Table 3. Impact of DII and selected variables.

	DII quartiles	Case/control	OR (95% CI)a	Adjusted OR (95% CI)b
Head and neck + Esophagus
	1 (−4.31 – −1.00)	174/771	1.00 (Reference)	1.00 (Reference)
	2 (−1.00 – −0.11)	243/770	1.41 (1.13–1.76)	1.33 (1.04–1.69)
	3 (0.11–0.58)	278/770	1.62 (1.31–2.01)	1.39 (1.10–1.76)
	4 (0.58–2.02)	333/770	1.96 (1.58–2.43)	1.73 (1.37–2.20)
	P for trend		<0.001	<0.001
Head and neck
	1 (−4.08 – −1.13)	96/443	1.00 (Reference)	1.00 (Reference)
	2 (−1.13 – −0.12)	129/447	1.35 (1.00–1.81)	1.33 (0.97–1.82)
	3 (−0.12 − 0.52)	168/449	1.76 (1.33–2.35)	1.57 (1.17–2.13)
	4 (0.52–1.99)	202/446	2.16 (1.63–2.87)	1.92 (1.42–2.59)
	P for trend		<0.001	<0.001
Esophagus
	1 (−4.31 – −1.03)	78/328	1.00 (Reference)	1.00 (Reference)
	2 (−1.03 – −0.11)	114/323	1.50 (1.08–2.10)	1.11 (0.99–1.25)
	3 (0.11–0.57)	110/321	1.45 (1.04–2.02)	1.42 (1.29–1.57)
	4 (0.57–2.02)	131/324	1.72 (1.24–2.38)	1.71 (1.54–1.90)
	P for trend		0.003	0.07

^aOdds ratios were estimated by conditional logistic models not adjusted for any covariates.

^bOdds ratios were estimated by conditional logistic models adjusted for smoking, ethanol consumption, flushing phenotype, teeth and occupation group.