Search

The article focuses on informed decision making in medical care when clinical evidence and conventional wisdom collide. It also discusses the U.S. Agency for Healthcare Research & Quality (AHRQ) involvement in exploring issues that impact clinical decision making under the AHRQ Effective Health Care (EHC) Program.

Background: Lyme disease is one of the most commonly reported infectious diseases in the United States (US), accounting for more than 90 % of all vector-borne diseases in North America. Objective: In this paper, self-reported tweets on Twitter were analyzed in order to predict potential Lyme disease cases and accurately assess incidence rates in the US. Methods: The study was done in three stages: (1) Approximately 1.3 million tweets were collected and pre-processed to extract the most relevant Lyme disease tweets with geolocations. A subset of tweets were semi-automatically labelled as relevant or irrelevant to Lyme disease using a set of precise keywords, and the remaining portion were manually labelled, yielding a curated labelled dataset of 77, 500 tweets. (2) This labelled data set was used to train, validate, and test various combinations of NLP word embedding methods and prominent ML classification models, such as TF-IDF and logistic regression, Word2vec and XGboost, and BERTweet, among others, to identify potential Lyme disease tweets. (3) Lastly, the presence of spatio-temporal patterns in the US over a 10-year period were studied. Results: Preliminary results showed that BERTweet outperformed all tested NLP classifiers for identifying Lyme disease tweets, achieving the highest classification accuracy and F1-score of 90 % . There was also a consistent pattern indicating that the West and Northeast regions of the US had a higher tweet rate over time. Conclusions: We focused on the less-studied problem of using Twitter data as a surveillance tool for Lyme disease in the US. Several crucial findings have emerged from the study. First, there is a fairly strong correlation between classified tweet counts and Lyme disease counts, with both following similar trends. Second, in 2015 and early 2016, the social media network like Twitter was essential in raising popular awareness of Lyme disease. Third, counties with a high incidence rate were not necessarily related with a high tweet rate, and vice versa. Fourth, BERTweet can be used as a reliable NLP classifier for detecting relevant Lyme disease tweets. [ABSTRACT FROM AUTHOR]

Background: In the United States, the National Alzheimer's Coordinating Center (NACC) and the Alzheimer's Disease Neuroimaging Initiative (ADNI) are two major data sharing resources for Alzheimer's Disease (AD) research. NACC and ADNI strive to make their data more FAIR (findable, interoperable, accessible and reusable) for the broader research community. However, there is limited work harmonizing and supporting cross-cohort interoperability of the two resources. Method: In this paper, we leverage an ontology-based approach to harmonize data elements in the two resources and develop a web-based query system to search patient cohorts across the two resources. We first mapped data elements across NACC and ADNI, and performed value harmonization for the mapped data elements with inconsistent permissible values. Then we built an Alzheimer's Disease Data Element Ontology (ADEO) to model the mapped data elements in NACC and ADNI. We further developed a prototype cross-cohort query system to search patient cohorts across NACC and ADNI. Results: After manual review, we found 172 mappings between NACC and ADNI. These 172 mappings were further used to construct common concepts in ADEO. Our data element mapping and harmonization resulted in five files storing common concepts, variables in NACC and ADNI, mappings between variables and common concepts, permissible values of categorical type data elements, and coding inconsistency harmonization, respectively. Our cross-cohort query system consists of three core architectural elements: a web-based interface, an advanced query engine, and a backend MongoDB database. Conclusions: In this work, ADEO has been specifically designed to facilitate data harmonization and cross-cohort query of NACC and ADNI data resources. Although our prototype cross-cohort query system was developed for exploring NACC and ADNI, its backend and frontend framework has been designed and implemented to be generally applicable to other domains for querying patient cohorts from multiple heterogeneous data sources. [ABSTRACT FROM AUTHOR]

Background: The US Veterans Administration (VA) has developed a robust and mature computational infrastructure in support of its electronic health record (EHR). Web technology offers a powerful set of tools for structuring clinical decision support (CDS) around clinical care. This paper describes informatics challenges and design issues that were confronted in the process of building three Web-based CDS systems in the context of the VA EHR.Methods: Over the course of several years, we implemented three Web-based CDS systems that extract patient data from the VA EHR environment to provide patient-specific CDS. These were 1) the VACS (Veterans Aging Cohort Study) Index Calculator which estimates prognosis for HIV+ patients, 2) Neuropath/CDS which assists in the medical management of patients with neuropathic pain, and 3) TRIM (Tool to Reduce Inappropriate Medications) which identifies potentially inappropriate medications in older adults and provides recommendations for improving the medication regimen.Results: The paper provides an overview of the VA EHR environment and discusses specific informatics issues/challenges that arose in the context of each of the three Web-based CDS systems. We discuss specific informatics methods and provide details of approaches that may be useful within this setting.Conclusions: Informatics issues and challenges relating to data access and data availability arose because of the particular architecture of the national VA infrastructure and the need to link to that infrastructure from local Web-based CDS systems. Idiosyncrasies of VA patient data, especially the medication data, also posed challenges. Other issues related to specific functional needs of individual CDS systems. The goal of this paper is to describe these issues so that our experience may serve as a useful foundation to assist others who wish to build such systems in the future. [ABSTRACT FROM AUTHOR]

Background: Prescription medication (PM) misuse/abuse has emerged as a national crisis in the United States, and social media has been suggested as a potential resource for performing active monitoring. However, automating a social media-based monitoring system is challenging-requiring advanced natural language processing (NLP) and machine learning methods. In this paper, we describe the development and evaluation of automatic text classification models for detecting self-reports of PM abuse from Twitter.Methods: We experimented with state-of-the-art bi-directional transformer-based language models, which utilize tweet-level representations that enable transfer learning (e.g., BERT, RoBERTa, XLNet, AlBERT, and DistilBERT), proposed fusion-based approaches, and compared the developed models with several traditional machine learning, including deep learning, approaches. Using a public dataset, we evaluated the performances of the classifiers on their abilities to classify the non-majority "abuse/misuse" class.Results: Our proposed fusion-based model performs significantly better than the best traditional model (F1-score [95% CI]: 0.67 [0.64-0.69] vs. 0.45 [0.42-0.48]). We illustrate, via experimentation using varying training set sizes, that the transformer-based models are more stable and require less annotated data compared to the other models. The significant improvements achieved by our best-performing classification model over past approaches makes it suitable for automated continuous monitoring of nonmedical PM use from Twitter.Conclusions: BERT, BERT-like and fusion-based models outperform traditional machine learning and deep learning models, achieving substantial improvements over many years of past research on the topic of prescription medication misuse/abuse classification from social media, which had been shown to be a complex task due to the unique ways in which information about nonmedical use is presented. Several challenges associated with the lack of context and the nature of social media language need to be overcome to further improve BERT and BERT-like models. These experimental driven challenges are represented as potential future research directions. [ABSTRACT FROM AUTHOR]

Background: Cancer is a leading cause of death in the United States. Primary care providers (PCPs) juggle patient cancer prevention and screening along with managing acute and chronic health problems. However, clinical decision support (CDS) may assist PCPs in addressing patients' cancer prevention and screening needs during short clinic visits. In this paper, we describe pre-implementation study design and cancer screening and prevention CDS changes made to maximize utilization and better fit a healthcare system's goals and culture. We employed the Consolidated Framework for Implementation Research (CFIR), useful for evaluating the implementation of CDS interventions in primary care settings, in understanding barriers and facilitators that led to those changes.Methods: In a three-arm, pragmatic, 36 clinic cluster-randomized control trial, we integrated cancer screening and prevention CDS and shared decision-making tools (SDMT) into an existing electronic medical record-linked cardiovascular risk management CDS system. The integrated CDS is currently being tested within a predominately rural upper Midwestern healthcare system. Prior to CDS implementation, we catalogued pre-implementation changes made from 2016 to 2018 based on: pre-implementation site engagement; key informant interviews with healthcare system rooming staff, providers, and leadership; and pilot testing. We identified influential barriers, facilitators, and changes made in response through qualitative content analysis of meeting minutes and supportive documents. We then coded pre-implementation changes made and associated barriers and facilitators using the CFIR.Results: Based on our findings from system-wide pre-implementation engagement, pilot testing, and key informant interviews, we made changes to accommodate the needs of the healthcare system based on barriers and facilitators that fell within the Intervention Characteristics, Inner Setting, and Outer Setting CFIR domains. Changes included replacing the expansion of medical assistant roles in one intervention arm with targeted SDMT, as well as altering cancer prevention CDS and study design elements.Conclusions: Pre-implementation changes to CDS may help meet healthcare systems' evolving needs and optimize the intervention by being responsive to real-world implementation barriers and facilitators. Frameworks like the CFIR are useful tools for identifying areas where pre-implementation barriers and facilitators may result in design changes, both to research studies and CDS systems.Trial Registration: NCT02986230. [ABSTRACT FROM AUTHOR]

