Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Textual analysis”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

269 records · Page 15Linked to original sources
PubMed2005

The epistemological-ontological divide in clinical radiology.

Medical ontologies like GALEN, the FMA or SNOMED represent a kind of "100% certain" medical knowledge which is not inherent to all medical sub-domains. Clinical radiology uses computerized imaging techniques to make the human body visible and interprets the imaging findings in a clinical context delivering a textual report. For clinical radiology few standardized vocabularies are available. We examined the definitions given in the glossary of terms for thoracic radiology published by the Fleischner Society. We further classified these terms with regard to their definitions in terms of (a) describing visible structures on the image itself, (b) referring to ontological entities of the body (anatomical or pathological), and (c) terms imposing knowledge on structures visible on the image, epistemologically representing ontological entities of the body. Each ontological/epistemological definition was rated on a scale of vague/weak-sound/strong and put in context with the evaluation comments for the use of the terms given in the glossary itself. The result of this distinction shows that clinical radiology uses many terms referring to ontological entities valid for representation in a medical ontology. However, many epistemological terms exist in the terminology which impose epistemological knowledge on ontological entities. The analysis of the evaluation comments reveals that terms classified as sound (ontologically) and strong (epistemologically) are evaluated higher than terms bearing vague or weak definitions. On the basis of this, we argue that the distinction between ontological and epistemological definitions is necessary in order to construct epistemologically-sensitive application ontologies for medical sub-domains, like clinical radiology, where knowledge is fragmented in terms of description, inferred from a description, concluded on the basis of imaging, or other additional information with varying degrees of certainty.

Electronic Data Processing↗
PubMed2005

An intelligent tutoring system that generates a natural language dialogue using dynamic multi-level planning.

OBJECTIVE: The objective of this research was to build an intelligent tutoring system capable of carrying on a natural language dialogue with a student who is solving a problem in physiology. Previous experiments have shown that students need practice in qualitative causal reasoning to internalize new knowledge and to apply it effectively and that they learn by putting their ideas into words. METHODS: Analysis of a corpus of 75 hour-long tutoring sessions carried on in keyboard-to-keyboard style by two professors of physiology at Rush Medical College tutoring first-year medical students provided the rules used in tutoring strategies and tactics, parsing, and text generation. The system presents the student with a perturbation to the blood pressure, asks for qualitative predictions of the changes produced in seven important cardiovascular variables, and then launches a dialogue to correct any errors and to probe for possible misconceptions. The natural language understanding component uses a cascade of finite-state machines. The generation is based on lexical functional grammar. RESULTS: Results of experiments with pretests and posttests have shown that using the system for an hour produces significant learning gains and also that even this brief use improves the student's ability to solve problems more then reading textual material on the topic. Student surveys tell us that students like the system and feel that they learn from it. The system is now in regular use in the first-year physiology course at Rush Medical College. CONCLUSION: We conclude that the CIRCSIM-Tutor system demonstrates that intelligent tutoring systems can implement effective natural language dialogue with current language technology.

Artificial Intelligence↗
PubMed2006

[E-learning in the education and training of physicians. Methods, results, evaluation].

E-learning has been established in the education and training of physicians in various types: linear sequential and hyper-textual forms of multimedia presentations and texts, tutorial systems and simulations. Case-based e-learning systems are of special importance in medicine because they allow for mediation of process and practical knowledge by presentation of authentic medical cases in a simulated environment. The integration into the medical education and advanced professional training is crucial for the long-term success of e-learning; in case-based systems this can be accomplished by blended learning approaches which combine elements of traditional teaching with e-learning. Learning management systems (LMS) support integration of traditional teaching and e-learning by serving as an organizational platform for content of teaching. Further, they provide means of communication for trainers and trainees, authoring tools, interactive components, course management and role-based sharing concept. The dissemination of e-learning can be fostered by attention to requirements and user analysis, early adoption to organizational structures, curricular integration and continuous cooperation with students. Summarized, didactic and organizational aspects determine the success of our own e-learning offers as well as they influence the general further development of e-learning more than technical features.

Computer-Assisted Instruction↗
PubMed2006

Font adaptive word indexing of modern printed documents.

