Understanding the differences in in-hospital mortality between Scotland and England

Aims: We describe differences in in-hospital mortality between Scotland and England and test whether these differences are robust to controlling for the case-mix of patients. In spite of Scotland and England having much in common in regard to their hospital systems and populations we observe trends in-hospital mortality – the percentage of elective and emergency Continuous Inpatient Spells (CIS) that ended in death – that are different: England’s in-hospital mortality rates have decreased faster than Scotland’s for both types of admissions. Data: Individual patient data from England (HES) and Scotland (SMR01) for the period 2003/04 – 2011/12. Episode data is linked into CIS. Sample: Elective and emergency admissions, including day cases and excluding maternity. Methods: Logit regression of in-hospital death on country and financial year dummies, and their interaction, controlling for age group, gender, deprivation decile, and HRG of the first episode; separately for elective and emergency admissions. Results: For elective admissions, England has a lower initial in-hospital mortality rate than Scotland, and this rate decreases in both countries but the decrease has been faster in England. For emergency admissions, England starts with a slightly higher in-hospital mortality rate and both countries in-hospital mortality rates reduce throughout the period but England’s does so faster. Conclusions: There are differences in in-hospital mortality between Scotland and England; these differences increase over time and persist when we account for patient characteristics. It is important to understand the causes and consequences of these differences and we make a number of suggestions for future research on this issue.