Background: Adoption of EHRs by U.S. ambulatory practices has been slow despite the perceived benefits of their use. Most evaluations of EHR implementations in the literature apply to large practice settings. While there are similarities relating to EHR implementation in large and small practice settings, the authors argue that scale is an important differentiator. Focusing on small ambulatory practices, this paper outlines the benefits and barriers to EHR use in this setting, and provides a "field guide" for these practices to facilitate successful EHR implementation. Discussion: The benefits of EHRs in ambulatory practices include improved patient care and office efficiency, and potential financial benefits. Barriers to EHRs include costs; lack of standardization of EHR products and the design of vendor systems for large practice environments; resistance to change; initial difficulty of system use leading to productivity reduction; and perceived accrual of benefits to society and payers rather than providers. The authors stress the need for developing a flexible change management strategy when introducing EHRs that is relevant to the small practice environment; the strategy should acknowledge the importance of relationship management and the role of individual staff members in helping the entire staff to manage change. Practice staff must create an actionable vision outlining realistic goals for the implementation, and all staff must buy into the project. The authors detail the process of implementing EHRs through several stages: decision, selection, pre-implementation, implementation, and post-implementation. They stress the importance of identifying a champion to serve as an advocate of the value of EHRs and provide direction and encouragement for the project. Other key activities include assessing and redesigning workflow; understanding financial issues; conducting training that is well-timed and meets the needs of practice staff; and evaluating the implementation process. Summary: The EHR implementation experience depends on a variety of factors including the technology, training, leadership, the change management process, and the individual character of each ambulatory practice environment. Sound processes must support both technical and personnel-related organizational components. Additional research is needed to further refine recommendations for the small physician practice and the nuances of specific medical specialties. [ABSTRACT FROM AUTHOR]

Background: Cardiovascular disease is the leading cause of death in the United States (US). Despite the well-recognized efficacy of statins, statin discontinuation rates remain high. Statin intolerance is a major cause of statin discontinuation. To accurately diagnose statin intolerance, healthcare professionals must distinguish between statin-associated and non-statin-associated muscle symptoms, because many muscle symptoms can be unrelated to statin therapy. Patients' feedback on muscle-related symptoms would help providers make decisions about statin treatment. Given the potential benefits and feasibility of existing apps for cardiovascular disease (CVD) management and the unmet need for an app specifically addressing statin intolerance management, the objectives of the study were 1) to describe the developmental process of a novel app designed for patients who are eligible for statin therapy to lower the risk of CVD; 2) to explore healthcare providers' feedback of the app; and 3) to explore patients' app usage experience. Methods: The app was developed by an interdisciplinary team. Healthcare provider participants and patient participants were recruited in the study. Providers were interviewed to provide their feedback about the app based on screenshots of the app. Patients were interviewed after a 30 days of app usage. Results: The basic features of the app included symptom logging, vitals tracking, patient education, and push notifications. Overall, both parties provided positive feedback about the app. Areas to be improved mentioned by both parties included: the pain question asked in symptom tracking and the patient education section. Both parties agreed that it was essential to add the trend report of the logged symptoms. Conclusions: The results indicated that providers were willing to use patient-reported data for disease management and perceived that the app had the potential to facilitate doctor-patient communication. Results also indicated that user engagement is the key to the success of app efficacy. To promote app engagement, app features should be tailored to individual patient's needs and goals. In the future, after it is upgraded, we plan to test the app usability and feasibility among a more diverse sample. [ABSTRACT FROM AUTHOR]

Background: The early detection and management of uncontrolled cardiovascular risk factors among prediabetes patients can prevent cardiovascular disease (CVD). Prediabetes increases the risk of CVD, which is a leading cause of death in the United States. CVD clinical decision support (CDS) in primary care settings has the potential to reduce cardiovascular risk in patients with prediabetes while potentially saving clinicians time. The objective of this study is to understand primary care clinician (PCC) perceptions of a CDS system designed to reduce CVD risk in adults with prediabetes.Methods: We administered pre-CDS implementation (6/30/2016 to 8/25/2016) (n = 183, 61% response rate) and post-CDS implementation (6/12/2019 to 8/7/2019) (n = 131, 44.5% response rate) independent cross-sectional electronic surveys to PCCs at 36 randomized primary care clinics participating in a federally funded study of a CVD risk reduction CDS tool. Surveys assessed PCC demographics, experiences in delivering prediabetes care, perceptions of CDS impact on shared decision making, perception of CDS impact on control of major CVD risk factors, and overall perceptions of the CDS tool when managing cardiovascular risk.Results: We found few significant differences when comparing pre- and post-implementation responses across CDS intervention and usual care (UC) clinics. A majority of PCCs felt well-prepared to discuss CVD risk factor control with patients both pre- and post-implementation. About 73% of PCCs at CDS intervention clinics agreed that the CDS helped improve risk control, 68% reported the CDS added value to patient clinic visits, and 72% reported they would recommend use of this CDS system to colleagues. However, most PCCs disagreed that the CDS saves time talking about preventing diabetes or CVD, and most PCCs also did not find the clinical domains useful, nor did PCCs believe that the clinical domains were useful in getting patients to take action. Finally, only about 38% reported they were satisfied with the CDS.Conclusions: These results improve our understanding of CDS user experience and can be used to guide iterative improvement of the CDS. While most PCCs agreed the CDS improves CVD and diabetes risk factor control, they were generally not satisfied with the CDS. Moreover, only 40-50% agreed that specific suggestions on clinical domains helped patients to take action. In spite of this, an overwhelming majority reported they would recommend the CDS to colleagues, pointing for the need to improve upon the current CDS.Trial Registration: NCT02759055 03/05/2016. [ABSTRACT FROM AUTHOR]

Background Since the introduction of electronic nursing documentation systems, its implementation in recent years has increased rapidly in Germany. The objectives of such systems are to save time, to improve information handling and to improve quality. To integrate IT in the daily working processes, the employee is the pivotal element. Therefore it is important to understand nurses' experience with IT implementation. At present the literature shows a lack of understanding exploring staff experiences within the implementation process. Methods A systematic review and meta-ethnographic synthesis of primary studies using qualitative methods was conducted in PubMed, CINAHL, and Cochrane. It adheres to the principles of the PRISMA statement. The studies were original, peer-reviewed articles from 2000 to 2013, focusing on computer-based nursing documentation in Residential Aged Care Facilities. Results The use of IT requires a different form of information processing. Some experience this new form of information processing as a benefit while others do not. The latter find it more difficult to enter data and this result in poor clinical documentation. Improvement in the quality of residents' records leads to an overall improvement in the quality of care. However, if the quality of those records is poor, some residents do not receive the necessary care. Furthermore, the length of time necessary to complete the documentation is a prominent theme within that process. Those who are more efficient with the electronic documentation demonstrate improved time management. For those who are less efficient with electronic documentation the information processing is perceived as time consuming. Normally, it is possible to experience benefits when using IT, but this depends on either promoting or hindering factors, e. g. ease of use and ability to use it, equipment availability and technical functionality, as well as attitude. Conclusions In summary, the findings showed that members of staff experience IT as a benefit when it simplifies their daily working routines and as a burden when it complicates their working processes. Whether IT complicates or simplifies their routines depends on influencing factors. The line between benefit and burden is semipermeable. The experiences differ according to duties and responsibilities. [ABSTRACT FROM AUTHOR]

Background: Notifiable disease surveillance in the United States is predominantly a passive process that is often limited by poor timeliness and low sensitivity. Interoperable tools are needed that interact more seamlessly with existing clinical and laboratory data to improve notifiable disease surveillance. Description: The Public Health Surveillance Knowledgebase (PHSkb±) is a computer database designed to provide quick, easy access to domain knowledge regarding notifiable diseases and conditions in the United States. The database was developed using Protégé ontology and knowledgebase editing software. Data regarding the notifiable disease domain were collected via a comprehensive review of state health department websites and integrated with other information used to support the National Notifiable Diseases Surveillance System (NNDSS). Domain concepts were harmonized, wherever possible, to existing vocabulary standards. The knowledgebase can be used: 1) as the basis for a controlled vocabulary of reportable conditions needed for data aggregation in public health surveillance systems; 2) to provide queriable domain knowledge for public health surveillance partners; 3) to facilitate more automated case detection and surveillance decision support as a reusable component in an architecture for intelligent clinical, laboratory, and public health surveillance information systems. Conclusions: The PHSkb provides an extensible, interoperable system architecture component to support notifiable disease surveillance. Further development and testing of this resource is needed. [ABSTRACT FROM AUTHOR]

Background: Healthcare is expected to increasingly integrate technologies enabled by artificial intelligence (AI) into patient care. Understanding perceptions of these tools is essential to successful development and adoption. This exploratory study gauged participants' level of openness, concern, and perceived benefit associated with AI-driven healthcare technologies. We also explored socio-demographic, health-related, and psychosocial correlates of these perceptions.Methods: We developed a measure depicting six AI-driven technologies that either diagnose, predict, or suggest treatment. We administered the measure via an online survey to adults (N = 936) in the United States using MTurk, a crowdsourcing platform. Participants indicated their level of openness to using the AI technology in the healthcare scenario. Items reflecting potential concerns and benefits associated with each technology accompanied the scenarios. Participants rated the extent that the statements of concerns and benefits influenced their perception of favorability toward the technology. Participants completed measures of socio-demographics, health variables, and psychosocial variables such as trust in the healthcare system and trust in technology. Exploratory and confirmatory factor analyses of the concern and benefit items identified two factors representing overall level of concern and perceived benefit. Descriptive analyses examined levels of openness, concern, and perceived benefit. Correlational analyses explored associations of socio-demographic, health, and psychosocial variables with openness, concern, and benefit scores while multivariable regression models examined these relationships concurrently.Results: Participants were moderately open to AI-driven healthcare technologies (M = 3.1/5.0 ± 0.9), but there was variation depending on the type of application, and the statements of concerns and benefits swayed views. Trust in the healthcare system and trust in technology were the strongest, most consistent correlates of openness, concern, and perceived benefit. Most other socio-demographic, health-related, and psychosocial variables were less strongly, or not, associated, but multivariable models indicated some personality characteristics (e.g., conscientiousness and agreeableness) and socio-demographics (e.g., full-time employment, age, sex, and race) were modestly related to perceptions.Conclusions: Participants' openness appears tenuous, suggesting early promotion strategies and experiences with novel AI technologies may strongly influence views, especially if implementation of AI technologies increases or undermines trust. The exploratory nature of these findings warrants additional research. [ABSTRACT FROM AUTHOR]

