AMIA

The Israel Penn International Transplant Tumor Registry is literally the world's premier repository of information on patients who have developed malignancies after organ transplants. The administrators of the Registry not only collect information but also provide consulting services based on the accumulated knowledge that the Registry contains. By creating a secure Web-based front end, we have made it possible for the Registry to keep pace with its burgeoning international caseload.

GELLO is a purpose-specific, object-oriented (OO) query and expression language [1]. GELLO is the result of a concerted effort of the Decision Systems Group (DSG) working with the HL7 Clinical Decision Support Technical Committee (CDSTC) to provide the HL7 ...

Medical care at mass casualty incidents and disasters requires rapid patient triage and assessment, acute care and disposition often in the setting of overwhelming numbers of victims, limited time, and little resources. Current systems rely on a paper triage tag on which rescuers and medical providers mark the patient’s triage status and record limited information on injuries and treatments administered in the field. In this manuscript, we describe the design, development and deployment of a wireless handheld device with an electronic medical record (EMR) for use by rescuers responding to mass casualty incidents (MCIs) and disasters. The components of this device, the WIISARD First Responder (WFR), includes a personal digital assistant (PDA) with 802.11 wireless transmission capabilities, microprocessor and non-volatile memory, and a unique EMR software that replicates the rapidity and ease of use of the standard paper triage tag. WFR also expands its functionality by recording real-time medical data electronically for simultaneous access by rescuers, mid-level providers and incident commanders on and off the disaster site. WFR is a part of the Wireless Information System for Medical Response in Disasters (WIISARD) architecture.

This paper uses empirical data emerging from a “participative action-research” project undertaken in early 2015, to test the hypothesis that the geopolitically-meaningful ancient mechanisms that were at work in Greco-Roman times, connecting the Jewish Diaspora with an active Jerusalemite center, continue to function today. Examples are drawn from recent Argentine and U.S. experiences. In one specific case related to the attempt to excommunicate Argentine Foreign Minister Hector Timerman from Jewish institutions, the attitudes of members of the public were studied through a “voluntary online panel.” The findings were linked to a major heuristic effort aimed at determining if the “Baron dialectic,” so-called in honor of Salo W. Baron (the most important historian of the Jewish people of the World War II generation) is still at work. This dialectic can be summarized as the complex trans-historical sociological mechanism whereby the Jewish religion is reinforced by the Jewish nationality, whilst the latter is supra-nationally rooted through the Diaspora communities. This process is likely one of the long-term causes of anti-Semitism. 

Este trabajo usa datos empíricos provenientes de una “investigación-acción participativa,” realizada a comienzos de 2015, para poner a prueba la hipótesis de que los mecanismos que tenían vigencia en tiempos greco-romanos, conectando las comunidades de la diáspora judía con un activo centro jerosolimitano, siguen funcionando en nuestros tiempos y siguen teniendo relevancia geopolítica. Se tomaron ejemplos provenientes de experiencias recientes de Argentina y Estados Unidos. En un caso específico, el de las presiones para excomulgar al canciller argentino Héctor Timerman de las instituciones de la comunidad judía, se estudiaron las actitudes de miembros del público a través de un “panel voluntario online.” Los hallazgos se vincularon a un significativo esfuerzo heurístico para determinar si la “dialéctica Baron,” así llamada en honor de Salo W. Baron (el más importante historiador del pueblo judío de la generación de la Segunda Guerra Mundial), sigue vigente. Esta dialéctica puede resumirse como el complejo mecanismo sociológico trans-histórico por el que la religión judía es reforzada por la nacionalidad judía, a la vez que ésta se arraiga supranacionalmente a través de las comunidades de la diáspora. Este proceso es, seguramente, una de las causas de largo plazo del antisemitismo.

