Population screening for colorectal cancer: the implications of an ageing population

Abstract

Population screening for colorectal cancer (CRC) has recently commenced in the United Kingdom supported by the evidence of a number of randomised trials and pilot studies. Certain factors are known to influence screening cost-effectiveness (e.g. compliance), but it remains unclear whether an ageing population (i.e. demographic change) might also have an effect. The aim of this study was to simulate a population-based screening setting using a Markov model and assess the effect of increasing life expectancy on CRC screening cost-effectiveness. A Markov model was constructed that aimed, using a cohort simulation, to estimate the cost-effectiveness of CRC screening in an England and Wales population for two timescales: 2003 (early cohort) and 2033 (late cohort). Four model outcomes were calculated; screened and non-screened cohorts in 2003 and 2033. The screened cohort of men and women aged 60 years were offered biennial unhydrated faecal occult blood testing until the age of 69 years. Life expectancy was assumed to increase by 2.5 years per decade. There were 407 552 fewer people entering the model in the 2033 model due to a lower birth cohort, and population screening saw 30 345 fewer CRC-related deaths over the 50 years of the model. Screening the 2033 cohort cost £96 million with cost savings of £43 million in terms of detection and treatment and £28 million in palliative care costs. After 30 years of follow-up, the cost per life year saved was £1544. An identical screening programme in an early cohort (2003) saw a cost per life year saved of £1651. Population screening for CRC is costly but enables cost savings in certain areas and a considerable reduction in mortality from CRC. This Markov simulation suggests that the cost-effectiveness of population screening for CRC in the United Kingdom may actually be improved by rising life expectancies.