In-hospital mortality patterns have been used internationally as one indicator of the quality of care. While there are a number of similar ways of doing this, the standardised mortality ratio (SMR) is the most commonly used approach for looking at hospital mortality patterns within a country.

The SMR is the ratio between the observed number of patients who die in hospital and the number that would be expected to die in hospital on the basis of the overall national rate. It is based on the primary reason a patient is admitted to hospital. Importantly this does not infer the cause of death.

The information comes from the Hospital In-Patient Enquiry (HIPE) system, which contains clinical and administrative data on patients who have been admitted to hospital and is routinely collected by all publicly funded acute hospitals. Personal information i.e. information which could be used to identify a patient such as name, address and date of birth, is not taken from the hospital system and is not used in this audit. Patient confidentiality and privacy is fully protected in this manner.
To ensure that “like is compared with like” across the diversity of hospitals, factors that potentially may directly influence the outcome are adjusted for in the analysis, for example, patient age and the presence of other serious illnesses to calculate the number of expected deaths.