Management of complex clinical trials involves coordinated-use of a myriad of software applications by trial personnel. The applications typically use distinct knowledge representations and generate enormous amount of information during the course of a trial. It becomes vital that the applications exchange trial semantics in order for efficient management of the trials and subsequent analysis of clinical trial data. Existing model-based frameworks do not address the requirements of semantic integration of heterogeneous applications. We have built an ontology-based architecture to support interoperation of clinical trial software applications. Central to our approach is a suite of clinical trial ontologies, which we call Epoch, that define the vocabulary and semantics necessary to represent information on clinical trials. We are continuing to demonstrate and validate our approach with different clinical trials management applications and with growing number of clinical trials.

Medical care at mass casualty incidents and disasters requires rapid patient triage and assessment, acute care and disposition often in the setting of overwhelming numbers of victims, limited time, and little resources. Current systems rely on a paper triage tag on which rescuers and medical providers mark the patient's triage status and record limited information on injuries and treatments administered in the field. In this manuscript, we describe the design, development and deployment of a wireless handheld device with an electronic medical record (EMR) for use by rescuers responding to mass casualty incidents (MCIs) and disasters. The components of this device, the WIISARD First Responder (WFR), includes a personal digital assistant (PDA) with 802.11 wireless transmission capabilities, microprocessor and non-volatile memory, and a unique EMR software that replicates the rapidity and ease of use of the standard paper triage tag. WFR also expands its functionality by recording ...

Information exchange, enabled by computable interoperability, is the key to many of the initiatives underway including the development of Regional Health Information Exchanges, Regional Health Information Organizations, and the National Health Information Network. These initiatives must include public health as a full partner in the emerging transformation of our nation's healthcare system through the adoption and use of information technology. An electronic health record - public health (EHR-PH)system prototype was developed to demonstrate the feasibility of electronic data transfer from a health care provider, i.e. hospital or ambulatory care settings, to multiple customized public health systems which include a Newborn Metabolic Screening Registry, a Newborn Hearing Screening Registry, an Immunization Registry and a Communicable Disease Registry, using HL7 messaging standards. Our EHR-PH system prototype can be considered a distributed EHR-based RHIE/RHIO model - a principal ...

In the patient-centered medical home, PCMH, patient care is overseen by a primary care physician leading a team of health care providers, who collaborate to optimize treatment. To facilitate interactions in PCMH, secure collaboration will be needed to: control access to information; dictate who can do what when; and promote sharing and concurrent access. This contrasts approaches such as the National Institute of Standard and Technology (NIST) role-based access control (RBAC), where the emphasis is on controlling access and separating responsibilities. This paper investigates secure collaboration within an application such as PCMH, through: a futuristic scenario for patient care; proposed collaboration extensions to the NIST RBAC standard with a fine-grained obligated mechanism and workflow; and a prototype of PCMH via the Google Wave real-time collaboration platform.

This paper describes the development and a preliminary usage analysis of a novel Internet-based support system that assists cancer patients in symptom monitoring and disease management. The system, called WebChoice, allows patients to monitor symptoms over time, and provides access to evidence-based self-management options tailored to their reported symptoms as well as a communication area where patients can ask questions to a clinical nurse specialist and exchange experiences with other cancer patients. A currently ongoing randomized clinical trial evaluates effects of WebChoice on patient outcomes. This paper describes essential features of WebChoice, steps we used to design it as user-friendly as possible; and preliminary findings on patients use.

Developing computer-interpretable clinical practice guidelines (CPGs) to provide decision support for guideline-based care is an extremely labor-intensive task. In the EON/ATHENA and SAGE projects, we formulated substantial portions of CPGs as computable statements that express declarative relationships between patient conditions and possible interventions. We developed query and expression languages that allow a decision-support system (DSS) to evaluate these statements in specific patient situations. A DSS can use these guideline statements in multiple ways, including: (1) as inputs for determining preferred alternatives in decision-making, and (2) as a way to provide targeted commentaries in the clinical information system. The use of these declarative statements significantly reduces the modeling expertise and effort required to create and maintain computer-interpretable knowledge bases for decision-support purpose. We discuss possible implications for sharing of such knowledge ...

