78 A critical interpretive synthesis of recommendations for de-intensification and de-implementation from population screening (dimples)

Objectives Screening has generally been met with enthusiasm, and, in some cases, has been associated with demonstrable reductions in morbidity and mortality. In other cases, the balance of benefits and harms may be less clear and may evolve over time. Recommendations to reduce (de-intensify) or stop screening altogether (de-implementation) have occurred, but have proved to be controversial. The goal of this study was to review existing recommendations to better understand the stated rationales for de-intensification or de-implementation of established population screening programs. In doing so we: Identify examples of population screening programs recommended for de–intensification or de–implementation and describe the characteristics of these program changes Describe the stated reasons for de–intensification or de–implementation Explore the information cited for justification of recommendations Method We conducted a Critical Interpretive Synthesis (CIS) of published literature. This included an extensive search (websites of major screening organizations, reference searches, and content expert input) in addition to an electronic search of standard data bases. Data were analysed, coded and labeled in an inductive manner, and thus allowed for further searches that explicitly sought out contrasting or conflicting evidence. The review has been registered with International Prospective Register of Systematic Reviews PROSPERO (CRD42016035279). Results Of 9570 titles total of 66 documents, were included for analyses. Of these, 55 covered adult screening conditions and were largely cancer-related. Infant conditions included congenital toxoplasmosis (1), Down syndrome (1), fetal movement count (2), hearing test (1), urinalysis (3) and neuroblastoma (3). Of 107 recommendations, 50 related to de-intensification (e.g. increase of start age, extension of screening interval), 33 called for de-implementation, and 24 called for changes to the screening modality (often due to reduction in invasiveness or in association with increased periodicity). Explicit reasons were provided for 49 recommendations, while only 41 recommendations cited specific information in the justification. Conclusion We identified examples of programs across jurisdictions, age ranges, and clinical areas. Cancer screening dominates the examples identified. Recommendations to de-intensify or de-implement programs varied in terms of the level of evidence cited, as well as the stated rationales. Only around half of the identified recommendations provided an explicit rationale, and less than half of all recommendations were supported by a specific citation of evidence. Given the contested nature of de-implementation decisions, there is a greater need for transparency regarding the rationale behind recommendations and clearer articulation of the evidence used to support specific recommendations.