Assessment of a cancer screening program

Several Asian countries are implementing nationwide cancer screening programs. Assessment of the effectiveness of these programs is critical to their success as this is the only way to ensure that the benefits of screening outweigh the harms. In this paper we focus on colorectal cancer (CRC) screening to illustrate the principles of screening program assessment. The International Agency for Research on Cancer (IARC) has defined organized screening, distinguishing it from opportunistic screening. The key advantage of organized screening is that it provides greater protection against the possible harms of screening. Since screening is a process, not simply a test, the effectiveness of a program depends on the quality of each step in the cancer screening process. The evaluation of long-term screening program outcomes (CRC incidence and mortality) will not be observable for many years, given the time it takes to plan, pilot and implement a program. However, early performance indicators of the impact of screening should be monitored to give an early indication whether the program is on track. The European Union (EU) has recommended a minimum dataset to be collected and reported regularly by a screening program. Using information from these data tables, early performance indicators can be generated (e.g., participation rate, proportion of screen-detected cancers that are early-stage). Subsequently, modeling the natural history of the disease can be very helpful to estimate long-term outcomes, making use of these directly measured early performance indicators. Modeling can also be used to estimate the cost-effectiveness of a screening program and the potential impact of changes in policy, as illustrated by its recent use in the Netherlands to change the definition of a positive fecal immunochemical test (FIT) for the CRC screening program. Programs should consider modeling as an important component of screening program evaluation.