The Veterans Administration (VA) has adopted an ambitious program to standardize its clinical terminology to comply with industry-wide standards. The VA is using commercially available tools and in-house software to create a high-quality reference terminology system. The terminology will be used by current and future applications with no planned disruption to operational systems. The first large customer of the group is the national VA Health Data Repository (HDR). Unique enterprise identifiers are assigned to each standard term, and a rich network of semantic relationships makes the resulting data not only recognizable, but highly computable and reusable in a variety of applications, including decision support and data sharing with partners such as the Department of Defense (DoD). This paper describes the specific methods and approaches that the VA has employed to develop and implement this innovative program in existing information system. The goal is to share with others our expe...

At NewYork-Presbyterian Hospital, we are committed to the delivery of high quality care. We have implemented a strategic planning process to determine the information technology initiatives that will best help us improve quality. The process began with the creation of a Clinical Quality and IT Committee. The Committee identified 2 high priority goals that would enable demonstrably high quality care: 1) excellence at data warehousing, and 2) optimal use of automated clinical documentation to capture encounter-related quality and safety data. For each high priority goal, a working group was created to develop specific recommendations. The Data Warehousing subgroup has recommended the implementation of an architecture management process and an improved ability for users to get access to aggregate data. The Structured Documentation subgroup is establishing recommendations for a documentation template creation process. The strategic planning process at times is slow, but assures that the...

Social networks affect both exposure to sexually transmitted infections (STIs) and associated risk behavior. Networks may also play a role in disparities in STI/HIV rates among African American youth. Accordingly, there is growing interest in the potential of social network-based interventions to reduce STI/HIV incidence in this group. However, any youth-focused network intervention must grapple with the role of technologies in the social lives of young people. We report results of 12 focus groups with 94 youth from one economically depressed city with a high STI/HIV prevalence. We examined how youth use information and communication technologies (ICTs) in order to socialize with others, and how this aligns with their communication about sexuality and HIV/STIs. The study resulted in the generation of five themes: distraction, diversification, dramatization, danger management and dialogue. We consider implications of these findings for future development of online, social network-bas...

Clinical information is often coded using different terminologies, and therefore is not interoperable. Our goal is to develop a general natural language processing (NLP) system, called Medical Text Extraction, Reasoning and Mapping System (MTERMS), which encodes clinical text using different terminologies and simultaneously establishes dynamic mappings between them. MTERMS applies a modular, pipeline approach flowing from a preprocessor, semantic tagger, terminology mapper, context analyzer, and parser to structure inputted clinical notes. Evaluators manually reviewed 30 free-text and 10 structured outpatient clinical notes compared to MTERMS output. MTERMS achieved an overall F-measure of 90.6 and 94.0 for free-text and structured notes respectively for medication and temporal information. The local medication terminology had 83.0% coverage compared to RxNorm's 98.0% coverage for free-text notes. 61.6% of mappings between the terminologies are exact match. Capture of duration w...

Misspellings are common in medical documents and can be an obstacle to information retrieval. We evaluated an algorithm to identify misspelled words through analysis of their prevalence in a representative body of text. We evaluated the algorithm's accuracy of identifying misspellings of 200 anti-hypertensive medication names on 2,000 potentially misspelled words randomly selected from narrative medical documents. Prevalence ratios (the frequency of the potentially misspelled word divided by the frequency of the non-misspelled word) in physician notes were computed by the software for each of the words. The software results were compared to the manual assessment by an independent reviewer. Area under the ROC curve for identification of misspelled words was 0.96. Sensitivity, specificity, and positive predictive value were 99.25%, 89.72% and 82.9% for the prevalence ratio threshold (0.32768) with the highest F-measure (0.903). Prevalence analysis can be used to identify and corre...

Although the pre-surgical patient-surgeon encounter is the opportunity to educate the patient, it is essential that the patient be given educational materials to complement the face-to-face exchange. This is virtually impossible to do well with brochures, because ...

