Presenter: M. Christopher Roebuck (CVS Caremark. Strategic Research)
Discussant: Yong LiIndiana University-Purdue University Indianapolis
To examine the relationship between adherence to medication therapy and the utilization and cost of health services in four chronic vascular disorders (congestive heart failure, hypertension, diabetes and dyslipidemia); while addressing the potential endogeneity of adherence. Effect estimates from prior studies may be biased due to unobservables correlated with both adherence and health services utilization and cost. An example is the “healthy user effect,” whereby individuals who engage in health-enhancing behaviors such as regular exercise and a proper diet also have better medication adherence rates.
DATA AND METHODS:
The sample consisted of 16,353 patients with congestive heart failure, 112,757 with hypertension, 42,080 with diabetes and 53,041 with dyslipidemia.
The study design was a retrospective, longitudinal claims analysis of patients with continuous pharmacy and medical insurance coverage from 1/1/2005 through 6/30/2008. Condition cohorts were constructed using ICD-9-CM diagnostic codes. Medical and pharmacy claims data were used to derive six annual measures of health services utilization and cost: number of inpatient hospital days, number of emergency department visits, number of outpatient physician visits, total pharmacy costs, total medical costs and total health care costs. Patients were defined as adherent where their condition-level Medication Possession Ratios were at least 0.80. Fixed effects models were estimated to control for the potential endogeneity of adherence.
Medication adherence was associated with fewer inpatient hospital days and emergency department visits, and a greater number of outpatient physician visits. Higher drug costs were more than offset by lower medical costs with annual total health care cost savings amounting to $7,823 for congestive heart failure, $3,908 for hypertension, $3,756 for diabetes and $1,258 for dyslipidemia. Benefit-cost ratios ranged from 2:1 for non-senior adults with dyslipidemia to over 13:1 for seniors with hypertension. Adherence effects were generally greater in magnitude for seniors than for non-seniors.
To fully understand the clinical and economic impacts of prescription drug use in community-based settings, researchers must also rely on carefully executed observational studies. In the absence of randomization, causality is difficult to establish, however advance econometric techniques can be used to reduce potential bias in effect estimates.
This study extends the literature on the impact of medication adherence on health services utilization and cost. Using a large panel dataset, fixed effects modeling controlled for the potential endogeneity of adherence. Results confirm that despite higher pharmacy costs, medication adherence provided substantial medical savings, likely due to reductions in hospitalization and emergency department use.
Given these findings, policies that improve medication adherence may be worthwhile investments for plan sponsors and patients. Disease management programs, pharmacist-consumer engagement, and physician education are among the alternatives. Value-based insurance designs, which reduce or eliminate patient out-of-pocket costs on maintenance medications, may be particularly attractive since economic incentives are well-known to affect patient behavior. As the drive for healthcare reform intensifies, policymakers searching for ways to bend the healthcare cost curve downward would do well to look to medication adherence for help.
The 3rd Biennial Conference of the American Society of Health Economists took place at Cornell University.
Software © 2010 iHEA - International Health Economics Association