Mealtime Insulin Administration: Pens Versus Vials
Published Online: December 13, 2013
Elizabeth L. Eby, MPH; Kristina S. Boye, RPh, MS, MPH, PhD; and Maureen J. Lage, PhD
Type 2 diabetes mellitus (T2DM) is estimated to affect between 23.2 million and 24.5 million Americans, or between 7.5% and 7.9% of the US population of 309 million individuals,1,2 with these numbers being predicted to double over the next 25 years.3 In addition to being an increasingly prevalent disease, T2DM is also costly: the US economic burden of T2DM is estimated at $245 billion annually, with direct costs of medical treatment totaling $176 billion.4
While treatment guidelines emphasize the importance of prescribing the right type of insulin at the appropriate time,5 recent studies have indicated that the method used to supply insulin to the body may also affect patient outcomes.6-13 The conventional way that exogenous insulin has been supplied to the bloodstream is via subcutaneous injection using a syringe and vial.14 However, patient aversion to self-injection using the conventional syringe method has been shown to have a negative impact on adherence to insulin therapy.15,16 To make insulin therapy less complicated and fear inducing, the insulin pen, which combines the insulin reservoir andsyringe into 1 unit, was introduced in 1985.8,17,18
Although the results of previous research suggest that patients prefer pens to vials,19 a recently published study of insulin delivery trends reported that the percentage of patients using a pen is still low in the United States, and that approximately half of T2DM patients still initiate insulin therapy with a conventional vial.20 Our present analysis examined the impact of mealtime insulin analogue initiation with a vial or prefi lled pen on patient adherence, compliance, and hypoglycemic events among commercially insured T2DM patients. In addition, for a subset of patients with valid glycated hemoglobin (A1C) test results, the present study also examined changes in A1C values between the pen and vial cohorts.
Data for this study were obtained from the Innovus InVision Data Mart database. This retrospective claims database captures person-specific information on patient characteristics, enrollment dates, inpatient and outpatient medical claims, and outpatient pharmaceutical claims from a geographically diverse distribution across the United States. In addition, laboratory test results are available for a subset of these patients. The database is fully compliant with the Health Insurance Portability and Accountability Act. This study examined asubset of individuals who received at least 1 recorded diagnosis of T2DM from January 1, 2006, through June 30, 2010.
For inclusion, patients must have received at least 1 outpatient prescription for mealtime insulin analogue, including rapid-acting insulin analogues and premixed insulin analogues. The date of the fi rst such prescription was identified as the index date. Note that while treatment guidelines vary regarding the type of insulin a patient with T2DM should initiate if they fail to achieve glycemic control with oral antidiabetic medication therapy,21,22 this analysis focuses on initiation of mealtime insulin independent of use of prior insulin therapies. Individuals were also required to have received at least 2 diagnoses of T2DM during the 6 months prior to the index date through the 12 months following the index date. Patients were excluded if they used reusable pens with cartridge insulin, 3 mL vials, inhaled insulin, or an insulin pump at any time during the study period. Patients were also excluded if they received a diagnosis of type 1 diabetes, secondary diabetes, or gestational diabetes during the study period, or if they switched method of insulin delivery in the post period. Finally, patients were required to have continuous insurance coverage throughout the study period and to have been at least 18 years old on the index date. Table 1 illustrates how each of these criteria impacted sample size.
The study considered 4 patient outcomes: adherence, persistence, hypoglycemic events, and glycemic control.
Adherence was proxied by the adjusted medication possession ratio (MPR). Typically, the MPR is calculated by dividing the number of days a medication was dispensed by the number of days in the time period of interest (in this case, the 1-year post period). However, package sizes are different for insulin pens and vials. Despite this fact, the majority of insulin prescriptions are coded as being for 30-days’ supply. To account for this, previous research6 has adjusted the MPR by multiplying the traditional MPR by the ratio of average number of days between refills divided by the average days’ supply. For example, an individual who requires less insulin may utilize their “30-day” supply of insulin over a 45-day period. Using the adjusted MPR, this individual would be considered to be fully adherent as long as they continued to refill their prescription every 45 days. Persistence was defined as the duration of time from initiation to discontinuation. Hypoglycemic events were calculated using a previously validated algorithm.23 This algorithm utilized International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes and identified patients with hypoglycemia if they received a specific diagnosis for hypoglycemia or were diagnosed with diabetes with other specified manifestations (ICD-9-CM of 250.8) in the absence of other codiagnoses.23 Finally, glycemic control was measured as change in A1C levels for the subset of patients who received at least 1 valid A1C test in the preperiod and another at least 90 days after the index date. For patients with multiple A1C values in the pre-period, the one closest to the index date was used, while the latest A1C value was used in the post period.
Multivariate analyses were used to examine the outcomes of interest. When the dependent variable was persistence, adjusted MPR, or change in A1C values, ordinary least squares regressions were utilized. Logistic regressions were estimated to examine the likelihood of achieving an adjusted MPR of at least 0.80, which is a level identified as “adherent” in previous studies examining pen versus vial use,7,10 and the likelihood of having a hypoglycemic event.
PDF is available on the last page.