Information exchange, enabled by computable interoperability, is the key to many of the initiatives underway including the development of Regional Health Information Exchanges, Regional Health Information Organizations, and the National Health Information Network. These initiatives must include public health as a full partner in the emerging transformation of our nation's healthcare system through the adoption and use of information technology. An electronic health record - public health (EHR-PH)system prototype was developed to demonstrate the feasibility of electronic data transfer from a health care provider, i.e. hospital or ambulatory care settings, to multiple customized public health systems which include a Newborn Metabolic Screening Registry, a Newborn Hearing Screening Registry, an Immunization Registry and a Communicable Disease Registry, using HL7 messaging standards. Our EHR-PH system prototype can be considered a distributed EHR-based RHIE/RHIO model - a principal ...

In view of the benefits yielded by the Terminology Server running in the Healthcare Information System of Hospital Italiano de Buenos Aires, the Institution implemented all the necessary changes in order to offer such services to other health-care institutions, using the Internet as communication vehicle. Megasalud is the most integrated healthcare network of Chile and decided to change their legacy system and develop their own Healthcare Information System and started to use Remote Terminology Services. After the implementation of these Terminology Services we tested the performance for identifying free text added in their electronic health record. Between 78% to 89% of text entered was recognized. The task of creating an institutional interface terminology provides an excellent service to the users, as they have the freedom to use free-text entry.

Personal health records (PHRs) are a rapidly expanding area in medical informatics due to the belief that they may improve healthcare delivery and control costs of care. To truly understand the full potential value of a technology, a cost analysis is critical. However, little evidence exists on the value potential of PHRs, and a cost model for PHRs does not currently exist in the literature.This paper presents a sample cost model for PHR systems, which include PHR infrastructure and applications. We used this model to examine the costs of provider-tethered, payer-tethered, third-party, and interoperable PHRs. Our model projects that on a perperson basis, third-party PHRs will be the most expensive followed by interoperable PHRs, and then provider-tethered PHRs and payer-tethered PHRs are the least expensive. Data interfaces are a major cost driver, thus these findings underscore the need for standards development and use in the implementation of PHR systems.

As health care systems and providers move towards meaningful use of electronic health records, the once distant vision of collaborative patient-centric, interdisciplinary plans of care, generated and updated across organizations and levels of care, may soon become a reality. Effective care planning is included in the proposed Stages 2–3 Meaningful Use quality measures. To facilitate interoperability, standardization of plan of care messaging, content, information and terminology models are needed. This degree of standardization requires local and national coordination. The purpose of this paper is to review some existing standards that may be leveraged to support development of interdisciplinary patient-centric plans of care. Standards are then applied to a use case to demonstrate one method for achieving patient-centric and interoperable interdisciplinary plan of care documentation. Our pilot work suggests that existing standards provide a foundation for adoption and implementation of patient-centric plans of care that are consistent with federal requirements.

A major portion of patient care planning occurs during the process of writing orders. Computerized order entry can present collections of predefined orders to the user during the ordering process. These order sets are useful for promoting standards of care, and provide one element of structured clinical knowledge to be used by Computerized Provider Order Entry (CPOE) systems at the

Knowledge of associations between biomedical entities, such as disease-symptoms, is critical for many automated biomedical applications. In this work, we develop automated methods for acquisition and discovery of medical knowledge embedded in clinical narrative reports. MedLEE, a Natural Language Processing (NLP) system, is applied to extract and encode clinical entities from narrative clinical reports obtained from New York-Presbyterian Hospital (NYPH), and associations between the clinical entities are determined based on statistical methods adjusted by volume tests. We focus on two types of entities, disease and symptom, in this study. Evaluation based on a random sample of disease-symptom associations indicates an overall recall of 90% and a precision of 92%. In conclusion, the preliminary study demonstrated that this method for knowledge acquisition of disease-symptom pairs from clinical reports is effective. The automated method is generalizable, and can be applied to detect o...

