Variation in routine electrocardiogram use in academic primary care practice

Area of Science:

• Cardiology
• Primary Care Medicine
• Health Services Research

Background:

• A lack of consensus exists regarding routine electrocardiogram (ECG) ordering in primary care settings.
• This ambiguity may lead to nonclinical variations in ECG utilization among healthcare providers.

Purpose of the Study:

• To investigate the extent of nonclinical variations in routine electrocardiogram (ECG) ordering among primary care providers.
• To identify patient and provider characteristics associated with ECG ordering.

Main Methods:

• Utilized computerized billing data from 10 internal medicine group practices (October 1996 - September 1997).
• Included 69,921 patients and 190,238 visits, excluding patients with known cardiac disease or symptoms.
• Adjusted ECG ordering rates for patient demographics and diagnoses; logistic regression identified predictors.

Main Results:

• Electrocardiograms were ordered in 4.4% of visits for patients without reported cardiac disease.
• Adjusted ECG ordering rates varied widely: 0.8%-8.6% across practices and 0.0%-24% among individual providers.
• Older patient age, male sex, comorbidities, Medicare payment, older male providers, and providers billing for ECG interpretation were significant predictors.

Conclusions:

• Patient characteristics do not explain the observed variations in ECG ordering.
• Significant nonclinical variations in screening ECG ordering necessitate the development of greater consensus and standardized guidelines in primary care.