Background: With the use of electronic health records (EHRs) increasing and causing unintended negative consequences, the medical scribe profession has burgeoned, but it has yet to be regulated. The purpose of this study was to describe scribe workflow as well as identify the threats and opportunities for the future of the scribe industry.Methods: The first phase of the study used ethnographic methods consisting of interviews and observations by a multi-disciplinary team of researchers at five United States sites. In April 2019, a two-day conference of experts representing different stakeholder perspectives was held to discuss the results from site visits and to predict the future of medical scribing. An interpretive content analysis approach was used to discover threats and opportunities for the future of medical scribes.Results: Threats facing the medical scribe industry were related to changes in the documentation model, EHR usability, different payment structures, the need to acquire disparate data during clinical encounters, and workforce-related changes relevant to the scribing model. Simultaneously, opportunities for medical scribing in the future included extension of their role to include workflow analysis, acting as EHR-related subject-matter-experts, and becoming integrated more effectively into the clinical care delivery team. Experts thought that if EHR usability increases, the need for medical scribes might decrease. Additionally, the scribe role could be expanded to allow scribes to document more or take on more informatics-related tasks. The experts also anticipated an increased use of alternative models of scribing, like tele-scribing.Conclusion: Threats and opportunities for medical scribing were identified. Many experts thought that if the scribe role could be expanded to allow scribes to document more or take on more informatics activities, it would be beneficial. With COVID-19 continuing to change workflows, it is critical that medical scribes receive standardized training as tele-scribing continues to grow in popularity and new roles for scribes as medical team members are identified. [ABSTRACT FROM AUTHOR]

Background: The patient ranking process for donor lung allocation in the United States is carried out by a classification-based, computerized algorithm, known as the match system. Experts have suggested that a continuous, points-based allocation framework would better serve waiting list candidates by removing hard boundaries and increasing transparency into the relative importance of factors used to prioritize candidates. We applied discrete choice modeling to match run data to determine the feasibility of approximating current lung allocation policy by one or more composite scores. Our study aimed to demystify the points-based approach to organ allocation policy; quantify the relative importance of factors used in current policy; and provide a viable policy option that adapts the current, classification-based system to the continuous allocation framework.Methods: Rank ordered logistic regression models were estimated using 6466 match runs for 5913 adult donors and 534 match runs for 488 pediatric donors from 2018. Four primary attributes are used to rank candidates and were included in the models: (1) medical priority, (2) candidate age, (3) candidate's transplant center proximity to the donor hospital, and (4) blood type compatibility with the donor.Results: Two composite scores were developed, one for adult and one for pediatric donor allocation. Candidate rankings based on the composite scores were highly correlated with current policy rankings (Kendall's Tau ~ 0.80, Spearman correlation > 90%), indicating both scores strongly reflect current policy. In both models, candidates are ranked higher if they have higher medical priority, are registered at a transplant center closer to the donor hospital, or have an identical blood type to the donor. Proximity was the most important attribute. Under a points-based scoring system, candidates in further away zones are sometimes ranked higher than more proximal candidates compared to current policy.Conclusions: Revealed preference analysis of lung allocation match runs produced composite scores that capture the essence of current policy while removing rigid boundaries of the current classification-based system. A carefully crafted, continuous version of lung allocation policy has the potential to make better use of the limited supply of donor lungs in a manner consistent with the priorities of the transplant community. [ABSTRACT FROM AUTHOR]

Background: Cardiovascular disease (CVD) is the leading cause of death in the United States (US). Better cardiovascular health (CVH) is associated with CVD prevention. Predicting future CVH levels may help providers better manage patients' CVH. We hypothesized that CVH measures can be predicted based on previous measurements from longitudinal electronic health record (EHR) data.Methods: The Guideline Advantage (TGA) dataset was used and contained EHR data from 70 outpatient clinics across the United States (US). We studied predictions of 5 CVH submetrics: smoking status (SMK), body mass index (BMI), blood pressure (BP), hemoglobin A1c (A1C), and low-density lipoprotein (LDL). We applied embedding techniques and long short-term memory (LSTM) networks - to predict future CVH category levels from all the previous CVH measurements of 216,445 unique patients for each CVH submetric.Results: The LSTM model performance was evaluated by the area under the receiver operator curve (AUROC): the micro-average AUROC was 0.99 for SMK prediction; 0.97 for BMI; 0.84 for BP; 0.91 for A1C; and 0.93 for LDL prediction. Model performance was not improved by using all 5 submetric measures compared with using single submetric measures.Conclusions: We suggest that future CVH levels can be predicted using previous CVH measurements for each submetric, which has implications for population cardiovascular health management. Predicting patients' future CVH levels might directly increase patient CVH health and thus quality of life, while also indirectly decreasing the burden and cost for clinical health system caused by CVD and cancers. [ABSTRACT FROM AUTHOR]

Background: Several studies highlight the effects of artificial intelligence (AI) systems on healthcare delivery. AI-based tools may improve prognosis, diagnostics, and care planning. It is believed that AI will be an integral part of healthcare services in the near future and will be incorporated into several aspects of clinical care. Thus, many technology companies and governmental projects have invested in producing AI-based clinical tools and medical applications. Patients can be one of the most important beneficiaries and users of AI-based applications whose perceptions may affect the widespread use of AI-based tools. Patients should be ensured that they will not be harmed by AI-based devices, and instead, they will be benefited by using AI technology for healthcare purposes. Although AI can enhance healthcare outcomes, possible dimensions of concerns and risks should be addressed before its integration with routine clinical care.Methods: We develop a model mainly based on value perceptions due to the specificity of the healthcare field. This study aims at examining the perceived benefits and risks of AI medical devices with clinical decision support (CDS) features from consumers' perspectives. We use an online survey to collect data from 307 individuals in the United States.Results: The proposed model identifies the sources of motivation and pressure for patients in the development of AI-based devices. The results show that technological, ethical (trust factors), and regulatory concerns significantly contribute to the perceived risks of using AI applications in healthcare. Of the three categories, technological concerns (i.e., performance and communication feature) are found to be the most significant predictors of risk beliefs.Conclusions: This study sheds more light on factors affecting perceived risks and proposes some recommendations on how to practically reduce these concerns. The findings of this study provide implications for research and practice in the area of AI-based CDS. Regulatory agencies, in cooperation with healthcare institutions, should establish normative standard and evaluation guidelines for the implementation and use of AI in healthcare. Regular audits and ongoing monitoring and reporting systems can be used to continuously evaluate the safety, quality, transparency, and ethical factors of AI-based services. [ABSTRACT FROM AUTHOR]

Background: Implementation theories, models and frameworks offer guidance when implementing and sustaining healthcare evidence-based interventions. However, selection can be challenging given the myriad of potential options. We propose to inform a decision support tool to facilitate the appropriate selection of an implementation theory, model or framework in practice. To inform tool development, this study aimed to explore barriers and facilitators to identifying and selecting implementation theories, models and frameworks in research and practice, as well as end-user preferences for features and functions of the proposed tool.Methods: We used an interpretive descriptive approach to conduct semi-structured interviews with implementation researchers and practitioners in Canada, the United States and Australia. Audio recordings were transcribed verbatim. Data were inductively coded by a single investigator with a subset of 20% coded independently by a second investigator and analyzed using thematic analysis.Results: Twenty-four individuals participated in the study. Categories of barriers/facilitators, to inform tool development, included characteristics of the individual or team conducting implementation and characteristics of the implementation theory, model or framework. Major barriers to selection included inconsistent terminology, poor fit with the implementation context and limited knowledge about and training in existing theories, models and frameworks. Major facilitators to selection included the importance of clear and concise language and evidence that the theory, model or framework was applied in a relevant health setting or context. Participants were enthusiastic about the development of a decision support tool that is user-friendly, accessible and practical. Preferences for tool features included key questions about the implementation intervention or project (e.g., purpose, stage of implementation, intended target for change) and a comprehensive list of relevant theories, models and frameworks to choose from along with a glossary of terms and the contexts in which they were applied.Conclusions: An easy to use decision support tool that addresses key barriers to selecting an implementation theory, model or framework in practice may be beneficial to individuals who facilitate implementation practice activities. Findings on end-user preferences for tool features and functions will inform tool development and design through a user-centered approach. [ABSTRACT FROM AUTHOR]

Background: Community-acquired pneumonia (CAP) is one of the leading causes of morbidity and mortality in the USA. Our objective was to assess the predictive value on critical illness and disposition of a sequential Bayesian Model that integrates Lactate and procalcitonin (PCT) for pneumonia.Methods: Sensitivity and specificity of lactate and PCT attained from pooled meta-analysis data. Likelihood ratios calculated and inserted in Bayesian/ Fagan nomogram to calculate posttest probabilities. Bayesian Diagnostic Gains (BDG) were analyzed comparing pre and post-test probability. To assess the value of integrating both PCT and Lactate in Severity of Illness Prediction we built a model that combined CURB65 with PCT as the Pre-Test markers and later integrated the Lactate Likelihood Ratio Values to generate a combined CURB 65 + Procalcitonin + Lactate Sequential value.Results: The BDG model integrated a CUBR65 Scores combined with Procalcitonin (LR+ and LR-) for Pre-Test Probability Intermediate and High with Lactate Positive Likelihood Ratios. This generated for the PCT LR+ Post-test Probability (POSITIVE TEST) Posterior probability: 93% (95% CI [91,96%]) and Post Test Probability (NEGATIVE TEST) of: 17% (95% CI [15-20%]) for the Intermediate subgroup and 97% for the high risk sub-group POSITIVE TEST: Post-Test probability:97% (95% CI [95,98%]) NEGATIVE TEST: Post-test probability: 33% (95% CI [31,36%]) . ANOVA analysis for CURB 65 (alone) vs CURB 65 and PCT (LR+) vs CURB 65 and PCT (LR+) and Lactate showed a statistically significant difference (P value = 0.013).Conclusions: The sequential combination of CURB 65 plus PCT with Lactate yielded statistically significant results, demonstrating a greater predictive value for severity of illness thus ICU level care. [ABSTRACT FROM AUTHOR]

