Skip to main content
. 2018 Sep 17;13(9):e0203926. doi: 10.1371/journal.pone.0203926

Table 4. Association of tooth loss to incident liver cancer and chronic liver disease (CLD) mortality by body mass index in the Nutrition Interventions Cohort, Linxian, China, 1986–2016.

    < 23.0 kg/m2 ≥ 23.0 kg/m2
      Liver cancer incidence CLD mortality   Liver cancer incidence CLD mortality
Patient characteristic Person-yrs N HRa 95% CI N HRa 95% CI Person-yrs N HRa 95% CI N HRa 95% CI
Ever lost a tooth                            
  No 127,833 64 1.00 referent 62 1.00 referent 59,789 23 1.00 referent 30 1.00 referent
  Yes 341,592 209 0.99 (0.73, 1.35) 228 0.99 (0.73, 1.34) 122,755 74 1.43 (0.87, 2.37) 88 1.04 (0.66, 1.62)
Number of teeth lost                            
  0 127,833 64 1.00 referent 62 1.00 referent 59,789 23 1.00 referent 30 1.00 referent
  1–3 57,397 28 0.88 (0.56, 1.38) 27 0.86 (0.54, 1.35) 24,492 15 1.61 (0.84, 3.10) 6 0.45 (0.19, 1.09)
  4–6 76,203 45 1.06 (0.72, 1.56) 36 0.81 (0.54, 1.24) 30,922 13 1.06 (0.53, 2.12) 21 1.17 (0.66, 2.07)
  7–10 63,315 34 0.91 (0.59, 1.40) 51 1.26 (0.86, 1.86) 25,253 16 1.57 (0.81, 3.06) 19 1.15 (0.63, 2.09)
  11–20 71,245 47 1.05 (0.70, 1.58) 51 1.03 (0.69, 1.53) 22,404 14 1.45 (0.71, 2.98) 25 1.48 (0.83, 2.66)
  21–32 73,432 55 1.10 (0.73, 1.67) 63 1.11 (0.74, 1.66) 19,684 16 1.78 (0.86, 3.69) 17 1.04 (0.53, 2.04)
  p-value for trendb     0.48     0.39       0.22     0.65  
Age-specific tooth lossc                            
  I 153,271 84 1.00 referent 81 1.00 referent 69,157 30 1.00 referent 39 1.00 referent
  II 93,436 65 1.11 (0.79, 1.54) 65 1.06 (0.76, 1.47) 36,972 27 1.47 (0.86, 2.51) 30 1.11 (0.68, 1.81)
  III 96,806 51 0.98 (0.69, 1.39) 67 1.21 (0.87, 1.68) 36,375 13 0.93 (0.48, 1.79) 20 0.94 (0.54, 1.63)
  IV 125,912 73 1.14 (0.82, 1.57) 77 1.10 (0.80, 1.52) 40,039 27 1.71 (1.00, 2.92) 29 1.20 (0.73, 1.97)
  p-value for trendb     0.58     0.42       0.14     0.62  

HR = hazard ratio; CI = confidence interval

aEstimated using cox-proportional hazards regression adjusted for age (age (at randomization) and age2 or loess smoothed age-specific predicted quartiles) sex, body mass index, education, cigarette smoking, alcohol use, and trial membership

bp-value of test for trend using either the number of teeth loss or the quartile of age-specific tooth loss as a continuous covariate

cAge-specific quartiles of tooth loss were estimated using loess regression