Background Numerous twin studies on smoking behavior have shown that genetic and environmental factors play significant and approximately equal roles in the determination of smoking initiation (SI) and smoking persistence (SP). However, estimates of heritability (h2), shared (c2) and unique environmental effects (e2) from the literature display considerable variability for SI and SP, due most probably to differences in statistical analysis models, age, gender, sample size, origin of cohorts and measurement of smoking behavior.
Methods A systematic literature search identified six studies for SI and 10 studies for SP. Data from these studies were obtained and re-analysed by meta-analytical techniques.
Results For SI, our results indicate that the parameters h2, c2 and e2 are (mean ± SEM): 0.37 ± 0.04, 0.49 ± 0.04 and 0.17 ± 0.03 in male adults, and 0.55 ± 0.04, 0.24 ± 0.06 and 0.16 ± 0.01 in female adults, respectively. These values were weighted by a combination of original estimates of variance from studies reporting variances plus estimated variances from studies where variances were not reported (called the combined variance method). Using the same approach for SP, we found that the parameters h2, c2 and e2 weighted by the combined variance method for the phenotype are (mean ± SEM): 0.59 ± 0.02, 0.08 ± 0.04 and 0.37 ± 0.03 in male adults, and 0.46 ± 0.12, 0.28 ± 0.08 and 0.24 ± 0.07 in female adults, respectively.
Conclusions Our results indicate that genetic factors play a more significant role for SI but a less significant role for SP in female adults compared to male adults. Significant gender difference was also detected in shared environmental factors for SI and SP. However, no significant gender difference was detected for e2 for either phenotype. These findings suggest that genetic and environmental factors may contribute differently to the determination of smoking initiation and persistence in male and female smokers.