My favorite cold treatment used to be zinc, the only medicine I know of that’s been clinically shown to actually reduce symptom duration.
When I recently offered a housemate some zinc for her cold, though, she told me the clinical review I cited (Cochrane) had been withdrawn. Yikes! Another victim of the replication crisis? Actually, it turns out the zinc meta-analysis was the victim of something even more titillating, claims of data theft:
All the mean (SD) values within the red rectangles in Analysis 1.1 (also published as Fig. 3), of Singh and Das (2013) are identical with those previously imputed by Hemilä (2011), compare the tables on pp. 6 and 7 of this document.
In the review by Hemilä (2011), the estimation of means (SDs) for 5 studies required the imputation of individual duration for certain patients. … Imputing cold duration for patients who had missing duration involves a substantial subjective element in a calculation. … it is not plausible that Singh and Das (2013) independently reached mean (SD) values for both zinc and placebo groups that are identical to 3 digit accuracy with the means (SDs) imputed 2 years earlier by Hemilä (2011), compare the tables on pp. 6 and 7 of this document.
(Hemilä also exhibits whole sentences lifted from his earlier review, in addition to many gripes with the Cochrane authors’ analysis.)
As someone who just wants his cold to get better, though, the important part is that Hemilä’s original paper—Zinc Lozenges May Shorten the Duration of Colds: A Systematic Review—also found that zinc (acetate lozenges of >75mg) shorten cold duration. Nobody seems to disagree about the conclusions; the retraction was just about the ethics.
So the next time a well-meaning relative forces echinacea or Vitamin C1 on your cold (or, God forbid, Airborne), you can reject it in good conscience and stay on your zinc high horse.
Vitamin C seems to be somewhat effective if you take it regularly before you get sick, but ineffective if you only start taking it once you have symptoms (source). ↩︎