Skip to main content

Table 1 Characteristics of the Project Cohort

From: Use of name recognition software, census data and multiple imputation to predict missing data on ethnicity: application to cancer registry records

   Cases Cases with missing ethnicity following linkage with HES records (% of cases) Chi2 statistic
(P-value)
Site Lower GI 24 446 4112(17)  
  Breast 28 795 6029(21)  
  Lung 24 060 5660(24)  
  Prostate 23 716 8814(37)  
  Upper GI 10 677 1727(16) 3500(p < 0.001)
Year of diagnosis 2001 15 102 4118(27)  
  2002 15 523 3840(25)  
  2003 15 731 3772(24)  
  2004 16 162 3681(23)  
  2005 16 317 3632(22)  
  2006 16 458 3470(21)  
  2007 16 401 3829(23) 206(p < 0.001)
Deprivation 1 (most deprived) 26 738 5149(19)  
(Income Domain of 2 22 104 4856(22)  
Index of Multiple 3 22 759 5425(24)  
Deprivation 2007) 4 22 465 5942(26)  
  5 (least deprived) 17 628 4970(28) 6300(p < 0.001)
Age < 40 1771 251(14)  
  40-49 5622 892(16)  
  50-59 15 338 2933(19)  
  60-69 27 759 5924(21)  
  70-79 35 258 8538(24)  
  80+ 25 946 7804(30) 1100(p < 0.001)
Sex Male 59 592 15 454(26)  
  Female 52 102 10 888(21) 391(p < 0.001)
Death Certificate No 106 217 23 577(22)  
Only registration Yes 5477 2765(50) 2300(p < 0.001)
Ever seen privately No 106 566 23 113(22)  
(cancer was diagnosed or treated outside the free National Health Service at least on one occasion) Yes 5128 3229(63) 4600(p < 0.001)
Surgery No 58 875 18 344(31)  
  Yes 52 819 7998(15) 4000(p < 0.001)
Radiotherapy No 73 520 19 009(26)  
  Yes 38 174 7333(19) 616(p < 0.001)
Chemotherapy No 93 778 24 650(26)  
  Yes 17 916 1692(9) 2400(p < 0.001)
Screen detected No 22 900 5012(22)  
breast cancer* Yes 5895 1017(17) 61(p < 0.001)
HES-linked No 19 694 19 694(100)  
  Yes 92 000 6648(7)  
Number of admissions 0 19 694 19 694(100)  
(includes non-cancer admissions) 1 8012 2071(26)  
  2 10 261 1414(14)  
  3 10 523 985(9)  
  4 9332 562(6)  
  5+ 53 872 1616(3) 6300(p < 0.001)**
  1. * Comparison of breast cancer cases who were and were not detected by population screening.
  2. ** Chi-square test excludes cases with no admissions as, by definition, none have ethnicity recorded.