Quarterly Hospital Standardised Mortality Ratios
The Scottish Patient Safety Programme (SPSP) aims to reduce hospital mortality by 15% by 2012. Hospital Standardised Mortality Ratios (HSMR) are used by hospitals participating in SPSP to monitor their progress against this target over time. From December 2009 Information Services Division (ISD) has produced quarterly HSMR for all Scottish hospitals participating in the SPSP.
What is HSMR?
HSMR is the ratio of the number of deaths in hospital within a given time period (observed deaths) to the number of deaths that might be predicted if the hospital had the same death rates as a reference population .
What is the reference population?
The reference population consists of all Scottish patients recorded on the Scottish Morbidity Record (SMR)01 in 2006/2007. This is used to assign the probability of death within 30 days from admission for each patient registered on SMR01 in subsequent years.
How is the probability of death calculated?
The predicted probability of death within 30 days from admission takes account of the following: patient's primary diagnosis; admission to a medical or surgical specialty; age; sex; where the patient was admitted from; whether admitted as an inpatient or day case and type of admission (elective / non-elective); the number and severity of prior conditions recorded in the previous (i) 12 months (ii) 5-years; the number of emergency admissions in the previous 12 months.
How timely are the data?
HSMR is published quarterly and the most recent publication in February 2012 reports on quarterly SMR01 data to September 2011.
What is does a high HSMR mean?
A high HSMR does not necessarily mean that poor quality care or unsafe services are being provided. It is not possible to distinguish between deaths which could potentially have been preventable and those which were not. Therefore, a high HSMR is regarded as a trigger for further, joint review of patient data by ISD and NHS Boards in collaboration with Healthcare Improvement Scotland.
How do we support NHS Boards in understanding their HSMR?
A review of HSMR data typically includes analysis of a patient population (for example HSMR data stratified by admission type and by diagnostic grouping). Case listings are made available to clinicians to facilitate a more in-depth review of individual patient records. In addition to better understanding the HSMR, the reviews may help to improve the quality of clinical coding on future records.
Further information on HSMR can be found at www.indicators.scot.nhs.uk
