Assessing ICD-9-CM and ICPC-2 Use in Primary Care. An Italian Case Study Elena Cardillo Institute of Informatics and Telematics, National Research Council Rende (CS), Italy (+39) 0984494957 elena.cardillo@iit.cnr.it Maria Teresa Chiaravalloti Institute for High-Performance Computing and Networking, National Research Council Rende (CS), Italy (+39) 0984494773 chiaravalloti@icar.cnr.it Erika Pasceri Institute of Informatics and Telematics, National Research Council Rende (CS), Italy (+39) 0984494955 erika.pasceri@iit.cnr.it ABSTRACT Controlled vocabularies and standardized coding systems play a fundamental role in the healthcare domain. The International Classification of Diseases (ICD) is one of the most widely used classification systems for clinical problems and procedures. In Italy the 9 th revision of the standard is used and recommended in primary care for encoding prescription documents. This paper describes a statistical and terminological study to assess ICD-9- CM use in primary care and its comparison to the International Classification of Primary Care (ICPC), specifically designed for primary care. The study has been conducted by analyzing the clinical records of about 199,000 patients provided by a set of 166 General Practitioners (GPs) in different Italian areas. The analysis has been based on several techniques for detecting coding practice and errors, like natural language processing and text-similarity comparison. Results showed that the selected GPs do not fully exploit the diseases and procedures descriptive capabilities of ICD-9-CM due to its complexity. Furthermore, compared to ICPC-2, it resulted less feasible in the primary care setting, particularly for the high granularity of the structure and for the lack of reasons for encounters. Categories and Subject Descriptors J.3 [Computer applications]: Life and Medical Sciences – health, medical information systems. General Terms Management, Measurement, Standardization, Languages. Keywords: Classification Systems, e-Health, Primary care, Terminology, ICD, ICPC. 1. INTRODUCTION Standardized coding and classification systems guarantee terminological coherence and unambiguousness, which are essential for healthcare information exchange efficiency. The amount of clinical information produced and stored every day by healthcare providers is useless if it cannot be reused in epidemiological studies, research, clinical decision support systems, and ultimately in delivering better services to patients. Moreover, the use of common terminology standards allows a coherent and efficient exchange of data among clinical systems. As Hammond and Cimino stated [12], a standard for coding patient data is nontrivial because it solves semantic issues related to definitions, use of qualifiers, different levels of granularity in the data, and synonymy, not to mention the breadth and depth that standard is required to have to be as such. Despite multiple benefits, two of the most commonly complained problems in using existing standards are about the strict terminological structure (the majority of medical terminologies are pre- coordinated hierarchical classification systems) and the level of the semantic coverage of the intended domain. As a result, standard terminologies are not widely accepted enough to guarantee interoperable exchange of coded clinical information. This paper proposes an analysis of the use of an international standard for coding diseases in primary care. In particular, it is analyzed how and how much the International Classification of Diseases (ICD) is used among General Practitioners (GPs) in Italy for diagnoses and problems encoding in prescriptions, exploring two use cases, Campania Region and the Autonomous Province of Trento (henceforth APT), and we compare it to the International Classification of Primary Care (ICPC), which is achieving a growing success among the health professionals of the domain. The attention is focused on ICD and ICPC coverage of the primary care domain and on the GPs attitude towards coding. The paper is structured as follows: in Section 2 an overview of ICD and ICPC is given in order to understand their scope, structure and use in the literature; in Section 3 the approach for analyzing the use of ICD and ICPC in primary care in Italy is presented and an evaluation of the results is shown in Section 4; finally Section 5 and 6 present some discussions and conclusions. 2. MATERIALS Primary care terminology is peculiar as it needs to cover all the branches of medicine, but it rarely reaches the articulate diagnosis expressions proper of a specialist. GPs are the first contact between patients and the healthcare system. They know about the general health condition of a patient, but to precisely define a diagnosis they often require further information. Therefore they usually codify a provisional diagnosis, based on symptoms, signs, abnormal findings and problems reported by the patient. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. DH '15, May 18 - 20, 2015, Florence, Italy Copyright is held by the owner/author(s). Publication rights licensed to ACM. ACM 978-1-4503-3492-1/15/05…$15.00 http://dx.doi.org/10.1145/2750511.2750525 95