Comparability of self-reported medication use and pharmacy claims data

Area of Science:

• Health Services Research
• Pharmacoepidemiology
• Geriatric Medicine

Background:

• Self-reported data is frequently used in medical research.
• The validity and reliability of self-reported medication use require rigorous assessment.

Purpose of the Study:

• To evaluate the accuracy of self-reported medication use against pharmacy claims data.
• To assess agreement levels for different medication classes in seniors.

Main Methods:

• Linked Canadian Community Health Survey and Ontario Drug Benefit Program data were analyzed.
• Kappa statistics were used to measure agreement between self-reports and claims data.
• Logistic regression modeled factors influencing agreement.

Main Results:

• Good to very good agreement (kappa 0.79-0.87) for oral diabetes medications.
• Moderate agreement (kappa 0.46-0.55) for antihypertensive medications.
• Agreement improved from 2001 to 2005, potentially due to survey question refinement.

Conclusions:

• Self-reports are a largely accurate data source for medication use.
• Self-reported antihypertensive medication use in the oldest and sickest seniors warrants cautious interpretation.