162-31: Using Contour Plots to Promote EHR Use in Physician Offices

The role of a quality improvement organization (QIO) is to assist healthcare providers in making successful and meaningful changes in the way care is delivered and in improving outcomes of that care. While traditional activities in this role addressed specific clinical indicators, the QIO role has evolved to include the promotion of Health Information Technology (HIT) and the use of Electronic Health Records (EHR), especially in the physician office setting. Contour plots provide a two-dimensional display of three-dimensional data. Traditional uses of contour plots include display of elevation, concentration, velocity, non-linear regression displays, and time histories of data in both geographic and non-geographic applications. While SAS provides the ability to create contour plots with SAS/GRAPH and SAS/IML, the new ODS graphics applications add value to this ability by adding this feature to the Kernel Density Estimation (KDE) procedure. The KDE procedure is a non-parametric technique for density estimation using a known density function (kernel) and averaging this across the observed data points to create a smooth approximation. This estimate is written to a SAS data set, which can then be used for plotting or analysis. In this application, the KDE procedure is combined with ODS graphics to create the contour display. Kernel density estimation allows the graphical display of distribution trends for physician office practices and practice size. This paper applies this application to Virginia. Examples were developed with version 9.1.3 of SAS executing on a Windows 2000 platform. SAS Version 9.1 is required for the ODS experimental graphics extensions. Examples are not platform-specific and can be adapted by both beginning and advanced SAS users. INTRODUCTION The Virginia Health Quality Center (VHQC), three-time winner of U.S. Senate Productivity and Quality Award for Virginia, functions as both a health care quality improvement organization and a patient safety organization. The VHQC assists healthcare providers in making successful and meaningful changes in the way care is delivered and in improving outcomes of that care, especially for the Medicare community of Virginia. Services provided by the VHQC include: 1) health care safety and quality improvement; 2) medical case review; 3) professional credentials verification; 4) health education; and 5) analytic services. The VHQC currently collaborates with all acute care and critical access hospitals and invites all physician offices, home health agencies, nursing homes, and other Virginia healthcare organizations to participate in the improvement of health care quality. In an effort to maximize available resources, spatial and longitudinal analyses are used to identify: 1) spatial distributions of providers; 2) spatial distribution of HIT and technology infrastructures; 3) patterns of HIT usage; and 4) patterns of healthcare quality within this framework. APPLICATION Quality improvement focus areas have been identified by the Centers for Medicare & Medicaid Services (CMS) for all settings of health care. For the physician office setting, QIOs are tasked with the promotion of the adoption of EHR systems and information technology in small-to-medium sized physician offices. The Medicare Prescription Drug and Modernization Act of 2003 (H.R. 1-263) encourages the use of health information technology to manage the clinical care of beneficiaries. The expectation is that greater reliance on technology will enhance access to patient information, decision support, and reference data, as well as improving patient-clinician communications. This information technology initiative offers an integrated approach to improving care for Medicare beneficiaries in the areas of diabetes, heart failure, coronary artery disease, hypertension, osteoarthritis, and preventive care. By educating physician offices on EHR system solutions and alternatives, as well as providing implementation and quality improvement assistance, physician offices will have the information necessary to migrate easily from paperbased health records to EHR systems that suit the needs of their office. It is essential that work in this area be cost-effective. It would be impossible to work with all physician office practices; therefore, practices must be selected carefully to allow for the diffusion of information and knowledge and for the recruiting of physician champions for this effort. Information about the current health care infrastructure, including trends of change over time, is essential to this effort. The applications presented in this paper are part of an effort to understand these complexities for the state of Virginia.