Delivery of safe, effective and appropriate health care is an imperative facing health care organizations globally. While many initiatives have been launched in a number of countries to address this need from a medical perspective, a similar focus for generating evidence-based nursing knowledge has been missing [1]. This paper reports on a collaborative evidence-based practice (EBP) research initiative that adds nursing knowledge into computerized care protocols. Here, a brief overview of the study’s aims, purpose and methodology is presented as well as results of data analysis and lessons learned. The research team examined nurses’ adoption patterns of EBP recommendations with respect to activity tolerance using four-month patient data collected from a pilot hospital. Study findings indicate a need for more focus on the system design and implementation process with the next rollout phase to promote evidence-based nursing practice.

Part-of-speech (POS) tagging is a fundamental step required by various NLP systems. The training of a POS tagger relies on sufficient quality annotations. However, the annotation process is both knowledge-intensive and time-consuming in the clinical domain. A promising solution appears to be for institutions to share their annotation efforts, and yet there is little research on associated issues. We performed experiments to understand how POS tagging performance would be affected by using a pre-trained tagger versus raw training data across different institutions. We manually annotated a set of clinical notes at Kaiser Permanente Southern California (KPSC) and a set from the University of Pittsburg Medical Center (UPMC), and trained/tested POS taggers with intra- and inter-institution settings. The cTAKES POS tagger was also included in the comparison to represent a tagger partially trained from the notes of a third institution, Mayo Clinic at Rochester.Intra-institution 5-fold cross-validation estimated an accuracy of 0.953 and 0.945 on the KPSC and UPMC notes respectively. Trained purely on KPSC notes, the accuracy was 0.897 when tested on UPMC notes. Trained purely on UPMC notes, the accuracy was 0.904 when tested on KPSC notes. Applying the cTAKES tagger pre-trained with Mayo Clinic’s notes, the accuracy was 0.881 on KPSC notes and 0.883 on UPMC notes. After adding UPMC annotations to KPSC training data, the average accuracy on tested KPSC notes increased to 0.965. After adding KPSC annotations to UPMC training data, the average accuracy on tested UPMC notes increased to 0.953. The results indicated: first, the performance of pre-trained POS taggers dropped about 5% when applied directly across the institutions; second, mixing annotations from another institution following the same guideline increased tagging accuracy for about 1%. Our findings suggest that institutions can benefit more from sharing raw annotations but less from sharing pre-trained models for the POS tagging task. We believe the study could also provide general insights on cross-institution data sharing for other types of NLP tasks.

The purpose of this study was to map the North American Nursing Diagnosis Association (NANDA) nursing diagnoses to the International Classification for Nursing Practice Version 1.0 (ICNP) and to compare the resulting representations and relationships to those within SNOMED Clinical Terms (CT). Independent reviewers reached agreement on 25 (i.e. 64%) of the 39 parent-child relationships identified via the mappings between NANDA entities. Other parent-child relationships were more questionable and are in need of further discussion. This work does not seek to promote one terminology over any other. Rather, this collaborative effort has the potential to mutually enhance all three terminologies involved in the study: ICNP, SNOMED CT and NANDA. In doing so it provides an example of the type of collaborative effort that is needed to facilitate the development of tools to support interoperability at a global level.

Handoffs have been recognized as a major healthcare challenge primarily due to the breakdowns in communication that occur during transitions in care. Consequently, they are characterized as being “remarkably haphazard”. To investigate the information ...

Part-of-speech (POS) tagging is a fundamental step required by various NLP systems. The training of a POS tagger relies on sufficient quality annotations. However, the annotation process is both knowledge-intensive and time-consuming in the clinical domain. A promising solution appears to be for institutions to share their annotation efforts, and yet there is little research on associated issues. We performed experiments to understand how POS tagging performance would be affected by using a pre-trained tagger versus raw training data across different institutions. We manually annotated a set of clinical notes at Kaiser Permanente Southern California (KPSC) and a set from the University of Pittsburg Medical Center (UPMC), and trained/tested POS taggers with intra- and inter-institution settings. The cTAKES POS tagger was also included in the comparison to represent a tagger partially trained from the notes of a third institution, Mayo Clinic at Rochester.Intra-institution 5-fold cross-validation estimated an accuracy of 0.953 and 0.945 on the KPSC and UPMC notes respectively. Trained purely on KPSC notes, the accuracy was 0.897 when tested on UPMC notes. Trained purely on UPMC notes, the accuracy was 0.904 when tested on KPSC notes. Applying the cTAKES tagger pre-trained with Mayo Clinic’s notes, the accuracy was 0.881 on KPSC notes and 0.883 on UPMC notes. After adding UPMC annotations to KPSC training data, the average accuracy on tested KPSC notes increased to 0.965. After adding KPSC annotations to UPMC training data, the average accuracy on tested UPMC notes increased to 0.953. The results indicated: first, the performance of pre-trained POS taggers dropped about 5% when applied directly across the institutions; second, mixing annotations from another institution following the same guideline increased tagging accuracy for about 1%. Our findings suggest that institutions can benefit more from sharing raw annotations but less from sharing pre-trained models for the POS tagging task. We believe the study could also provide general insights on cross-institution data sharing for other types of NLP tasks.