We propose an approach for the word-level indexing of modern printed documents which are difficult to recognize using current OCR engines. By means of word-level indexing, it is possible to retrieve the position of words in a document, enabling queries involving proximity of terms. Web search engines implement this kind of indexing, allowing users to retrieve Web pages on the basis of their textual content. Nowadays, digital libraries hold collections of digitized documents that can be retrieved either by browsing the document images or relying on appropriate metadata assembled by domain experts. Word indexing tools would therefore increase the access to these collections. The proposed system is designed to index homogeneous document collections by automatically adapting to different languages and font styles without relying on OCR engines for character recognition. The approach is based on three main ideas: the use of Self Organizing Maps (SOM) to perform unsupervised character clustering, the definition of one suitable vector-based word representation whose size depends on the word aspect-ratio, and the run-time alignment of the query word with indexed words to deal with broken and touching characters. The most appropriate applications are for processing modern printed documents (17th to 19th centuries) where current OCR engines are less accurate. Our experimental analysis addresses six data sets containing documents ranging from books of the 17th century to contemporary journals.

Abstracting and Indexing↗
PubMed2006

Building an ontology of pulmonary diseases with natural language processing tools using textual corpora.

Pathologies and acts are classified in thesauri to help physicians to code their activity. In practice, the use of thesauri is not sufficient to reduce variability in coding and thesauri are not suitable for computer processing. We think the automation of the coding task requires a conceptual modeling of medical items: an ontology. Our task is to help lung specialists code acts and diagnoses with software that represents medical knowledge of this concerned specialty by an ontology. The objective of the reported work was to build an ontology of pulmonary diseases dedicated to the coding process. To carry out this objective, we develop a precise methodological process for the knowledge engineer in order to build various types of medical ontologies. This process is based on the need to express precisely in natural language the meaning of each concept using differential semantics principles. A differential ontology is a hierarchy of concepts and relationships organized according to their similarities and differences. Our main research hypothesis is to apply natural language processing tools to corpora to develop the resources needed to build the ontology. We consider two corpora, one composed of patient discharge summaries and the other being a teaching book. We propose to combine two approaches to enrich the ontology building: (i) a method which consists of building terminological resources through distributional analysis and (ii) a method based on the observation of corpus sequences in order to reveal semantic relationships. Our ontology currently includes 1550 concepts and the software implementing the coding process is still under development. Results show that the proposed approach is operational and indicates that the combination of these methods and the comparison of the resulting terminological structures give interesting clues to a knowledge engineer for the building of an ontology.

France↗
PubMed2006

Adaptive prediction trees for image compression.

This paper presents a complete general-purpose method for still-image compression called adaptive prediction trees. Efficient lossy and lossless compression of photographs, graphics, textual, and mixed images is achieved by ordering the data in a multicomponent binary pyramid, applying an empirically optimized nonlinear predictor, exploiting structural redundancies between color components, then coding with hex-trees and adaptive runlength/Huffman coders. Color palettization and order statistics prefiltering are applied adaptively as appropriate. Over a diverse image test set, the method outperforms standard lossless and lossy alternatives. The competing lossy alternatives use block transforms and wavelets in well-studied configurations. A major result of this paper is that predictive coding is a viable and sometimes preferable alternative to these methods.

Algorithms↗
PubMed1976

[Structural interpretation of a supposed symptomatic schizophrenia].

A short text spoken by a patient with a supposed symptomatic schizophrenia in lupus erythematodes is analyzed by means of a structural interpretation. While the text appears schizophrenic, it is only the confusion and intermingling of five quite distinct textual levels which create this impression. These five levels are presented in three distinctive modes of narrative. (1) Deviation and periphrase: while psychopathologically these breaks in the narrative are deviations from the main thread of argument, stylistically they are periphrasic statements, all having identical structures. They are all related emotionally as well as thematically to the central subject of the text. Their informational quality is impaired by their density and their interlocked form. (2) Narrative: the text contains three meaningful and well-arranged stories. In one case a multilevel narrative structure is employed. Each story represents a response to the initial question. (3) Metalanguage: on a metatextual level, the patient repeatedly makes reference to her linguistic peculiarities. These statements coincide with objective evaluation. Delusions and misidentification of individuals and situations used in classical psychopathological diagnosis can be given a different meaning (in the sense used by Weinrich) if interpretation has deepened the knowledge of the text. In this case, the similarity to schizophrenic texts is only superficial. A more detailed analysis illustrates the difference from schizophrenia on every level.

Adult↗
PubMed2002

Restoring accents in unknown biomedical words: application to the French MeSH thesaurus.

