Search

Usability has been identified as a key issue in health informatics. Worldwide numerous projects have been carried out in an attempt to increase and optimize health system usability. Usability testing, involving observing end users interacting with systems, has been widely applied and numerous publications have appeared describing such studies. However, to date, fewer works have been published describing methodological approaches to analyzing the rich data stream that results from usability testing. This includes analysis of video, audio and screen recordings. In this paper we describe our work in the development and application of a coding scheme for analyzing the usability of health information systems. The phases involved in such analyses are described.

Standardized quality assurance (QA) plays an import role to maintain and develop success of cardiovascular procedures (CP). Well-established QA models from Germany could be shared in a form repository for world-wide reuse and exchange. Therefore, we collected the complete set of all quality QA forms for CP, which is obligatory to be filled out by all German health service providers. Original forms were converted into standardized study forms according to ODM (Operational Data Model) and translated into English. Common medical concepts and clusters of medical concepts were identified based on UMLS coding of form items. All forms are available on the web as multilingual ODM documents. UMLS concept coverage analysis indicates 88% coverage with few but critically important definition gaps, which need to be addressed by UMLS.

The Austrian Procedure Catalogue is used for procedure coding in Austria. Its architecture and content has some major weaknesses. The aim of this study is the presentation of a new potential content model for this classification system consisting of main characteristics of health interventions. It is visualized using a UML class diagram. Based on this proposition, an implementation of an ontology for procedure coding is planned.

The generation of a semantic clinical infostructure requires linking ontologies, clinical models and terminologies [1]. Here we describe an approach that would permit data coming from different sources and represented in different standards to be queried in a homogeneous and integrated way. Our assumption is that data providers should be able to agree and share the meaning of the data they want to exchange and to exploit. We will describe how Clinical Element Model (CEM) and OpenEHR datasets can be jointly exploited in Semantic Web environments.

We started a multi-year project to collect discharge summaries from multiple hospitals and create a big text database to build a common document vector space, and develop various applications such as the autoselection of the disease. As the first step, we extracted discharge summary from two hospitals. Using a text mining method, we carried out a DPC selection. There was a difference in term structure and number of terms between the discharge summaries from both hospitals. Nevertheless, the selection rate of the disease is resembled closely.

Semantic interoperability is an unsolved problem which occurs while working with medical forms from different information systems or institutions. Standards like ODM or CDA assure structural homogenization but in order to compare elements from different data models it is necessary to use semantic concepts and codes on an item level of those structures. We developed and implemented a web-based tool which enables a domain expert to perform semi-automated coding of ODM-files. For each item it is possible to inquire web services which result in unique concept codes without leaving the context of the document. Although it was not feasible to perform a totally automated coding we have implemented a dialog based method to perform an efficient coding of all data elements in the context of the whole document. The proportion of codable items was comparable to results from previous studies.

Incomplete coding is a known problem in hospital information systems. In order to detect non-coded secondary diseases we developed a text classification system which scans discharge summaries for drug names. Using a drug knowledge base in which drug names are linked to sets of ICD-10 codes, the system selects those documents in which a drug name occurs that is not justified by any ICD-10 code within the corresponding record in the patient database. Treatment episodes with missing codes for diabetes mellitus, Parkinson's disease, and asthma/COPD were subject to investigation in a large German university hospital. The precision of the method was 79%, 14%, and 45% respectively, roughly estimated recall values amounted to 43%, 70%, and 36%. Based on these data we predict roughly 716 non-coded diabetes cases, 13 non-coded Parkinson cases, and 420 non-coded asthma/COPD cases among 34,865 treatment episodes.

: The WHO International Classification of Diseases is used in many national applications to plan, manage and fund through case mix health care systems and allows international comparisons of the performance of these systems. There is no such measuring tool for health interventions or procedures. To fulfil this requirement the WHO-FIC Network recommended in 2006 to develop an International Classification of Health Interventions (ICHI). This initiative is aimed to harmonise the existing national classifications and to provide a basic system for the countries which have not developed their own classification systems. It is based on the CEN/ISO ontology framework standard named Categorial Structure defined from a non formal bottom up ontology approach. The process of populating the framework is ongoing to start from a common model structure encompassing the ICD 9CM Volume 3 granularity.

Aim: The purpose of this study is to review current in-hospital based coding techniques and explore various means by which clinicians and HIM (Healthcare Information Management) staff members (which include Coding Technologist or "Coders") miss potential technical fee reimbursement due to poor dictation and/or lack of knowledge of the newer MS-DRG coding system in the U.S., Methods: An analysis of the 2008 implemented MS-DRG reimbursement system was conducted to compare it to the older DRG system in use prior. With the help of a practicing Cardiologist, five randomly selected inpatient charts were reviewed to identify whether or not there was an opportunity to push deeper in to the patient diagnosis and treatment plan to determine if an opportunity existed to code each patient's diagnosis plan at a higher level due to poor documentation practices., Results: The review of five in-hospital patients charts seen by a Cardiologist, found that by correctly understanding the newer MS-DRG coding specifications and documenting patient diagnosis and comorbidities to appropriate reimbursement levels the hospital could have increased technical fee payments by about $50,000., Conclusions: By educating clinicians and HIM staff members to be rigorous in their dictation and interpretation of the patients MS-DRG diagnosis, comorbidities and treatment protocol, hospitals will be able to reclaim potentially millions of dollars annually. As more facilities adopt EMRs (electronic medical records), the introduction of "reminders" or drop down lists within an EMR could aid on the whole coding completeness, thereby increasing overall technical fee reimbursement levels while ensuring possibility of not overcharging accounts.

This paper presents: (1) a framework of formal representation of ICD10, which functions as a bridge between ontological information and natural language expressions; and (2) a methodology to use formally described ICD10 for computer-assisted ICD coding. First, we analyzed and structurized the meanings of categories in 15 chapters of ICD10. Then we expanded the structured ICD10 (S-ICD10) by adding subordinate concepts and labels derived from Japanese Standard Disease Names. The information model to describe formal representation was refined repeatedly. The resultant model includes 74 types of semantic links. We also developed an ICD coding module based on S-ICD10 and a 'Coding Principle,' which achieved high accuracy (>70%) for four chapters. These results not only demonstrate the basic feasibility of our coding framework but might also inform the development of the information model for formal description framework in the ICD11 revision.

ICD-coding is a complex and difficult task. Coding results vary a great deal depending on each coder's ability. Although the Japanese Standard Disease-Code Master facilitates the coding tasks, it also engenders post-coordination problems derived from combinations of basic diseases (with ICD code) and modifiers. Post-coordination sometimes alters the original ICD code dramatically. To solve this problem, this paper presents a proposal for using internal structures of disease names to correct the ICD code. First, we built an internal structure analyzer, which achieved high (83.7%) accuracy. Results demonstrated that the analyzed output is helpful for precise ICD-coding tasks.