Effect of feedback letters to physicians and pharmacists on the appropriate use of medication in the treatment of asthma.

Background: Suboptimal medication treatment of asthma has been reported. More specifically, short-acting beta 2-agonists are overused, while inhaled corticosteroids are underused. This can be related in part to poor adherence by patients to the prescribed regimen and to professionals' failure to comply with practice guidelines. Feedback seems to have an effect on professional practices related to medication use.

Objective: To assess the impact of feedback letters to physicians and pharmacists on their patients' appropriate use of asthma medication.

Methods: Two randomized trials were set up in the province of Quebec, Canada: one with physicians and another with pharmacists. A sample of voluntary physicians and pharmacists was randomly assigned to either the experimental group or to the control group. Those in the experimental groups received three consecutive feedback letters over a 9-month period summarizing the asthma medications acquired by their patients over the preceding year. The feedback focused on short-acting beta 2-agonists, long-acting beta 2-agonists and antileukotrienes and provided information on compliance with five appropriate-use criteria. Pharmacists received aggregate profiles and individual profiles with patients' names, while most physicians received aggregate profiles for all their eligible patients. Each mailing also included a pamphlet that summarized practice guidelines on asthma treatment.

Results: Seventy-one physicians and 60 pharmacists participated in the study. Physicians who received the feedback letters did not differ from those in the control group in terms of their proportion of prescriptions compliant with the criteria, either before the feedback or after it (p > 0.05). The before-after difference was also similar between groups. The same was true for pharmacists. However, although the before-after difference for criteria 1 (frequency of use of short-acting beta 2-agonists) and 2 (frequency of use of long-acting beta 2-agonists) did not reach the usual statistical significance threshold of 0.05, the p value was under 0.10.

Conclusions: As designed in this study, feedback provided to physicians did not improve the appropriate use of asthma medication. However, feedback to pharmacists is promising, especially when including patients' names so that pharmacists can intervene more specifically.