Health care procedures.

BACKGROUND: ICD is used for coding medical diagnoses across the world, but there is no globally accepted coding system for health care procedures. The need for the introduction of a common international medical procedure classification has been addressed by the Australian NCCH, which proposed the International Classification of Health Interventions (ICHI) as the basis of an international procedure classification. In parallel, the French multiaxial Classification Commune des Actes Médicaux (CCAM) has been established. OBJECTIVES: The aim is to compare ICHI to the CCAM architecture and to assess their appropriateness for supporting international comparability of procedure data and give a recommendation for the further development of international procedure classifications. METHODS: The architecture of both ICHI and CCAM was thoroughly analyzed. ICHI classes were mapped to the classes of the multiaxial CCAM basic coding tables. This was done manually by domain experts, which analyzed the exact wording of each ICHI title. The result was assessed in terms of representability and granularity. RESULTS: 78.4% of ICHI classes could be mapped directly to CCAM. The anatomical site could be represented in 99.3%. Numerous ICHI classes combined anatomical sites requiring more than one CCAM code. Problems arouse due to imprecise ICHI descriptions. CONCLUSIONS: CCAM appeared as the more elaborate and mature system whereas ICHI had some drawback regarding ambiguity and varying granularity. It is recommended to improve the structure of ICHI by the beneficial aspects of the CCAM and to avoid semantic ambiguities by applying ontological principles and logic-based representation languages.

Immunity to adeno-associated virus vectors in animals and humans: a continued challenge.

Recombinant vectors based on adeno-associated virus (AAV) have been shown to stably express many genes in vivo without mounting immune responses to vectors or transgenes. Thus, AAV vectors have rapidly become the reagents of choice for therapeutic gene transfer. Yet one of the first translations of AAV gene therapy into humans unexpectedly resulted in only short-term expression of the therapeutic gene accompanied by transient but significant toxicity. Immune responses to the vector capsid were held accountable for these results, confirming that a detailed understanding of the interaction of AAV vectors with the immune system is of great importance for the safety and success of gene therapy applications. Most humans display naturally acquired immunity to AAV; circumventing neutralizing antibodies and memory T-cell responses is challenging, but not impossible. This review will evaluate the strategies that have been proposed to overcome such responses and summarize recent findings about the mechanisms and circumstances that lead to the activation of innate and adaptive immune responses to AAV vector components.

[The French Common Classification of Procedures CCAM. An option for Germany]. / Die französische Prozedurenklassifikation CCAM. Eine Option für Deutschland.

A working group of the National Board for Classification in Health Care at the Federal Ministry for Health (KKG) concerned itself between 1996 and 2003 with the topic of a possible follow-up procedure classification system for the actually used "Operation and Procedure Coding System" (OPS-301), which was introduced to Germany in the mid 1990's. In the context of a feasibility study the American "Procedure Coding system" (PCS) and the French "Classification commune des actes médicaux "(CCAM) were examined and evaluated as possible follow-up classifications. In this study it could be shown that the CCAM has great advantages over the pure multiaxial PCS. The strength for that is its medical content, its construction, the methods used and in particular the usability by physicians, the ease of statistic evaluations and the support to maintain the CCAM. In this article the CCAM is presented in detail with its basic principles and its classification methods together with some of the most important coding rules. It is discussed whether the CCAM can be a follow-up classifications system for the German OPS as it is, or if it would make more sense to reconstruct the existing medical content of the OPS with the architecture and methodology of the CCAM.

[Electronic distribution of patient records and images through PACS: implementation of data protection at the University Hospital in Freiburg]. / Elektronische Befund- und Bildverteilung aus einem PACS: Umsetzung der datenschutzrechtlichen Aspekte am Beispiel des Universitätsklinikums Freiburg.

In large installations of a Picture Archiving and Communication System (PACS), data protection becomes an issue. Above all, not every employee should have access to all data stored in the system. At the Freiburg University Hospital, we developed a concept of dynamic assignment of authorization, assuring that every physician is authorized to the data of all patients under his care. We describe the technical implementation of data protection developed for the integration of PACS and electronic patient records in Freiburg. The system was installed in January 2001 and the experience of its two years in operation is reported.

Automated coding of diagnoses--three methods compared.

