medical diagnosis %D 2015 %7 27.03.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: The Internet has greatly enhanced health care, helping patients stay up-to-date on medical issues and general knowledge. Many cancer patients use the Internet for cancer diagnosis and related information. Recently, cloud computing has emerged as a new way of delivering health services but currently, there is no generic and fully automated cloud-based self-management intervention for breast cancer patients, as practical guidelines are lacking. Objective: We investigated the prevalence and predictors of cloud use for medical diagnosis among women with breast cancer to gain insight into meaningful usage parameters to evaluate the use of generic, fully automated cloud-based self-intervention, by assessing how breast cancer survivors use a generic self-management model. The goal of this study was implemented and evaluated with a new prototype called “CIMIDx”, based on representative association rules that support the diagnosis of medical images (mammograms). Methods: The proposed Cloud-Based System Support Intelligent Medical Image Diagnosis (CIMIDx) prototype includes two modules. The first is the design and development of the CIMIDx training and test cloud services. Deployed in the cloud, the prototype can be used for diagnosis and screening mammography by assessing the cancers detected, tumor sizes, histology, and stage of classification accuracy. To analyze the prototype’s classification accuracy, we conducted an experiment with data provided by clients. Second, by monitoring cloud server requests, the CIMIDx usage statistics were recorded for the cloud-based self-intervention groups. We conducted an evaluation of the CIMIDx cloud service usage, in which browsing functionalities were evaluated from the end-user’s perspective. Results: We performed several experiments to validate the CIMIDx prototype for breast health issues. The first set of experiments evaluated the diagnostic performance of the CIMIDx framework. We collected medical information from 150 breast cancer survivors from hospitals and health centers. The CIMIDx prototype achieved high sensitivity of up to 99.29%, and accuracy of up to 98%. The second set of experiments evaluated CIMIDx use for breast health issues, using t tests and Pearson chi-square tests to assess differences, and binary logistic regression to estimate the odds ratio (OR) for the predictors’ use of CIMIDx. For the prototype usage statistics for the same 150 breast cancer survivors, we interviewed 114 (76.0%), through self-report questionnaires from CIMIDx blogs. The frequency of log-ins/person ranged from 0 to 30, total duration/person from 0 to 1500 minutes (25 hours). The 114 participants continued logging in to all phases, resulting in an intervention adherence rate of 44.3% (95% CI 33.2-55.9). The overall performance of the prototype for the good category, reported usefulness of the prototype (P=.77), overall satisfaction of the prototype (P=.31), ease of navigation (P=.89), user friendliness evaluation (P=.31), and overall satisfaction (P=.31). Positive evaluations given by 100 participants via a Web-based questionnaire supported our hypothesis. Conclusions: The present study shows that women felt favorably about the use of a generic fully automated cloud-based self- management prototype. The study also demonstrated that the CIMIDx prototype resulted in the detection of more cancers in screening and diagnosing patients, with an increased accuracy rate. %M 25830608 %R 10.2196/medinform.3709 %U http://medinform.jmir.org/2015/1/e12/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 1 %P e11 %T From Data to Optimal Decision Making: A Data-Driven, Probabilistic Machine Learning Approach to Decision Support for Patients With Sepsis %+ Department of Computer Science and Genome CenterUniversity of California, DavisOne Shields AveDavis, CA, CA 95616United States1 53075277071 5307526747iliast@ucdavis.edu  %A Tsoukalas,Athanasios %A Albertson,Timothy %A Tagkopoulos,Ilias %K sepsis %K clinical decision support tool %K probabilistic modeling %K Partially Observable Markov Decision Processes %K POMDP %K CDSS %D 2015 %7 24.02.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: A tantalizing question in medical informatics is how to construct knowledge from heterogeneous datasets, and as an extension, inform clinical decisions. The emergence of large-scale data integration in electronic health records (EHR) presents tremendous opportunities. However, our ability to efficiently extract informed decision support is limited due to the complexity of the clinical states and decision process, missing data and lack of analytical tools to advice based on statistical relationships. Objective: Development and assessment of a data-driven method that infers the probability distribution of the current state of patients with sepsis, likely trajectories, optimal actions related to antibiotic administration, prediction of mortality and length-of-stay. Methods: We present a data-driven, probabilistic framework for clinical decision support