Tools & Standards * CoDe * About
The CoDe (”Coding of Death in HIV”) Project

A significant proportion of deaths in HIV-1 infected persons are now caused by non-AIDS events. It is important to closely monitor the causes of death in this population in order to target interventions appropriately, should specific causes of death emerge or become predominant. It is possible that deaths from diseases related to an accelerated aging process will become more frequent. The same applies for causes of death related to co-infections (e.g. hepatitis) or other co-morbidities (e.g. sequelae of intravenous drug use). Furthermore, it is important to be able to evaluate the risk factors for such emerging diseases, including their possible relationship with immunodeficiency.

Until now there has not been a uniform classification system for causes of death in HIV patients. Studies have either created their own coding systems based o­n frequent and/or ‘important’ causes (e.g. rare but important adverse events such as lactic acidosis and pancreatitis), or have used ICD9 or ICD10 codes from death certificates. In many cases, the ICD system cannot be directly adapted to HIV infected persons. Many AIDS defining illnesses are poorly identified in the ICD system, and some diseases (e.g. CNS diseases) have a different aetiology in HIV patients and are therefore not covered by the ICD system, or at great risk of mis-classification.

In July 2004, a meeting was held in Copenhagen with the participation from executive committees of a large number of pivotal observational studies and clinical trials that routinely collect data o­n causes of death. At this meeting, it became clear that there was a need for a harmonization and standardization of the approach taken when collecting data o­n cause of death and when reviewing these deaths. As a result, the CoDe Project was initiated.

The CoDe Project is a uniform coding system that can be applied to studies of individuals with HIV infection, including a detailed data collection o­n the causes of death and contributing factors, as well as a centralised review process of the data collected.

To view a diagram illustrating the internal CoDe process, please follow this link.