Table 3. Association between oral lesions and COVID-19 infection and differences due to effect modification by smoking status using multi-level binary logistic regression.

Reported presence of oral lesions	All AOR (95% CI)	Smokers AOR (95% CI)	Non-smokers AOR (95% CI)	p of difference due to effect modification
Stained teeth	0.95 [0.71–1.26]	1.24 [0.74–2.07]	1.00 [0.79–1.26]	0.02*
Dental caries	0.99 [0.79–1.23]	1.13 [0.66–1.92]	1.03 [0.82–1.29]	0.59
Gingival inflammation	1.00 [0.79–1.25]	1.21 [0.71–2.07]	1.02 [0.81–1.28]	0.29
Dry mouth	1.03 [0.82–1.29]	1.26 [0.74–2.15]	1.03 [0.82–1.31]	0.09
Change in taste	1.16 [0.93–1.46]	1.13 [0.65–1.98]	1.22 [0.96–1.54]	0.86
Leukoplakia	1.05 [0.83–1.33]	1.06 [0.60–1.89]	1.06 [0.83–1.35]	0.90
Burns and ulcers	1.01 [0.80–1.28]	1.00 [0.56–1.79]	1.03 [0.81–1.31]	0.95
Hairy tongue	1.03 [0.81–1.30]	1.06 [0.60–1.90]	1.03 [0.81–1.31]	0.85

Notes:

An asterisk (*) indicates statistically significant at p < 0.05. Overall model was adjusted for gender, having medical problems and smoking status with country entered as random effect variable.

AOR, adjusted odds ratio; CI, confidence interval.