Injuries occurring among admitted patients

Data source

Hospitalisations data are sourced from the Australian Institute of Health and Welfare’s (AIHW) National Hospital Morbidity Database (NHMD). This database includes data on virtually all hospital admissions in Australia, in both public and private hospitals. Data are collected for each episode of hospital care (called a hospital separation), which starts when a patient is admitted to hospital and ends when the hospital stay ends, or there is change in the type of care (for example, from acute care to rehabilitation hospital care).

Most diseases or conditions are recorded as a diagnosis following these criteria:

• Diseases are recorded as principal diagnosis when they are considered to be the primary reason for the patient being hospitalised.
• Diseases that coexist with principal diagnosis, or arise during the episode of care, are recorded as additional diagnoses when they affect the management of patients in terms of requiring therapeutic treatment, diagnostic procedures, or increased nursing care and/or monitoring.

For more information about data contained in the NHMD refer to the AIHW Hospitals technical notes and glossary.

Identification of hospital-onset injuries

Injury case identification

Hospital-onset injury refers to injuries that occur after a patient has already been admitted to hospital, rather than injuries that were present on admission.

Hospital-onset injuries are defined as:

• an additional diagnosis of injury defined as ICD-10-AM codes S00-T98 using ‘Chapter 19 Injury, poisoning and certain other consequences of external causes’, and
• a condition flagged as onset during the episode of admitted patient care, and
• meeting the denominator criteria

Injuries due to Complications of surgical and medical care (T80–T88) and Sequelae of injuries, of poisoning and of other consequences of external causes (T90–T98) are not described in detail as these classifications are reported elsewhere (ACSQHC 2026, AIHW 2025a, AIHW 2025b).

The condition onset flag (COF) indicates the onset of the condition relative to the beginning of the admitted patient episode of care. The same COF should be allocated to the external cause, place of occurrence and activity codes to match that of the corresponding injury or disease code. See METEOR 796636 for more information.

Analysis cohorts

All hospital separations in the NHMD from 2015–16 to 2024–25, excluding separations with any of the following:

• Care type is ‘Newborn with only unqualified days’ (Care type=7.3)
• Care type is ‘Hospital boarder’ (Care type=10)
• Care type is ‘Organ procurement – posthumous’ (Care type=9)
• Same-day chemotherapy – DRG: R63Z and admission date=separation date
• Same-day haemodialysis – DRG: L61Z and admission date=separation date

Sequencing of injury diagnosis and external cause codes

The external cause code should be sequenced directly after an injury, poisoning or other adverse event diagnosis code, followed by place of occurrence (for external cause codes V00-Y84) and activity codes (for V00-Y34) (see METEOR 746659)

Multiple cause codes may be assigned if they are required to classify the clinical concept, and multiple diagnoses may arise from the same cause. Therefore, relationships between codes can be inferred incorrectly as there is no diagnosis cluster indicator up to ICD-10-AM twelfth addition.

Injury diagnosis codes were mapped to their corresponding external cause code as follows:

1. An external cause code immediately following an injury code in sequence was linked.
2. For multiple consecutive injury diagnosis codes, the external cause code following the last diagnosis in the sequence was linked to the preceding injury codes.
3. If there were multiple external cause codes for an injury (or series of injuries), the first reported external cause was taken. This is usually the one that most closely describes the cause or intent related to the diagnosis.

A limitation of this method is that injury diagnosis codes appearing within a sequence containing both injury and non-injury codes may not be associated with an external cause, unless the injury code is the final entry in the sequence.

With ICD-10-AM thirteenth edition a diagnosis cluster identifier has been introduced (see METEOR 799088) which will improve the quality of data available to report on additional diagnoses in future.

Methods

Rates and proportions

As this report focused on admitted patients, the analysis did not use Estimated Resident Population (ERP) data (which are often used in other reporting) to derive rates. Rather, this report used more specific denominators that better reflected the cohorts of interest.

One episode of admitted care could include multiple injury diagnoses, therefore the total number of diagnoses will be greater than the number of hospital separations.

Socioeconomic status

Comparisons between areas that are more or less socioeconomically disadvantaged are based on the Socioeconomic Indexes for Areas (SEIFA) which uses the Index of Relative Socioeconomic Disadvantage (IRSD).

The IRSD classifies individuals according to the socioeconomic characteristics of the area in which they live. It scores each area by summarising attributes of the population, such as income, educational attainment, unemployment rate and jobs in relatively unskilled occupations. The index does not show how individuals living in the same area differ from each other in their socioeconomic group.

Refer to the ABS Socioeconomic indexes for areas (SEIFA): Technical paper 2021 for more information.

Remoteness

Comparisons between remoteness areas use the Australian Statistical Geography Standard (ASGS) Remoteness Areas structure, which is based on area of residence. Australia is divided into 5 classes of remoteness based on a measure of relative access to services. The 5 remoteness areas are Major cities, Inner regional, Outer regional, Remote and Very remote. Refer to the ABS Remoteness structure for more information.

Data suppression and confidentiality

Data (cells) in tables may be suppressed to maintain the privacy or confidentiality of a person or organisation; or because a proportion, rate (numerator or denominator) or other measure is related to a small number of events (and may therefore not be reliable). Data may also be suppressed to avoid attribute disclosure. The abbreviation ‘n.p.’ (not published) has been used in tables to denote these suppressions. In these tables, the suppressed information is included in the totals.