Identifying Hypertension-Related Comorbidities From Administrative Data: What's the Optimal Approach?

Abstract

The objective was to determine the best strategy for identifying outpatients with hypertension-related diagnoses using Veterans Affairs (VA) administrative databases. We reviewed 1176 outpatient charts from 10 VA sites in 1999, taking the presence of 11 diagnoses relevant to hypertension management as the "gold standard" for identifying the comorbidity. We calculated agreement, sensitivity, and specificity for the chart versus several administrative data-based algorithms. Using 1999 data and requiring 1 administrative diagnosis, observed agreement ranged from 0.98 (atrial fibrillation) to 0.85 (hyperlipidemia), and kappas were generally high. Sensitivity varied from 38% (tobacco use) to 97% (diabetes); specificity exceeded 91% for 10 of 11 diagnoses. Requiring 2 years of data and 2 diagnoses improved most measures, with minimal sensitivity decrease. Agreement between the database and charts was good. Administrative data varied in its ability to identify all patients with a given diagnosis but identified accurately those without. The best strategy for case-finding required 2 diagnoses in a 2-year period.

Renal Insufficiency and Use of Revascularization among a National Cohort of Men with Advanced Lower Extremity Peripheral Arterial Disease

“…ICD-9 codes for PAD in the VA system are only moderately sensitive (8). Furthermore, we required two critical limb ischemia codes within 1 yr of cohort entry to maximize specificity at the cost of sensitivity.…”

“…The NPCD is a comprehensive data source that includes encounter information for all inpatient and outpatient VA care (7). The NPCD has excellent specificity and moderate sensitivity for a variety of different comorbid conditions (including PAD) (8). To maximize specificity, we required that each patient have at least one subsequent critical limb ischemia diagnostic code entry within 1 yr of their initial code entry.…”

Diagnosis and Treatment of Incident Hypertension Among Patients with Diabetes: a U.S. Multi-Disciplinary Group Practice Observational Study

“…22 An established algorithm was used to identify stroke/transient ischemic attack. 23 Elixhauser algorithms 24 were utilized to identify depression, other neurologic disorders, collagen vascular disease, and solid tumors (without metastasis); other published algorithms were used to identify hyperlipidemia, 25 chronic kidney disease, 26 eye disease, 27 congestive heart failure, 25 peripheral vascular disease, 27 and dementia. 28 Due to low prevalence, comorbidities were categorized into the following categories for analysis: microvascular disease (chronic kidney disease and/or eye disease), noncardiac cerebrovascular complications (stroke/transient ischemic attack and/or other neurologic conditions), ischemic heart disease/congestive heart failure (congestive heart failure and/or ischemic heart disease), and other comorbidities (dementia, collagen vascular disease, and/or malignancy).…”

Impact of Renal Insufficiency on Mortality in Advanced Lower Extremity Peripheral Arterial Disease

“…A recent study comparing the chart diagnosis of PAD with ICD-9 diagnostic code entries in the National Patient Care Database reported excellent specificity (95%) but only moderate sensitivity (64%) for these codes. These authors also found that specificity could be enhanced further by requiring two diagnostic codes during a 1-yr period (13). To maximize specificity, we thus required that cohort patients have at least two diagnostic code entries for critical limb ischemia during a 1-yr period.…”