Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website

RRP 09-139 – HSR&D Study

New | Current | Completed | DRA | DRE | Portfolios/Projects | Centers | Career Development Projects

RRP 09-139
Process Oriented, Validated Electronic Performance Measures Pilot Study
Christopher L. Bryson MD MS
VA Puget Sound Health Care System Seattle Division, Seattle, WA
Seattle, WA
Funding Period: October 2009 - September 2011

BACKGROUND/RATIONALE:
Ischemic heart disease (IHD) affects a large proportion of VA patients; more than 500,000 VA patients have a diagnosis of IHD. Currently, outpatient performance measures for control and prevention of IHD are based solely on surrogate measures of disease control, such as achievement of a target low-density lipoprotein (LDL) level and achievement of a target blood pressure, rather than initiation and maintenance of appropriate therapy. In addition, measurement data are abstracted manually despite the wealth of electronic data available in the VA.

OBJECTIVE(S):
The major goal of this study was to develop and validate pilot data for performance measures for treatment of hyperlipidemia and hypertension in patients with IHD. Specific research aims were to: 1) identify and validate a cohort of patients with IHD using an electronic algorithm applied to administrative data, and 2) develop electronic algorithms to determine prescription of and adherence to appropriate pharmacologic therapy for hyperlipidemia.

METHODS:
This pilot study reviewed retrospective records of IHD patients in VISN 20 using data warehouse records from FY 2007 through FY 2010. We defined a cohort of IHD patients and constructed measures of risk reduction using electronic data on diagnosis and treatment of patients. For the performance measures, we used rolling, quarterly adherence measures for a cohort of patients defined by utilization data from the prior 12 months. We developed an algorithm using inpatient and outpatient ICD-9 and CPT codes to identify the patient cohort and used outpatient pharmacy data to construct medication prescription and adherence performance measures. To validate the patient identification algorithm, we reviewed 199 charts (102 with IHD and 97 without IHD). We also conducted an anonymous short telephone survey (49 statin takers and 34 non-statin takers identified from VA pharmacy data) to assess the degree of statin filled in non-VA settings, or the completeness of using VA pharmacy data for statin adherence.

FINDINGS/RESULTS:
The algorithm identified 40,207 IHD patients (FY07: 19,226 and FY10: 20,981) and 85,411 non-IHD patients (FY07: 39,682 and FY10: 45,729). The validation result based on the chart reviews shows that overall 84.4% of the cohort was classified correctly as IHD or non-IHD using this algorithm. Among IHD patients, 91.8% were classified correctly (sensitivity), while, among non-IHD patients, 79.0% were classified correctly (specificity). Based on patient responses to the telephone interview asking where they fill their statin medication prescriptions, we found 2 out of 49 identified statin takers and 4 out of 34 non-statin takers currently fill a statin medication prescription in non-VA settings. For the adherence performance measures, 74.4% of the 3,029 patients who were adherent to statin medications achieved target LDL levels compared to 47.7% of 1,089 patients who were not adherent.

IMPACT:
While the validation tests of the algorithm showed moderately high sensitivity, this ICD-9 and CPT algorithm is not sensitive enough alone to develop a high quality cohort for performance metrics. This study does provide pilot data for future studies to fully develop performances measures for hyperlipidemia and hypertension in a national cohort. Ultimately, these studies will help improve both the quality of care, by introducing more accurate and effective performance measures, and the efficiency of performance measurement for IHD patients in the VA.

PUBLICATIONS:
None at this time.


DRA: Cardiovascular Disease
DRE: none
Keywords: Cardiovasc’r disease, Quality assurance, improvement, QUERI Implementation
MeSH Terms: none