Fig. 4.

Effect of vitamin C on the total days confined indoors by subgroups in the Anderson (1972) trial. The scale of days is used in this figure since that is more useful in practical terms, and the population is homogeneous with respect to viruses, participants, and outcome definitions. For example, over the 3-month follow-up in participants who had contact with young children, vitamin C decreases days confined indoors by 1.12 days, and in those who did not have contact with children by 0.26 days. Thus, the effect was 0.86 days greater in the former (P = 0.016 for the difference in the effect between the subgroups). MD, mean difference