Hospital Standardised Mortality Ratio 2015

The Ministry of Health has recently released figures that show Gisborne Hospital has one of the highest Hospital Standardised Mortality Ratio (HSMR) in the country. HSMRs have been introduced as a tool for quality improvement in hospitals. The measure compares the observed number of deaths in a hospital with the number predicted - adjusting for a range of known risk factors.

The patients that Gisborne Hospital sees are very different compared to other hospitals in the country. While the HSMR takes factors like age, gender and clinical conditions into account, it is telling that Gisborne, Whakatane and Northland all have higher HSMRs than other hospitals.

We are all demographically, geographically and socio-economically with similar. We are remote, relatively deprived, with high Māori populations.

It is apparent that there are factors that affect Tairawhiti that the HSMR doesn’t take into account:

  • co-morbidity’s and how acutely unwell our patients are
  • the fact that we supply hospice services (terminally ill patients are not always recorded as such) a
  • we admit people in the ED who stay longer than 3 hours.

We are continually looking at the care we are providing, the outcomes achieved and how we can improve on them. We do this case by case and as a whole for the DHB. We review each death that has occurred in the hospital in the previous month and have set up a specific committee to lead this process.

The purpose of this is to find instances where we can improve and implement those changes where we can.

In 2013 there were 9 people who died in the palliative care beds. In other DHBs they would have probably died in a hospice facility or the community. In 2013 there were 11 people admitted who also died in ED. In other DHBs those deaths would have been excluded (because they don’t admit in ED and the deaths would then have been excluded).

These two areas therefore account for 20 of the 35 “excess” deaths in the Ministry of Health modelling for the 2013 year. It is likely the remaining excess relates to our morbid population or other factors we have responded to such as very late in life transfers to the hospital of people from residential care.

Whatever the actual cause for the variation it exists and the important work is to keep looking at all the deaths for any individual or modifiable factors, keep up the quality and safety culture of the organisation, promote clinical leadership, utilise quality and safety markers to keep up our safety record and utilise quality improvement tools like global trigger tools which identify areas for improvement.

We also need to look at the trend of the rate rather than the absolute rate. We want a trend down as we address the modifiable factors.

You can be confident that we provide safe care for Tairāwhiti people. We have comprehensive systems for reporting and investigating incidents in care. Our performance on other indicators of quality in care - for example infections in hospital or falls, do not show any issues of concern.