in sepsis-related cases. We first define states, actions, observations and rewards based on clinical practice, expert knowledge and data representations in an EHR dataset of 1492 patients. We then use Partially Observable Markov Decision Process (POMDP) model to derive the optimal policy based on individual patient trajectories and we evaluate the performance of the model-derived policies in a separate test set. Policy decisions were focused on the type of antibiotic combinations to administer. Multi-class and discriminative classifiers were used to predict mortality and length of stay. Results: Data-derived antibiotic administration policies led to a favorable patient outcome in 49% of the cases, versus 37% when the alternative policies were followed (P=1.3e-13). Sensitivity analysis on the model parameters and missing data argue for a highly robust decision support tool that withstands parameter variation and data uncertainty. When the optimal policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25). Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82). Conclusions: A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy. This work provides a solid basis for a scalable probabilistic clinical decision support framework for sepsis treatment that can be expanded to other clinically relevant states and actions, as well as a data-driven model that can be adopted in other clinical areas with sufficient training data. %M 25710907 %R 10.2196/medinform.3445 %U http://medinform.jmir.org/2015/1/e11/ %0 Journal Article %@ 1438-8871 %I JMIR Publications Inc. %V 17 %N 2 %P e41 %T FRAT-up, a Web-based Fall-Risk Assessment Tool for Elderly People Living in the Community %+ Department of Computer Science and Engineering - DISIUniversity of BolognaViale Risorgimento 2Bologna, 40136Italy39 051 209308639 051 2093869federico.chesani@unibo.it  %A Cattelani,Luca %A Palumbo,Pierpaolo %A Palmerini,Luca %A Bandinelli,Stefania %A Becker,Clemens %A Chesani,Federico %A Chiari,Lorenzo %K accidental falls %K odds ratio %K risk assessment %K risk factors %K ROC curve %K aged %D 2015 %7 18.02.2015 %9 Original Paper %J J Med Internet Res %G English %X Background: About 30% of people over 65 are subject to at least one unintentional fall a year. Fall prevention protocols and interventions can decrease the number of falls. To be effective, a prevention strategy requires a prior step to evaluate the fall risk of the subjects. Despite extensive research, existing assessment tools for fall risk have been insufficient for predicting falls. Objective: The goal of this study is to present a novel web-based fall-risk assessment tool (FRAT-up) and to evaluate its accuracy in predicting falls, within a context of community-dwelling persons aged 65 and up. Methods: FRAT-up is based on the assumption that a subject’s fall risk is given by the contribution of their exposure to each of the known fall-risk factors. Many scientific studies have investigated the relationship between falls and risk factors. The majority of these studies adopted statistical approaches, usually providing quantitative information such as odds ratios. FRAT-up exploits these numerical results to compute how each single factor contributes to the overall fall risk. FRAT-up is based on a formal ontology that enlists a number of known risk factors, together with quantitative findings in terms of odds ratios. From such information, an automatic algorithm generates a rule-based probabilistic logic program, that is, a set of rules for each risk factor. The rule-based program takes the health profile of the subject (in terms of exposure to the risk factors) and computes the fall risk. A Web-based interface allows users to input health profiles and to visualize the risk assessment for the given subject. FRAT-up has been evaluated on the InCHIANTI Study dataset, a representative population-based study of older persons living in the Chianti area (Tuscany, Italy). We compared reported falls with predicted ones and computed performance indicators. Results: The obtained area under curve of the receiver operating characteristic was 0.642 (95% CI 0.614-0.669), while the Brier score was 0.174. The Hosmer-Lemeshow test indicated statistical significance of miscalibration. Conclusions: FRAT-up is a web-based tool for evaluating the fall risk of people aged 65 or up living in the community. Validation results of fall risks computed by FRAT-up show that its performance is comparable to externally validated state-of-the-art tools. A prototype is freely available through a web-based interface. Trial Registration: ClinicalTrials.gov NCT01331512 (The InChianti Follow-Up Study); http://clinicaltrials.gov/show/NCT01331512 (Archived by WebCite at http://www.webcitation.org/6UDrrRuaR). %M 25693419 %R 10.2196/jmir.4064 %U http://www.jmir.org/2015/2/e41/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 1 %P e8 %T On-Admission Pressure Ulcer Prediction Using the Nursing Needs Score %+ Department of Medical InformaticsInstitute of Health BiosciencesThe University of Tokushima Graduate SchoolKuramoto3-18-15Tokushima, 7708503Japan81 