An investigation of discrepancies in outcome reporting and selective reporting bias in interrupted time series studies of health interventions: a methodological study

Abstract Background Selective reporting bias occurs when authors report outcomes or results based on the P value, magnitude or direction of the results. Selective reporting has not been examined in interrupted time series (ITS) studies. Therefore, we investigated (1) discrepancies between outcomes specified in protocols and those reported in the corresponding results reports, (2) completeness of results reporting, and (3) whether there was evidence of selective reporting bias in ITS studies. Met