Skip to main content

Table 2 Multivariate analyses for overall survival and progression-free survival

From: Prior oral contraceptive use in ovarian cancer patients: assessing associations with overall and progression-free survival

Model Oral contraceptive variable N (events) for overall survival Adjusted HR (95 % CI) for overall survival P-Value N (events) for progression-free survival Adjusted HR (95 % CI) for progression – free survival P-Value
All variablesa Ever vs Never Users 880 (393) 0.82 (0.66, 1.03) 0.09 821 (481) 0.78 (0.64, 0.96) 0.02
Duration of use       
1–48 months vs never 857 (383) 0.91 (0.70, 1.19) 0.27 800 (467) 0.71 (0.56, 0.91) 0.03
>48 months vs never   0.80 (0.61, 1.05)    0.83 (0.66, 1.06)  
Duration of use as a continuous variable 857 (383) 0.998 (0.996, 1.00) 0.06 800 (467) 0.999 (0.997, 1.001) 0.22
Only the Statistically Significant Univariate Variablesb Ever vs Never Users 950 (429) 0.89 (0.72, 1.10) 0.28 879 (520) 0.80 (0.66, 0.98) 0.03
Duration of use       
1–48 months vs never 926 (418) 1.01 (0.78, 1.30) 0.35 858 (506) 0.74 (0.59, 0.94) 0.04
>48 months vs never   0.84 (0.65, 1.10)    0.86 (0.68, 1.08)  
Duration of use as a continuous variable 926 (418) 0.998 (0.996, 1.000) 0.06 858 (506) 0.999 (0.997, 1.001) 0.17
Only the Statistically Significant Univariate Variables, Excluding Tumor Gradec Ever vs Never Users 1204 (511) 0.95 (0.78, 1.15) 0.57 1111 (619) 0.84 (0.70, 0.996) 0.046
Duration of use 1173 (497)   0.80 1083 (601)   0.052
1–48 months vs never   1.00 (0.79, 1.27)    0.77 (0.62, 0.95)  
>48 months vs never   0.93 (0.74, 1.18)    0.89 (0.72, 1.09)  
Duration of use as a continuous variable 1173 (497) 0.999 (0.997, 1.001) 0.19 1083 (601) 0.999 (0.998, 1.000) 0.15
  1. aAdjusted for all variables from Table 1 except for age at menarche, age at menopause, and platinum-based chemotherapy within 3 months, all of which were missing in approximately 450 patients
  2. bStatistically significant variables in univariate analyses include tumor stage, tumor type, tumor grade, debulking status after surgery, age at cancer diagnosis, and number of live births
  3. cTumor grade was excluded because it was missing in approximately 300 patients
  4. NOTE: When age was excluded, all the oral contraceptive use models reached statistical significance (p < 0.05), except in the progression-free survival model that used continuous duration of oral contraceptive use as the key variable