In reply
Using data from Table 2 and Table 3 in our study,1 Lapane and Cooke compared the crude incidence rate of depression among women consuming 2 or more cups of caffeinated coffee per day with the incidence rate among women consuming a similar amount of decaffeinated coffee. Since they obtained similar rates, they suggest that the camaraderie and other social benefits that may accompany coffee consumption, not caffeine itself, could be the reason for the lower risk among coffee drinkers. However, these rates are not directly comparable because several differences between women who drink caffeinated coffee and those who drink decaffeinated coffee could affect their risk of depression. The most conspicuous of these differences is in the frequency of current smoking, itself a risk factor for depression, which was 2 times higher (18.6% vs 9.2%) among women consuming 2 or more cups of caffeinated coffee per day compared with those drinking 2 or more cups of decaffeinated coffee per day. In our multivariate models, we controlled for different important confounders, such as smoking status, marital status, physical activities, Mental Health Index–5 score, and also a social networks–social ties variable substantially similar to that proposed by Lapane and Cooke (eg, involved in church, volunteer activity, or community group ≥1 hours per week). Although we cannot entirely exclude the possibility of some residual confounding associated with those factors described by Lapane and Cooke, the adjusted results, which included several social ties variables, indeed support our conclusion that the inverse association between coffee intake and depression is most likely explained by caffeine intake.