Reply

Abstract

Doctors Inadomi and Somsouk provided a very well-written summary of our recent study on the association between long-term proton pump inhibitor (PPI) therapy and the risk of hip fracture. We completely agree with their take-home message. Nevertheless, we would like to clarify 1 misunderstanding regarding the results of our dosage analysis. The 1.75 daily dose was not the average daily dose among the study population who were prescribed PPIs. In fact, the vast majority of the long-term (ie, ≥1 year) PPI users in our study were once daily users. Therefore, the point estimate of our primary analysis (ie, ≥1 year of PPI therapy) primarily reflected the effect of single daily dose use of PPI therapy but not high-dose PPI therapy. We used 1.75 average daily dose as a cut-point to differentiate those who had received ≥75% of their prescriptions at a dose of twice daily from those who used mostly once-daily dose. In our dosage-specific analysis, the odds ratio associated with ≥1 year of PPI therapy at <1.75 average daily dose was 1.40, and >90% of PPI users in this dosage category had an average daily dose of <1.1. This odds ratio was almost the same as the overall odds ratio of 1.44 associated with ≥1 year of PPI therapy. On the other hand, the odds ratio associated with long-term high-dose (ie, ≥1 year and ≥1.75 average daily dose) PPI therapy was much higher at 2.65. These results suggest a clear dose–response effect and highlight the importance of using the lowest effective dose when prescribing PPI therapy, as suggested by us and Drs. Inadomi and Somsouk.