Objective: To examine the association between the medical imaging utilization and information related to patients' socioeconomic, demographic and clinical factors during the patients' ED visits; and to develop predictive models using these associated factors including natural language elements to predict the medical imaging utilization at pediatric ED.Methods: Pediatric patients' data from the 2012-2016 United States National Hospital Ambulatory Medical Care Survey was included to build the models to predict the use of imaging in children presenting to the ED. Multivariable logistic regression models were built with structured variables such as temperature, heart rate, age, and unstructured variables such as reason for visit, free text nursing notes and combined data available at triage. NLP techniques were used to extract information from the unstructured data.Results: Of the 27,665 pediatric ED visits included in the study, 8394 (30.3%) received medical imaging in the ED, including 6922 (25.0%) who had an X-ray and 1367 (4.9%) who had a computed tomography (CT) scan. In the predictive model including only structured variables, the c-statistic was 0.71 (95% CI: 0.70-0.71) for any imaging use, 0.69 (95% CI: 0.68-0.70) for X-ray, and 0.77 (95% CI: 0.76-0.78) for CT. Models including only unstructured information had c-statistics of 0.81 (95% CI: 0.81-0.82) for any imaging use, 0.82 (95% CI: 0.82-0.83) for X-ray, and 0.85 (95% CI: 0.83-0.86) for CT scans. When both structured variables and free text variables were included, the c-statistics reached 0.82 (95% CI: 0.82-0.83) for any imaging use, 0.83 (95% CI: 0.83-0.84) for X-ray, and 0.87 (95% CI: 0.86-0.88) for CT.Conclusions: Both CT and X-rays are commonly used in the pediatric ED with one third of the visits receiving at least one. Patients' socioeconomic, demographic and clinical factors presented at ED triage period were associated with the medical imaging utilization. Predictive models combining structured and unstructured variables available at triage performed better than models using structured or unstructured variables alone, suggesting the potential for use of NLP in determining resource utilization. [ABSTRACT FROM AUTHOR]

Background: The use of post-acute care (PAC) for cardiovascular conditions is highly variable across geographical regions. Although PAC benefits include lower readmission rates, better clinical outcomes, and lower mortality, referral patterns vary widely, raising concerns about substandard care and inflated costs. The objective of this study is to identify factors associated with PAC referral decisions at acute care discharge.Methods: This study is a retrospective Electronic Health Records (EHR) based review of a cohort of patients with coronary artery bypass graft (CABG) and valve replacement (VR). EHR records were extracted from the Cerner Health-Facts Data warehouse and covered 49 hospitals in the United States of America (U.S.) from January 2010 to December 2015. Multinomial logistic regression was used to identify associations of 29 variables comprising patient characteristics, hospital profiles, and patient conditions at discharge.Results: The cohort had 14,224 patients with mean age 63.5 years, with 10,234 (71.9%) male and 11,946 (84%) Caucasian, with 5827 (40.96%) being discharged to home without additional care (Home), 5226 (36.74%) to home health care (HHC), 1721 (12.10%) to skilled nursing facilities (SNF), 1168 (8.22%) to inpatient rehabilitation facilities (IRF), 164 (1.15%) to long term care hospitals (LTCH), and 118 (0.83%) to other locations. Census division, hospital size, teaching hospital status, gender, age, marital status, length of stay, and Charlson comorbidity index were identified as highly significant variables (p- values < 0.001) that influence the PAC referral decision. Overall model accuracy was 62.6%, and multiclass Area Under the Curve (AUC) values were for Home: 0.72; HHC: 0.72; SNF: 0.58; IRF: 0.53; LTCH: 0.52, and others: 0.46.Conclusions: Census location of the acute care hospital was highly associated with PAC referral practices, as was hospital capacity, with larger hospitals referring patients to PAC at a greater rate than smaller hospitals. Race and gender were also statistically significant, with Asians, Hispanics, and Native Americans being less likely to be referred to PAC compared to Caucasians, and female patients being more likely to be referred than males. Additional analysis indicated that PAC referral practices are also influenced by the mix of PAC services offered in each region. [ABSTRACT FROM AUTHOR]

Background: The purpose of this study is to assess the Business-to-Consumer (B2C) model for telemonitoring patients with Chronic Heart Failure (CHF) by analysing the value it creates, both for organizations or ventures that provide telemonitoring services based on it, and for society.Methods: The business model assessment was based on the following categories: caveats, venture type, six-factor alignment, strategic market assessment, financial viability, valuation analysis, sustainability, societal impact, and technology assessment. The venture valuation was performed for three jurisdictions (countries) - Singapore, the Netherlands and the United States - in order to show the opportunities in a small, medium-sized, and large country (i.e. population).Results: The business model assessment revealed that B2C telemonitoring is viable and profitable in the Innovating in Healthcare Framework. Analysis of the ecosystem revealed an average-to-excellent fit with the six factors. The structure and financing fit was average, public policy and technology alignment was good, while consumer alignment and accountability fit was deemed excellent. The financial prognosis revealed that the venture is viable and profitable in Singapore and the Netherlands but not in the United States due to relatively high salary inputs.Conclusions: The B2C model in telemonitoring CHF potentially creates value for patients, shareholders of the service provider, and society. However, the validity of the results could be improved, for instance by using a peer-reviewed framework, a systematic literature search, case-based cost/efficiency inputs, and varied scenario inputs. [ABSTRACT FROM AUTHOR]

Background: This study's goal is to understand the factors that drive individuals to adopt Electronic Health Record (EHR) portals and to estimate if there are differences between countries with different healthcare models.Methods: We applied a new adoption model using as a starting point the extended Unified Theory of Acceptance and Use of Technology (UTAUT2) by incorporating the Concern for Information Privacy (CFIP) framework. To evaluate the research model we used the partial least squares (PLS) - structural equation modelling (SEM) approach. An online questionnaire was administrated in the United States (US) and Europe (Portugal). We collected 597 valid responses.Results: The statistically significant factors of behavioural intention are performance expectancy ([Formula: see text] total = 0.285; P < 0.01), effort expectancy ([Formula: see text] total = 0.160; P < 0.01), social influence ([Formula: see text] total = 0.198; P < 0.01), hedonic motivation ([Formula: see text] total = -0.141; P < 0.01), price value ([Formula: see text] total = 0.152; P < 0.01), and habit ([Formula: see text] total = 0.255; P < 0.01). The predictors of use behaviour are habit ([Formula: see text] total = 0.145; P < 0.01), and behavioural intention ([Formula: see text] total = 0.480; P < 0.01). Social influence, hedonic motivation, and price value are only predictors in the US group. The model explained 53% of the variance in behavioural intention and 36% of the variance in use behaviour.Conclusions: Our study identified critical factors for the adoption of EHR portals and significant differences between the countries. Confidentiality issues do not seem to influence acceptance. The EHR portals usage patterns are significantly higher in US compared to Portugal. [ABSTRACT FROM AUTHOR]

Background: It is becoming increasingly common for individuals and organizations to use social media platforms such as Facebook. These are being used for a wide variety of purposes including disseminating, discussing and seeking health related information. U.S. Federal health agencies are leveraging these platforms to 'engage' social media users to read, spread, promote and encourage health related discussions. However, different agencies and their communications get varying levels of engagement. In this study we use statistical models to identify factors that associate with engagement.Methods: We analyze over 45,000 Facebook posts from 72 Facebook accounts belonging to 24 health agencies. Account usage, user activity, sentiment and content of these posts are studied. We use the hurdle regression model to identify factors associated with the level of engagement and Cox proportional hazards model to identify factors associated with duration of engagement.Results: In our analysis we find that agencies and accounts vary widely in their usage of social media and activity they generate. Statistical analysis shows, for instance, that Facebook posts with more visual cues such as photos or videos or those which express positive sentiment generate more engagement. We further find that posts on certain topics such as occupation or organizations negatively affect the duration of engagement.Conclusions: We present the first comprehensive analyses of engagement with U.S. Federal health agencies on Facebook. In addition, we briefly compare and contrast findings from this study to our earlier study with similar focus but on Twitter to show the robustness of our methods. [ABSTRACT FROM AUTHOR]

