A meta-study of 31 placebo-controlled trials of the gastric acid secretion inhibitor drug Cimetidine in the treatment of gastric or duodenal ulcers found that placebo treatments, in many cases, were as effective as active drugs: of the 1692 patients treated in the 31 trials, 76% of the 916 treated with the drug were "healed", and 48% of the 776 treated with placebo were "healed".[78] (It is now known gastic secretion is irrelevant since most ulcers are due to the the bacterium Helicobacter pylori).) These results were confirmed by the direct post-treatment endoscopy. It was also found that German placebos were "stronger" than others; and that, overall, different physicians evoked quite different placebo responses in the same clinical trial (p.15). Moreover, that in many of these trials the gap between the active drugs and the placebo controls was "not because [the trials' constituents] had high drug effectiveness, but because they had low placebo effectiveness" (p.13).
In some trials, placebos were effective in 90% of the cases, whilst in others the placebos were only effective in 10% of the cases. It was argued that "what is demonstrated in [these] studies is not enhanced healing in drug groups, but reduced healing in placebo groups" (p.14). It was also noted the results of two studies (one conducted in Germany, the other in Denmark), which examined "ulcer relapse in healed patients" showed that the rate of relapse amongst those "healed" by the active drug treatment was five times that of those "healed" by the placebo treatment (pp.14–15).