A Hospital Incident Reporting Ontology (HIRO) is being developed in Protégé-OWL to demonstrate feasibility and clinical value of using an ontology to combine, compare, and analyze data from across many public and private reporting systems collecting adverse events and near misses for patient safety. The HIRO is based on the JCAHO Patient Safety Event Taxonomy (PSET) and de-identified hospital incident reports.

Physicians have many questions when caring for patients, and frequently need to seek answers for their questions. Information retrieval systems (e.g., PubMed) typically return a list of documents in response to a user's query. Frequently the number of returned documents is large and makes physicians' information seeking "practical only 'after hours' and not in the clinical settings". Question answering techniques are based on automatically analyzing thousands of electronic documents to generate short-text answers in response to clinical questions that are posed by physicians. The authors address physicians' information needs and described the design, implementation, and evaluation of the medical question answering system (MedQA). Although our long term goal is to enable MedQA to answer all types of medical questions, currently, we implemented MedQA to integrate information retrieval, extraction, and summarization techniques to automatically generate par...

Think-aloud usability analysis provides extremely useful data but is very time-consuming and expensive to perform because of the extensive manual video analysis that is required. We describe a simple method for automated detection of usability problems from ...

We have developed a novel approach, the Digital Crumb Investigator, for using data collected as a byproduct of Electonic Health Record (EHR) use to help define care teams and care processes. We are developing tools and methods to utilize these routinely collected data to visualize and quantify care networks across acute care and ambulatory settings We have chosen a clinical care domain where clinicians use EHRs in their offices, on the maternity wards and in the neonatal intensive care units as a test paradigm for this technology. The tools and methods we deliver should readily translate to other health care settings that collect behind-the-scenes electronic metadata such as audit trails. We believe that by applying the methods of social networking to define clinical relationships around a patient's care we will enable new areas of research into the usage of EHRs to promote patient safety and other improvements in care.

Telemedicine has the potential to transcend geographic and socio-cultural barriers to the delivery of high quality health care to the medically underserved populations. However, there are significant cognitive and usability barriers. This paper presents a multifaceted cognitive evaluation of the IDEATel diabetes education and telemedicine program. The evaluation included a cognitive walkthrough analysis to characterize task complexity and identify potential problems as well as field usability testing in patients' homes. The study revealed dimensions of the interface that impeded optimal access to system resources. In addition, we found significant obstacles corresponding to perceptual-motoric skills, mental models of the system, and health literacy. The objective of this work is to contribute to a design framework so that participants with a wide range of skills can better manage their chronic illnesses.

The transition from paper to electronic documentation systems in acute care settings is often gradual and characterized by a period in which paper and electronic processes coexist. Intermediate technologies are needed to "bridge" the gap between paper and electronic systems as a means to improve work flow efficiency through data acquisition at the point of care in structured formats to inform decision support and facilitate reuse. The purpose of this paper is to report on the findings of a study conducted on three acute care units at Brigham and Women's Hospital and Massachusetts General Hospital in Boston, MA to evaluate the feasibility of digital pen and paper technology as a means to capture vital sign data in the context of acute care workflows and to make data available in a flow sheet in the electronic medical record.