Recent trends in incidence of nonmelanoma skin cancers in the East of Scotland, 1992-2003

Background Historically, ascertainment of nonmelanoma skin cancer (NMSC) by cancer registries in the U.K. has been shown to be incomplete in several studies. However, recent evidence suggesting that almost all clinically diagnosed NMSCs are verified histologically, coupled with the increasing availability of electronic histopathology data to cancer registries, raises the possibility that this situation may have improved. Objectives To assess recent trends in incidence of the main types of NMSC and carcinoma in situ (CIS) of the skin in Scotland. Methods The study was restricted to selected health board areas in the East of Scotland for which pathology data have been used routinely to support cancer registration since the early 1990s. Incident cases of squamous cell carcinoma (SCC) of the skin, CIS of the skin, and first ever basal cell carcinoma (BCC) were extracted from the Scottish Cancer Registry covering the period of diagnosis 1992-2003. Sex-specific, age-standardized and age-specific incidence rates were calculated for four consecutive 3-year periods of diagnosis. Estimated annual percentage changes (EAPCs) in incidence were calculated by Poisson regression modelling, with adjustment for age. The percentage distribution of SCC, BCC and CIS of the skin by anatomical site and sex was calculated for the period of diagnosis 1997-2003. Results The crude incidence of SCC for the period 1995-97 was 34.7 per 100 000, comparable with the best existing Scottish estimate of 32.2 derived from a prospective survey in Glasgow during March 1995. Age-adjusted rates of SCC, first ever BCC, and CIS of the skin have all increased significantly in both sexes between 1992 and 2003 (all P < 0.001), with EAPCs ranging in magnitude from +1.4% (first ever BCC in males) to +5.1% (CIS in males). The majority of lesions arose on the head and neck area, with the exception of CIS, which in females was more commonly located on the limbs. Conclusions Ascertainment of NMSC has probably improved since the advent and use of electronic pathology data. Ongoing increases in age-adjusted incidence, combined with ageing of the population, will have major implications for the clinical workload associated with NMSC for the foreseeable future.