88633917881 886339411yokotamu19@gmail.com  %A Nakamura,Yoko %A Ghaibeh,A. Ammar %A Setoguchi,Yoko %A Mitani,Kazue %A Abe,Yoshiro %A Hashimoto,Ichiro %A Moriguchi,Hiroki %K pressure ulcer %K nursing needs score %K prediction %K logistic regression %K imbalanced data %D 2015 %7 11.02.2015 %9 Original Paper %J JMIR Med Inform %G English %X Background: Pressure ulcers (PUs) are considered a serious problem in nursing care and require preventive measures. Many risk assessment methods are currently being used, but most require the collection of data not available on admission. Although nurses assess the Nursing Needs Score (NNS) on a daily basis in Japanese acute care hospitals, these data are primarily used to standardize the cost of nursing care in the public insurance system for appropriate nurse staffing, and have never been used for PU risk assessment. Objective: The objective of this study was to predict the risk of PU development using only data available on admission, including the on-admission NNS score. Methods: Logistic regression was used to generate a prediction model for the risk of developing PUs after admission. A random undersampling procedure was used to overcome the problem of imbalanced data. Results: A combination of gender, age, surgical duration, and on-admission total NNS score (NNS group B; NNS-B) was the best predictor with an average sensitivity, specificity, and area under receiver operating characteristic curve (AUC) of 69.2% (6920/100), 82.8% (8280/100), and 84.0% (8400/100), respectively. The model with the median AUC achieved 80% (4/5) sensitivity, 81.3% (669/823) specificity, and 84.3% AUC. Conclusions: We developed a model for predicting PU development using gender, age, surgical duration, and on-admission total NNS-B score. These results can be used to improve the efficiency of nurses and reduce the number of PU cases by identifying patients who require further examination. %M 25673118 %R 10.2196/medinform.3850 %U http://medinform.jmir.org/2015/1/e8/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 3 %N 1 %P e4 %T Adoption of Clinical Decision Support in Multimorbidity: A Systematic Review %+ Centre for Health InformaticsInstitute of Population HealthThe University of ManchesterVaughan HousePortsmouth StManchester, M13 9GBUnited Kingdom44 161 275 113244 161 275 1132paolo.fraccaro@postgrad.manchester.ac.uk  %A Fraccaro,Paolo %A Arguello Casteleiro,Mercedes %A Ainsworth,John %A Buchan,Iain %K decision support systems, management %K systematic review %K multiple chronic diseases %K multiple pathologies %K multiple medications %D 2015 %7 07.01.2015 %9 Review %J JMIR Med Inform %G English %X Background: Patients with multiple conditions have complex needs and are increasing in number as populations age. This multimorbidity is one of the greatest challenges facing health care. Having more than 1 condition generates (1) interactions between pathologies, (2) duplication of tests, (3) difficulties in adhering to often conflicting clinical practice guidelines, (4) obstacles in the continuity of care, (5) confusing self-management information, and (6) medication errors. In this context, clinical decision support (CDS) systems need to be able to handle realistic complexity and minimize iatrogenic risks. Objective: The aim of this review was to identify to what extent CDS is adopted in multimorbidity. Methods: This review followed PRISMA guidance and adopted a multidisciplinary approach. Scopus and PubMed searches were performed by combining terms from 3 different thesauri containing synonyms for (1) multimorbidity and comorbidity, (2) polypharmacy, and (3) CDS. The relevant articles were identified by examining the titles and abstracts. The full text of selected/relevant articles was analyzed in-depth. For articles appropriate for this review, data were collected on clinical tasks, diseases, decision maker, methods, data input context, user interface considerations, and evaluation of effectiveness. Results: A total of 50 articles were selected for the full in-depth analysis and 20 studies were included in the final review. Medication (n=10) and clinical guidance (n=8) were the predominant clinical tasks. Four studies focused on merging concurrent clinical practice guidelines. A total of 17 articles reported their CDS systems were knowledge-based. Most articles reviewed considered patients’ clinical records (n=19), clinical practice guidelines (n=12), and clinicians’ knowledge (n=10) as contextual input data. The most frequent diseases mentioned were cardiovascular (n=9) and diabetes mellitus (n=5). In all, 12 articles mentioned generalist doctor(s) as the decision maker(s). For articles reviewed, there were no studies referring to the active involvement of the patient in the decision-making process or to patient self-management. None of the articles reviewed adopted mobile technologies. There were no rigorous evaluations of usability or effectiveness of the CDS systems reported. Conclusions: This review shows that multimorbidity is underinvestigated in the informatics of supporting clinical decisions. CDS