Skip to main content

Table 3 Adjusteda odds ratio of re-excision estimatesb for stage I, II and III breast cancer patients whose initial surgery was BCS

From: Re-excision and survival following breast conserving surgery in early stage breast cancer patients: a population-based study

  Adjusted1 Odds Ratios of Re-excision (95% CI)
  Stage I Stage II Stage III
Total Patients 3476 1840 343
Age at Diagnosis P < 0.001 P = 0.43 P = 0.32
  < 50 1.00 1.00 1.00
 50-59 0.68 (0.54, 0.87) 1.07 (0.80, 1.41) 1.06 (0.57, 1.97)
 60-69 0.58 (0.44, 0.75) 1.09 (0.80, 1.49) 0.84 (0.39, 1.80)
 70-79 0.56 (0.42, 0.76) 0.85 (0.56, 1.29) 0.68 (0.24, 1.90)
 80+ 0.35 (0.21, 0.60) 0.67 (0.37, 1.20) 0.21 (0.04, 1.22)
Geography of Surgery P = 0.004 P = 0.04 P = 0.09
 Calgary 1.00 1.00 1.00
 Edmonton 1.74 (1.18, 2.57) 1.64 (1.09, 2.45) 2.14 (1.07, 4.29)
 Central 2.26 (1.27, 4.03) 1.62 (0.86, 3.05) 4.13 (1.07, 15.9)
 South 1.74 (1.00, 3.04) 1.55 (0.87, 2.75) 2.13 (0.77, 5.89)
 North 2.89 (1.47, 5.68) 2.58 (1.34, 4.97) 3.79 (0.94, 15.3)
Surgeon Volume P = 0.80 P = 0.95 P = 0.16
 Very High (60+) 1.00 1.00 1.00
 High (20-59) 0.96 (0.63, 1.45) 0.99 (0.65, 1.51) 1.40 (0.69, 2.83)
 Medium (13-19) 0.96 (0.58, 1.60) 1.16 (0.67, 2.00) 1.88 (0.71, 5.01)
 Low (5-12) 0.83 (0.42, 1.66) 1.22 (0.59, 2.50) 0.24 (0.04, 1.55)
 Very Low (< 5) 0.51 (0.16, 1.62) 1.26 (0.45, 3.54) 1.62 (0.18, 14.2)
Year of Diagnosis P = 0.42 P = 0.18 P = 0.27
 2002-2005 1.00 1.00 1.00
 2006-2009 1.08 (0.89, 1.32) 0.85 (0.67, 1.07) 0.74 (0.43, 1.26)
Tumor Size   P < 0.001 P < 0.001
 T0/T1 1.00 1.00 1.00
 T2 1.39 (1.04, 1.86) 1.40 (0.15, 12.8)
 T3 6.33 (2.82, 14.2) 1.33 (0.12, 14.2)
 T4 1.83 (0.17, 19.9)
Nodal Status   P < 0.001 P = 0.75
 N0 1.00 1.00 1.00
 N1 1.74 (1.30, 2.33) 1.40 (0.15, 12.8)
 N2 1.33 (0.12, 14.2)
 N3 1.83 (0.17, 19.9)
RE Parameter Estimates (SDs)
 Hospital 0 (−) 0 (−) 0.15 (0.00, 9.09)
 Surgeon 0.43 (0.30, 0.62) 0.39 (0.25, 0.61) 0 (−)
  1. Abbreviations: BCS breast conserving surgery, CI confidence interval, RE random effects, SD standard deviation
  2. aAdjusted for the variables in the table
  3. bMulti-level logistic regression with hospitals and surgeons as crossed random effects