Hospital Standardised Mortality Ratio (HSMR)

Identifying Attributes

Care Settings
Hospital Care
Country
Australia
Publishing Organisation
Australian Commission on Safety and Quality in Health Care (ACSQHC): Core Hospital-Based Outcome Indicators
Type of Quality Indicator
Outcome
IOM Quality Dimension
Safety
Domain
Mortality

Defining Attributes

Definition

The ratio of the observed number of hospital separations that end in the patient's death, to the number of separations expected to end in death based on the patient's characteristics, for principal diagnoses accounting for 80% of in-hospital mortality.

Numerator

Observed number of in-hospital deaths x 100

Denominator

Expected number of in-hospital deaths (the sum of the estimated probabilities of death for all separations meeting the denominator criteria, calculated using national risk adjustment coefficients).

Exclusions

Missing: admission mode, sex.

Use of Risk Adjustment
Yes
Risk Adjustments

Age at admission (years) • Sex • Principal diagnosis code (mapped to national in-hospital mortality risk deciles) • Admission urgency status: emergency, elective • Length of stay (including leave days) categorised as 1 day, 2 days, 3-9 days, 10-15 days, 16-21 days and 22-365 days • Additional (comorbid) diagnoses (Charlson index) categorised into 0 – Charlson Index score of 0; 1 – Charlson index score of 1; 2 Charlson index score ≥2 • Admission mode (inward transfer status) = admitted patient transferred from another hospital.

Stratifications

By hospital peer group.

Collection and Reporting Attributes

Type of Data Collection
Administrative data
Data Collection Methods

Core Hospital-Based Outcome Indicators

Frequency of Data Collection
Frequency of Data Collection in Days
Reporting Methods

Internal state/territory reports

Reporting Frequency
Reporting Frequency in Days
Indicator Has Recommended Targets
No

Source and Reference Attributes

Link to Measurement Tools
Quality Indicator Confirmed to be Part of a Program Used to Monitor Quality and Safety of Care Among Older People at a Population-Level between 2012-2022
Yes
Assessed by the Australian Consortium for Aged Care Collaborators as Generally Containing Good Properties (Importance and Scientific Acceptability)
No
Australian Consortium for Aged Care Endorsed
No
Identified by PHARMA-Care Project
No
Upload Date
02 December 2025