Hospitalisations and deaths data are coded using different versions of the International Statistical Classification of Diseases and Related Health Problems (ICD), namely ICD-10-AM and ICD-10. There is limited availability of substance specific codes within these classification systems which often restricts the reporting of other drug-related hospitalisations and deaths data to the drug class or sub-class level.

An indicator on alcohol hospitalisations can be measured by:

Alcohol-related hospitalisations: counting those where alcohol (use, addiction or diseases) was the direct reason for the hospital admission (principal diagnosis) or related to the treatment while in hospital (other diagnosis). This indicator counts whole admissions for patients who have alcohol use problems or alcohol-related diseases.

Alcohol attributable hospitalisations: counting a proportion of those where, according to evidence from the peer-reviewed literature, alcohol directly or indirectly contributed to the morbidity burden (diseases, disorders or conditions). This indicator counts whole (alcohol use disorders) or partial admissions where alcohol played a role in causing the condition or disease, even if the patient themselves did not have alcohol use problems or alcohol-related diseases. Many alcohol-related diseases are not 100% attributable to alcohol.

An indicator on alcohol deaths can be measured by:

Alcohol-induced and alcohol-related deaths: counting those deaths where alcohol (addiction or alcohol-related diseases) was the direct reason for the underlying cause of death or related to the death of the person as an associated cause of death. This indicator counts whole deaths for people who died of alcohol use problems or alcohol-related diseases.

Alcohol attributable deaths: counting a proportion of deaths where, according to evidence from the peer-reviewed literature, alcohol directly or indirectly contributed to the fatality burden (diseases, disorders or conditions). This indicator counts whole (alcohol use disorders) or partial deaths where alcohol played a role in causing the condition or disease, even if the person who died themselves did not have alcohol use problems or alcohol-related diseases (such as a car passenger killed in a crash involving a drunk driver). Many alcohol-related diseases are not 100% attributable to alcohol.

Other drug-related emergency department presentations are identified using a combination of diagnosis codes and key word searches applied to free text fields.

The non-survey data comes from routinely collected administrative data. Routinely collected data are useful for examining harms at the state level and trends over time, however, there may be challenges identifying specific groups at higher risk. For example, detailed information is not routinely collected on gender identity or sexual preference in administrative hospital data, which means that information cannot be reported for lesbian, gay, bisexual, transgender, intersex and queer (LGBTIQ+) communities.

Survey data is based on self-reported alcohol and other drug consumption in NSW. The data are collected predominantly through Computer Assisted Telephone Interviewing as part of the NSW Population Health Survey. In 2024, a trial of a push-to-web method was undertaken using a posted invitation to households to complete an online survey. Data from that trial have been included in 2024 reporting. The survey data are adjusted to the actual NSW population so that the statistics provide an estimate of the percentage of people who drank alcohol and used specified substances for non-medical reasons in NSW.

For specific considerations for each topic area, refer to data notes in each individual indicator listed below.

Alcohol and other drugs

Data for NSW