Aim: This systematic review and meta-analysis aimed to critically appraised data from comparable studies leading to quantitative assessment of any independent association between use of oral smokeless tobacco in any form, of betel quid without tobacco and of areca nut with incidence of oral cancer in South Asia and the Pacific.

Methods: Studies (case control and/or cohort) were identified by searching Pub Med, CINAHL and Cochrane databases through June 2013 using the keywords oral cancer: chewing tobacco; smokeless tobacco; betel quid; betel quid without tobacco; areca nut; Asia, the Pacific and the reference lists of retrieved articles. A random effects model was used to compute adjusted summary OR(RE) for the main effect of these habits along with their corresponding 95% confidence intervals. To quantify the impact of between-study heterogeneity on adjusted main-effect summary OR(RE), Higgins' H and I2 statistics along with their 95% uncertainty intervals were used. Funnel plots and Egger's test were used to evaluate publication bias.

Results: Meta-analysis of fifteen case-control studies (4,553 cases; 8,632 controls) and four cohort studies (15,342) which met our inclusion criteria showed that chewing tobacco is significantly and independently associated with an increased risk of squamous-cell carcinoma of the oral cavity (adjusted main-effect summary for case- control studies OR(RE) = 7.46; 95% CI = 5.86-9.50, P<0.001), (adjusted main-effect summary for cohort studies RR = 5.48; 95% CI = 2.56-11.71, P<0.001). Furthermore, meta-analysis of fifteen case control studies (4,648 cases; 7,847 controls) has shown betel quid without tobacco to have an independent positive association with oral cancer, with OR = 2.82 (95% CI = 2.35-3.40, P<0.001). This is presumably due to the carcinogenicity of areca nut. There was no significant publication bias.

Conclusion: There is convincing evidence that smokeless (aka chewing) tobacco, often used as a component of betel quid, and betel quid without tobacco, are both strong and independent risk factors for oral cancer in these populations. However, studies with better separation of the types of tobacco and the ways in which it is used, and studies with sufficient power to quantify dose-response relationships are still needed.