The impact of insulin type on severe hypoglycaemia events requiring inpatient and emergency department care in patients with type 2 diabetes

• Published in: Diabetes research and clinical practice 2013-12, Vol.102 (3), p.175-182
• Main Authors: Solomon, Matthew D, Vijan, Sandeep, Forma, Felicia M, Conrad, Ryan M, Summers, Nicholas T, Lakdawalla, Darius N
• Format: Article
• Language: English
• Subjects: Adult; Aged; Cohort Studies; Diabetes Complications - chemically induced; Diabetes Mellitus, Type 2 - complications; Diabetes Mellitus, Type 2 - drug therapy; Emergency Service, Hospital - statistics & numerical data; Endocrinology & Metabolism; Female; Humans; Hypoglycaemia; Hypoglycemia - chemically induced; Hypoglycemic Agents - therapeutic use; Inpatients - statistics & numerical data; Insulin; Insulin - classification; Insulin - therapeutic use; Male; Middle Aged; Prognosis; Type 2 diabetes; Young Adult
• ISSN: 0168-8227; 1872-8227
• DOI: 10.1016/j.diabres.2013.09.013

Abstract Aims To evaluate the risk from different insulin types on severe hypoglycaemia (SHG) events requiring inpatient (IP) or emergency department (ED) care in patients with type 2 diabetes. Methods Type 2 diabetes patients newly started on insulin in a large commercial claims database were evaluated for SHG events. Patients were classified into an insulin group based on their most frequently used insulin type. Multivariable Cox models assessed the association between insulin type and the risk of SHG events. Results We identified 8626 patients (mean age 53.5 years; 55% female) with type 2 diabetes followed for an average of 4.0 years after insulin initiation. Of these, 161 (1.9%) had a SHG event at an average of 3.1y after insulin initiation. Patients with SHG events were slightly older (56.4 vs. 53.4 years), used a similar number of OADs (1.1 vs. 1.2) but had more co-morbidities compared with those without SHG events. In multivariate Cox models, premixed insulin (HR 2.12; p < 0.01), isophane insulin (NPH) (HR 2.02; p < 0.01), and rapid acting insulin (HR 2.75; p < 0.01) had significantly higher risks of SHG events compared with glargine. No statistically significant difference in SHG events was seen with detemir (HR 1.20; p = 0.73). Conclusions Among patients with type 2 diabetes, the use of newer basal insulin analogues was associated with lower rates of SHG events requiring IP or ED care compared with users of other insulin formulations. Future research should examine the impact of hypoglycaemia events of different severity levels.