Previous studies have evaluated the association between cigarette smoking and incidence of non-Hodgkin lymphoma (NHL) with inconclusive results. Our main objective was to evaluate this relationship using a meta-analysis of observational studies. A literature search was undertaken through October 2011 looking for pertinent studies. Seven prospective cohort and 17 case-control studies were included in this meta-analysis. Outcomes were calculated using the random-effects model and are reported as odds ratio (OR). Meta-regression was used to evaluate the dose-response of intensity and duration of smoking in NHL incidence. Our study found an OR of 1.40 (95% confidence interval [CI] 1.14-1.73; p = 0.001) in current female smokers seen only in case-control studies. No increased odds of NHL was seen in men. There was no association between smoking and the most common NHL subtypes, with the exception of a statistical trend toward a higher incidence of T-cell lymphoma. In conclusion, there appears to be an increase in the odds of NHL in current female smokers.