Abstract
Objective: To evaluate the annual change in HbA1c values among patients with type 2 diabetes
in primary care practices comparing different insulin regimens.
Research Design and Methods: Longitudinal data from 666 nationwide general and internal medicine practices in
Germany (Disease Analyser, IMS HEALTH) from 7/2004 to 6/2006 were analysed, including
348 patients (mean age±SD: 61±11 years) with continuous short-acting (regular insulin
or analogues), 1 906 with biphasic (72±9 years), 439 with basal-bolus (68±10 years),
and 1 719 with basal insulin therapy (65±10 years). The mean of the individual relative
changes in HbA1c (level in 2006 divided by level in 2005) were compared between insulin
groups, adjusting for age, sex, BMI, diabetes duration, oral antidiabetics, comorbidity,
health insurance, visits, hospitalisations, and practice type using general linear
models.
Results: There was only a small difference in baseline HbA1c values (range 7.3–7.5%) between
the four insulin groups (p=0.008). Substantial group differences were observed for
age, diabetes duration, and additional prescriptions of oral antidiabetics (p<0.0001).
After adjusting for potential confounders, the relative annual increase in HbA1c was
highest for biphasic insulin (1.021; 95%CI 1.016–1.025), followed by basal-bolus therapy
(1.017; 1.012–1.022), and basal insulin (monotherapy) (1.012; 1.002–1.021). No significant
change in HbA1c was found for short-acting insulin (1.006; 0.996–1.016).
Conclusions: Although many options for insulin therapy are now available, a progression of glycemia
still occurs in the majority of insulin-treated type 2 diabetic patients in primary
care.
Key words
type 2 diabetes - insulin treatment - general practice - HbA1c - glycemic control
References
Correspondence
rathmann@ddz.uni-duesseldorf.de
