Skip to main content

Table 5 Associations between candidate correlating factors and percent change in CA125 levels among postmenopausal women (N = 1116) by univariable and multivariable linear regressions

From: Factors associated with serum CA125 level in women without ovarian cancer in the United States: a population-based study

Variables Univariable linear regressionsa Multivariable linear regressionsb  
Effect estimate CA125 change P value Effect estimate Standard error CA125 change P value
Smoking
 Current −0.096 −9.15% 0.258     
 Former −0.014 −1.39% 0.835     
 None ref       
EtOH
  > 2 drinks/day −0.178 −16.31% 0.076 −0.267 0.129 −23.43% 0.055
 1–2 drinks/day 0.021 2.12% 0.545 0.050 0.072 5.13% 0.499
 None ref       
Caffeine −0.0002 −0.02% 0.034 −0.00002 0.0002 −0.002% 0.887
Age of menarche
  > 12 0.020 2.02% 0.687     
  ≤ 12 ref       
Breastfeedc 0.020 2.02% 0.689     
OCP duration
 10+ years 0.009 0.90% 0.914     
 5–10 years −0.002 −0.20% 0.978     
 2–5 years 0.070 7.25% 0.543     
  < 2 years ref       
CRP 0.057 5.87% 0.074 −0.028 0.069 −2.76% 0.696
Ferritin −0.0003 −0.03% 0.138 0.00062 0.0006 0.06% 0.473
Osteoporosis −0.017 −1.69% 0.750     
Osteoarthritis −0.048 −4.69% 0.333     
Hysterectomy −0.084 −8.06% 0.050 −0.095 0.120 −9.06% 0.441
Oophorectomy −0.104 −9.88% 0.061 −0.050 0.102 −4.88% 0.628
Salpingectomy 0.044 4.50% 0.381     
Endometriosis −0.205 −18.54% 0.126 −0.068 0.112 −6.57% 0.554
Fibroids −0.242 −21.49% 0.046 −0.161 0.132 −14.87% 0.242
CADd 0.131 14.00% 0.134 0.249 0.115 28.27% 0.047
HLD 0.054 5.55% 0.364     
Non-ovarian cancer −0.048 −4.69% 0.299     
  1. EtOH alcohol, CHC current hormonal contraception, OCP oral contraceptive pill, CRP C reactive protein, CAD coronary artery disease, HLD hyperlipidemia
  2. a All univariable regressions were adjusted for baseline variables including age, race/ethnicity, BMI, parity and current hormonal replacement therapy (HRT) use
  3. b Multivariable regressions included the baseline variables (age, race/ethnicity, BMI, parity, HRT) and eligible variables from univariable analyses with cut-off p value < 0.25
  4. c For the “Breastfeed” variable, the reference for parity is one instead of none
  5. d This variable demonstrated significant association with CA125 level in the multivariable linear regression model