Background Control of blood pressure is frequently inadequate in spite of availability of several classes of well tolerated and effective antihypertensive drugs. Several factors, including the use of suboptimal doses of drugs, inadequate or ineffective treatments and poor drug compliance may be the reason for this phenomenon. The aim of the current non- randomized study was to evaluate the effectiveness of a Home-Based Telemedicine service in patients with uncontrolled hypertension. Methods 74 patients were enrolled in a Home Based Telemedicine group and 94 patients in the Usual Care group. At baseline and at the end of the study, patients in both groups were seen in a cardiology office. Patients in Home Based Telemedicine group additionally were followed by a physician-nurse, through scheduled and unscheduled telephone appointments. These patients also received a blood pressure measuring device that could transmit the readings to a central data monitor via secure data connection. Results During the study period (80 ± 25 days), a total of 17401 blood pressure measurements were taken in the Home Based Telemedicine group corresponding to 236 ± 136 readings per patient and a mean daily measurement of 3 ± 1.7. The scheduled telephone contacts (initiated by the nurse) equaled to 5.2 ± 4.3/patient (370 in total) and the unscheduled telephone contacts (initiated by the patients) were 0.4 ± 0.9/patient (30 in total). The mean systolic blood pressure values decreased from 153 ± 19 mmHg to 130 ± 15 mmHg (p < 0.0001) at the end of the study and diastolic blood pressure values decreased from 89 ± 10 mmHg to 76 ± 11 mmHg (p < 0.0001). In the Usual Care group, the mean systolic blood pressure values decreased from 156 ± 16 mmHg to 149 ± 17 mmHg (p < 0.05) at the end of the study and diastolic blood pressure values decreased from 90 ± 8 mmHg to 86 ± 9 mmHg (p < 0.05). The changes in drug therapy initiated following telephone contacts were 1.81 ± 1.73 per patient. Conclusions The addition of a structured physician-nurse approach supported by remote telemonitoring of blood pressure is likely to improve outcome in patients with uncontrolled hypertension. [ABSTRACT FROM AUTHOR]

Background To demonstrate that subject selection based on sufficient laboratory results and medication orders in electronic health records can be biased towards sick patients. Methods Using electronic health record data from 10,000 patients who received anesthetic services at a major metropolitan tertiary care academic medical center, an affiliated hospital for women and children, and an affiliated urban primary care hospital, the correlation between patient health status and counts of days with laboratory results or medication orders, as indicated by the American Society of Anesthesiologists Physical Status Classification (ASA Class), was assessed with a Negative Binomial Regression model. Results Higher ASA Class was associated with more points of data: compared to ASA Class 1 patients, ASA Class 4 patients had 5.05 times the number of days with laboratory results and 6.85 times the number of days with medication orders, controlling for age, sex, emergency status, admission type, primary diagnosis, and procedure. Conclusions Imposing data sufficiency requirements for subject selection allows researchers to minimize missing data when reusing electronic health records for research, but introduces a bias towards the selection of sicker patients. We demonstrated the relationship between patient health and quantity of data, which may result in a systematic bias towards the selection of sicker patients for research studies and limit the external validity of research conducted using electronic health record data. Additionally, we discovered other variables (i.e., admission status, age, emergency classification, procedure, and diagnosis) that independently affect data sufficiency. [ABSTRACT FROM AUTHOR]

Background Hospital-based Emergency Departments are struggling to provide timely care to a steadily increasing umber of unscheduled ED visits. Dwindling compensation and rising ED closures dictate that meeting this challenge demands greater operational efficiency. Methods Using techniques from operations research theory, as well as a novel event-driven algorithm for processing priority queues, we developed a flexible simulation platform for hospital-based EDs. We tuned the parameters of the system to mimic U.S. nationally average and average academic hospitalbased ED performance metrics and are able to assess a variety of patient flow outcomes including patient door-to-event times, propensity to leave without being seen, ED occupancy level, and dynamic staffing and resource use. Results The causes of ED crowding are variable and require site-specific solutions. For example, in a nationally average ED environment, provider availability is a surprising, but persistent bottleneck in patient flow. As a result, resources expended in reducing boarding times may not have the expected impact on patient throughput. On the other hand, reallocating resources into alternate care pathways can dramatically expedite care for lower acuity patients without delaying care for higher acuity patients. In an average academic ED environment, bed availability is the primary bottleneck in patient flow. Consequently, adjustments to provider scheduling have a limited effect on the timeliness of care delivery, while shorter boarding times significantly reduce crowding. An online version of the simulation platform is available at http://spark.rstudio.com/klopiano/EDsimulation/. Conclusion In building this robust simulation framework, we have created a novel decision-support tool that ED and hospital managers can use to quantify the impact of proposed changes to patient flow prior to implementation. [ABSTRACT FROM AUTHOR]

Background: Health care news stories have the potential to inform and educate news consumers and health-care consumers about the tradeoffs involved in health-care decisions about treatments, tests, products, and procedures. These stories have the potential to influence not only individual decision making but also the broader public dialogue about health-care reform. For the past 7 years, a Web-based project called http://HealthNewsReview.org has evaluated news stories to try to improve health-care journalism and the quality and flow of information to consumers. Analysis: http://HealthNewsReview.org applies 10 standardized criteria to the review of news stories that include claims about medical interventions. Two or three reviewers evaluate each story. The team has evaluated more than 1,800 stories by more than a dozen leading U.S. news organizations. About 70% have received unsatisfactory scores based on application of these criteria: reporting on costs, quantifying potential benefits, and quantifying potential harms. Conclusions: Inaccurate, imbalanced, incomplete news stories may drown out more careful scrutiny of the evidence by many influential news organizations. Unquestioned claims and assertions about the benefits of medical interventions are passed on to the American public daily by journalists who are supposed to vet independently any such claims. Communication about these issues is, in itself, a major health-policy issue. [ABSTRACT FROM AUTHOR]

Background: In 2009, the U.S. Preventive Services Task Force released new mammography screening guidelines that sparked a torrent of criticism. The subsequent conflict was significant and pitted the Task Force against other health organizations, advocacy groups, the media, and the public at large. We argue that this controversy was driven by the systematic removal of uncertainty from science communication. To increase comprehension and adherence, health information communicators remove caveats, limitations, and hedging so science appears simple and more certain. This streamlining process is, in many instances, initiated by researchers as they engage in dissemination of their findings, and it is facilitated by public relations professionals, journalists, public health practitioners, and others whose tasks involve using the results from research for specific purposes. Analysis: Uncertainty is removed from public communication because many communicators believe that it is difficult for people to process and/or that it is something the audience wants to avoid. Uncertainty management theory posits that people can find meaning and value in uncertainty. We define key terms relevant to uncertainty management, describe research on the processing of uncertainty, identify directions for future research, and offer recommendations for scientists, practitioners, and media professionals confronted with uncertain findings. Conclusions: Science is routinely simplified as it is prepared for public consumption. In line with the model of information overload, this practice may increase short-term adherence to recommendations at the expense of long-term message consistency and trust in science. [ABSTRACT FROM AUTHOR]

Background: In accordance with the People's Republic of China's (China) National Health Reform Plan of 2009, two of the nation's leading hospitals, located in Beijing, have implemented electronic medical record (EMR) systems from different vendors. To inform future EMR adoption and policy in China, as well as informatics research in the US, this study compared the United State's Hospital Meaningful Use (MU) Objectives (phase 1) objectives to the EMR functionality of two early hospital EMR adopters in China. Methods: At both hospitals, the researchers observed a physician using the EMR and noted MU functionality that was seen and functionality that was not seen yet was available in the EMR. The information technology department was asked about the availability of functionality neither observed nor known to the physician. Results and conclusions: Approximately half the MU objectives were available in each EMR. Some differences between the EMRs in the study and MU objectives were attributed to operational differences between the health systems and the cultures in the two countries. [ABSTRACT FROM AUTHOR]

Background: Blindness due to diabetic retinopathy (DR) is the major disability in diabetic patients. Although early management has shown to prevent vision loss, diabetic patients have a low rate of routine ophthalmologic examination. Hence, we developed and validated sparse learning models with the aim of identifying the risk of DR in diabetic patients. Methods: Health records from the Korea National Health and Nutrition Examination Surveys (KNHANES) V-1 were used. The prediction models for DR were constructed using data from 327 diabetic patients, and were validated internally on 163 patients in the KNHANES V-1. External validation was performed using 562 diabetic patients in the KNHANES V-2. The learning models, including ridge, elastic net, and LASSO, were compared to the traditional indicators of DR. Results: Considering the Bayesian information criterion, LASSO predicted DR most efficiently. In the internal and external validation, LASSO was significantly superior to the traditional indicators by calculating the area under the curve (AUC) of the receiver operating characteristic. LASSO showed an AUC of 0.81 and an accuracy of 73.6% in the internal validation, and an AUC of 0.82 and an accuracy of 75.2% in the external validation. Conclusion: The sparse learning model using LASSO was effective in analyzing the epidemiological underlying patterns of DR. This is the first study to develop a machine learning model to predict DR risk using health records. LASSO can be an excellent choice when both discriminative power and variable selection are important in the analysis of high-dimensional electronic health records. [ABSTRACT FROM AUTHOR]

