Rising rates and widening socioeconomic disparities in DKA in Scotland

This study evaluated trends in diabetic ketoacidosis (DKA) occurrence, recurrence and outcomes between 1 January 2004 and 31 December 2018 in Scotland. Using the national SCI-Diabetes database, a total of 37 939 individuals with type 1 diabetes were identified, in whom there were 30 427 hospital admissions for DKA (occurring in 10 397 individuals [27.4% of the cohort]) over the study period. Of the admissions, 1490 (4.9%) were at the time of diabetes diagnosis. The crude DKA event rate was 7858 per 100 000 person-years.

The standardised DKA event rate increased gradually over the study period, from 6182 to 8261 events per 100 000 person-years. This pattern was seen in both sexes and across all socioeconomic groups; however, the most deprived quintile saw both the highest event rates and the largest increase. Notably, in this quintile, the event rate (per 100 000 person-years) rose from around 12 000 to 18 500 between 2010 and 2014, although it has since fallen steadily to around 15 000 in 2018. There was a significant association between deprivation and DKA event rates, and this remained significant after adjustment for HbA_1c (which was higher in more deprived groups) and insulin pump use and structured education (which were less common in these groups).

Age band stratification suggested that the greatest elevation in event rates was in people aged ≥60 years, while those aged 10–19 years saw a fall in event rates. The increased rates in older people, coupled with the fact that DKA was more likely to result in death in this age group, is concerning given the rise in the number of older people with type 1 diabetes.

Women were more likely to experience DKA than men; this may have been associated with the higher HbA_1c in women that has previously been observed in Scotland, as well as a high prevalence of disordered eating in younger women. However, DKA was more likely to be fatal in men.

Higher HbA_1c and prescriptions for antidepressants and methadone (as proxies for depression and opioid addiction) were all associated with increased DKA risk and case fatality rates. In contrast, completion of a structured education course and use of an insulin pump were associated with reduced DKA event rates, although not reduced case fatality rates. These factors offer potential routes to improving DKA rates, particularly in those in the most deprived socioeconomic groups; however, the fact that deprivation was still associated with DKA after adjustment for these factors suggests that other causes might be at play.

Click here to read the study in full.