Table 1. Characteristic of case-control studies included in the meta-analysis.

Study	Year	Country	Study design	No. of cases	No. of controls	Sex	Outcome ascertainment	Assessment of tooth loss	Follow-up(yrs)	Adjustment for covariates
Demirer T	1990	Turkey	A case-control study	100	200	W and M	Cancers were histologically proven.	Questionnaire	1	Unadjust
Watabe K	1998	Japan	A case-control study	242	484	W and M	Cancers were pathologically confirmed.	Questionnaire	1	Unadjust
Hiraki A	2008	Japan	A Hospital-based case-control study	702	1,404	W and M	Cancers were confirmed by the national cancer registery.	Self-administered questionnaire	4	Adjusted for age, sex, smoking and drinking status (never, former, current), vegetable and fruit intake, BMI, and regular exercise.
Shakeri A	2013	Iran	A case-control study	309	613	W and M	Cancers were histologically proven.	Physical examination by dentists	7	Adjusted for age, ethnicity, education fruit and vegetable use, socioeconomic status, ever opium or tobacco use, and denture use.

NA, not available; M, male; W, female.