The objective of this meta-analysis is to assess the data regarding changes in herbal cannabis potency over time (from 1970 to 2009). Systematic searches of 17 electronic scientific databases identified studies on this topic, within which 21 case series studies satisfied our inclusion criteria of reporting the mean tetrahydrocannabinol (THC) value per number of samples per year. No language, publication date, publication type or status restrictions were imposed. The study selection and data extraction processes were performed independently but uniformly by two authors, included screening, determination of eligibility and inclusion of the eligible studies in the systematic review, and a meta-analysis of the results on THC content in herbal cannabis samples. We considered papers and not monographic scientific publications, rejecting all studies that were not focused on the subject of this review. Meta-analysis by year was performed on 21 studies containing 75 total mean THC observations from 1979 to 2009 using the random effects model. The results revealed much variability between studies. Further, there was a significant correlation between year and mean THC in herbal cannabis. The combined data indicated the correlation between year and mean THC in herbal cannabis, revealing a temporal trend of increasing potency (5% above the mean THC value in the Poisson regression analysis). The results of the analysis suggest that there has been a recent and consistent increase in cannabis potency worldwide.