In languages with diacritic marks, such as French, there remain instances of textual or terminological resources that are available in electronic form without diacritic marks, which hinders their use in natural language interfaces. In a specialized domain such as medicine, it is often the case that some words are not found in the available electronic lexicons. The issue of accenting unknown words then arises: it is the theme of this work. We propose two internal methods for accenting unknown words, which both learn on a reference set of accented words the contexts of occurrence of the various accented forms of a given letter. One method is adapted from part-of-speech tagging, the other is based on finite state transducers. We show experimental results for letter e on the French version of the Medical Subject Headings thesaurus. With the best training set, the tagging method obtains a precision-recall breakeven point of 84.2+/-4.4% and the transducer method 83.8+/-4.5% (with a baseline at 64%) for the unknown words that contain this letter. A consensus combination of both increases precision to 92.0+/-3.7% with a recall of 75%. We perform an error analysis and discuss further steps that might help improve over the current performance.

Algorithms↗
PubMed2006

Patient subjective experience and satisfaction during the perioperative period in the day surgery setting: a systematic review.

This systematic review used the Joanna Briggs Institute Qualitative Assessment and Review Instrument to manage, appraise, analyse and synthesize textual data in order to present the best available information in relation to how patients experience nursing interventions and care during the perioperative period in the day surgery setting. Some of the significant findings that emerged from the systematic review include the importance of pre-admission contact, provision of relevant, specific education and information, improving communication skills and maintaining patient privacy throughout their continuum of care.

Ambulatory Surgical Procedures↗
PubMed1993

Response analysis in histopathology external quality assessment schemes.

AIMS: To develop a computerised method for analysing the results of histopathology external quality assessment (EQA) schemes which can provide confidential personal reports to individual participating pathologists. METHODS: A program was developed using the OMNIS database system, running on Apple Macintosh or IBM compatible computers. RESULTS: The program produces a general report of participants' responses to each case, and a choice of two types of personal report. One of these provides a list of the participant's diagnoses with a list of the most popular (Consensus) diagnoses for comparison. The other provides automatically calculated scores for the pathologist's performance along with simple statistical evaluation. The scores can be calculated by comparison with the consensus of the group or with correct diagnoses if they are known. A histogram indicating the distribution of performance within the group can be produced. The program can accept uncertainty in the form of differential diagnosis lists from participants. Potentially dangerous diagnostic errors can be identified and handled separately. Participants are identified only by code numbers and confidentiality can easily be enforced. The program is currently being used in the national renal pathology EQA scheme and in the local general histopathology scheme in the East Midlands. CONCLUSIONS: This program offers solutions to problems which have bedevilled the organisers of histopathology EQA schemes. It offers confidential advice to pathologists and will help to identify areas where an individual might benefit from continuing career grade medical education. It raises the possibility of the development of nationally agreed standards of performance in the reporting of pathological specimens, and it may be applicable to other specialties where textual reports are produced.

Confidentiality↗
PubMed2004

When redundant on-screen text in multimedia technical instruction can interfere with learning.

It is frequently assumed that presenting the same material in written and spoken form benefits learning and understanding. The present work provides a theoretical justification based on cognitive load theory, and empirical evidence based on controlled experiments, that this assumption can be incorrect. From a theoretical perspective, it is suggested that if learners are required to coordinate and simultaneously process redundant material such as written and spoken text, an excessive working memory load is generated. Three experiments involving a group of 25 technical apprentices compared the effects of simultaneously presenting the same written and auditory textual information as opposed to either temporally separating the two modes or eliminating one of the modes. The first two experiments demonstrated that nonconcurrent presentation of auditory and visual explanations of a diagram proved superior, in terms of ratings of mental load and test scores, to a concurrent presentation of the same explanations when instruction time was constrained. The 3rd experiment demonstrated that a concurrent presentation of identical auditory and visual technical text (without the presence of diagrams) was significantly less efficient in comparison with an auditory-only text. Actual or potential applications of this research include the design and evaluation of multimedia instructional systems and audiovisual displays.

Adolescent↗
PubMed2002

Experimental comparisons of data entry by automated speech recognition, keyboard, and mouse.

In a series of experiments isolated-word automated speech recognition (ASR) was compared with keyboard and mouse interfaces for three data entry tasks: textual phrase entry, selection from a list, and numerical data entry. To effect fair comparisons, the tasks were designed to minimize the transaction cycle for each input mode and data type, and the main comparisons used times from only correct data entries. With the hardware and software employed the results indicate that for inputting short phrases, ASR competes only if the typist's speed is below 45 words per minute. For selecting an item from a list, ASR offers an advantage only if the list length exceeds 15 items. For entering numerical data, ASR offers no advantage over keypad or mouse. An extrapolation to latency-free ASR suggests that even as hardware and software become faster, human factors will dominate and the results would shift only slightly in favor of ASR.

