[Meta-analysis of the prevalence of peptic ulcer in relation to tobacco consumption].

The reported studies of the association between smoking and the incidence of peptic ulcer have had nonconclusive results. Therefore, we have carried out a quantitative review including and evaluating all the factors related to that association. To this end, the techniques of meta-analysis suggested by Rosenthal have been used, and it has been found that the type of study design acted as a moderating variable. A correlation of 0.22 was found for studies with a case-control design, and 0.07 for studies with a sampling survey design. The small power of this association can be attributed to the great variety of factors which, in addition to smoking, may play a role in the etiology and pathogenesis of peptic ulcer.