Direct comparison
Prospective vs. Retrospective Study
Prospective studies collect data forward from enrollment; retrospective studies reuse existing records. Compare bias, cost, and evidence strength.
Side-by-side comparison
| Dimension | Prospective Study | Retrospective Study |
|---|---|---|
| Direction of data collection | Forward — data collected as events occur, from enrollment onward | Backward — draws on data already recorded before the study began |
| Outcome status at study start | Has not yet occurred | Has already occurred |
| Typical data source | Purpose-built data collection — case report forms, structured follow-up visits | Existing records — medical charts, registries, claims/administrative data |
| Strength of causal inference | Generally stronger — temporal sequence (exposure before outcome) is directly observed | Generally weaker — temporal sequence must be reconstructed from historical records |
| Recall bias risk | Low — data captured close to when it occurs | Higher when based on participant recall; lower (but not zero) when based on contemporaneous records |
| Cost and duration | Higher cost, longer timeline — requires active follow-up over the study period | Lower cost, faster — data usually already exist |
| Control over data quality | High — researchers define what's measured, how, and when | Limited — bound by what was recorded for another purpose, often clinical care |
| Suited to rare outcomes | Impractical — would require a very large sample and long follow-up | Well suited — existing cases can be identified directly |
| Common study designs | Prospective cohort studies; most interventional/randomized trials | Retrospective cohort studies; case-control studies; chart reviews |
| Consent / IRB pathway | Typically requires prospective informed consent from participants | May qualify for waiver of consent and HIPAA authorization when using only existing records |
Common questions
FAQ
Is a retrospective cohort study the same as a case-control study?+
No. A retrospective cohort study still starts from exposure status and follows the cohort's outcomes forward using historical records. A case-control study starts from outcome status — comparing cases to controls — and looks backward for differences in exposure. Both are retrospective in the sense of using existing data, but they organize the comparison differently.
Can a cohort study be either prospective or retrospective?+
Yes. "Cohort study" describes the comparison logic (following exposed vs. unexposed groups over time); "prospective" or "retrospective" describes when the data were collected relative to when the study began. The same cohort design can be executed either way, with different implications for bias and cost.
Does "prospective" here mean the same thing as prospective clinical trial registration?+
No — these are distinct concepts that share a word. Prospective trial registration (an ICMJE requirement for interventional trials) is about when a trial's existence is registered on a public registry, at or before enrollment begins. The prospective/retrospective distinction on this page is about when study data are collected relative to the outcome, and applies to observational research as well as trials.
Which design provides stronger evidence?+
Prospective designs generally provide stronger evidence for causal inference because exposure and outcome are measured consistently and recall bias is minimized. Retrospective designs remain valuable for studying rare outcomes, generating hypotheses cheaply, and situations where prospective follow-up would be impractical or unethical to withhold.
Going deeper
