*3.2. ICD-11*

The 11th revision of the ICD was adopted by the World Health Assembly in 2019 and is now available for implementation. ICD-11 was designed as an electronic classification. All entities including diseases, disorders, injuries and symptoms are stored in the ICD-11 foundation with each defined in a standard way using a structured content model. All entities also have their own URI, with a web-service API enabling direct links to other electronic infrastructure and health information systems [13]. These new capabilities go beyond the usual advancements in medical and scientific knowledge associated with an ICD revision and will make the classification a more integrated part of future health information infrastructure.

ICD-11 also enables capture of additional information about diseases and conditions using extension codes. In the mortality use case, extension codes may capture additional information on non-proprietary names of drugs for drug related deaths, or risk factors for external cause deaths. The possibilities associated with extension codes are extensive and may only be limited by the information available when coding and compiling data. Concepts such as post-coordination and clustering of codes have been proposed to provide structure to more complex ICD-11 datasets, although methods for structuring and using groups of codes will need further consideration.

#### *3.3. Automated Coding Solutions*

Auto-coding systems are critical for processing the large number of deaths that occur in Australia, allowing codes to be assigned to individual entities and automated rules to be applied for the selection of underlying causes of death. Australia uses the Iris mortality auto-coding system, with this product enabling auto-coding of around 65% of doctor certified deaths each year. A project is now underway to develop an ICD-11 version of Iris. This project seeks to realise the benefits of the extended vocabulary and concepts of ICD-11, to integrate with ICD-11 tools, to interface with healthcare systems and eMCCDs and use advanced techniques such as machine leaning to detect certification errors and increase auto-coding rates [14].

The new electronic components of the CRVS system hold the key to transforming information from death certificates into usable epidemiological data in a way that is rapid, automated, reliable and accurate, and consistent across institutions and countries [14], and will ensure mortality data can meet future information demands including those highlighted by COVID-19.