Computer Peripherals↗
PubMed1995

Computer-assisted trauma care prototype.

Each year, civilian accidental injury results in 150,000 deaths and 400,000 permanent disabilities in the United States alone. The timely creation of and access to dynamically updated trauma patient information at the point of injury is critical to improving the state of care. Such information is often non-existent, incomplete, or inaccurate, resulting in less than adequate treatment by medics and the loss of precious time by medical personnel at the hospital or battalion aid station as they attempt to reassess and treat the patient. The Trauma Care Information Management System (TCIMS) is a prototype system for facilitating information flow and patient processing decisions in the difficult circumstances of civilian and military trauma care activities. The program is jointly supported by the United States Advanced Research Projects Agency (ARPA) and a consortium of universities, medical centers, and private companies. The authors' focus has been the human-computer interface for the system. We are attempting to make TCIMS powerful in the functions it delivers to its users in the field while also making it easy to understand and operate. To develop such a usable system, an approach known as user-centered design is being followed. Medical personnel themselves are collaborating with the authors in its needs analysis, design, and evaluation. Specifically, the prototype being demonstrated was designed through observation of actual civilian trauma care episodes, military trauma care exercises onboard a hospital ship, interviews with civilian and military trauma care providers, repeated evaluation of evolving prototypes by potential users, and study of the literature on trauma care and human factors engineering. This presentation at MedInfo '95 is still another avenue for soliciting guidance from medical information system experts and users. The outcome of this process is a system that provides the functions trauma care personnel desire in a manner that can be easily and accurately used in urban, rural, and military field settings. his demonstration will focus on the user interfaces for the hand-held computer device included in TCIMS, the Field Medic Associate (FMA). The FMA prototype is a ruggedized, water-resistant personal computer, weighing approximately 5 lbs. It has an LCD graphical user interface display for patient record input and output, pen-based and audio input, audio output, and wireless communications capabilities. Automatic recording and dynamic, graphical display of time-stamped trends in patient vital signs will be simulated during the demonstration. Means for accessing existing patient record information (e.g., allergies to particular medications) and updating the record with the nature of the injury, its cause, and the treatments that were administered will be shown. These will include use of an electronic pen to mark up anatoglyphs (standard drawings of human body appearing on computer screen) to show where injuries occurred and where treatments were applied, and to input textual descriptions of the nature of the injury, its cause, what treatments were administered, etc. Computer recognition of handwritten inputs will be shown. Likewise, voice annotation and audio playback of patient record information by medics and hospital personnel will be illustrated. These latter technologies free the care providers' hands to treat the patient; they can therefore provide inputs to the patient record while information is fresh in their minds. The audio playback option allows hospital personnel to select more detailed voice annotations of specific portions of the patient record by simply touching the electronic pen to a particular place where an electronic pen marking was made by a medic in the field and then listening to the medic's corresponding audio commentary. Finally, the FMA's means for assisting the medic in simultaneously managing several injured patients will be shown. (abstract truncated)

Emergency Medical Services↗
PubMed2001

Comparison of literal, inferential, and intentional text comprehension in children with mild or severe closed head injury.

BACKGROUND: Children with head injury have impairments in pragmatic language at the level of both single words and texts. Text comprehension deficits are likely to be the more consequential for everyday and academic function, yet the relative magnitudes of literal and nonliteral text comprehension deficits have not been measured. DESIGN: We compared the magnitude of the impairment in three forms of text comprehension for children with mild or severe head injury relative with controls: literal language (understanding literal text information), inferential language (making pragmatic inferences, textual coherence inferences, or enriching inferences), and the language of mental states and intentions (eg, producing speech acts, appreciating irony, and understanding deception). MEASURES: Effect sizes were used to measure the magnitude of the difference between children with head injury and age-matched controls. RESULTS: Children with severe closed-head injury were significantly impaired on tasks of literal text understanding, inferencing, and intentionality. Children with mild head injury were impaired on some inferencing and all intentionality tasks, although they had no literal text comprehension deficits. CONCLUSIONS: For both groups, the greatest deficits (ie, the largest effect sizes) were on tasks requiring understanding of the language of mental states and intentions. The data bear on the long-term effects of childhood closed-head injury on text- and discourse-level language and also on the nature and timing of language rehabilitation in children with head injury.

Adolescent↗
PubMed2004

Recognizing names in biomedical texts: a machine learning approach.

