Certain Behavioral, Social Factors Associated With DKA Hospitalizations in T1D

Researchers of a new study have identified certain social and behavioral factors associated with repeat hospital admissions for diabetes ketoacidosis (DKA) among patients with type 1 diabetes (T1D).

Over the 2-year study period (2019-2011), 26% of 243 patients admitted to a single center for DKA were admitted at least twice for the complication. Having multiple readmissions was more likely to occur among patients with substance use disorder or who previously were incarcerated.

The findings published in Journal of the Endocrine Society, wrote the researchers, help identify specific patient groups for whom interventions can be implemented in order to lower the recurrence of DKA episodes. Interventions, they underscored, should include perspectives from patients about what interventions could be most effective.

Hospital admission desk

Image credit: Maks_Lab-stock.adobe.com

Rates of substance use disorder were nearly twice as high (60.9% vs 33.0%; P < .001) among patients with multiple admissions than among patients admitted just once over the study period. Patients with multiple admissions were also significantly more likely to have been incarcerated (26.6% vs 14.5%; P = .007). This association, said the researchers, offers a new insight to previous research exploring the effect of previous incarceration on DKA admissions.

“While there has been much research looking at the increased risk of DKA and hyperosmolar hyperglycemic syndrome in patients who are in prison, few studies have looked at these events in those following release,” described the researchers. “Randall et al, in their 2011 study of 164 minority patients with DKA at a similar large community hospital, did not find a difference between those with a single vs multiple DKA episodes. However, they similarly did see a significant increase in incarceration history as the number of admissions increased.”

The group highlighted a similar pattern among research on the effect of substance abuse, noting that although various studies have explored the link between substance abuse/disorder and DKA readmission, few have gone deeper into differences based on type of substance. In the current study, multivariate analysis shows that among substances, which included cannabis, tobacco, and psychoactive substances, only cannabis was independently associated with multiple admissions.

Across other psychosocial factors, the researchers found no significant differences between patients who were admitted for DKA once vs those who had multiple DKA admissions. These factors included laboratory values, length of hospital stay, coverage type, history of homelessness, employment status, living alone, independence of daily activities, and barriers to discharge.

From a clinical perspective, there were no differences in diagnostic diabetes and DKA lab values between the 2 patient groups, nor were there statistical differences in length of hospital stay.

“A lipid panel is typically not a standard laboratory test that is run on patients admitted for DKA. Likely for that reason, only 83 patients admitted for DKA had a lipid test drawn throughout their admission,” explained the researchers. “However, there was a significant increase in the level of HDL-C [high-density lipoprotein cholesterol] in those with readmissions for DKA. The association between T1D and HDL-C level has been explored in multiple studies; however, none have shown an association between HDL-C levels and incidence of DKA or recurrent DKA in this population. Our findings are unlikely to be related to selection bias among individuals who had lipid measurements taken, as there were no demographic differences between those with and without lipid measurements.”

Mean cholesterol levels, low-density lipoprotein cholesterol, and serum triglycerides were comparable between the two groups.

Reference

Peedikayil J, Reddy S, Nair R, et al. Social and metabolic characteristics associated with multiple DKA admissions at a large county hospital. J Endocr Soc. Published online January 19, 2024. doi:10.1210/jendso/bvad173