Implementation of warnings from Dear Doctor Letters (Rote-Hand-Briefe): an analysis of medication data from a large cohort of elderly patients.

BACKGROUND: Dear Doctor Letters (also known as Direct Healthcare Professional Communications) inform physicians about significant newly discovered drug risks and about measures to take to reduce these risks. How far these warnings actually influence prescribing behavior is unclear.

METHOD: The Geriatrics in Bavaria-Database (GiB-DAT, Geriatrie in Bayern Datenbank) collects data from more than 50 centers offering inpatient geriatric health care. Based on GiB-DAT data, the discharge medication of 76 568 patients (81 ± 8 years, 67% women) was recorded in a standardized manner and analyzed for the implementation of information contained in two Dear Doctor Letters about the risks of dose-related or drug interaction-related QT interval prolongation caused by citalopram or escitalopram.

RESULTS: Patients were discharged with a median of 8 drugs. In the four quarters of 2012 following release of the Dear Doctor Letters, in comparison to the four quarters before their publication, a marked drop from 9.8% to 4.1% was seen in prescriptions for >20 mg/day citalopram--a dosage no longer recommended in elderly patients--and a similar drop from 23.6% to 12.8% in prescriptions for >10 mg/day escitalopram (p<0.0001). Co-prescription of either of these two drugs with other QT interval-prolonging drugs, which was now contraindicated, remained almost unchanged (citalopram: 19.3% [95% confidence interval (CI): 17.9-20.9%] versus 18.4% [95% CI: 17.0-19.8%]; escitalopram: 17.6% [95% CI: 15.8-19.6%] versus 17.1% [95% CI: 14.5-19.9%]).

CONCLUSION: Simple information in Dear Doctor Letters, such as a reduction of the maximum daily dose, was better implemented than complicated information regarding contraindicated co-medication. Quality assurance systems such as the GiB-DAT network make it possible to identify problems of this kind. Licensing authorities and pharmaceutical companies should should improve the clinical usability, e.g., by providing official reference lists of drugs with safety warnings.