MOTIVATION: With an overwhelming amount of textual information in molecular biology and biomedicine, there is a need for effective and efficient literature mining and knowledge discovery that can help biologists to gather and make use of the knowledge encoded in text documents. In order to make organized and structured information available, automatically recognizing biomedical entity names becomes critical and is important for information retrieval, information extraction and automated knowledge acquisition. RESULTS: In this paper, we present a named entity recognition system in the biomedical domain, called PowerBioNE. In order to deal with the special phenomena of naming conventions in the biomedical domain, we propose various evidential features: (1) word formation pattern; (2) morphological pattern, such as prefix and suffix; (3) part-of-speech; (4) head noun trigger; (5) special verb trigger and (6) name alias feature. All the features are integrated effectively and efficiently through a hidden Markov model (HMM) and a HMM-based named entity recognizer. In addition, a k-Nearest Neighbor (k-NN) algorithm is proposed to resolve the data sparseness problem in our system. Finally, we present a pattern-based post-processing to automatically extract rules from the training data to deal with the cascaded entity name phenomenon. From our best knowledge, PowerBioNE is the first system which deals with the cascaded entity name phenomenon. Evaluation shows that our system achieves the F-measure of 66.6 and 62.2 on the 23 classes of GENIA V3.0 and V1.1, respectively. In particular, our system achieves the F-measure of 75.8 on the "protein" class of GENIA V3.0. For comparison, our system outperforms the best published result by 7.8 on GENIA V1.1, without help of any dictionaries. It also shows that our HMM and the k-NN algorithm outperform other models, such as back-off HMM, linear interpolated HMM, support vector machines, C4.5, C4.5 rules and RIPPER, by effectively capturing the local context dependency and resolving the data sparseness problem. Moreover, evaluation on GENIA V3.0 shows that the post-processing for the cascaded entity name phenomenon improves the F-measure by 3.9. Finally, error analysis shows that about half of the errors are caused by the strict annotation scheme and the annotation inconsistency in the GENIA corpus. This suggests that our system achieves an acceptable F-measure of 83.6 on the 23 classes of GENIA V3.0 and in particular 86.2 on the "protein" class, without help of any dictionaries. We think that a F-measure of 90 on the 23 classes of GENIA V3.0 and in particular 92 on the "protein" class, can be achieved through refining of the annotation scheme in the GENIA corpus, such as flexible annotation scheme and annotation consistency, and inclusion of a reasonable biomedical dictionary. AVAILABILITY: A demo system is available at http://textmining.i2r.a-star.edu.sg/NLS/demo.htm. Technology license is available upon the bilateral agreement.

Abstracting and Indexing↗
PubMed1999

The United States Supreme Court and psychiatry in the 1990s.

In the 1990s, the Supreme Court has decided several cases that have had an impact on psychiatry and psychiatric patients in the criminal justice system, on psychiatric hospitalization, and on psychotherapist-patient privilege. Of the seven cases discussed in this article, Chief Justice Rehnquist and Justice Scalia voted similarly in all seven cases. Since joining the court, Justice Thomas has voted with them. Justice Scalia interprets the Constitution, using what has been termed "textualism": avoid reference to legislative history, and interpret the Constitution according to the plain language meaning of the relevant section. Chief Justice Rehnquist and Justices Scalia and Thomas are inclined to protect states' rights from court decisions that expand US Constitutional power in cases involving civil plaintiffs and criminal defendants. They seek to protect states from being sued in federal courts, and, if there is doubt, lean toward not interfering with state prerogatives. They tend to not find unenumerated rights and prefer clear-cut rules over amorphous standards. Justices Kennedy and O'Connor, at times joined by Justice Souter in the middle of the court, provide the deciding votes in many cases. They seem to prefer a case-by-case pragmatism over a global jurisprudential philosophy. Approaching cases one at a time, they usually avoid broad philosophic pronouncements when they join with Chief Justice Rehnquist. Justice Stevens, joined by Justices Breyer and Ginsburg since they have been appointed to the court, is more likely to favor a broader reading of the 14th Amendment's Due Process and Equal Protection clauses. Of the seven cases, Kennedy and O'Connor voted with the majority in five cases, the dissent in one case (Zinermon v Burch), and split their votes in one case (Foucha v Louisiana, with O'Connor siding with the Court and Kennedy with the dissent). Commager, a noted historian, believed that political issues can be explored, explained, and debated and that the people of the new American democracy, armed with knowledge and freedom to defend, argue, and choose, will make the right decisions for their common welfare. This theory applies equally to the court: Whenever questions involving psychiatry and psychiatric patients are brought to the court, American psychiatry must make its views known in that forum. To do so requires awareness and knowledge of the cases that involve psychiatry and psychiatric patients that the court has decided, including those decided in the 1990s. To participate effectively, psychiatrists must understand the political landscape in which the cases arrive at the court's doorstep and the composition and leanings of the court and examine carefully the fact patterns (understanding that some fact patterns are more sympathetic than others). This awareness should result in amicus briefs that are scholarly, rely on empiric data, and are scrupulously honest about the limitations of our knowledge. In this way, psychiatrists may fully participate in the debate and aid the court in its exploration and analysis of the issues involving psychiatry and psychiatric patients.

