Oral antihyperglycemic medication nonadherence and subsequent hospitalization among individuals with type 2 diabetes

College of Pharmacy, University of Michigan, Ann Arbor, Michigan, United States
Diabetes Care (Impact Factor: 8.57). 10/2004; 27(9):2149-53. DOI: 10.2337/diacare.27.9.2149
Source: PubMed

ABSTRACT This study examines the association between oral antihyperglycemic medication nonadherence and subsequent hospitalization among individuals with type 2 diabetes.
Using administrative claims data (2000-2001) from a managed care organization in the Midwestern U.S., this study analyzed 900 enrollees, aged 18 years and over, with type 2 diabetes who were taking oral antihyperglycemic agents both years but who did not use insulin. Nonadherence was defined as a medication possession ratio (MPR) <80%. Multivariate logistic regression analyses were performed where hospitalization in 2001 was regressed on nonadherence to the oral antihyperglycemic drug regimen in 2000, while controlling for nonadherence to drugs for hypertension and dyslipidemia and for hospitalization in 2000, age, sex, intensity of the diabetes drug regimen, and comorbidities.
The proportion of enrollees who were nonadherent to the antihyperglycemic drug regimen in 2001 was 28.9%, whereas 18.8 and 26.9% were nonadherent to antihypertensive and lipid-modifying drugs, respectively. The increase in the hospitalization rate for 2001 was most apparent where the antihyperglycemic MPR for 2000 dropped to <80%. Enrollees who were nonadherent to oral diabetes medications in 2000 were at higher risk of hospitalization in 2001 (odds ratio 2.53; 95% CI 1.38-4.64), whereas nonadherence to drugs for hypertension and dyslipidemia were not significantly associated with hospitalization.
Patients with type 2 diabetes who do not obtain at least 80% of their oral antihyperglycemic medications across 1 year are at a higher risk of hospitalization in the following year.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Abstract: The prevalence of diabetes is higher in non-Hispanic Blacks, and Hispanics as compared to non-Hispanic Whites. Adherence to medications is required in achieving glycemic control as it has a well established relationship with improved treatment outcomes and is associated with decreased utilization of medical resources. This study examined the racial differences in medication adherence and associated healthcare costs and utilization. A modified version of Health Belief Model and Aday-Andersen model for healthcare utilization was used as the theoretical model. This retrospective cohort study which used North Carolina Medicaid database compared 1517 African Americans, 1115 Whites and 505 patients of other races newly starting metformin, sulfonylureas or thiazolidiendiones. The patients were followed for one year after the start of medication to gather the data on medication utilization, healthcare costs, hospitalization and emergency room visits. Demographic, clinical and medication related information was extracted from the database. Multiple log-linear regression analysis was employed to model medication adherence and healthcare costs, while multiple logistic regression analysis was utilized for hospitalization, and emergency room visits. Adherence levels of this population were considerably lower than the generally accepted level of eighty percent. The study found that there were racial differences in medication adherence, healthcare costs, likelihood of hospitalizations and emergency room visits. Being an African American was associated with decreased medication adherence, increased healthcare costs, increased likelihood of hospitalization and emergency room visit as compared to Whites. Factors such as age, comorbidities, and number of medications consumed were also significant predictors of outcomes. Medication adherence was mainly associated with hospitalization and ER visit, which reinforces the importance of adherence in avoiding complications in diabetes patients. The predictive power of the models explaining adherence was moderate, while models explained variance in healthcare costs considerably well. Third-party payers such as Medicaid need to target at-risk patients based on the above mentioned factors. Disease management programs remain the most important tool to target at-risk patients, consider their clinical profile and medication management issues for optimal utilization of healthcare resources. 0.80 MB Title from first page of PDF file. Thesis (Ph. D.)--Ohio State University, 2006. Includes bibliographical references (p. 111-126). Available online via OhioLINK's ETD Center System requirements: World Wide Web browser and PDF viewer.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Patients with chronic illness often have many medications available for treatment. Medications are changed, and added when the patient experiences side effects or the disease progresses so that a medication is no longer effective. It is the purpose of this project to use survival data mining techniques to investigate prescribed medications for patients with Type II diabetes. We examine the Medical Expenditure Panel Survey for the years 2000-2004 and validate the results using a second database for a cohort of patients with diabetes. We preprocess the data by filtering to medications listed for the ICD9 code for diabetes. Then we remove medications not specifically related to glucose control, eliminating misspellings and name variations. In the MEPS database, this leaves a total of 73,000 prescriptions. We use sequential market basket analysis to examine combinations of medications used. We also use survival data mining techniques since medication switching is quite similar to a customer switching to a new provider. Since switching can occur multiple times, there are multiple time points to investigate. A total of 30% of the prescriptions were for insulin, with 24% for Metformin. Most of the switching from one drug to another occurred within five months or less. Changing and adding medications tends to occur within short periods of time. In addition, it appears to be far more common to switch medications rather than to add. Also, patients averaged 9 prescriptions each, when 12 prescriptions (30 days each) indicates compliance.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: There is limited information in the primary literature regarding the relationship of medication adherence to attainment of glycosylated hemoglobin A1c (A1c) goals. The 2 oral antihyperglycemic medications, sulfonylurea and/or metformin, were chosen for retrospective analysis because they are the 2 most common oral medications used by patients with diabetes. To describe the relationship between adherence with 1 or both of 2 oral antihyperglycemic medications (sulfonylurea and metformin) and A1c goal attainment for health maintenance organization (HMO) patients enrolled in a diabetes disease management program. This was a retrospective, descriptive evaluation of patients enrolled in a managed care diabetes disease management program in a 188,000-member independent practice association model HMO located in the Southeast. The dataset in this analysis contained demographic, enrollment, pharmacy claims, and clinical laboratory data. Continuously enrolled patients were included if there was a documented A1c value obtained at least 90 days after the initial oral antihyperglycemic medication (sulfonylurea or metformin) prescription index date. The medication possession ratio (MPR) was calculated from the pharmacy claim records and correlated with the A1c value. A total of 42% of patients on sulfonylurea therapy and 46% of those on metformin reached an A1c goal of < or = 7.0%. For patients taking a sulfonylurea, the mean MPR for those who reached the predetermined A1c goal (< or = 7.0) was 0.82 (0.29) compared with 0.72 (0.31) for those patients who did not reach the A1c target goal (P < 0.001). For patients taking metformin, the mean MPR for those who reached the predetermined A1c goal was 0.77 (0.3) versus 0.62 (0.3) for those patients who did not reach the A1c target goal (P < 0.001). A Pearson correlation analysis revealed a positive relationship between the MPR and A1c for sulfonylurea (r = -0.295, P < 0.001) and for metformin (r = -0.285, P < 0.001). For those patients taking both sulfonylurea and metformin, the Pearson correlation analysis showed a positive relationship between the 2 MPRs (r = 0.65, P < 0.001). Medication adherence as measured by the MPR was higher for patients taking a sulfonylurea or metformin who reached the target A1c goal of d7.0% compared with patients taking these drugs who did not reach the target A1c goal.
    Journal of managed care pharmacy: JMCP 11/2005; 12(6):466-71. · 2.68 Impact Factor
Show more