Compliance and Persistence With Concomitant Statin and Oral Antihyperglycemic Therapy

In patients with type 2 diabetes, compliance and persistence were generally low for both statin and antihyperglycemic therapy, but they were significantly lower with statin therapy.
Published Online: November 07, 2011
Qiaoyi Zhang, MD, PhD; Changgeng Zhao, MS; Michael J. Davies, PhD; Larry Radican, PhD; and Thomas Seck, MD

Objectives: To compare compliance and persistence with statin and oral antihyperglycemic therapies in patients with type 2 diabetes who received concomitant therapy.
 

Study Design: Retrospective cohort study using a large US commercial claims database.

 

Methods: Patients with type 2 diabetes and dispensed prescriptions for both statin and oral antihyperglycemic therapies on the same date in 2006 (index date = first date of such dispensing) were included in the analysis (N = 52,414). Patients were required to have continuous enrollment in the database for 1 year prior to (baseline) and 2 years after (follow-up) index date. The 2-year medication possession ratio (MPR) was compared between statin and oral antihyperglycemic therapy. For the persistence analysis, treatment discontinuation was defined by a gap >30 days between the last date of supply from previous dispensing and subsequent refill. The likelihood of discontinuation of statin versus oral antihyperglycemic therapy was estimated by fitting a robust Cox proportional hazards regression model, adjusted for baseline variables.

 

Results: The 2-year MPR was 70% for statin and 78% for oral antihyperglycemic therapy (P <.0001). The proportion of patients with a 2-year MPR >80% was 52% for statin and 63% for oral antihyperglycemic therapy (P <.0001). The median time to discontinuation of statin was significantly shorter compared with oral antihyperglycemic therapy (284 vs 495 days, P <.001). There was a greater risk to discontinue statin than oral antihyperglycemic therapy (adjusted hazard ratio: 1.47 [95% confidence interval 1.45-1.48]).
 

Conclusions: Compliance and persistence with statin therapy significantly lagged behind oral antihyperglycemic therapy in patients with type 2 diabetes who were treated concomitantly with both therapies.
 

(Am J Manag Care. 2011;17(11):746-752)

  • In patients with type 2 diabetes who received concomitant treatment with statin therapy and oral antihyperglycemic therapy, compliance and persistence were significantly lower with statin therapy compared with oral antihyperglycemic therapy.

 

  • Integrative approaches to improve compliance and persistence with these agents may reduce cardiovascular risks and improve clinical outcomes in this population.
Patients with type 2 diabetes are at an increased risk for microvascular and macrovascular complications and mortality.1-3 Statins are recommended for patients with type 2 diabetes, regardless of low-density lipoprotein (LDL)-cholesterol levels, to reduce their elevated cardiovascular risk.4,5 Improvement in glycemic control, including the use of antihyperglycemic therapies, reduces the risk of microvascular outcomes in patients with type 2 diabetes.6-8 Despite the clinical benefits of these agents for diabetes related vascular complications, many patients are not at recommended cholesterol and glycemic treatment targets and the use of statins and antihyperglycemic agents is suboptimal.9-12 Moreover, compliance (or adherence) and persistence are important factors for effective lipid and glycemic control.13,14 Importantly, nonadherence to statins and antihyperglycemic agents has been shown to be associated with an increased risk for hospitalization and mortality, as well as increased costs, in patients with type 2 diabetes.15-20

Adherence and persistence with statin and antihyperglycemic treatments ranges from 36% to 93%.15,21-25 Previous studies on estimates of compliance (or adherence) and persistence mainly focus on individual drug classes across patient cohorts, and estimates have not been published, to date, for concomitant use of statin and antihyperglycemic therapy. Given that statin therapy is recommended for patients with type 2 diabetes, it was of interest to assess the compliance (ie, act of conforming to the recommendations of the provider for the prescribed therapy26) and persistence (ie, act of continuing a treatment for a prescribed duration26) with statin and oral antihyperglycemic agents in patients with type 2 diabetes who received these therapies concomitantly. Therefore, using a large US claims database, the present study compared compliance and persistence with statin and oral antihyperglycemic therapies in patients with type 2 diabetes who received concomitant therapy.

METHODS

Data Source

Data for this analysis are from a large US commercial claims database (Thomson Reuters MarketScan Commercial Claims and Encounters Database and Medicare Supplemental and Coordination of Benefits Database). The commercial database contains de-identified, personspecific health data including clinical utilization, expenditures, insurance enrollment/plan benefit, inpatient, outpatient, and outpatient prescription information. The data cover more than 20 million individuals annually, and include private sector health data from approximately 100 payers. The data can be linked to track detailed patient information across sites and types of providers, and over time. The supplemental database provides similar data for retirees with Medicare supplemental insurance paid for by large employers.