interventions that systematize clinical practice guidelines without considering the interactions of different conditions and care processes may lead to unhelpful or harmful clinical actions. To improve patient safety in multimorbidity, there is a need for more evidence about how both conditions and care processes interact. The data needed to build this evidence base exist in many electronic health record systems and are underused. %M 25785897 %R 10.2196/medinform.3503 %U http://medinform.jmir.org/2015/1/e4/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 2 %N 2 %P e29 %T Clinical Decision Support System to Enhance Quality Control of Spirometry Using Information and Communication Technologies %+ Hospital Clinic - IDIBAPS - CiberesRespiratory Diagnostic CenterUniversity of BarcelonaSotano porta 6Villarroel, 170Barcelona, 08036Spain34 93227554034 932275455fburgos@ub.edu  %A Burgos,Felip %A Melia,Umberto %A Vallverdú,Montserrat %A Velickovski,Filip %A Lluch-Ariet,Magí %A Caminal,Pere %A Roca,Josep %K spirometry %K telemedicine %K information communication technologies %K primary care %K quality control %D 2014 %7 21.10.2014 %9 Original Paper %J JMIR Med Inform %G English %X Background: We recently demonstrated that quality of spirometry in primary care could markedly improve with remote offline support from specialized professionals. It is hypothesized that implementation of automatic online assessment of quality of spirometry using information and communication technologies may significantly enhance the potential for extensive deployment of a high quality spirometry program in integrated care settings. Objective: The objective of the study was to elaborate and validate a Clinical Decision Support System (CDSS) for automatic online quality assessment of spirometry. Methods: The CDSS was done through a three step process including: (1) identification of optimal sampling frequency; (2) iterations to build-up an initial version using the 24 standard spirometry curves recommended by the American Thoracic Society; and (3) iterations to refine the CDSS using 270 curves from 90 patients. In each of these steps the results were checked against one expert. Finally, 778 spirometry curves from 291 patients were analyzed for validation purposes. Results: The CDSS generated appropriate online classification and certification in 685/778 (88.1%) of spirometry testing, with 96% sensitivity and 95% specificity. Conclusions: Consequently, only 93/778 (11.9%) of spirometry testing required offline remote classification by an expert, indicating a potential positive role of the CDSS in the deployment of a high quality spirometry program in an integrated care setting. %M 25600957 %R 10.2196/medinform.3179 %U http://medinform.jmir.org/2014/2/e29/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 2 %N 2 %P e20 %T Exploring a Clinically Friendly Web-Based Approach to Clinical Decision Support Linked to the Electronic Health Record: Design Philosophy, Prototype Implementation, and Framework for Assessment %+ Center for Medical InformaticsYale University School of Medicine300 George Street, Suite 501New Haven, CT, 06511United States1 203 737 29031 203 737 5708perry.miller@yale.edu  %A Miller,Perry %A Phipps,Michael %A Chatterjee,Sharmila %A Rajeevan,Nallakkandi %A Levin,Forrest %A Frawley,Sandra %A Tokuno,Hajime %K Internet %K clinical decision support systems %K electronic health records %K neuropathic pain %K therapeutics %D 2014 %7 18.08.2014 %9 Original Paper %J JMIR Med Inform %G English %X Background: Computer-based clinical decision support (CDS) is an important component of the electronic health record (EHR). As an increasing amount of CDS is implemented, it will be important that this be accomplished in a fashion that assists in clinical decision making without imposing unacceptable demands and burdens upon the provider’s practice. Objective: The objective of our study was to explore an approach that allows CDS to be clinician-friendly from a variety of perspectives, to build a prototype implementation that illustrates features of the approach, and to gain experience with a pilot framework for assessment. Methods: The paper first discusses the project’s design philosophy and goals. It then describes a prototype implementation (Neuropath/CDS) that explores the approach in the domain of neuropathic pain and in the context of the US Veterans Administration EHR. Finally, the paper discusses a framework for assessing the approach, illustrated by a pilot assessment of Neuropath/CDS. Results: The paper describes the operation and technical design of Neuropath/CDS, as well as the results of the pilot assessment, which emphasize the four areas of focus, scope, content, and presentation. Conclusions: The work to date has allowed us to explore various design and implementation issues relating to the approach illustrated in Neuropath/CDS, as well as the development and pilot application of a framework for assessment. %M 25580426 %R 10.2196/medinform.3586 %U http://medinform.jmir.org/2014/2/e20/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 2 %N 2 %P e14 %T