Define the population and index event
Describe the care setting, country, index event and inclusion criteria. Set the lookback and follow-up periods and identify exclusions such as overlapping treatment episodes. This helps institutions distinguish a potentially usable cohort from a broad inventory estimate.
Name the variables needed for the question
Separate essential fields from preferred detail. A useful dataset may depend on linked outcomes, a diagnostic reference standard or repeat testing. State whether a structured result is sufficient or whether reports, imaging or specimen metadata are necessary.
Verify how outcomes were observed
Establish who recorded the outcome, at what interval and in which system. Check missing follow-up, changes to code sets and differences across sites. For comparisons, ask whether the underlying populations can be characterized and whether data capture varies systematically.
Review authorization before detailed exchange
Confirm the proposed research or commercial use, institution approvals and handling safeguards. Review links and detailed timestamps for disclosure risk. Plan diligence using aggregate counts and approved documentation before considering any record access.
Write a brief the source can answer
List minimum volume, geography, formats, timing and rights requirements. Ask what can be supplied today, what requires approval and where coverage or permitted use is uncertain.