Can outpatient pharmacy data identify persons with undiagnosed COPD?

Area of Science:

• Pulmonary Medicine
• Health Informatics
• Pharmacoepidemiology

Background:

• Chronic obstructive pulmonary disease (COPD) poses a significant public health challenge, often diagnosed late.
• Early identification of undiagnosed COPD is crucial for timely intervention and improved patient outcomes.
• Outpatient pharmacy data represents a rich, underutilized resource for health condition surveillance.

Purpose of the Study:

• To develop and validate a predictive algorithm for identifying individuals with undiagnosed COPD using outpatient pharmacy utilization.
• To assess the feasibility and accuracy of using administrative pharmacy data for COPD case-finding.

Main Methods:

• A case-control study design was employed using managed care administrative data.
• Conditional logistic regression identified drugs associated with COPD diagnosis, which informed a predictive algorithm.
• The algorithm was validated internally and externally, followed by clinical validation using spirometry and questionnaires.

Main Results:

• COPD patients exhibited significantly higher utilization of antibiotics, cardiac, and respiratory medications in the two years preceding diagnosis.
• The developed algorithm achieved 60% sensitivity and 70% specificity in identifying COPD cases.
• Clinical validation confirmed that 25% of individuals flagged by the algorithm had previously undiagnosed COPD.

Conclusions:

• Pharmacy utilization patterns can serve as an early indicator for potential COPD diagnosis.
• Predictive algorithms leveraging pharmacy data offer an efficient strategy for identifying at-risk populations for undiagnosed COPD.