OWLing Clinical Data Repositories With the Ontology Web Language %+ Hospital ClínicUnit of Medical InformaticsUniversity of BarcelonaVillarroel 170Barcelona, 08036Spain34 93227920634 932279240rlozano@clinic.ub.es  %A Lozano-Rubí,Raimundo %A Pastor,Xavier %A Lozano,Esther %K biomedical ontologies %K data storage and retrieval %K knowledge management %K data sharing %K electronic health records %D 2014 %7 01.08.2014 %9 Original Paper %J JMIR Med Inform %G English %X Background: The health sciences are based upon information. Clinical information is usually stored and managed by physicians with precarious tools, such as spreadsheets. The biomedical domain is more complex than other domains that have adopted information and communication technologies as pervasive business tools. Moreover, medicine continuously changes its corpus of knowledge because of new discoveries and the rearrangements in the relationships among concepts. This scenario makes it especially difficult to offer good tools to answer the professional needs of researchers and constitutes a barrier that needs innovation to discover useful solutions. Objective: The objective was to design and implement a framework for the development of clinical data repositories, capable of facing the continuous change in the biomedicine domain and minimizing the technical knowledge required from final users. Methods: We combined knowledge management tools and methodologies with relational technology. We present an ontology-based approach that is flexible and efficient for dealing with complexity and change, integrated with a solid relational storage and a Web graphical user interface. Results: Onto Clinical Research Forms (OntoCRF) is a framework for the definition, modeling, and instantiation of data repositories. It does not need any database design or programming. All required information to define a new project is explicitly stated in ontologies. Moreover, the user interface is built automatically on the fly as Web pages, whereas data are stored in a generic repository. This allows for immediate deployment and population of the database as well as instant online availability of any modification. Conclusions: OntoCRF is a complete framework to build data repositories with a solid relational storage. Driven by ontologies, OntoCRF is more flexible and efficient to deal with complexity and change than traditional systems and does not require very skilled technical people facilitating the engineering of clinical software systems. %M 25599697 %R 10.2196/medinform.3023 %U http://medinform.jmir.org/2014/2/e14/ %0 Journal Article %@ 2291-9694 %I Gunther Eysenbach %V 2 %N 2 %P e8 %T A Validation of an Intelligent Decision-Making Support System for the Nutrition Diagnosis of Bariatric Surgery Patients %+ Pontifical Catholic University of Paraná (PUCPR)St Imaculada Conceição, 1155Curitiba, 80215901Brazil55 41 3242654355 41 32426543nutriclinmagda@bol.com.br  %A Cruz,Magda RR %A Martins,Cristina %A Dias,João %A Pinto,José S %K bariatric surgery %K nutrition diagnosis %K artificial intelligence %K Bayesian networks %K decision-making %K support system %D 2014 %7 08.07.2014 %9 Original Paper %J JMIR Med Inform %G English %X Background: Bariatric surgery is an important method for treatment of morbid obesity. It is known that significant nutritional deficiencies might occur after surgery, such as, calorie-protein malnutrition, iron deficiency anemia, and lack of vitamin B12, thiamine, and folic acid. Objective: The objective of our study was to validate a computerized intelligent decision support system that suggests nutritional diagnoses of patients submitted to bariatric surgery. Methods: There were fifteen clinical cases that were developed and sent to three dietitians in order to evaluate and define a nutritional diagnosis. After this step, the cases were sent to four bariatric surgery expert dietitians who were aiming to collaborate on a gold standard. The nutritional diagnosis was to be defined individually, and any disagreements were solved through a consensus. The final result was used as the gold standard. Bayesian networks were used to implement the system, and database training was done with Shell Netica. For the system validation, a similar answer rate was calculated, as well as the specificity and sensibility. Receiver operating characteristic (ROC) curves were projected to each nutritional diagnosis. Results: Among the four experts, the rate of similar answers found was 80% (48/60) to 93% (56/60), depending on the nutritional diagnosis. The rate of similar answers of the system, compared to the gold standard, was 100% (60/60). The system sensibility and specificity were 95.0%. The ROC curves projection showed that the system was able to represent the expert knowledge (gold standard), and to help them in their daily tasks. Conclusions: The system that was developed was validated to be used by health care professionals for decision-making support in their nutritional diagnosis of patients submitted to bariatric surgery. %M 25601419 %R 10.2196/medinform.2984 %U http://medinform.jmir.org/2014/2/e8/