Managed Care - February 2008 - (Page 49) strate similar problems in the pediatric and elderly populations (Mangione-Smith 2007, Jencks 2000, Schoen 2006). These striking and disturbing results continue to pave the way for the implementation and refinement of pay-for-performance initiatives and evolving physicianlevel measurement (Rosenthal 2005). Several studies have demonstrated that the implementation of evidenced-based, guideline-driven care results in improved clinical and economic outcomes (Peterson 2006, Walsh 2006, Yurk 2004). Such strategies address physician performance, yet, even with half of patients with chronic conditions receiving appropriate evidence-based care, many still have poor outcomes. These outcomes are often related to patients’ poor self-management, including nonadherence to medication. Medication adherence, as used in this paper, is the extent to which a patient acts in accordance with the prescribed interval and dosage regimen as recommended by the health care provider (ISPOR). Nonadherence to medications may occur for a variety of reasons. One that is frequently discussed in the literature is cost. Patients who cannot afford the cost or even the copayments will frequently be nonadherent by not filling or not refilling prescriptions, by reducing doses to achieve longer supply, or by skipping doses (Kessler 2007, Tseng 2004). There are many other concerns that can be linked to reduced medication adherence. A Boston Consulting Group/Harris Interactive survey found that 24 percent of nonadherence is related to forgetting to take or to refill prescriptions and that 20 percent is related to concern about side effects. Seventeen percent stated that cost was the primary issue, 14 percent didn’t feel they needed the medication, and the remaining 21 percent stated a variety of other reasons (BCG 2003). In this paper, we present results from analyses of claims data from a national health plan. We sought to determine if findings from earlier studies, with study populations that were smaller than the present study, would be confirmed on a national scale and for multiple common primary care conditions. These results illustrate widespread medication adherence problems regardless of condition, and these problems contribute to higher health care utilization, increased costs, and poor outcomes. Finally, we compare our findings with current literature, offer suggestions for improving medication adherence, and discuss the results’ implications on health care costs and outcomes. METHODS Data A retrospective analysis was conducted based on claims data from a large national health plan (UnitedHealthcare) with more than 4 million members in regions across the United States. The data consist of pharmacy and medical claims as well as eligibility information and a subset of laboratory results. The data set does not contain any personal identifiers and is compliant with the Health Insurance Portability and Accountability Act of 1996 (HIPAA). Study population Study subjects were identified as having at least 2 medical claims during the period from Oct. 1, 2002 through Dec. 31, 2004 who had a diagnosis of asthma, chronic heart failure (CHF), coronary artery disease (CAD), depression, diabetes, hyperlipidemia, hypertension, or migraine, using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9CM) codes. A separate analyses for each dis- ease condition was conducted; subjects were not mutually exclusive. The index date for all subjects was set as Sept. 30, 2004 to standardize the study follow-up period using the most currently available claims data. Subjects were required to be age 18 or older for all conditions except pediatric asthma, to be continuously enrolled with medical benefits for at least 12 months before the index date, and to be continuously enrolled with pharmacy benefits for at least 6 months before the index date. For asthma, additional eligibility criteria to stratify pediatric asthma (ages 5–17) versus adult asthma (age 18 or older) was applied and excluded subjects with chronic obstructive pulmonar y disease (COPD). In addition, subjects were identified as having persistent asthma if they met any of the following criteria: at least 1 hospitalization for asthma as the principal diagnosis in the past 12 months, at least 1 emergency room visit for asthma as the principal diagnosis in the past 12 months, 6 or more outpatient visits for asthma in the past 12 months, 2 or more prescription fills for oral corticosteroids in the past 6 months, or 2 or more prescription fills for any asthma medications in the past 12 months. EBM Connect software was used to select patients and to measure guideline and medication adherence. The software, used to measure guideline adherence through application of a series of clinical rules and algorithms, was developed through the review of published literature, review by clinicians, and validation analyses of claims data. Utilizing national, peer-reviewed guidelines and expert coding, the rules and algorithms in the software are regularly updated to reflect the most current evidence. FEBRUARY 2008 / MANAGED CARE 49
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.