The Effect of Vitamin C on the Common Cold

Proponents of evidence-based medicine (EBM) emphasize that conclusions about medical interventions should be based on controlled trials with clinically relevant outcomes, so that the greatest weight should be given to the systematic reviews of the trials. One of the most comprehensive EBM databases is the Cochrane Library, which contains thousands of systematic reviews on medical interventions. The Cochrane review on vitamin C and the common cold analyzed 29 trial comparisons on the preventive effects of vitamin C, 29 comparisons on the effect of regular vitamin C on common cold duration and severity, and 7 comparisons on the therapeutic effect of vitamin C. The review concludes: ‘‘Implications for practice: The prophylaxis trials found a reduction in common cold duration of 8% in adults and 13% in children . . . . So far, therapeutic supplementation has not been shown to be effective. Nevertheless, given the consistent effect of vitamin C on duration and severity in the regular supplementation studies, and the low cost and safety, it may be worthwhile for common cold patients to test on an individual basis whether therapeutic vitamin C is beneficial for them.’’ In their review discussing the effect of dietary supplements on the common cold, Mathes and Bellanger refer to 9 papers and state that ‘‘various trials have shown no difference in either duration, or severity of symptoms, when vitamin C was compared to placebo. In addition, numerous trials have shown no difference regarding the most commonly studied endpoint: frequency of symptoms.’’ Since the conclusions of the Cochrane review diverge considerably from the conclusions of Mathes and Bellanger, it is useful to look at the findings of the studies they refer to. Some of the findings are inconsistent with their statements. Pitt and Costrini found statistically significant (P[2-tail]1⁄4 .024) yet small reduction in common cold severity among US Marines who were administered 2 g/day vitamin C. Van Straten and Joslin reported a highly significant reduction in the number of participants who had 2 colds during their trial (2/84 vs 16/84; P 1⁄4 .0004) by 1 g/day vitamin C. Ludvigsson et al found a 14% reduction (P 1⁄4 .016) in the duration of ‘‘absence from school because of upper respiratory tract infection’’ in their main trial, and a 39% reduction (P 1⁄4 .004) in the duration of ‘‘upper respiratory tract infection’’ in their smaller trial when schoolchildren were administered 1 g/day of vitamin C. Terrence Anderson carried out 3 vitamin C trials in Canada in the 1970s. In their first trial, Anderson et al administered 1 g/day vitamin C regularly and 3 g/day extra when a participant caught the common cold. They found a 30% (P 1⁄4 .01) decrease in the number of days confined indoors in the vitamin C group. Furthermore, staying free of illness during the study period is also a relevant measure of effect and Anderson et al found 8 percentage point increase in the proportion of participants who were ‘‘not ill during the trial,’’ ‘‘not confined to the house,’’ and ‘‘not off work’’ in the vitamin C group (P 1⁄4 .01). In their second trial, Anderson et al were ambitious and they had 2 placebo arms and 6 vitamin C arms with different dosages. There were technical problems such as the mislabeling of bottles of tablets, and therefore no firm conclusions can be drawn from the second trial. Nevertheless, in the vitamin C arm which was administered 8 g of vitamin C only on the first day, 6.6 percentage points (P 1⁄4 .046) more participants had colds lasting for only 1 day, compared with the arm administered 4 g of vitamin C, which suggests dose dependency. In their third trial, Anderson et al administered 1 to 1.5 g/day of vitamin C therapeutically so that supplementation was initiated when a person caught a cold. They found a reduction in ‘‘days spent indoors per subject’’ by 25% (P 1⁄4 .048) among the vitamin C participants. These 7 controlled trials described above do not support Mathes and Bellanger’s statements although the studies are provided as references. Furthermore, Mathes and Bellanger do not describe their rationalization to select the 9 references out of several dozens. In systematic reviews, it is important to describe the selection criteria, in order to prevent bias caused by selecting studies that are consistent with the reviewers’ preconceptions. If Mathes and Bellanger were not ready to carry out their own systematic review, they should have taken a look at the Cochrane review which used explicit selection criteria and included all studies fulfilling those criteria. The main findings of the Cochrane review are as follows. In 5 trials with participants under heavy acute physical stress (3 with marathon runners), vitamin C reduced the incidence of colds by 52% (95% CI: 36% to 65%), but there is no evidence that vitamin C prevents colds in the general population. In regular