Are patients reliable when self-reporting medication use? Validation of structured drug interviews and home visits by drug analysis and prescription data in acutely hospitalized patients

Abstract

The medication history among hospitalized patients often relies on patients' self-reports due to insufficient communication between health care professionals. The aim of the present study was to estimate the reliability of patients' self-reported medication use. Five hundred patients admitted to an acute medical department at a Danish university hospital were interviewed on the day of admission about their recent medication use. Blood samples drawn immediately after admission were screened for contents of 5 drugs (digoxin, bendroflumethiazide, amlodipine, simvastatin, glimepiride), and the results were compared to the patients' self-reported medication history. Information on prescribed drugs dispensed from any Danish pharmacy was collected from nationwide real-time pharmacy records. The authors performed home visits in a subgroup of 115 patients 4 weeks after their discharge. Stored drugs were inspected, and patients were interviewed about their drug use. Additional blood samples were drawn for drug analysis. The median age of included patients was 72 years, and 298 patients (60%) were women. Patients reported use of 3 (median) prescription-only medications (range, 0-14) during the structured interview. The congruence between self-report and drug analysis was high for all 5 drugs measured (all kappa >0.8). However, 9 patients (2%) reported use of drugs that were not detected in their blood samples. In 29 patients (6%), the blood samples contained drugs not reported during the structured interview, but 14 of these drugs were registered in either hospital files or pharmacy records. Overall, the sensitivity of information from hospital files, structured interviews, and pharmacy records in identifying drug users was 87% to 93%, with no significant differences between methods. In conclusion, patients' self-reports are reliable when estimating recent use of cardiovascular and antidiabetic drugs.