[Standardised in-hospital mortality decreasing over time].

OBJECTIVE

To study the changes in standardised in-hospital mortality in the period 2005 to 2010.

DESIGN

Retrospective, descriptive study.

METHOD: Data from the National Medical Registration (LMR) for the period 2005 to 2010 were used to calculate a time-series of the standardised in-hospital mortality rate. Predicted mortality was calculated for 50 principle diagnosis groups using logistic regression models with 8 explanatory variables. This was used, in combination with actual in-hospital mortality, to calculate the Time-Series Standardised Hospital Mortality Ratio (TSHMR).

RESULTS

The TSHMR decreased significantly each year in the period 2005 to 2010. The greatest decrease was from 2009 to 2010, which was partly due to a registration effect. Total age- and sex-standardised mortality in the population also decreased in the period 2005 to 2010, mortality out of hospital decreasing less than in-hospital mortality. Standardised hospital mortality decreased significantly in all adult age categories (≥ 25 years of age). The TSHMR fell most sharply in the diagnosis groups 'gastrointestinal haemorrhage' and 'deficiencies and other anaemias'. The diagnosis groups 'heart valve disorders' and 'aspiration pneumonia due to food or vomitus' showed the smallest decrease in TSHMR, and this decrease was not significant.

CONCLUSION

The decrease in standardised in-hospital mortality shows that the risk of dying in hospital has decreased substantially in the past few years. The size of the decrease in TSHMR differed per diagnosis group. There can be many reasons for this decrease in in-hospital mortality, including improved quality of care; however, it may also be due to, for instance, changes in hospital admission and discharge policies.