In Germany, new legal requirements have raised the importance of the accurate encoding of admission and discharge diseases for in- and outpatients. In response to emerging needs for computer-supported tools we examined three methods for automated coding of German-language free-text diagnosis phrases. We compared a language-independent lexicon-free n-gram approach with one which uses a dictionary of medical morphemes and refines the query by a mapping to SNOMED codes. Both techniques produced a ranked output of possible diagnoses within a vector space framework for retrieval. The results did not reveal any significant difference: The correct diagnosis was found in approximately 40% for three-digit codes, and 30% for four-digit codes. The lexicon-based method was then modified by substituting the vector space ranking by a heuristic approach that capitalizes on the semantic structure of SNOMED, thus raising the number of correct diagnoses significantly (approximately 50% for three-digit codes, and 40% for four-digit codes). As a result, we claim that lexicon-based retrieval methods do not perform better than the lexicon-free ones, unless conceptual knowledge is added.

Using ICD-10 for case groups.

The definition of German case groups now uses an adaptation of the ICD-10: ICD-10-SGBV. For the transformation of the existing ICD-9-based definition into the ICD-10-SGBV an ICD-9/ICD-10 conversion table was used. The derived raw definitions were manually refined in detail. Due to the transformation of the classification system, the number of definitions increased immensely, as the ICD-10 SGBV is by far more detailed than the formerly used ICD-9. In our opinion, ICD-10 SGBV is not ideal for the definition of case groups, because this classification system is designed to support statistics on morbidity and mortality, but definitions of case groups are oriented to cost factors. Therefore the definition of case groups should be based on a specially designed classification system or one should reassess the necessity of a parallel definition by codes of classifications as well as by text. In both alternatives the introduction of a detailed and systematic medical documentation with expressive terminology systems will offer the advantage of classifying patients into medical oriented classifications as well as case groups.

Realization of a CPR using HL7 queries from communication server to information server.

A communication server is often used in Germany to interface legacy systems. It handles the translation, duplication, and transmission of messages to all systems that might need this information. Based on the principles of "for every data there is only one master" and "information on demand" we have built an information server that reduces the messages required and simplifies the handling of messages, the update of data already sent compared to a communication server. It also allows to implement a security policy as required by the German data privacy legislature.

Towards a multilingual morpheme thesaurus for medical free-text retrieval.

We introduce a methodology for the segmentation of complex compounds into medically plausible morphemes. A tool for thesaurus compilation and management is presented, and design principles for a multilingual morpheme thesaurus are outlined. Our goal is to enhance the quality of medical free-text retrieval by replacing lexically based through morpheme-based search procedures.

Conversion problems concerning automated mapping from ICD-10 to ICD-9.

The increasing parallel use of ICD-9 and ICD-10 complicates the comparability of coded diagnoses. This is the reason why we developed a symmetric table for interactive conversion between ICD-9 and ICD-10, based on a vector space text-retrieval method that resulted in unambiguous mapping from ICD-9 to ICD-10 in 64%, from ICD-10 to ICD-9 in 87% of all three- and four-character classes of the tabular list. Out of the remaining 13% of multi-valued relations, a table for automated mapping from ICD-10 to ICD-9 was created. In 9% of cases, the selection offered no problems. A compromise between preserving information content and maintaining the logical integrity had to be found in 2.4%; in 1.6% automated mapping was impossible because of newly defined concepts and structural differences between ICD-9 and ICD-10 that are not counterbalanced by a consistent system of residual categories. We recommend that in a future revision of the ICD, compatibility with the then existing classification system should be considered.

[Procedure Coding System (PCS): evaluation by the Ministry of Health]. / Procedure Coding System (PCS): Bewertungsverfahren beim BMG.

It can be assumed that the German surgical procedure coding system OPS-301 must be replaced by a new coding system at the beginning of the next millennium. The "Procedure Coding System (ICD-10-PCS)" which is presently being developed in the USA was considered the most promising system by the German Curators for questions of classification in Public Health (KKG). Therefore, a working group with members representing all organisations of the KKG was established to evaluate the PCS with special reference to the German conditions. The working group found that PCS is at the moment the best available procedure coding system. The German version of PCS has been licensed by DIMDI in December 1997 and will therefore be in the public domain. The German version will be developed in co-operation with all involved organisations for pilot studies and accompanying scientific evaluation, thus a maximum of consensus should be achieved.

Critical success factors for a hospital-wide PACS.

Since 1996 a picture archiving and communications system (PACS) is installed at the university hospital of Freiburg. The PACS is integrated in the hospital information system (HIS) and several modalities of different vendors are attached to it. During the implementation phase three critical factors to the success of our PACS installation where identified: the support of the workflow, the interface of the radiological information system (RIS) to the modalities, and the security policy to allow hospital-wide access to the images and results in the PACS.

