Abstract
ObjectiveTo validate self-reported use of medications for secondary prevention of coronary heart disease (CHD) in a population-based health study by comparing self-report with pharmacy dispensing data, and explore different methods for defining medication use in prescription databases. Study design and settingSelf-reported medication use among participants with CHD (n = 1483) from the seventh wave of the Tromsø Study was linked with the Norwegian Prescription Database (NorPD). Cohen’s kappa, sensitivity, specificity, and positive and negative predictive values were calculated, using NorPD as the reference standard. Medication use in NorPD was defined in three ways; fixed-time window of 180 days, and legend-time method assuming a daily dose of one dosage unit or one defined daily dose (DDD). ResultsKappa-values for antihypertensive drugs, lipid-lowering drugs and acetylsalicylic acid all showed substantial agreement (kappa ≥0.61). Validity varied depending on the method used for defining medication use in NorPD. Applying a fixed-time window gave higher agreement, positive predictive values and specificity compared with the legend-time methods. ConclusionSelf-reported use of medication for secondary prevention of CHD shows high validity when compared with pharmacy dispensing data. For CHD medications, fixed-time window appears to be the most appropriate method for defining medication use in prescription databases.
Highlights
Medication use is an important factor in many epidemiological studies, either as exposure or outcome
Applying a fixed-time window gave higher agreement, positive predictive values and specificity compared with the legend-time methods
Using a fixed-time window to define current medication use gave higher agreement, positive predictive values and specificity compared with the legend-time methods
Summary
Medication use is an important factor in many epidemiological studies, either as exposure or outcome. Poor measurement of medication use can lead to over- or underestimation of true associations and risks [1]. There are several ways to measure medication use, where self-reported use, e.g. questionnaires or interviews, and pharmacy dispensing data are common methods. Self-reported use may be biased by poor recall and underreporting of socially stigmatized medication classes [2,3]. It may be prone to selection bias as some data sources include only reimbursed medications, and others are based on claims from selected insurance companies or pharmacies [4,5,6,7,8,9,10]. A few countries, like the Scandinavian countries, have complete prescription registries that include all prescription-based medications dispensed from pharmacies [11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
