The Influence of Insulin Use on Glycemic Control How well do adults follow prescriptions for insulin? JOYCE A. CRAMER 1 MARY JO PUGH, PHD 2,3 OBJECTIVE — The purposes of this study were to determine the relationship between insu- lin self-management and glycemic control and to identify patient characteristics associated with better control. RESEARCH DESIGN AND METHODS — A Department of Veterans Affairs regional database was used to identify patients with diabetes on chronic insulin therapy (n = 6,222) with dose defined as number of units and doses. The rate of insulin use during a 2-year period was calculated using pharmacy data. Regression analyses were used 1) to predict compliance with prescribed insulin regimens using demographic variables, HbA 1c levels, and a measure of dia- betes management intensity and 2) to predict HbA 1c levels using demographic variables and rates of insulin use. RESULTS — Insulin use was 77.44  17% of prescribed amounts, including wastage; HbA 1c levels were 7.98  1.66%. Concomitant oral hypoglycemic agent use (84.89  16%) was higher than insulin use (P  0.0001) but correlated with insulin use (r = 0.189, P  0.0001). Ordinary least-squares regression showed that race, HbA 1c levels, and intensity of diabetes management were significant predictors of insulin use. Age, race, and insulin use were significant predictors of HbA 1c levels. CONCLUSIONS — Adults prescribed a specific insulin regimen averaged using 77% of prescribed doses, demonstrating good intention to follow the prescription. However, HbA 1c higher than the recommended level suggested that the rate of insulin use, the prescribed regi- men, or both were inadequate to achieve good glycemic control in patients with long-term insulin use. Diabetes Care 28:78 – 83, 2005 A lthough the initial treatment in most individuals with type 2 diabe- tes is an oral hypoglycemic agent (OHA), the progression to insulin is com- mon (28 –39% incidence among older men) (1). Insulin may be added to or sub- stitute for an OHA to achieve adequate glycemic control. Regimens range from one to four daily doses (or more), depend- ing on whether an individual can manage with conventional dosing or requires in- tensive therapy. Despite extensive pre- scribing of insulin, many patients fail to achieve goals for glycemic control based on HbA 1c levels (2). Numerous barriers to use of insulin have been described (e.g., fear of injec- tions and hypoglycemic events, burden of injections, timing in relation to meals, etc.) (3,4). OHA dosing has been studied by observation with electronic monitors and by prescription refill records, show- ing that patients take 67– 85% of OHAs as prescribed (5). Assessing whether pa- tients follow prescribed insulin regimens is more complicated than assessment of OHA because of the inability to monitor injections and the units needed for each dose and the wastage when filling sy- ringes (6). The effectiveness of insulin treatment may be assessed using an ad- ministrative database of prescription records to define the amount of insulin dispensed to patients and laboratory data listing HbA 1c levels. These two objective measures can be used to evaluate how well physicians are managing patient’s glycemic control. Morris et al. (7) used prescription records in Tayside, Scotland, to determine how much insulin was ob- tained for use by children and adolescents as a surrogate for insulin self-manage- ment. The age of this cohort indicates a preponderance of patients with type 1 di- abetes, most of whom probably were re- sponsible for their own injections. They demonstrated a relationship between the amount of insulin obtained and HbA 1c levels, adverse events, and hospitaliza- tions. A study of adult type 2 diabetes pa- tients used a health insurance database in the U.S. to determine that patients ob- tained 63  24% of insulin refills (8). We proposed to extend these assess- ments by reviewing the use of insulin by adults followed at Department of Veter- ans Affairs Medical Centers. We hypoth- esized that patients who took insulin regularly, as prescribed, would have bet- ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● From the 1 Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut; the 2 Center for Health Quality, Outcomes, and Economic Research, VA Medical Center, Bedford, Massachusetts; and the 3 Boston University School of Public Health, Boston, Massachusetts. Address correspondence and reprint requests to Joyce A. Cramer, Yale University School of Medicine, 950 Campbell Ave. (Room 7-127, G7E), West Haven, CT 06516-2770. E-mail: joyce.cramer@yale.edu. Received for publication 28 July 2004 and accepted in revised form 13 September 2004. J.A.C. is a member of an advisory board for, serves as a consultant to, and has received honoraria from Novo Nordisk. M.J.P. is currently affiliated with the Audie L. Murphy VA Hospital, San Antonio, Texas, and was with the Edith Nourse Rogers VA Hospital in Bedford, Massachusetts, during this project. The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. Abbreviations: OHA, oral hypoglycemic agent. A table elsewhere in this issue shows conventional and Syste `me International (SI) units and conversion factors for many substances. © 2005 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Epidemiology/Health Services/Psychosocial Research O R I G I N A L A R T I C L E 78 DIABETES CARE, VOLUME 28, NUMBER 1, JANUARY 2005