ITT for beginners

Dear Heyerick et al.

Thank you for your lovely study regarding the use of hops extract in menopausal women. Please note, however, that when you claim to use an Intention To Treat analysis, it’s considered obligatory to actually use the data from all patients, including the ones who drop out mid study. You don’t get to eliminate their data then go on to claim that your response may have been due to them dropping out.

This omission particularly suprises me, since as far as I can tell, you would have obtained a statistically significant response for your trial if you had analysed according to the ITT you claim to be using. Gaining a greater response for your active arm compared to placebo is much easier if you include the almost 25% of your placebo group who dropped out due to lack of effect.

Yrs in perplexity,

Stealthflower

Reference: Heyerick, et al. A first prospective, randomized, double-blind, placebo-controlled study on the use of a standardised hop extract to alleviate menopausal discomforts. Maturitas 2006; 54:164-75.