. 2025 Aug 14;84:44368. doi: 10.2340/aos.v84.44368

Table 4.

Multiplicative and additive interaction of ANRIL genotype and obesity on periodontitis under recessive model.

Groups	N	OR (95% CI), p	n	OR (95% CI), p	Multiplicative interaction^a	Additive interaction measurements^b
Groups	N	OR (95% CI), p	n	OR (95% CI), p	p	RERI OR (95%CI), p	AP OR (95%CI), p
Stage II vs. no periodontitis/stage I		GT/TT n = 2,435		GG n = 646
No obesity	1,867	1.00	482	0.93 (0.75, 1.15), 0.494	0.238	0.35 (–0.17, 0.87), 0.192	0.26 (–0.06, 0.57), 0.117
Obesity	568	1.12 (0.90, 1.40), 0.314	164	1.36 (0.95, 1.94), 0.097	0.238	0.35 (–0.17, 0.87), 0.192	0.26 (–0.06, 0.57), 0.117
Stage III–IV vs. no periodontitis/stage I		GT/TT n = 1,566		GG n = 386
No obesity	1,190	1.00	275	0.84 (0.57, 1.24), 0.381	0.005	1.51 (0.26, 2.77), 0.018	0.61 (0.35, 0.87), <0.001
Obesity	376	1.12 (0.80, 1.58), 0.502	111	2.47 (1.48, 4.12), 0.001	0.005	1.51 (0.26, 2.77), 0.018	0.61 (0.35, 0.87), <0.001

OR: odds ratio; CI: confidence interval; HDL-C: high-density lipoprotein cholesterol; hs-CRP: high-sensitivity C-reactive protein; RERI: relative excess risk due to interaction; AP: attributable proportion. Bold values denote statistical significance at the p < 0.05 level.

Multiplicative interaction: binominal logistic regression model included ANRIL genotype, obesity, product term ANRIL genotype*obesity, age, sex, education, smoking, toothbrushing frequency, low HDL-C, and hs-CRP.

Interaction on an additive scale means that the combined effect of two exposures is larger (or smaller) than the sum of the individual effects of the two exposures. ORs are adjusted for age, sex, education, smoking, toothbrushing frequency, low HDL-C, and hs-CRP. RERI and AP were calculated with the formulas: RERI = OR11 – OR10 – OR01 + 1, AP = RERI/OR11. The delta method introduced by Hosmer and Lemeshow was used to calculate the 95% CI of RERI and AP.