Table 4.

Absolute and relative socioeconomic inequality in plaque-free, periodontal health, and deep pockets in adults aged 25–54 from the Dutch National Oral Health Survey 1995–2013

Estimate (95% confidence interval)	1995 (n = 917)	2002 (n = 795)	2007 (n = 669)	2013 (n = 702)
Plaque-free
Regression-based absolute effect index a	2.6 (− 1.7 to 6.9)	1.9 (− 2.1 to 6.0)	8.4 (2.1 to 14.7)	8.6 (1.5 to 15.7)
Regression-based relative effect index b	1.308 (0.819 to 2.090)	1.292 (0.674 to 2.477)	1.698 (1.124 to 2.565)	1.561 (1.080 to 2.257)
Periodontal health
Regression-based absolute effect index a	3.7 (− 3.5 to 10.8)	10.0 (1.6 to 18.4)	4.0 (− 3.8 to 11.8)	9.6 (1.8 to 17.4)
Regression-based relative effect index b	1.163 (0.873 to 1.549)	1.516 (1.080 to 2.126)	1.188 (0.852 to 1.656)	1.516 (1.092 to 2.106)
Deep pockets
Regression-based absolute effect index a	− 4.2 (− 7.1 to − 1.3)	− 2.3 (− 5.9 to 1.3)	− 3.1 (− 5.4 to 0.7)	− 2.6 (− 5.6 to 0.3)
Regression-based relative effect index b	0.433 (0.216 to 0.869)	0.730 (0.382 to 1.396)	0.618 (0.325 to 1.176)	0.585 (0.290 to 1.177)

All logistic regression models were adjusted for including age, gender, ethnicity, SES, time since the last dental visit, number of teeth present, and oral hygiene behaviors (toothbrushing frequency, floss usage, and toothpick usage). Low-SES was regarded as reference

^aRegression-based absolute effect index was estimated according to the rate of oral health outcomes. Rate difference was defined as the difference between the predicted probability of the oral health outcome in high-SES group and that in low-SES group. Absolute inequality was considered significant if the 95% confidence interval did not cross zero. Bold indicates P value < 0.05

^bRegression-based relative effect index was estimated according to the adjusted odds ratio of SES for oral health outcomes. Relative inequality was considered significant if the 95% confidence interval did not cross one. Bold indicates P value < .05