Validity of gout diagnoses in administrative data

Area of Science:

• Rheumatology
• Epidemiology
• Health Services Research

Background:

• Administrative claims data are frequently used for epidemiologic studies.
• The accuracy of gout diagnoses in administrative data requires validation.

Purpose of the Study:

• To assess the validity of gout diagnoses recorded in administrative claims data.
• To determine the utility of administrative data for gout epidemiologic studies.

Main Methods:

• A random sample of 200 patients with at least two gout claims was selected from 800,000 members.
• Medical records were reviewed, and gout diagnoses were independently assessed by two rheumatologists.
• Physician assessments were compared against American College of Rheumatology (ACR), Rome, and New York gout criteria.

Main Results:

• The positive predictive value of two or more coded gout diagnoses was 61%.
• There was low concordance between physician assessments and established gout criteria (kappa values ranging from 0.16 to 0.20).

Conclusions:

• Relying solely on administrative data for gout research may lead to significant misclassification.
• Medical record review may not serve as an adequate gold standard for validating gout diagnoses in claims data.