Skip to main content
. 2005 Dec 6;94(1):36–42. doi: 10.1038/sj.bjc.6602894

Table 2. Surgeon volume and treatment and histological prognostic factors.

  High (6) (⩾50) Medium (4) (11–49) Low (15) (⩽10) Total (25) Test, P-value
Surgeon volumea, N
 Therapeutic procedures, N (%)         Cusick's test for trend
  BCT 1133 (54) 299 (56) 29 (38) 1461 (54) z=1.15, P=0.25
  Radiotherapy 724 (35) 169 (32) 615 (20) 908 (34) z=−2.6, P=0.014
  Tamoxifen 1129 (54) 211 (39) 21 (28) 1361 (50) z=−7.23, P<0.01
  Chemotherapy 74 (3.5) 9 (1.7) 4 (5.3) 88 (3.3) z=−1.13, P=0.26
           
 Invasive status, N (%)         Wilcoxin's rank sum
  Invasive tumours 1636 (78) 422 (79) 63 (83) 2121 (78) P=0.97
  In situ tumours 448 (21) 112 (21) 12 (16) 572 (21)  
  Missingb 8 (0.38) 2 (0.37) 1 (1.32) 11 (0.41)  
  All tumours 2092 536 76 2704  
           
Invasive tumours onlyc, N (%): 2121 (100)
 Malignant nodesa         Spearman's rank
  None 1221 (75) 312 (74) 37 (59) 1570 (74) correlation
  Few (1–4) 299 (18) 79 (19) 18 (29) 396 (19) P=0.59
  Medium (5–9) 60 (3.7) 14 (3.3) 3 (4.8) 77 (3.6)  
  Lots (⩾10) 18 (1.1) 9 (2.1) 0 27 (1.3)  
  Missingb 38 (2.3) 8 (1.9) 5 (7.9) 51 (2.4)  
           
 Size (mm)a, N 1620 417 60 2097 Spearman's rank
  Median (IQR) 13 (9–18) 13 (9–19) 15 (10.5–22) 13 (9–18) correlation
  Range 1–99 1–99 2–60 1–99 P=0.53
  Missingb 16 (0.8) 5 (0.9) 3 (3.9) 24 (0.9)  
           
           
 Grade          
  1 528 (32) 115 (27) 10 (16) 653 (31) Cusick's test
  2 711 (43) 181 (43) 27 (43) 919 (43) for trend
  3 224 (14) 55 (13) 9 (14) 288 (14) P=0.85
  Missingb 173 (11) 71 (17) 17 (27) 261 (12)  
           
 NPIa         Spearman rank
  Good 936 (57) 220 (52) 19 (30) 1175 (55) Correlation
  Moderate 391 (24) 106 (25) 19 (30) 516 (24) P=0.94
  Poor 87 (5.3) 17 (4.0) 4 (6.4) 108 (5.1)  
  Missingb 222 (14) 79 (19) 21 (33) 322 (15)  

BCT=breast conservation therapy; IQR=interquartile range.

All tumours, N (%): 2704 (100), one patient with missing surgeon ID.

a

Surgeon volume, malignant nodes, size and Nottingham Prognostic Index analysed as continuous variables.

b

Missing values not included in analyses.

c

These data not available for noninvasive tumours.