Commitment of Persons with Psychiatric Disorders↗
PubMed2006

Documentation and coding of ED patient encounters: an evaluation of the accuracy of an electronic medical record.

OBJECTIVE: The aim of the study was to describe a paper-based, template-driven and an electronic medical record used for capturing emergency care clinical information and to compare the accuracy of these documentation systems for coding patient encounters using the American Medical Association Current Procedural Terminology-2004 (AMA CPT-2004) evaluation and management codes intended for provider reimbursement. METHODS: A retrospective, cross-sectional study of 4-consecutive-day samples of ED patient encounter records from 2 similar community hospitals was done. For clinical documentation, hospital A uses an electronic medical record, whereas hospital B uses a paper-based template-driven record. Using a simple analytic model, expert coders A and B, respectively, coded the records from hospitals A and B for completeness. First, power analysis determined the acceptability of the patient record sample sizes (1 - beta = .90 at 1% significance level), and the frequency of AMA CPT-2004 primary evaluation and management codes 99281 through 99285 was calculated. Second, the completeness discrepancy rates for hospitals A and B were compared to determine the accuracy of both the paper-based, template-driven record and the electronic medical record in documenting and representing the clinical encounter. Third, interrater reliability between expert coders A and B was calculated to assess the level of agreement between each expert coder in determining the completeness discrepancy rates between hospitals A and B. Finally, the frequency of primary evaluation and management codes was analyzed to determine if there was a statistically significant difference between the paper-based, template-driven record and the electronic medical record representation of the clinical information, and if that difference could be attributable to the differing clinical documentation systems used in hospitals A and B. RESULTS: First, descriptive display demonstrated a difference in the frequency of the primary evaluation and management codes 99283 and 99284 within hospital A (expert coder A assessment, 36.1% vs 39.1%; expert coder B assessment, 36.6% vs 38.7%) and hospital B (expert coder A assessment, 47.8% vs 21.9%; expert coder B assessment, 48.6% vs 21.4%) was noted with the median, primary evaluation, and management code for hospital A of 99284 and the median, primary evaluation, and management code for hospital B of 99283. Second, Fisher exact test compared the completeness discrepancy rates between hospitals A and B as assessed by each expert coder and demonstrated no statistically significant difference in the completeness discrepancy rates (accuracy) between the paper-based, template-driven record and the electronic medical record documentation and coding system when assessed by either expert coder A (P = .370) or expert coder B (P = .819). Third, interrater reliability between expert coders A and B was evaluated using Cohen's kappa statistic. When evaluated both individually and jointly with respect to hospitals A and B, expert coders A and B had a good strength of agreement in their assessments of the accuracy of the documentation and coding system for hospital A (kappa = 0.6200) and hospital B (kappa = 0.6906) as well as for both hospitals evaluated together (kappa = 0.6616). Finally, interhospital differences in the frequency of primary evaluation and management codes were evaluated using Pearson chi(2) test with 3 df. The results for expert coder A (chi(2) = 47.4160; P < .001) and expert coder B (chi(2) = 46.5946; P < .001) recognize that there is a statistically significant degree of difference between hospitals A and B in the frequency distribution of primary evaluation and management codes, probably because of the dispersion of codes 99283 and 99284. CONCLUSIONS: A keystroke-driven, electronic medical record that resides on a knowledge platform that incorporates a clinical structured terminology, administrative coding schemata, AMA CPT-2004 codes and uses object-oriented, open-ended, branching chain clinical algorithms that "force" physician documentation of the clinical elements provides an equally accurate capture and representation of ED clinical encounter data as a paper-based, template-driven documentation system both in terms of the presence or absence of both the medically necessary, discrete data elements and the textual documentation-dependent, medical decision-making elements.

Chi-Square Distribution↗