[Coding Causes of Death with IRIS Software. Impact in Navarre Mortality Statistic]

Area of Science:

• Public Health
• Medical Informatics
• Epidemiology

Background:

• Limited research exists on mortality statistics changes due to automated coding systems like IRIS compared to manual methods.
• This study assesses the impact of implementing the IRIS software on Navarre's mortality statistics.

Purpose of the Study:

• To evaluate the effect of using the IRIS (Infostat) software for automatic cause-of-death coding on mortality statistics in Navarre.
• To compare the accuracy and outcomes of IRIS coding against traditional manual coding methods.

Main Methods:

• A double-coding approach was applied to 5,060 death certificates from Navarre residents in 2014.
• Coincidence rates between IRIS and manual coding were calculated for ICD-10 chapters and the Spanish National Statistics Institute (INE-102) list.
• Changes in mortality rates were estimated based on the coding method.

Main Results:

• IRIS automatically coded 90% of death certificates with high concordance: 79.1% at ICD-10 4-character level and 92.0% at chapter level.
• Concordance with the INE-102 list was 88.3%, with higher accuracy for individuals under 65.
• Compared to manual coding, IRIS showed increased mortality rates for endocrine diseases (31%), mental disorders (19%), and nervous system diseases (9%), with a decrease in genitourinary diseases (21%).

Conclusions:

• IRIS demonstrates strong agreement (9 out of 10 deaths) with manual coding at the ICD-10 chapter level, consistent with other studies.
• Implementation of IRIS led to notable increases in reported deaths from endocrine diseases (e.g., diabetes, hyperlipidemia) and mental disorders (e.g., dementias).