Proceedings. AMIA Symposium最新文献

The Connecticut Hospital Association has developed a data resource for inpatient pharmacy information. The RxData project collects drug dispensing information from member hospitals and joins it to administrative discharge data in the Chime database. The resulting dataset is useful for descriptive epidemiology of drug use patterns as well as surveillance, quality improvement, and some hypothesis testing. The drug identity is derived from the National Drug Code submitted by hospital pharmacies. A drug reference file is used together with these codes for hierarchical analysis. Data are accepted from participating facilities in a variety of formats and mapped to a common schema. The program uses locally developed roll-up logic to overcome the lack of consistent standards for recording inpatient drug order and dispense information. Dispensing records at different levels of aggregation are collected from source pharmacy information systems and converted to a standard "regimen" based on continuous dispensing of the same drug. The resulting record structure allows direct comparison of data from dissimilar systems. Data are currently available for eleven acute care hospitals and most of their associated emergency and outpatient surgery facilities. The program is expected to expand to cover most if not all Connecticut hospitals over the next two to three years.

Abstract As the home becomes an increasingly important site for health care, an increasing number of technology applications or devices are being introduced to support health at home. However, introducing new technology into a household raises a number of issues that must be considered prior to, during, and after the technology is implemented. This paper reviews the experiences of the UW-Madison Advanced Technologies for Health@Home Project, summarizing our assessment of household requirements that should be analyzed prior to introducing new technology. The overall goal of the Health@Home project is to improve the functionality and content of information technology innovations for the home. Using Venkatesh and Mazumdar's framework this article will summarize the relevant social, behavioral, technological, and physical dimensions of households that must be carefully assessed and understood to help ensure that the technology fits the needs of home residents.

We present a method to obtain an end-to-end characterization of the performance of an application over a network. This method is not dependent on any specific application or type of network. The method requires characterization of network parameters, such as latency and packet loss, between the expected server or client endpoints, as well as characterization of the application's constraints on these parameters. A subjective metric is presented that integrates these characterizations and that operates over a wide range of applications and networks. We believe that this method may be of wide applicability as research and educational applications increasingly make use of computation and data servers that are distributed over the Internet.

In general, it is very straightforward to store concept identifiers in electronic medical records and represent them in messages. Information models typically specify the fields that can contain coded entries. For each of these fields there may be additional constraints governing exactly which concept identifiers are applicable. However, because modern terminologies such as SNOMED CT are compositional, allowing concept expressions to be pre-coordinated within the terminology or post-coordinated within the medical record, there remains the potential to express a concept in more than one way. Often times, the various representations are similar, but not equivalent. This paper describes an approach for retrieving these pre- and post-coordinated concept expressions: (1) Create concept expressions using a logically-well-structured terminology (e.g., SNOMED CT) according to the rules of a well-specified information model (in this paper we use the HL7 RIM); (2) Transform pre- and post-coordinated concept expressions into a normalized form; (3) Transform queries into the same normalized form. The normalized instances can then be directly compared to the query. Several implementation considerations have been identified. Transformations into a normal form and execution of queries that require traversal of hierarchies need to be optimized. A detailed understanding of the information model and the terminology model are prerequisites. Queries based on the semantic properties of concepts are only as complete as the semantic information contained in the terminology model. Despite these considerations, the approach appears powerful and will continue to be refined.

Computer-based prompting and reminder systems have been shown to be highly effective in increasing rates of preventive services delivery. However, there are many more recommended preventive services than can be practically included in a typical clinic visit. Therefore prioritization of preventive services prompts is necessary. We describe two approaches to prioritizing preventive services prompts based on expected value decision making. One method involves a static, global prioritization across all preventive services and has been used in a production system for almost 7 years. The second method uses influence diagrams to prioritize prompts dynamically, based on individual patient data. The latter approach is still under development. Both methods are labor intensive and require a combination of epidemiologic data and expert judgment. Compromises in strictly normative process were necessary to achieve user satisfaction.

Objective: To identify controlled evidence from the child health literature on patient conditions and clinical procedures that resulted in unacceptable adverse outcomes.

Methods: Systematic searches of MEDLINE (1966 to 2001), and Cochrane Database of Systematic Reviews (2001) were done. Studies that met the eligibility criteria, were verified for quality of methodology and lack of conflicting studies. A knowledge base of Child Health Safety Modules was then developed. The knowledge base could be used to transfer controlled evidence on potentially harmful interventions into clinical decision support systems conforming with Arden Syntax, a widely applied computer standard.

Results: The searches identified knowledge to create 41 Child Health Safety Modules for medications and procedures in child health care, from 29 randomized controlled trials and 12 non-randomized controlled studies. The modules are focused on 28 medication interventions and 13 other clinical procedures. Eighty five percent of the studies were published between 1997-2001.

Conclusion: An increasing amount of controlled evidence on risks of adverse outcomes in child health is available to alert clinicians when potential planning errors are about to be overlooked.

The purpose of this paper is to explore the application of knowledge management concepts to an information systems (IS) knowledge base, as opposed to a clinical one. The field of Medical Informatics is committed to helping others manage medical information and knowledge through the application of information technology. At Partners HealthCare, a wide variety of clinical information management systems have been built and implemented in complex environments, creating an extensive applied informatics knowledge base. How should healthcare IS departments manage this intellectual capital? That's the question that Partners HealthCare is asking its senior and middle IS managers. This paper reports on an internal survey addressing Knowledge Management (KM) requirements, the potential application of this technology in our organization, and discusses where we are today and where to go from here.

Clinical guidelines aim to eliminate clinician errors, reduce practice variation, and promote best medical practices. Computer-interpretable guidelines (CIGs) can deliver patient-specific advice during clinical encounters, which makes them more likely to affect clinician behavior than narrative guidelines. To reduce the number of errors that are introduced while developing narrative guidelines and CIGs, we studied the process used by the ACP-ASIM to develop clinical algorithms from narrative guidelines. We analyzed how changes progressed between subsequent versions of an algorithm and between a narrative guideline and its derived clinical algorithm. We recommend procedures that could limit the number of errors produced when generating clinical algorithms. In addition, we developed a tool for authoring CIGs in GLIF3 format and validating their syntax, data type matches, cardinality constraints, and structural integrity constraints. We used this tool to author guidelines and to check them for errors.

Prediction of microscopic spread of tumor cells is becoming critically important in the decision making process in planning radiation therapy for cancer. Until recently, radiation treatment of head and neck cancer has been conservative, treating large regions to insure eradication of disease. However, if it is known that regional spread is confined, a more focused treatment can be considered, with the payoff of reducing or eliminating morbidity due to irradiating healthy tissue in the vicinity of node groups. Knowledge about the occurrence of micrometastases comes mainly from pathology reports in connection with surgery. As the data accrue, it will be possible and necessary to represent this knowledge in a symbolic computational model. Our work reports on the feasibility of modeling this knowledge using published data.

Several clinical terminologies now utilize description logic to model the logical definitions of concepts. Recent editions of the Systematized Nomenclature of Medicine (SNOMED) have been developed using the description logic Ontylog. A significant design criterion for SNOMED is to keep concept expressions simple enough to be broadly usable by clinicians, while maintaining faithful representation of concept meaning. Motivated by this criterion, "role grouping" has been developed as an extension to the description logic Ontylog. This paper describes the problems that motivated the creation of role grouping, outlines the semantics of role grouping, illustrates the benefits of this construct with examples from SNOMED Clinical Terms, and provides an algorithm for determining normal forms for expressions involving role groups.