Research into item-method directed forgetting has demonstrated that participants can intentionally forget studied items when they are asked to do so immediately following initial presentation. In spite of its long research history, it is still largely unclear whether the impaired retention of to-be-forgotten (TBF) materials compared to to-be-remembered (TBR) materials is a lasting effect and whether time-dependent forgetting varies for the two item types. To assess this, we investigated memory performance of both item types at four retention intervals spanning from 3 min up to 1 week. The resulting recall (Experiment 1) and recognition (Experiment 2) rates were then fitted with power functions of time. Memory performance of the TBR items exceeded that of the TBF items in both experiments and at all retention intervals, indicating sustained effects of directed forgetting. Across both experiments, retention data were well described by the power function. Crucially, the function's forgetting rate parameter was found to differ significantly for the two item types, with increased forgetting of TBF compared to TBR items. The findings support the assumption of an involvement of rehearsal processes in item-method directed forgetting, indicating that TBR and TBF items differ in memory strength and do so over prolonged retention intervals.