Background: The rate of elective surgeries varies dramatically by geography in the United States. For many of these surgeries, there is not clear evidence of their relative merits over alternate treatment choices and there are significant tradeoffs in short- and long-term risks and benefits of selecting one treatment option over another. Conditions and symptoms for which there is this lack of a single clear evidence-based treatment choice present great opportunities for patient and provider collaboration on decision making; back pain and joint osteoarthritis are two such ailments. A number of decision aids are in active use to encourage this shared decision-making process. Decision aids have been assessed in formal studies that demonstrate increases in patient knowledge, increases in patient-provider engagement, and reduction in surgery rates. These studies have not widely demonstrated the added benefit of health coaching in support of shared decision making nor have they commonly provided strong evidence of cost reductions. In order to add to this evidence base, we undertook a comparative study testing the relative impact on health utilization and costs of active outreach through interactive voice response technology to encourage health coaching in support of shared decision making in comparison to mailed outreach or no outreach. This study focused on individuals with back pain or joint pain. Methods: We conducted four waves of stratified randomized comparisons for individuals with risk for back, hip, or knee surgery who did not have claims-based evidence of one or more of five chronic conditions and were eligible for population care management services within three large regional health plans in the United States. An interactive voice response (IVR) form of outreach that included the capability for individuals to directly connect with health coaches telephonically, known as AutoDialogW, was compared to a control (mailed outreach or natural levels of inbound calling depending on the study wave). In total, the study include 24,167 adults with commercial and Medicare Advantage private coverage at three health plans and at risk for lumbar back surgery, hip repair/ replacement, or knee repair/replacement. Results: Interactive voice response outreach led to 10.7 (P-value < .0001) times as many inbound calls within 30 days as the control. Over 180 days, the IVR group ("intervention") had 67 percent (P-value < .0001) more health coach communications and agreed to be sent 3.2 (P-value < .0001) time as many DVD- and/or booklet-based decision aids. Targeted surgeries were reduced by 6.7 percent (P-value = .6039). Overall costs were lower by 4.9 percent (P-value = .055). Costs that were not related to maternity, cancer, trauma and substance abuse ("actionable costs") were reduced by 6.5 percent (P-value = .0286). Conclusions: IVR with a transfer-to-health coach-option significantly increased levels of health coaching compared to mailed or no outreach and lead to significantly reduced actionable medical costs. Providing high levels of health coaching to individuals with these types of risks appears to have produced important levels of actionable medical cost reductions. We believe this impact resulted from more informed and engaged health care decision making. [ABSTRACT FROM AUTHOR]

Background: In recent years, there have been numerous initiatives undertaken to describe critical information needs related to the collection, management, analysis, and dissemination of data in support of biomedical research (J Investig Med 54:327-333, 2006); (J Am Med Inform Assoc 16:316-327, 2009); (Physiol Genomics 39:131-140, 2009); (J Am Med Inform Assoc 18:354-357, 2011). A common theme spanning such reports has been the importance of understanding and optimizing people, organizational, and leadership factors in order to achieve the promise of efficient and timely research (J Am Med Inform Assoc 15:283-289, 2008). With the emergence of clinical and translational science (CTS) as a national priority in the United States, and the corresponding growth in the scale and scope of CTS research programs, the acuity of such information needs continues to increase (JAMA 289:1278-1287, 2003); (N Engl J Med 353:1621-1623, 2005); (Sci Transl Med 3:90, 2011). At the same time, systematic evaluations of optimal people, organizational, and leadership factors that influence the provision of data, information, and knowledge management technologies and methods are notably lacking. Methods: In response to the preceding gap in knowledge, we have conducted both: 1) a structured survey of domain experts at Academic Health Centers (AHCs); and 2) a subsequent thematic analysis of public-domain documentation provided by those same organizations. The results of these approaches were then used to identify critical factors that may influence access to informatics expertise and resources relevant to the CTS domain. Results: A total of 31 domain experts, spanning the Biomedical Informatics (BMI), Computer Science (CS), Information Science (IS), and Information Technology (IT) disciplines participated in a structured surveyprocess. At a high level, respondents identified notable differences in theaccess to BMI, CS, and IT expertise and services depending on the establishment of a formal BMI academic unit and the perceived relationship between BMI, CS, IS, and IT leaders. Subsequent thematic analysis of the aforementioned public domain documents demonstrated a discordance between perceived and reported integration across and between BMI, CS, IS, and IT programs and leaders with relevance to the CTS domain. Conclusion: Differences in people, organization, and leadership factors do influence the effectiveness of CTS programs, particularly with regard to the ability to access and leverage BMI, CS, IS, and IT expertise and resources. Based on this finding, we believe that the development of a better understanding of how optimal BMI, CS, IS, and IT organizational structures and leadership models are designed and implemented is critical to both the advancement of CTS and ultimately, to improvements in the quality, safety, and effectiveness of healthcare. [ABSTRACT FROM AUTHOR]

Background: The practice of evidence-based medicine requires efficient biomedical literature search such as PubMed/MEDLINE. Retrieval performance relies highly on the efficient use of search field tags. The purpose of this study was to analyze PubMed log data in order to understand the usage pattern of search tags by the end user in PubMed/MEDLINE search. Methods: A PubMed query log file was obtained from the National Library of Medicine containing anonymous user identification, timestamp, and query text. Inconsistent records were removed from the dataset and the search tags were extracted from the query texts. A total of 2,917,159 queries were selected for this study issued by a total of 613,061 users. The analysis of frequent co-occurrences and usage patterns of the search tags was conducted using an association mining algorithm. Results: The percentage of search tag usage was low (11.38% of the total queries) and only 2.95% of queries contained two or more tags. Three out of four users used no search tag and about two-third of them issued less than four queries. Among the queries containing at least one tagged search term, the average number of search tags was almost half of the number of total search terms. Navigational search tags are more frequently used than informational search tags. While no strong association was observed between informational and navigational tags, six (out of 19) informational tags and six (out of 29) navigational tags showed strong associations in PubMed searches. Conclusions: The low percentage of search tag usage implies that PubMed/MEDLINE users do not utilize the features of PubMed/MEDLINE widely or they are not aware of such features or solely depend on the high recall focused query translation by the PubMed's Automatic Term Mapping. The users need further education and interactive search application for effective use of the search tags in order to fulfill their biomedical information needs from PubMed/MEDLINE. [ABSTRACT FROM AUTHOR]

Background: Death records are a rich source of data, which can be used to assist with public surveillance and/or decision support. However, to use this type of data for such purposes it has to be transformed into a coded format to make it computable. Because the cause of death in the certificates is reported as free text, encoding the data is currently the single largest barrier of using death certificates for surveillance. Therefore, the purpose of this study was to demonstrate the feasibility of using a pipeline, composed of a detection rule and a natural language processor, for the real time encoding of death certificates using the identification of pneumonia and influenza cases as an example and demonstrating that its accuracy is comparable to existing methods. Results: A Death Certificates Pipeline (DCP) was developed to automatically code death certificates and identify pneumonia and influenza cases. The pipeline used MetaMap to code death certificates from the Utah Department of Health for the year 2008. The output of MetaMap was then accessed by detection rules which flagged pneumonia and influenza cases based on the Centers of Disease and Control and Prevention (CDC) case definition. The output from the DCP was compared with the current method used by the CDC and with a keyword search. Recall, precision, positive predictive value and F-measure with respect to the CDC method were calculated for the two other methods considered here. The two different techniques compared here with the CDC method showed the following recall/ precision results: DCP: 0.998/0.98 and keyword searching: 0.96/0.96. The F-measure were 0.99 and 0.96 respectively (DCP and keyword searching). Both the keyword and the DCP can run in interactive form with modest computer resources, but DCP showed superior performance. Conclusion: The pipeline proposed here for coding death certificates and the detection of cases is feasible and can be extended to other conditions. This method provides an alternative that allows for coding free-text death certificates in real time that may increase its utilization not only in the public health domain but also for biomedical researchers and developers. [ABSTRACT FROM AUTHOR]

The article presents a study analyzing the working of OpenNotes, an online section of medical reports. It states that OpenNotes is a free online access and is an effective way of engaging patients with medical care. As mentioned, a mixed method approach was designed for conducting the study in healthcare systems of Boston, Massachusetts, Pennsylvania and Seattle, Washington (D.C).

The article focuses on the recommended practices for computerized clinical decision support and knowledge management in community settings in the U.S. It states that in the U.S. efforts to encourage widespread use of clinical information systems by hospitals and health care providers are likely to succeed only if they meet the needs of the major stakeholders.

The article focuses on a study conducted to evaluate disease surveillance using a biosurveillance application which combined data from both populations in the U.S. Using the ESSENCE biosurveillance system, including one infectious disease outbreak, a retrospective analysis of NC-VAMC/Lovell FHCC and other Chicago-area VAMC data was performed.

Background: Traditional information retrieval techniques typically return excessive output when directed at large bibliographic databases. Natural Language Processing applications strive to extract salient content from the excessive data. Semantic MEDLINE, a National Library of Medicine (NLM) natural language processing application, highlights relevant information in PubMed data. However, Semantic MEDLINE implements manually coded schemas, accommodating few information needs. Currently, there are only five such schemas, while many more would be needed to realistically accommodate all potential users. The aim of this project was to develop and evaluate a statistical algorithm that automatically identifies relevant bibliographic data; the new algorithm could be incorporated into a dynamic schema to accommodate various information needs in Semantic MEDLINE, and eliminate the need for multiple schemas. Methods: We developed a flexible algorithm named Combo that combines three statistical metrics, the Kullback- Leibler Divergence (KLD), Riloff's RlogF metric (RlogF), and a new metric called PredScal, to automatically identify salient data in bibliographic text. We downloaded citations from a PubMed search query addressing the genetic etiology of bladder cancer. The citations were processed with SemRep, an NLM rule-based application that produces semantic predications. SemRep output was processed by Combo, in addition to the standard Semantic MEDLINE genetics schema and independently by the two individual KLD and RlogF metrics. We evaluated each summarization method using an existing reference standard within the task-based context of genetic database curation. Results: Combo asserted 74 genetic entities implicated in bladder cancer development, whereas the traditional schema asserted 10 genetic entities; the KLD and RlogF metrics individually asserted 77 and 69 genetic entities, respectively. Combo achieved 61% recall and 81% precision, with an F-score of 0.69. The traditional schema achieved 23% recall and 100% precision, with an F-score of 0.37. The KLD metric achieved 61% recall, 70% precision, with an F-score of 0.65. The RlogF metric achieved 61% recall, 72% precision, with an F-score of 0.66. Conclusions: Semantic MEDLINE summarization using the new Combo algorithm outperformed a conventional summarization schema in a genetic database curation task. It potentially could streamline information acquisition for other needs without having to hand-build multiple saliency schemas. [ABSTRACT FROM AUTHOR]

