Purpose: Although drinking and smoking have been associated with gamma-glutamyl transferase (GGT) levels, studies on the effects of smoking and drinking on GGT levels are scarce. The objective of this study was to assess the individual and combined effects of smoking, drinking on GGT levels in the Shaanxi province of Northwest China. Participants and Methods: A questionnaire survey was conducted in a population that underwent health examination at the First Affiliated Hospital of Xi’an Jiaotong University and included employees of enterprises or public institutions. The survey was used to collect the baseline characteristics, smoking status, and drinking status of the participants. This information was collected from January 2019 to December 2019. Data related to the physical examinations were exported using the hospital information system (HIS). A linear regression model was employed to explore the effects of smoking and drinking on GGT levels. The restricted cubic spline model was applied to assess the dose–response relationship between amount of smoking, alcohol consumption and GGT levels. Results: A total of 10,177 participants were included in the study. Linear regression indicated that smoking (β=3.37, 95% CI: 2.57–4.17) and drinking (β=5.55 L, 95% CI: 4.40–6.71) individually, and collectively (β=9.30, 95% CI: 7.83–10.76) had a positive effect on GGT levels. The restricted cubic spline presented a linear dose–response relationship between the amount of daily smoking and GGT levels (P for non-linearity=0.148, P for overall association <0.001, OR=2.49, 95% CI: 1.27–4.90), and that between weekly alcohol consumption and GGT levels (P for non-linearity=0.231, P for overall association <0.001, OR=4.79, 95% CI: 1.72–13.32). In case of females, stratified analysis showed that in comparison to the reference group, only current drinkers had a significant effect on GGT levels (OR=3.37, 95% CI: 0.19–6.55). Conclusion: Smoking and drinking have a dose-dependent and a synergistic effect on GGT levels. They should be controlled concurrently, especially among males.