Target trial emulation
What would a specified treatment-strategy trial estimate, and can available data emulate it?
When to use this project
Use target trial emulation to structure a causal comparison of treatment strategies with observational data. First specify the hypothetical trial, then map each element to what the data can actually support.
How it differs
It is a design framework for observational causal studies, often implemented as a comparative cohort. Explicit eligibility, strategy assignment, time zero and follow-up prevent avoidable design errors, but do not create randomization or eliminate unmeasured confounding.
Decisions before you start
- Specify eligibility, strategies, assignment, follow-up, outcomes, causal contrast and analysis before data mapping.
- Align entry and follow-up and document any grace period, repeated eligibility or strategy deviations.
- Use sequential trials, cloning or weighting only when the chosen strategies and estimand require them.
How the flows work together
Each flow below links to its diagram, steps and tables. Flow order organizes the project; it does not imply that all work is strictly sequential. Data assessment, analysis planning and protocol refinement may inform one another.
1. Emulation overview
Overview · Question through causal effect interpretation
Use this overview to state the purpose and link to detailed decisions. Keep a versioned protocol and decision register, and add evidence references for the rationale.
2. Target trial specification
Design · Define hypothetical trial and estimands
Specify the hypothetical trial before looking for convenient proxies. Separate its assignment mechanism from observed treatment choice, and name the causal contrast rather than assuming every comparison estimates an intention-to-treat effect.
3. Protocol-to-data mapping
Data · Map trial requirements onto sources
Show the target requirement and recorded implementation side by side. Document unavailable variables and deviations, including their likely effects on identification or interpretation. Availability is not validation.
4. Eligibility and strategy assignment
Selection · Align entry strategies and time-zero
Eligibility, assignment and follow-up start must align. Grace periods, repeated eligibility and cloning are optional design choices that need explicit implementations; never require future survival to qualify entry.
5. Exposure and follow-up
Conduct · Track strategy adherence and censoring
Specify initiation versus sustained strategies and handling of deviations. Censoring at treatment change can introduce selection, and adjustment needs measured predictors and support. Preserve competing-event definitions.
6. Adjustment and estimation plan
Statistics · Specify identification and estimation methods
State the identifying assumptions and choose an estimator compatible with baseline or time-varying confounding. Weighting, standardization and related methods require diagnostics and cannot guarantee exchangeability.
7. Sensitivity and feasibility assessment
Statistics · Evaluate deviations and identifying assumptions
Assess support, protocol-to-data deviations, measurement error and residual confounding. Restrict the target question when the available data cannot support it. TARGET reporting guidance does not certify that causal assumptions hold.
Equations and statistical interpretation
These explanations describe notation and assumptions. Flow does not calculate these values.
Causal risk difference
Yᵃ(t) indicates whether an outcome would occur by t under strategy a in the specified target population. The contrast compares risks under the two strategies, not merely observed treatment groups.
Identification requires consistency, exchangeability and positivity, with appropriate interference and censoring assumptions. Sustained strategies may require methods for time-varying confounding affected by prior treatment.
Published example
Inspect the explicit target-trial design and how alternative analytic choices introduce bias. The example illustrates design discipline; emulation still relies on observational identification assumptions.
This publication illustrates a study design, not an endorsement of every method or a reproduction of its results. Follow its protocol, supplement and update history when studying the example.
Hypothetical scenario
A health system wants to estimate five-year cardiovascular risk under starting a preventive medicine now versus not starting at the same eligible visit.
- Emulation overview: Define a feasible strategy contrast and target patient population.
- Target trial specification: Write the hypothetical protocol and choose an initiation or sustained-strategy effect.
- Protocol-to-data mapping: Map eligibility, prescriptions, outcomes and confounders to recorded information, documenting gaps.
- Eligibility and strategy assignment: Align eligibility, strategy classification and follow-up start without conditioning on later survival.
- Exposure and follow-up: Specify treatment changes, loss to follow-up and competing events for the chosen estimand.
- Adjustment and estimation plan: Select an identification and estimation approach with overlap and model diagnostics.
- Sensitivity and feasibility assessment: Assess sensitivity to residual confounding, measurement error and deviations from the target protocol.
Deliver a populated protocol, decision log, data specification, and planned outputs. Enter numerical findings only after analysis and verification. Keep decisions made before seeing results distinguishable from later amendments.
Sources
- TARGET Statement (2025): reporting observational target trial emulations
- Hernán et al. (2016): specifying a target trial and preventing immortal-time bias
- STROBE checklist for cohort studies
- RECORD checklist for studies using routinely collected health data
- HMA–EMA Catalogue of RWD studies: registration and protocol guidance
- Hernán and Robins: Causal Inference—What If
Source check: 2026-09-17. Record the version of each reporting or appraisal tool used in your actual protocol.
Compare all Observational and RWE studies project types · Open editor in a new tab