The minimum basic data set as the core of the Freiburg University Hospital information system.

At the University Hospital Freiburg (1,927 beds; 53,000 inpatients/year), we developed and ran a Medical Information, Retrieval, and Archives System (MIRA) by integrating the Minimum Basic Data Set into the self developed patient administration system and all the other parts of our hospital information system. Especially for the physicians, MIRA offers access to basic medical data of all patients treated in all medical departments and separate buildings since 1986. MIRA is mainly used in cases of readmission and for scientific and managerial purposes. The privacy controlled access to the central patient data bases (records of 450,000 patients) of MIRA is possible via 610 online PCs and dialog terminals. Because of the many easy-to-use standard functions and special extensions for different clinical departments, MIRA became one of the best accepted edp systems for physicians, nurses and administrators.

[The hospital service misuse problem in orthopedics]. / Die Fehlbelegungsproblematik in der Orthopädie.

In a representative study for all German acute hospitals the amount of inappropriate (medically not necessary) patient care was estimated. The data material, sampling design, weighting technic and the appropriateness expert rating (retrospective, consent judging) was controlled. An overall proportion of inappropriate days of care of 18.45% were found but for orthopedic diseases this rate was 31.9% and even 37.7% for the younger than 60 orthopedic patients. Similar high rates for orthopedic diseases were found in US hospital studies of inappropriateness and should lead to an analysis of their reasons.

[Discriminatory value of bronchoalveolar lavage parameters in differential diagnosis--exogenous allergic alveolitis, sarcoidosis and patients with healthy lungs]. / Die diskriminatorische Wertigkeit von BAL-Parametern bei der Differentialdiagnose--Exogen-allergische Alveolitis, Sarkoidose und Lungengesunde.

For this study, the results of 487 BAL investigations involving 126 patients with sarcoidosis, 34 patients with exogenous-allergic alveolitis, 18 subjects with healthy lungs, and 309 patients with other pulmonary disorders, were analysed. The diagnoses had been established independently of the BAL results. After elimination of interdependent variables, the discriminatory usefulness of the BAL parameters: total number of cells, relative percentage of lymphocytes, natural killer cells, and T4/T8 ratio, was analysed with the aid of ROC curves and the Wilcoxon test. The ROC curves permit a rapid overview of the differential significance of the various BAL variables, and identify the relationship between sensitivity and specificity.

A multiuser system for whole body plethysmographic measurements and interpretation.

A multiuser system for whole body plethysmographic measurements and interpretation which has been developed under clinical conditions is described. The following measurements can be carried out in a rapid way and in one session with the patient: specific airway resistance during spontaneous breathing, determination of functional residual capacity, static lung volumes, and maximal forced expiratory data. Each section is normally measured twice and can be repeated up to ten times. The final results are displayed and printed together with a consistent system of normal reference values. All values and selected original curves are stored automatically in an integrated data base system. Obstructive patients are measured again after the inhalation of a bronchodilator. All results are evaluated by an automatic interpretation program. This analyzes and graduates airway obstruction, lung volumes, and pharmacological airway reversibility using standardized texts which are written below all numerical printouts and graphical plots. The interpretation algorithm is tree structured and uses the normal reference values as a knowledge base. The system supports up to four online laboratories with their own A/D converter and up to 20 video terminals, printers, plotters, and modems. Our laboratory performs 8,297 such complete measurements on 4,671 different patients per year with one body box.

Medical expert systems: design and applications in pulmonary medicine.

Expert systems are computer applications with a built-in knowledge of a special field to solve problems like a human expert on this subject. An expert system consists of an inference component for problem specification and solving, a knowledge base for storage of data, facts, rules and heuristics, an interface for knowledge acquisition, another for the interaction with the user, and a component for reasoning. Pulmonary expert systems are used for automatic interpretations of lung function data which are measured on-line and directly computed. There are also some expert systems for consultation in difficult pulmonary cases and for protection against overlooking rare diseases. Perhaps the most successful applications of pulmonary expert systems are for educational purposes and a large variety of teachware is available. Pulmonary expert systems are presently limited to small applications; often it is necessary to type in large amounts of data and to follow a deeply structured cumbersome dialog. Also, the problem of responsibility for computer decisions has to be mentioned. But pulmonary expert systems as integrated supplements of normal pulmonary measuring devices, as decision support, and as parts of pulmonary teachware will be helpful tools for the pneumological physician.