Table 4.

Association of serum 25(OH)D, using season-specific quartiles, with incident liver cancer and chronic liver disease mortality among selected subgroups (presented as ORs and 95% CIs from unconditional logistic regression models)

Incident liver cancerChronic liver disease mortality
OverallQ1Q2Q3Q4PtrendQ1Q2Q3Q4Ptrend
# of cases / controls72/10656/10841/10533/10872/10663/10852/10538/108
Multivariable-adjusteda1.87 (1.12–3.12)1.52 (0.90–2.57)1.12 (0.64–1.96)1.00 (ref)0.0071.69 (1.00–2.88)1.27 (0.75–2.17)1.25 (0.73–2.16)1.00 (ref)0.06
Without diabetes
 # of cases / controls63/10249/10136/10231/10569/10260/10151/10236/105
 Multivariable-adjusteda1.81 (1.07–3.07)1.55 (0.90–2.66)1.09 (0.62–1.93)1.00 (ref)0.011.71 (1.00–2.95)1.30 (0.75–2.25)1.26 (0.72–2.21)1.00 (ref)0.06
Alcohol, <11.3 g/dayb,cPinteraction = 0.41Pinteraction = 0.03
 # of cases / controls23/5325/5120/5312/4713/5310/515/5314/47
 Multivariable-adjusteda1.49 (0.65–3.41)1.93 (0.85–4.39)1.40 (0.60–3.28)1.00 (ref)0.290.80 (0.32–2.00)0.61 (0.23–1.64)0.31 (0.10–0.98)1.000.86
Alcohol, ≥11.3 g/dayb,c
 # of cases / controls42/4829/4918/4819/5851/4848/4940/4821/58
 Multivariable-adjusteda2.13 (1.06–4.28)1.44 (0.69–3.00)0.87 (0.39–1.93)1.00 (ref)0.012.64 (1.32–5.27)2.07 (1.04–4.11)2.16 (1.07–4.35)1.00 (ref)0.01
Among the subset with information on hepatitis B and Cd
 # of cases / controls63/9247/8634/8723/8265/9260/8650/8733/82
 Multivariable-adjusteda2.06 (1.14–3.73)1.74 (0.94–3.22)1.15 (0.60–2.18)1.00 (ref)0.0051.60 (0.89–2.87)1.38 (0.77–2.47)1.32 (0.73–2.38)1.00 (ref)0.13
HBV- and HCV-e
 # of cases / controls55/8541/8327/7718/7762/8552/8348/7731/77
 Multivariable-adjusteda2.33 (1.23–4.41)1.81 (0.93–3.51)1.25 (0.62–2.54)1.00 (ref)0.0031.75 (0.96–3.20)1.28 (0.69–2.35)1.48 (0.80–2.75)1.00 (ref)0.12
Lag analyses, ≥5 years
 # of cases / controls64/10448/10536/10430/10450/10450/10542/10431/104
 Multivariable-adjusteda1.88 (1.10–3.21)1.40 (0.81–2.44)1.06 (0.59–1.89)1.00 (ref)0.0091.53 (0.86–2.73)1.31 (0.74–2.33)1.29 (0.72–2.31)1.00 (ref)0.17
Lag analyses, ≥10 years
 # of cases / controls39/9035/9629/9225/10031/9034/9625/9221/100
 Multivariable-adjusteda1.60 (0.87–2.93)1.32 (0.72–2.43)1.22 (0.64–2.30)1.00 (ref)0.131.58 (0.80–3.12)1.36 (0.70–2.64)1.29 (0.64–2.59)1.00 (ref)0.20
Lag analyses, ≥15 years
 # of cases / controls21/7026/8319/8419/9413/7022/8314/8412/94
 Multivariable-adjusteda1.55 (0.75–3.19)1.50 (0.76–2.99)1.17 (0.57–2.44)1.00 (ref)0.171.44 (0.58–3.57)1.80 (0.79–4.11)1.32 (0.55–3.19)1.00 (ref)0.29
  • aAdjusted for age, date of blood draw, BMI, history of diabetes, number of years smoked, and daily intake of alcohol and coffee.

  • bMedian alcohol intake in controls was 11.3 g/day.

  • cAlcohol information was not available in 14 liver cancer cases, 23 chronic liver disease deaths, and 20 controls.

  • dHepatitis B and C information was available for 167 liver cancer cases, 208 chronic liver disease deaths, and 347 controls.

  • e147 men diagnosed with incident liver cancer, 193 men who died from chronic liver disease, and 322 controls were negative for both HBV and HCV.