Study Design and Patient Selection

In a retrospective cohort study, patients with type 2 diabetes and 18 years and older were eligible if they were dispensed prescriptions for both a statin and any oral antihyperglycemic agent on the same date in 2006. The same date was used in order to have a similar starting point for each patient. Patients

with type 2 diabetes were identified by The International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9- CM) diagnosis codes (250.x2). The first date in 2006 of such dispensing was considered the index date. Patients needed to have medical records for at least 1 year prior to (baseline) and 2 years after (follow-up) the index date. Patients with type 1 diabetes or secondary diabetes or pregnant women were excluded. Patients on only injectable antihyperglycemic agents were excluded from the analysis. Insulin was not included because comparisons between oral and injectable agents are difficult. Persistency measures may vary substantially with insulin compared with oral antihyperglycemic agents because patients may have severe adverse events if they stop their insulin. Furthermore, insulin prescription refill is variable because of constant dose adjustments to maintain glucose control. Exenatide, a glucagon-like peptide-1 agonist, was relatively new to the diabetes market in 2006 and thus would not have had a meaningful impact on the overall findings.

End Points

Statin and oral antihyperglycemic therapy utilization were assessed as separate entities for each patient. The follow-up period was censored at the end of 2 years. Compliance was estimated as the medication possession ratio (MPR) over 2 years (MPR = total days of supplies of medicine dispensed during the 2-year follow-up/730 days). The proportion of patients with an MPR of at least 80% was calculated.27,28 A sensitivity analysis was conducted for compliance using a 1-year followup period. To control for different therapy start dates, compliance was also assessed in a subgroup of patients who had their first prescriptions (ie, new to both treatments) for statin and oral antihyperglycemic therapy on the same date in 2006.

For the persistence analysis, treatment discontinuation was defined as a gap of at least 30 days between the last date of supply from previous dispensing and the subsequent refill. A switch in the type of statin or oral antihyperglycemic agent was not considered a nonpersistent event as long as there was not a gap of at least 30 days. Persistence was also assessed using a minimum of 60- and 90-day gaps to define treatment discontinuation because down-titration or dose-splitting may have increased the time between prescription refills.

Statistical Analyses

Overall MPR and the proportion of patients with an MPR of at least 80% were compared between statin and oral antihyperglycemic therapies. Discontinuation and time to discontinuation were estimated and compared between therapies. The likelihood of discontinuation of statin versus oral antihyperglycemic therapy was estimated by fitting a robust Cox proportional hazards regression model, adjusted for patient characteristics and comorbidities at baseline. ICD-9-CM and Current Procedural Terminology (CPT) codes were used to determine comorbid disease conditions.

RESULTS

In total, 52,414 patients with type 2 diabetes met the inclusion criteria. The mean age of the cohort was 62 years, 54% were men, 23% had preexisting cardiovascular conditions, and 14% had preexisting microvascular conditions (Table 1). Based on the study design, there were 4 treatment scenarios in terms of the timing of statin and oral antihyperglycemic therapies on the index date: 1) existing statin existing oral antihyperglycemic agent users (68.3% of entire cohort); 2) new statin new oral antihyperglycemic agent users (9.5%); 3) new statin existing oral antihyperglycemic agent users (15.7%); and 4) existing statin new oral antihyperglycemic agent users (6.5%). The patients in this new new users cohort were generally younger and had fewer preexisting comorbid conditions relative to the overall cohort (data not shown).

Overall compliance, assessed using MPR, was significantly lower (P <.0001) with statin therapy compared with oral antihyperglycemic therapy after 2 years of follow-up (Table 2). The proportion of patients with MPR of at least 80% after 2 years was significantly lower (P <.0001) with statin than with oral antihyperglycemic therapy (Table 2). In an analysis focused on compliance to concomitant therapy with a statin and oral antihyperglycemic therapy, the proportion of patients with an MPR of at least 80% for both therapies after 2 years was 43%. In sensitivity analyses, the between-treatment difference in MPR results after 1 year was generally consistent with the 2-year results (Table 2).

In an analysis focused on the cohort which initiated both statin and oral antihyperglycemic therapy for the first time on the same date in 2006, MPR values were generally lower compared with the entire cohort, and the difference between statin and oral antihyperglycemic therapy was significant (Table 2). Between-treatment differences were generally consistent after 1 and 2 years of follow-up in this cohort (Table 2).

The median time to discontinuation for statin therapy was significantly shorter (284 days [95% confidence interval (CI) 280-289] vs 495 days [484-505]; P<.0001) compared with oral antihyperglycemic therapy (Figure). A lower proportion of patients was persistent (ie, no treatment gaps >30 days) with statin therapy (28%) than with oral antihyperglycemic therapy (41%) over the 2-year period. In sensitivity analyses using no treatment gaps of a minimum of 60 or 90 days, the overall trend in persistency for both classes was numerically higher compared with the gap of at least 30 days, but the between-treatment difference remained significant (data not shown).

PDF is available on the last page.
Feature
Recommended Articles