Background: Respiratory Syncytial Virus (RSV), a major cause of bronchiolitis, has a large impact on the census of pediatric hospitals during outbreak seasons. Reliable prediction of the week these outbreaks will start, based on readily available data, could help pediatric hospitals better prepare for large outbreaks. Methods: Naïve Bayes (NB) classifier models were constructed using weather data from 1985-2008 considering only variables that are available in real time and that could be used to forecast the week in which an RSV outbreak will occur in Salt Lake County, Utah. Outbreak start dates were determined by a panel of experts using 32,509 records with ICD-9 coded RSV and bronchiolitis diagnoses from Intermountain Healthcare hospitals and clinics for the RSV seasons from 1985 to 2008. Results: NB models predicted RSV outbreaks up to 3 weeks in advance with an estimated sensitivity of up to 67% and estimated specificities as high as 94% to 100%. Temperature and wind speed were the best overall predictors, but other weather variables also showed relevance depending on how far in advance the predictions were made. The weather conditions predictive of an RSV outbreak in our study were similar to those that lead to temperature inversions in the Salt Lake Valley. Conclusions: We demonstrate that Naïve Bayes (NB) classifier models based on weather data available in real time have the potential to be used as effective predictive models. These models may be able to predict the week that an RSV outbreak will occur with clinical relevance. Their clinical usefulness will be field tested during the next five years. [ABSTRACT FROM AUTHOR]

Background: Patients with a prolonged intensive care unit (ICU) length of stay account for a disproportionate amount of resource use. Early identification of patients at risk for a prolonged length of stay can lead to quality enhancements that reduce ICU stay. This study developed and validated a model that identifies patients at risk for a prolonged ICU stay. Methods: We performed a retrospective cohort study of 343,555 admissions to 83 ICUs in 31 U.S. hospitals from 2002- 2007. We examined the distribution of ICU length of stay to identify a threshold where clinicians might be concerned about a prolonged stay; this resulted in choosing a 5-day cut-point. From patients remaining in the ICU on day 5 we developed a multivariable regression model that predicted remaining ICU stay. Predictor variables included information gathered at admission, day 1, and ICU day 5. Data from 12,640 admissions during 2002-2005 were used to develop the model, and the remaining 12,904 admissions to internally validate the model. Finally, we used data on 11,903 admissions during 2006-2007 to externally validate the model. Results: The variables that had the greatest impact on remaining ICU length of stay were those measured on day 5, not at admission or during day 1. Mechanical ventilation, PaO2: FiO2 ratio, other physiologic components, and sedation on day 5 accounted for 81.6% of the variation in predicted remaining ICU stay. In the external validation set observed ICU stay was 11.99 days and predicted total ICU stay (5 days + day 5 predicted remaining stay) was 11.62 days, a difference of 8.7 hours. For the same patients, the difference between mean observed and mean predicted ICU stay using the APACHE day 1 model was 149.3 hours. The new model's r² was 20.2% across individuals and 44.3% across units. Conclusions: A model that uses patient data from ICU days 1 and 5 accurately predicts a prolonged ICU stay. These predictions are more accurate than those based on ICU day 1 data alone. The model can be used to benchmark ICU performance and to alert physicians to explore care alternatives aimed at reducing ICU stay. [ABSTRACT FROM AUTHOR]

Background: Individuals claiming a disability benefit after long-term sickness absence, have to undergo medical disability assessments. These assessments, often carried out by specialized physicians, can be complicated by wrong expectations or defensive attitudes of disability benefit claimants. It is hypothesized that empowerment of these claimants will enhance the physician-patient relationship by shifting claimants from a passive role to a more active and constructive role during disability assessments. Furthermore, empowerment of claimants may lead to a more realistic expectation and acceptance of the assessment outcome among claimants and may lead to a more accurate assessment by the physician. Methods/Design: In a two-armed randomized controlled trial (RCT), 230 claimants will be randomized to either the intervention or control group. For the intervention group, an interactive website was designed http:// www.wiagesprek.nl using an Intervention Mapping procedure. This website was tested during a pilot study among 51 claimants. The final version of the website consists of five interactive modules, in which claimants will be prepared and empowered step-by-step, prior to their upcoming disability assessment. Other website components are a forum, a personal health record, a personal diary, and information on disability assessment procedures, return to work, and coping with disease and work disability. Subjects from the control group will be directed to a website with commonly available information only. Approximately two weeks prior to their disability assessment, disability claimants will be recruited through the Dutch Workers Insurance Authority (UWV). Outcomes will be assessed at five occasions: directly after recruitment (baseline), prior to disability assessment, directly after disability assessment as well as 6 and 16 weeks after the assessment. The study's primary outcome is empowerment, measured with the Vrijbaan questionnaire. Secondary outcomes include claimants' satisfaction, perceived justice, coping strategy, and knowledge. A process evaluation will also be conducted. Discussion: This study evaluates the effectiveness of an interactive website aimed at empowerment of disability claimants. It is hypothesized that by increasing empowerment, the physician-patient relationship may be enhanced and claimants' satisfaction and perceived justice can be improved. Results are expected in 2010. [ABSTRACT FROM AUTHOR]

Background: Text-based patient medical records are a vital resource in medical research. In order to preserve patient confidentiality, however, the U.S. Health Insurance Portability and Accountability Act (HIPAA) requires that protected health information (PHI) be removed from medical records before they can be disseminated. Manual de-identification of large medical record databases is prohibitively expensive, time-consuming and prone to error, necessitating automatic methods for large-scale, automated de-identification.Methods: We describe an automated Perl-based de-identification software package that is generally usable on most free-text medical records, e.g., nursing notes, discharge summaries, X-ray reports, etc. The software uses lexical look-up tables, regular expressions, and simple heuristics to locate both HIPAA PHI, and an extended PHI set that includes doctors' names and years of dates. To develop the de-identification approach, we assembled a gold standard corpus of re-identified nursing notes with real PHI replaced by realistic surrogate information. This corpus consists of 2,434 nursing notes containing 334,000 words and a total of 1,779 instances of PHI taken from 163 randomly selected patient records. This gold standard corpus was used to refine the algorithm and measure its sensitivity. To test the algorithm on data not used in its development, we constructed a second test corpus of 1,836 nursing notes containing 296,400 words. The algorithm's false negative rate was evaluated using this test corpus.Results: Performance evaluation of the de-identification software on the development corpus yielded an overall recall of 0.967, precision value of 0.749, and fallout value of approximately 0.002. On the test corpus, a total of 90 instances of false negatives were found, or 27 per 100,000 word count, with an estimated recall of 0.943. Only one full date and one age over 89 were missed. No patient names were missed in either corpus.Conclusion: We have developed a pattern-matching de-identification system based on dictionary look-ups, regular expressions, and heuristics. Evaluation based on two different sets of nursing notes collected from a U.S. hospital suggests that, in terms of recall, the software out-performs a single human de-identifier (0.81) and performs at least as well as a consensus of two human de-identifiers (0.94). The system is currently tuned to de-identify PHI in nursing notes and discharge summaries but is sufficiently generalized and can be customized to handle text files of any format. Although the accuracy of the algorithm is high, it is probably insufficient to be used to publicly disseminate medical data. The open-source de-identification software and the gold standard re-identified corpus of medical records have therefore been made available to researchers via the PhysioNet website to encourage improvements in the algorithm. [ABSTRACT FROM AUTHOR]

Background: Strokes are a leading cause of morbidity and the first cause of adult disability in the United States. Currently, no biomarkers are being used clinically to diagnose acute ischemic stroke. A diagnostic test using a blood sample from a patient would potentially be beneficial in treating the disease.Results: A classification approach is described for differentiating between proteomic samples of stroke patients and controls, and a second novel predictive model is developed for predicting the severity of stroke as measured by the National Institutes of Health Stroke Scale (NIHSS). The models were constructed by applying the Logical Analysis of Data (LAD) methodology to the mass peak profiles of 48 stroke patients and 32 controls. The classification model was shown to have an accuracy of 75% when tested on an independent validation set of 35 stroke patients and 25 controls, while the predictive model exhibited superior performance when compared to alternative algorithms. In spite of their high accuracy, both models are extremely simple and were developed using a common set consisting of only 3 peaks.Conclusion: We have successfully identified 3 biomarkers that can detect ischemic stroke with an accuracy of 75%. The performance of the classification model on the validation set and on cross-validation does not deteriorate significantly when compared to that on the training set, indicating the robustness of the model. As in the case of the LAD classification model, the results of the predictive model validate the function constructed on our support-set for approximating the severity scores of stroke patients. The correlation and root mean absolute error of the LAD predictive model are consistently superior to those of the other algorithms used (Support vector machines, C4.5 decision trees, Logistic regression and Multilayer perceptron). [ABSTRACT FROM AUTHOR]

Background: The National Institute of Diabetes and Digestive and Kidney Diseases have established central repositories for the collection of DNA, biological samples, and clinical data to be catalogued at a single site. Here we present an overview of the site which stores the clinical data and links to biospecimens. Description: The NIDDK Data repository is a web-enabled resource cataloguing clinical trial data and supporting information from NIDDK supported studies. The Data Repository allows for the co-location of multiple electronic datasets that were created as part of clinical investigations. The Data Repository does not serve the role of a Data Coordinating Center, but rather as a warehouse for the clinical findings once the trials have been completed. Because both biological and genetic samples are collected from many of the studies, a data management system for the cataloguing and retrieval of samples was developed. Conclusion: The Data Repository provides a unique resource for researchers in the clini cal areas supported by NIDDK. In addition to providing a warehouse of data, Data Repository staff work with the users to educate them on the datasets as well as assist them in the acquisition of multiple data sets for cross-study analysis. Unlike the majority of biological databases, the Data Repository acts both as a catalogue for data, biosamples, and genetic materials and as a central processing point for the requests for all biospecimens. Due to regulations on the use of clinical data, the ultimate release of that data is governed under NIDDK data release policies. The Data Repository serves as the conduit for such requests. [ABSTRACT FROM AUTHOR]

