We read the article by Lucas et al1 on the inverse association between coffee consumption and depression with interest. If this were a randomized clinical trial, it would be highly likely that the treatment arm would be some level of multiple cups of coffee per day with caffeine and it would be more likely that the comparison arm would be some “placebo juice”—perhaps a similar number of decaffeinated cups of coffee per day. This would be done to create comparability in the treatment experience and to create comparability with respect to the elements of the “drug regimen” that may also influence the outcome under study. It is necessary to isolate the effects of the caffeine from other facets of the coffee drinking experience such as taking time out of the day to drink the coffee, perhaps with coworkers, family, or friends over an interesting conversation in which shared life experiences ease the day's burden, as it may well be this and not the actual coffee that provides the beneficial effect. The role of social support in preventing major depressive episodes should not be minimized.2 Furthermore, emotionally supportive social relationships have been shown to be substantially more protective against major depression for women than for men.3