Yugo Floristán Floristán1, Josu Delfrade Osinaga2, Jesus Carrillo Prieto3, Jesus Aguirre Perez4, Conchi Moreno-Iribas5. 1. Instituto de Salud Pública y Laboral de Navarra. Instituto de Investigación Sanitaria de Navarra (IdiSNA). Pamplona. España.yugo.floristan.floristan@cfnavarra.es. 2. Instituto de Salud Pública y Laboral de Navarra. Instituto de Investigación Sanitaria de Navarra (IdiSNA). Pamplona. España and CIBER de Epidemiología y Salud Pública (CIBERESP). 3. Instituto Nacional de Estadística (INE). Madrid. España. 4. Instituto de Salud Pública y Laboral de Navarra. Instituto de Investigación Sanitaria de Navarra (IdiSNA). Pamplona. España. 5. Instituto de Salud Pública y Laboral de Navarra. Instituto de Investigación Sanitaria de Navarra (IdiSNA). Pamplona. España and Red de Investigación en Servicios Sanitarios en Enfermedades Crónicas (REDISSEC). Pamplona.
Abstract
BACKGROUND: There are few studies that analyze changes in mortality statistics derived from the use of IRIS software, an automatic system for coding multiple causes of death and for the selection of the underlying cause of death, compared to manual coding. This study evaluated the impact of the use of IRIS in the Navarre mortality statistic. METHODS: We proceeded to double coding 5,060 death certificates corresponding to residents in Navarra in 2014. We calculated coincidence between the two encodings for ICD10 chapters and for the list of causes of the Spanish National Statistics Institute (INE-102) and we estimated the change on mortality rates. RESULTS: IRIS automatically coded 90% of death certificates. The coincidence to 4 characters and in the same chapter of the CIE10 was 79.1% and 92.0%, respectively. Furthermore, coincidence with the short INE-102 list was 88.3%. Higher matches were found in death certificate of people under 65 years. In comparison with manual coding there was an increase in deaths from endocrine diseases (31%), mental disorders (19%) and disease of nervous system (9%), while a decrease of genitourinary system diseases was observed (21%). CONCLUSIONS: The coincidence at level of ICD10 chapters coding by IRIS in comparison to manual coding was 9 out of 10 deaths, similar to what is observed in other studies. The implementation of IRIS has led to increased of endocrine diseases, especially diabetes and hyperlipidaemia, and mental disorders, especially dementias.
BACKGROUND: There are few studies that analyze changes in mortality statistics derived from the use of IRIS software, an automatic system for coding multiple causes of death and for the selection of the underlying cause of death, compared to manual coding. This study evaluated the impact of the use of IRIS in the Navarre mortality statistic. METHODS: We proceeded to double coding 5,060 death certificates corresponding to residents in Navarra in 2014. We calculated coincidence between the two encodings for ICD10 chapters and for the list of causes of the Spanish National Statistics Institute (INE-102) and we estimated the change on mortality rates. RESULTS: IRIS automatically coded 90% of death certificates. The coincidence to 4 characters and in the same chapter of the CIE10 was 79.1% and 92.0%, respectively. Furthermore, coincidence with the short INE-102 list was 88.3%. Higher matches were found in death certificate of people under 65 years. In comparison with manual coding there was an increase in deaths from endocrine diseases (31%), mental disorders (19%) and disease of nervous system (9%), while a decrease of genitourinary system diseases was observed (21%). CONCLUSIONS: The coincidence at level of ICD10 chapters coding by IRIS in comparison to manual coding was 9 out of 10 deaths, similar to what is observed in other studies. The implementation of IRIS has led to increased of endocrine diseases, especially diabetes and hyperlipidaemia, and mental disorders, especially dementias.
Entities:
Keywords:
Aging; Biostatistics; Causes of Death; Clinical Coding; Endocrine System Diseases; Epidemiology; ICD-10-CM; Mental disorders; Mortality; Nervous System; Spain