Background: Public health departments in the United States are beginning to gain timely access to health data, often as soon as one day after a visit to a health care facility. Consequently, new approaches to outbreak surveillance are being developed. When cases cluster geographically, an analysis of their spatial distribution can facilitate outbreak detection. Our method focuses on detecting perturbations in the distribution of pair-wise distances among all patients in a geographical region. Barring outbreaks, this distribution can be quite stable over time. We sought to exemplify the method by measuring its cluster detection performance, and to determine factors affecting sensitivity to spatial clustering among patients presenting to hospital emergency departments with respiratory syndromes. Methods: The approach was to (1) define a baseline spatial distribution of home addresses for a population of patients visiting an emergency department with respiratory syndromes using historical data; (2) develop a controlled feature set simulation by inserting simulated outbreak data with varied parameters into authentic background noise, thereby creating semisynthetic data; (3) compare the observed with the expected spatial distribution; (4) establish the relative value of different alarm strategies so as to maximize sensitivity for the detection of clustering; and (5) measure factors which have an impact on sensitivity. Results: Overall sensitivity to detect spatial clustering was 62%. This contrasts with an overall alarm rate of less than 5% for the same number of extra visits when the extra visits were not characterized by geographic clustering. Clusters that produced the least number of alarms were those that were small in size (10 extra visits in a week, where visits per week ranged from 120 to 472), diffusely distributed over an area with a 3 km radius, and located close to the hospital (5 km) in a region most densely populated with patients to this hospital. Near perfect alarm rates were found for clusters that varied on the opposite extremes of these parameters (40 extra visits, within a 250 meter radius, 50 km from the hospital). Conclusion: Measuring perturbations in the interpoint distance distribution is a sensitive method for detecting spatial clustering. When cases are clustered geographically, there is clearly power to detect clustering when the spatial distribution is represented by the M statistic, even when clusters are small in size. By varying independent parameters of simulated outbreaks, we have demonstrated empirically the limits of detection of different types of outbreaks. [ABSTRACT FROM AUTHOR]

Background: American public policy makers recently established the goal of providing the majority of Americans with electronic health records by 2014. This will require a National Health Information Infrastructure (NHII) that is far more complete than the one that is currently in its formative stage of development. We describe a conceptual framework to help measure progress toward that goal. Discussion: The NHII comprises a set of clusters, such as Regional Health Information Organizations (RHIOs), which, in turn, are composed of smaller clusters and nodes such as private physician practices, individual hospitals, and large academic medical centers. We assess progress in terms of the availability and use of information and communications technology and the resulting effectiveness of these implementations. These three attributes can be studied in a phased approach because the system must be available before it can be used, and it must be used to have an effect.As the NHII expands, it can become a tool for evaluating itself. Summary: The NHII has the potential to transform health care in America - improving health care quality, reducing health care costs, preventing medical errors, improving administrative efficiencies, reducing paperwork, and increasing access to affordable health care. While the President has set an ambitious goal of assuring that most Americans have electronic health records within the next 10 years, a significant question remains "How will we know if we are making progress toward that goal?" Using the definitions for "nodes" and "clusters" developed in this article along with the resulting measurement framework, we believe that we can begin a discussion that will enable us to define and then begin making the kinds of measurements necessary to answer this important question. [ABSTRACT FROM AUTHOR]

Background: Design and test the reliability of a web-based system for multicenter, real-time collection of data in the emergency department (ED), under waiver of authorization, in compliance with HIPAA. Methods: This was a phase I, two-hospital study of patients undergoing evaluation for possible pulmonary embolism. Data were collected by on-duty clinicians on an HTML data collection form (prospective e-form), populated using either a personal digital assistant (PDA) or personal computer (PC). Data forms were uploaded to a central, offsite server using secure socket protocol transfer. Each form was assigned a unique identifier, and all PHI data were encrypted, but were password-accessible by authorized research personnel to complete a follow-up e-form. Results: From April 15, 2003-April 15 2004, 1022 prospective e-forms and 605 follow-up e-forms were uploaded. Complexities of PDA use compelled clinicians to use PCs in the ED for data entry for most forms. No data were lost and server log query revealed no unauthorized entry. Prospectively obtained PHI data, encrypted upon server upload, were successfully decrypted using password-protected access to allow follow-up without difficulty in 605 cases. Non-PHI data from prospective and follow-up forms were available to the study investigators via standard file transfer protocol. Conclusions: Data can be accurately collected from on-duty clinicians in the ED using real-time, PC-Internet data entry in compliance with the Privacy Rule. Deidentification-reidentification of PHI was successfully accomplished by a password-protected encryption-deencryption mechanism to permit follow-up by approved research personnel. [ABSTRACT FROM AUTHOR]

Background: Large biomedical data sets have become increasingly important resources for medical researchers. Modern biomedical data sets are annotated with standard terms to describe the data and to support data linking between databases. The largest curated listing of biomedical terms is the National Library of Medicine's Unified Medical Language System (UMLS). The UMLS contains more than 2 million biomedical terms collected from nearly 100 medical vocabularies. Many of the vocabularies contained in the UMLS carry restrictions on their use, making it impossible to share or distribute UMLS-annotated research data. However, a subset of the UMLS vocabularies, designated Category 0 by UMLS, can be used to annotate and share data sets without violating the UMLS License Agreement. Methods: The UMLS Category 0 vocabularies can be extracted from the parent UMLS metathesaurus using a Perl script supplied with this article. There are 43 Category 0 vocabularies that can be used freely for research purposes without violating the UMLS License Agreement. Among the Category 0 vocabularies are: MESH (Medical Subject Headings), NCBI (National Center for Bioinformatics) Taxonomy and ICD-9-CM (International Classification of Diseases-9-Clinical Modifiers). Results: The extraction file containing all Category 0 terms and concepts is 72,581,138 bytes in length and contains 1,029,161 terms. The UMLS Metathesaurus MRCON file (January, 2003) is 151,048,493 bytes in length and contains 2,146,899 terms. Therefore the Category 0 vocabularies, in aggregate, are about half the size of the UMLS metathesaurus. A large publicly available listing of 567,921 different medical phrases were automatically coded using the full UMLS metatathesaurus and the Category 0 vocabularies. There were 545,321 phrases with one or more matches against UMLS terms while 468,785 phrases had one or more matches against the Category 0 terms. This indicates that when the two vocabularies are evaluated by their fitness to find at least one term for a medical phrase, the Category 0 vocabularies performed 86% as well as the complete UMLS metathesaurus. Conclusion: The Category 0 vocabularies of UMLS constitute a large nomenclature that can be used by biomedical researchers to annotate biomedical data. These annotated data sets can be distributed for research purposes without violating the UMLS License Agreement. These vocabularies may be of particular importance for sharing heterogeneous data from diverse biomedical data sets. The software tools to extract the Category 0 vocabularies are freely available Perl scripts entered into the public domain and distributed with this article. [ABSTRACT FROM AUTHOR]

Background: Improving health in our nation requires strengthening four major domains of the health care system: personal health management, health care delivery, public health, and health-related research. Many avoidable shortcomings in the health sector that result in poor quality are due to inaccessible data, information, and knowledge. A national health information infrastructure (NHII) offers the connectivity and knowledge management essential to correct these shortcomings. Better health and a better health system are within our reach. Discussion: A national health information infrastructure for the United States should address the needs of personal health management, health care delivery, public health, and research. It should also address relevant global dimensions (e.g., standards for sharing data and knowledge across national boundaries). The public and private sectors will need to collaborate to build a robust national health information infrastructure, essentially a 'paperless' health care system, for the United States. The federal government should assume leadership for assuring a national health information infrastructure as recommended by the National Committee on Vital and Health Statistics and the President's Information Technology Advisory Committee. Progress is needed in the areas of funding, incentives, standards, and continued refinement of a privacy (i.e., confidentiality and security) framework to facilitate personal identification for health purposes. Particular attention should be paid to NHII leadership and change management challenges. Summary: A national health information infrastructure is a necessary step for improved health in the U.S. It will require a concerted, collaborative effort by both public and private sectors. [ABSTRACT FROM AUTHOR]

Background: The storage and distribution of electrocardiogram data is based on different formats. There is a need to promote the development of standards for their exchange and analysis. Such models should be platform-/ system- and application-independent, flexible and open to every member of the scientific community. Methods: A minimum set of information for the representation and storage of electrocardiogram signals has been synthesised from existing recommendations. This specification is encoded into an XML-vocabulary. The model may aid in a flexible exchange and analysis of electrocardiogram information. Results: Based on advantages of XML technologies, ecgML has the ability to present a system-, application- and format-independent solution for representation and exchange of electrocardiogram data. The distinction between the proposal developed by the U.S Food and Drug Administration and ecgML model is given. A series of tools, which aim to facilitate ecgML-based applications, are presented. Conclusions: The models proposed here can facilitate the generation of a data format, which opens ways for better and clearer interpretation by both humans and machines. Its structured and transparent organisation will allow researchers to expand and test its capabilities in different application domains. The specification and programs for this protocol are publicly available. [ABSTRACT FROM AUTHOR]