Indicators to identify cancer screening providers with suboptimal case detection: A scoping review

Abstract

Several international guidelines consider sensitivity (of test, episode, or programme) and related measures of the detection of prevalent cases of target disease to be among key performance indicators for quality control of routine cancer screening programmes and use them to identify suboptimal providers. We aimed to describe the variability encountered in real‐world settings around the measurement of these quantities in cervical and colorectal cancer screening, where the target for disease detection includes preinvasive disease. We performed a scoping review of grey literature, including international guidelines, annual statistical reports, and other official documents from European cervical and colorectal screening programmes. From the reviewed material, we extracted information on 20 measures used for this purpose. Some measures have been adopted in several programmes, but none have been used in all, not even within the same cancer type. While many of the methods might appear plausible for the intended use, our analysis showed that when applied to routinely collected data they may provide misleading or uninterpretable estimates of the ability of individual providers and the service as a whole to detect prevalent cases. Screening programmes should be cautious in their choice and interpretation of these measures. Further methodological development is required to better support policymakers and quality control managers in prioritising measures that are fit for purpose in routine cancer screening programmes.