Impact of Adherence and Weight Loss on Glycemic Control in Patients with Type 2 Diabetes: Cohort Analyses of Integrated Medical Record, Pharmacy Claims, and Patient-Reported Data

Abstract

BACKGROUND: Managed care organizations put great effort into managing the population of patients with type 2 diabetes mellitus (T2DM) because of the health and economic burden of this disease. In patients with T2DM, weight loss and glycemic control are primary treatment aims to help improve patient outcomes, but these goals are not easily achieved. While achieving these aims requires a multifaceted approach of drug therapy management and lifestyle modification, truly understanding the role of medication adherence in achieving these outcomes is important for both patient and population management. This study expands on existing evidence that weight loss is associated with improved glycemic control by examining the role of medication adherence in achieving these goals in a managed care setting. This study is unique in that these associations are evaluated using multiple sources of data, including medical records for treatment outcomes, pharmacy claims, and patient-reported data to assess medication adherence. These data sources represent those typically available to payers or providers.OBJECTIVE: To describe the relationships between medication and adherence, weight change, and glycemic control in patients with T2DM.METHODS: This historical cohort study included adult patients with T2DM in a large integrated health system and was based on electronic health record and pharmacy claims data from November 1, 2010, through October 31, 2011, as well as data from a self-reported adherence survey conducted in March 2012. Included patients received a diabetes medication from a therapeutic class not previously received, between November 1, 2010, and April 30, 2011 (index date), who had blood glucose (HbA1c) and weight values at index date and 6 months follow-up, participated in an adherence survey, and had ≥ 1 prescription claim for the index-date drug. Associations between the dual outcomes of weight loss (≥ 3%) and HbA1c control (< 7.0%), while controlling for medication adherence and other demographic, treatment, and clinical variables, were evaluated using structural equation models (SEM). Separate models adjusted for different measures of medication adherence-self-reported using the 5-item Medication Adherence Rating Scale (MARS-5) and a modified medication possession ratio (mMPR) from pharmacy claims data. RESULTS:The study included 166 patients with a mean age of 61.1 (standard deviation = 12.1) years; 56.0% were female. Medication adherence was high, with 72.2% adherent using MARS-5 and 77.1% using mMPR measures. The SEMs found that only self-reported medication adherence is associated with weight loss (MARS-5: OR = 1.70, 95% CI = 1.11-2.60), while both self-reported and claims-based medication adherence were associated with HbA1c < 7.0% (MARS-5: OR = 1.59, 95% CI = 1.09-2.34; mMPR: OR 2.71, 95% CI = 1.22-5.98). Further, weight loss is significantly associated with HbA1c < 7.0% (MARS-5: OR = 3.60, 95% CI = 2.39-5.46; mMPR: OR 2.99, 95% CI = 1.45-6.17).CONCLUSIONS: This study has provided additional evidence ...

Correlation between adherence to antiepileptic drugs and quality of life in patients with epilepsy: A longitudinal study

“…Fifthly, we used the inclusion criteria published in 1989 but not the updated version published in 2010. Finally, it needs to be noted that MARS-5 has not been commonly used in the neurology clinic because of its ineffectiveness in patients with 'silent' disease, such as hypertension, COPD, and type 2 diabetes mellitus [18,19,44,45]. However, the drive of medication adherence is stronger in patients with epilepsy in order to manage their seizures' symptoms; therefore, they tend to provide more accurate self-reported MARS-5 results.…”

Considerations for the analysis of longitudinal electronic health records linked to claims data to study the effectiveness and safety of drugs

“…Our search returned 542 unique articles/abstracts, among which nine journal articles and six abstracts were selected. Most of the articles were excluded because they did not investigate drug effectiveness or safety ( n = 350), some for using only EHR ( n = 47) or claims data ( n = 10), some for using both types of datasets without data linkage ( n = 55).…”

“…If not, the researcher will need to supplement the data with EHR. Some example outcomes in this scenario include change in glycosylated hemoglobin (HbA1c), BMI, or blood pressure (BP), smoking cessation, semen volume and sperm concentration, left ventricular ejection fractions assessed by an echocardiogram, or QT interval prolongation assessed by an electrocardiogram. When choosing EHR as the additional source of outcome information, researchers should be mindful of how the information was recorded and how patients with the outcome information recorded in EHR differ from those missing such information.…”

Persistence with rapid-acting insulin and its association with A1C level and severe hypoglycemia among elderly patients with type 2 diabetes

“…Continuing treatment with antidiabetes medication, irrespective of treatment class, is known to positively influence outcomes. Numerous recent studies have reported associations between persistence/adherence and glycemic control, resource use, mortality, and cost, in patients with T2D 23–28 . Our results suggest that policy makers should consider implementing interventions targeted to increase persistence to RAI, which may translate into better outcomes.…”