This is a revision of a previously submitted proposal to use matrix decomposition to predict likelihood of future adverse events based on prior adverse reactions. Several clarifications were requested, including more detail on adverse event definition, methods used, impact of excluding those who experience no adverse events, and so forth. Minor changes were made to the proposal. For a methods-oriented project, I would have preferred the investigators to provide more detail. But I believe the proposal as written is satisfactory.
Recommendation for this data request: Approve