Table 2.

Estimates of the association between prediabetic conditions and bleeding on probing

(a) Using logistic regression models (Outcome: BOP < median vs BOP ≥ median)
Independent variable (separate models)	Age adjusted OR (95% CI)	OR (95% CI)†
Fasting glucose (10 mg/dL)‡	1.4 (0.8 –2.3)	1.5 (0.9 –2.6)
1-h Glucose (10 mg/dL)§	1.1 (1.0 –1.3)	1.1 (1.0 –1.3)*
2-h Glucose (10 mg/dL)¶	1.1 (0.9 –1.3)	1.2 (0.8 –1.4)
IFG (yes/no)	4.0 (1.0 –16.4)	5.5 (1.2 –25.3)*
IGT (1-h)	3.3 (1.3 –8.8)	3.7 (1.3 –10.6)*
IGT (2-h)	1.2 (0.2 –5.7)	1.3 (0.2 –7.3)
Fasting insulin	1.0 (1.0 – 1.2)	1.0 (1.0 –1.1)
HOMA	1.1 (0.9 –1.4)	1.2 (0.9 –1.5)
Prediabetes (yes/no)	2.1 (0.8 – 8.3)	3.6 (1.0 –13.2)*

(b) Using ANCOVA (multiple linear regression or generalized linear model) (Outcome: Percentage of bleeding: continuous form)
Independent variable (separate models)	Age adjusted least squares means ± SE	Least squares means ± SE†
Normal (No IFG)	27.0 ± 2.4	26.8 ± 2.3*
IFG	43.2 ± 5.9	43.8 ± 6.0
Normal GT (1-h)	27.2 ± 2.7	27.0 ± 2.7
IGT (1-h)	33.9 ± 4.0	34.1 ± 4.1
Normal GT (2-h)	29.4 ± 2.3	29.1 ± 2.3
IGT (2-h)	26.8 ± 8.3	29.8 ± 8.3
Normal insulin: (<14)	29.5 ± 3.3	28.2 ± 3.5
High insulin: (≥14)	29.0 ± 3.1	30.1 ± 3.3
Normal HOMA (<3)	30.9 ± 3.3	30.1 ± 3.4
High HOMA (≥3)	27.8 ± 3.1	28.4 ± 3.2
Normal (no IFG, no IGT 2-h)	27.4 ± 2.5	27.0 ± 2.4*
Prediabetics	37.7 ± 5.2	39.0 ± 5.3

^*P-value ≤ 0.05.

^†Adjusted for age, sex, smoking status (never, former, current), alcohol consumption (current vs never or former), waist circumference (in category), and number of missing teeth.

^‡Fasting glucose.

^§1-h glucose.

^¶2-h glucose in 10 mg/dL increments.

ANCOVA, analysis of covariance; BOP, bleeding on probing; CI, confidence interval; GT, glucose tolerance; HOMA, homeostasis model assessment; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; OR, odds ratio; SE, standard error.