Article Text
Abstract
Introduction To compare regional outcomes of Faecal Occult Blood testing(FOBT)screening
Method National Bowel Cancer Audit Programme data from three NHS Hospitals,for colorectal cancers(CRC)in the screening age group(60–74 years)over a 2 year period(August 2011–2013)were linked for their FOBT screening status(BCSP database/Eastern Hub).Patient locations were divided into 4 regions(based on home postcodes):Mansfield(Mfd)postcodes(NG17-21);Newark(Nw)postcodes(NG22-25);Nottingham(Ngh)postcodes(NG1-NG16);Derby(Drby)postcodes(DE1-DE75).Patient/tumour demographics and survival for Interval cancers(IC-screening patients developing cancers within 2 years of a negative FOB test),screen-detected cancers(SDC)and cancers in those who declined the screening programme(DSP)were analysed.All regions and centres involved were in incident rounds of screening.Tumours at and beyond splenic flexure were considered left-sided.1-year mortality was looked at as all patients had a minimum of 1 year follow-up.
Results Of the 496 cancers detected,123 (25%)were IC,219 (44%)were DSP and 154 (31%)were SDC. Demographic parameters were comparable between regions including male: female ratio (Mfd (51:27); Nw (20:18); Ngh (134:88); Drby (102:58):p=0.52, X2=2.22); age (mean) at cancer diagnosis (years) (Mfd(65.9); Nw (65.9); Ngh (66.2); Drby (67.9): p=0.46, X2=4.67). Tumour characteristics including left:right-sided (Mfd(59:19); Nw(27:11); Ngh (168:54); Drby (104:54): p=0.17,X2=5.02) and Duke’s staging Low(A/B): High (C/D) (Mfd(39:39); Nw (24:14); Ngh (109:113); Drby (67:91): p=0.126,X2=5.70) were comparable.Ethnic demographics,deprivation index,d patient follow-up were comparable between the four regions.Mfd region had a higher proportion of IC compared to other regions (Mfd(n=78); IC-25 (32%), DSP-25 (32%), SDC-28 (36%); Nw (n=38); IC-7 (18%), DSP-22 (58%), SDC-9(24%); Ngh (n=222); IC-44 (20%), DSP-111 (50%), SDC-67(30%); Drby (n=158); IC-47(30%), DSP-61 (38%), SDC-50(32%): p= 0.025,X2=14.41 ). Mfd had a significantly higher 1 year mortality compared to other regions (Mfd 13:IC-5.DSP-3.SDC-5; Nw 2: IC-1. DSP-1.SDC-0; Ngh 22: IC-6, DSP-16,SDC-0; Drby 12:IC-7, DSP-5, SDC-0: p= 0.002, X2=20.07). The median time (months) between cancer diagnosis and death was shortest for the Mfd region (Mfd-7; Nw-13.5; Ngh-15; Drby-15.5)
Conclusion Our findings for the first time highlight regional variations in CRC mortality. We cannot explain our findings based on existing literature or on the known behaviour of colorectal cancer. Though there remains the possibility that our findings may be incidental, this also raises the possibility of the existence of hitherto undetected regional variations in colorectal cancer phenotypes.
Disclosure of Interest None Declared
- None