Impact of a health alert and its implementation on flutamide prescriptions for women: an interrupted time series analysis.

<bold>Background: </bold>Off-label drug use among ambulatory patients is often based on little or no scientific support. This paper reports the impact of a health warning about the risks of off-label flutamide use by women and the actions subsequently implemented by the public health service targeting such use.<bold>Method: </bold>The study was undertaken in a region in north-west Spain. We designed a segmented regression model of an interrupted time series, in which the dependent variable was the monthly value of defined daily doses of flutamide per 1000 inhabitants/day (DDD/TID), both total and stratified by sex. The following two data sources were used: flutamide prescriptions billed to the Spanish National Health Service; and flutamide deliveries made by wholesale drug distributors to pharmacies. The intervention assessed consisted of the issue of an official health warning and the actions subsequently taken to implement it.<bold>Results: </bold>There was an immediate reduction of 49.33% in DDD/TID billed to the Spanish National Health Service in respect of women; the mean value of the population percentage of DDD/TID of flutamide billed in respect of women fell from 34.4% pre-intervention to 23.72% post-intervention. There was an immediate reduction of 19.92% (95%CI: 6.68-33.15%) in total DDD/TID invoiced. There were no significant changes in DDD/TID billed in respect of men or in flutamide use in the private medical sector.<bold>Conclusions: </bold>Off-label drug misuse is a reality among ambulatory patients, even after actions are implemented following a toxicity warning issued by the competent Health Authority. [ABSTRACT FROM AUTHOR]