A cost comparison of long-acting insulin analogs vs NPH insulin-based treatment in patients with type 2 diabetes using routinely collected primary care data from the UK. (April 2015)

• Record Type: Journal Article
• Authors: Idris, Iskandar; Gordon, Jason; Tilling, Carl; Vora, Jiten
• Is Part Of: Journal of medical economics. Volume 18:Number 4(2015)
• Page Start: 273
• Page End: 282
• Publication Date: 2015-04
• Subjects: Medical care -- Cost control -- Periodicals; Medical economics -- Periodicals; 362.10941
• DOI: 10.3111/13696998.2014.991788
• Languages: English
• ISSNs: 1369-6998

Abstract

<abstract> <title>Abstract</title> <sec id="ss1"> <title>Aim:</title> <p>The aim of this analysis was to investigate total healthcare costs, HbA1c, and weight changes over a 36-month period in patients with type 2 diabetes initiated on NPH or long-acting insulin analogs.</p> </sec> <sec id="ss2"> <title>Methods:</title> <p>Electronic patient data from 479 general practices in the UK (THIN database) were examined for new users of glargine (<italic>n</italic> = 794), detemir (<italic>n</italic> = 252), or NPH insulin (<italic>n</italic> = 430). Annualized healthcare costs and clinical outcomes in years 1, 2, and 3 following insulin initiation were quantified and compared with baseline, using ANOVA and linear regression models.</p> </sec> <sec id="ss3"> <title>Results:</title> <p>A significant difference (<italic>p</italic> < 0.05) in total healthcare costs increases at year 1 vs baseline was observed between glargine and detemir, detemir and NPH, but not between glargine and NPH (increase: +£486, +£635, and +£420 for glargine, detemir, and NPH users, respectively). However, increases by year 3 were not significantly different between the insulins. A propensity score analysis comparing analog and NPH insulin showed that, following insulin initiation, increases in costs were higher with insulin analogs at year one (+£220), but this difference decreased over time in each year following insulin initiation (+£168 and +£146, respectively, for years 2 and 3). HbA1c reductions were not significantly different between the groups at all time points. Differences in weight gain between glargine and NPH were statistically significant at year 1 (0.87 kg vs 1.11 kg) and year 3 (1.15 kg vs 1.57 kg), but other estimates of between-group differences in weight gain were non-significant.</p> </sec> <sec id="ss4"> <title>Conclusions:</title> <p>Following insulin initiation, the difference in healthcare costs of long-acting analogs compared to NPH insulin was transient. By year 3, the cost differences were not significantly different between the two cohorts, driven by an observed reduction in the cost of self-monitoring of blood glucose (SMBG) in the analog group and an increase in the cost of bolus insulin in the NPH group.</p> </sec> </abstract>