Skip to main content
. 2011 Jun 13;9:42. doi: 10.1186/1477-7525-9-42

Table 4.

Multivariate association between Oral Health-Related Quality of Life* and cohort attributes

Attributes Oral Health-Related Quality of Life (Adjusted Odds Ratios)

Speaking Swallowing Chewing Social Pain
Characteristics
 Males 1 1 1 1
 Females 0.69[p = 0.00] 1.26[p = 0.00] 1.20[p = 0.00] 1.25[p = 0.00]
Age (years)
 15-29 1 1 1 1 1
30-49 1.52 [p = 0.00] 1.53[p = 0.00] 1.77[p = 0.00] 1.09[p = 0.00] 0.92[p = 0.00]
50+ 2.02 [p = 0.00] 2.22[p = 0.00] 2.85[p = 0.00] 0.77[p = 0.00] 0.99[p = 0.90]
Income (Baht per month)
 < 3000 1.40 [p = 0.00] 1.56[p = 0.01] 1.02 [p = 0.57] 1.30[p = 0.00] 1.31[p = 0.00]
 3001-7000 1.16 [p = 0.04] 1.70[p = 0.00] 1.09 [p = 0.00] 1.26[p = 0.00] 1.20[p = 0.00]
 7001-10000 1.07 [p = 0.45] 1.39[p = 0.01] 1.03 [p = 0.38] 1.17[p = 0.00] 1.13[p = 0.00]
 > 10000 1 1 1 1 1
Education
 Up to high school 0.82[p = 0.00] 1.04[p = 0.72] 1.09[p = 0.00] 0.99[p = 0.79]
 Diploma/certificate 0.98[p = 0.80] 1.33[p = 0.03] 1.18 [p = 0.00] 1.07[p = 0.08]
 University degree 1 1 1 1
Household assets (Baht)
 0-30,000 1.20[p = 0.00] 1.24[p = 0.00] 1.27[p = 0.00]
 30,001-60,000 1.14[p = 0.00] 1.16[p = 0.00] 1.17[p = 0.00]
 > 60,000 1 1 1
Lifecourse urbanization
 Lifetime rural residents 1 1 1 1
 Rural-urban residents 1.10[p = 0.09] 1.07[p = 0.00] 1.06[p = 0.00] 1.05[p = 0.07]
 Lifetime urban residents 1.40[p = 0.01] 1.01[p = 0.64] 1.07[p = 0.00] 0.93[p = 0.03]
Health behaviours
Smoking
 Not a regular smoker 1 1 1 1 1
 Ever smoker 1.30[p = 0.00] 1.31[p = 0.04] 1.36[p = 0.00] 1.25[p = 0.00] 1.25[p = 0.00]
 A regular smoker 1.32[p = 0.00] 1.44[p = 0.01] 1.51[p = 0.00] 1.39[p = 0.00] 1.40[p = 0.00]
Alcohol drinking
 Not a regular drinker 1 1 1 1
 A regular drinker 1.23 [p = 0.04] 1.26[p = 0.00] 1.24[p = 0.00] 1.27[p = 0.00]
Number of teeth
 Remaining teeth ≥ 20 1 1 1 1 1
 Remaining teeth < 20 6.43[p = 0.00] 6.27[p = 0.00] 3.26[p = 0.00] 2.61[p = 0.00] 1.54[p = 0.00]

Table displays results of one model for each of the five OHRQoL dimensions. Each model is based on backward stepwise multivariate logistic regression predicting the yes-no outcome for each of the five dimensions (Stata 10). Variables shown in each model were those retained as statistically significant at p < 0.05.