Why English? | Reporting CONSORT-Surrogate Trials Through Target Outcomes, Validation Evidence, Measurement, Treatment Effects, Harms and Uncertainty

ORIENTATION

What this guide owns

Current fact check

CONSORT-Surrogate was published in 2024 after a consensus-driven process. It contains nine modified items and two new items for trials using surrogate endpoints as primary outcomes.

Canonical boundary

This article owns reporting of surrogate endpoints in randomised trial reports. The Gold Standard surrogate-endpoints article owns broader causal and design theory, while the SPIRIT/CONSORT Outcomes article owns complete outcome specification across protocols and reports.

Worked example

Every chapter returns to a fictional trial using a six-month imaging marker as the primary surrogate for a later mobility outcome that matters directly to participants. It is fictional, low-stakes and educational; it does not report a real school, clinical decision or guaranteed outcome.

Warm up with eduKate’s guides to explaining scientific evidence and asking better questions in class.

CHAPTER 1 OF 13

1. Identify the Surrogate Endpoint in the Title and Abstract

Back to contents

Why this move matters

State that a primary outcome is a surrogate and name the corresponding target outcome early.

Pair the surrogate with the target outcome

State that a primary outcome is a surrogate and name the corresponding target outcome early. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The abstract identifies the imaging marker as a surrogate for later mobility rather than calling it simply the clinical endpoint. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

A technical outcome name may not tell readers that it substitutes for something else. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Identify the Surrogate Endpoint in the Title and Abstract”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 2 OF 13

2. Define the Target Outcome That Matters Directly

Back to contents

Why this move matters

Describe the event, function, symptom, quality of life or survival outcome the surrogate is intended to predict or substitute for.

Pair the surrogate with the target outcome

Describe the event, function, symptom, quality of life or survival outcome the surrogate is intended to predict or substitute for. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The target is independent mobility at two years, not the six-month scan itself. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Different stakeholders may value different target outcomes. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Define the Target Outcome That Matters Directly”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 3 OF 13

3. Explain Why a Surrogate Was Used

Back to contents

Why this move matters

Report ethical, practical, statistical or developmental reasons and what time, sample size or feasibility advantage it offers.

Pair the surrogate with the target outcome

Report ethical, practical, statistical or developmental reasons and what time, sample size or feasibility advantage it offers. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The target outcome develops slowly, so the marker permits an earlier phase decision. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Convenience does not establish validity. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Explain Why a Surrogate Was Used”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 4 OF 13

4. Describe the Biological or Causal Rationale

Back to contents

Why this move matters

Explain the pathway linking intervention, surrogate and target outcome, including plausible routes by which treatment could affect the target outside the surrogate.

Pair the surrogate with the target outcome

Explain the pathway linking intervention, surrogate and target outcome, including plausible routes by which treatment could affect the target outside the surrogate. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The marker reflects tissue repair, while pain and balance may influence mobility through separate pathways. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

A plausible pathway is not empirical validation. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Describe the Biological or Causal Rationale”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 5 OF 13

5. Report the Validation Evidence and Its Context

Back to contents

Why this move matters

Summarise evidence supporting the surrogate–target relationship, population, intervention class, analysis level, uncertainty and limits.

Pair the surrogate with the target outcome

Summarise evidence supporting the surrogate–target relationship, population, intervention class, analysis level, uncertainty and limits. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: Validation comes from several trials in the same disease and intervention class, but not in adolescents. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Individual-level correlation alone does not prove that treatment effects on the surrogate predict treatment effects on the target. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Report the Validation Evidence and Its Context”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 6 OF 13

6. Specify How and When the Surrogate Was Measured

Back to contents

Why this move matters

Give instrument, assay or imaging method, timing, processing, thresholds, blinding, quality control and handling of repeated measures.

Pair the surrogate with the target outcome

Give instrument, assay or imaging method, timing, processing, thresholds, blinding, quality control and handling of repeated measures. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: Scans are centrally read using a versioned protocol at baseline and six months. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Measurement error can attenuate or distort both surrogate and treatment-effect estimates. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Specify How and When the Surrogate Was Measured”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 7 OF 13

7. Prespecify the Surrogate Analysis and Estimand

Back to contents

Why this move matters

State analysis population, time point, metric, model, covariates, missing-data method, multiplicity and the treatment effect being estimated.

Pair the surrogate with the target outcome

State analysis population, time point, metric, model, covariates, missing-data method, multiplicity and the treatment effect being estimated. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The primary contrast is adjusted mean change at six months in all randomised participants. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Changing the surrogate definition after seeing results can create a favourable but unstable story. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Prespecify the Surrogate Analysis and Estimand”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 8 OF 13

8. Report the Treatment Effect on the Surrogate With Precision

Back to contents

Why this move matters

Give estimates, confidence intervals, denominators, distributions and clinically interpretable units rather than only a p-value.

Pair the surrogate with the target outcome

Give estimates, confidence intervals, denominators, distributions and clinically interpretable units rather than only a p-value. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The report presents the adjusted marker difference with a confidence interval and the observed values by group. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Statistical significance does not show that the surrogate change is large enough to matter. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Report the Treatment Effect on the Surrogate With Precision”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 9 OF 13

9. Report Available Target Outcomes and Other Direct Outcomes

Back to contents

Why this move matters

Present target outcomes, symptoms, function and quality of life collected during follow-up even when immature or secondary.

Pair the surrogate with the target outcome

Present target outcomes, symptoms, function and quality of life collected during follow-up even when immature or secondary. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: Early mobility data are shown with uncertainty and clearly labelled as incomplete. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Immature direct outcomes should not be treated as definitive, but hiding them is worse. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Report Available Target Outcomes and Other Direct Outcomes”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 10 OF 13

10. Report Harms Without Assuming the Surrogate Captures Them

Back to contents

Why this move matters

Use the current harms guidance to show adverse events, withdrawals and burden separately from the surrogate benefit pathway.

Pair the surrogate with the target outcome

Use the current harms guidance to show adverse events, withdrawals and burden separately from the surrogate benefit pathway. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: An improved scan does not erase a higher rate of fatigue or treatment discontinuation. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

A surrogate designed for benefit may contain little information about harm. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Report Harms Without Assuming the Surrogate Captures Them”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 11 OF 13

11. Explain Uncertainty in Translating to the Target Outcome

Back to contents

Why this move matters

State what validation supports, what remains untested and how large or directionally stable the target effect could be.

Pair the surrogate with the target outcome

State what validation supports, what remains untested and how large or directionally stable the target effect could be. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The discussion says the marker improvement supports further study but cannot quantify mobility benefit reliably. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Language such as improves outcomes can collapse surrogate and target into one claim. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Explain Uncertainty in Translating to the Target Outcome”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 12 OF 13

12. Avoid Causal and Regulatory Overclaim

Back to contents

Why this move matters

Distinguish association, predictive validation, causal mediation, qualification and decisions by particular regulators or health systems.

Pair the surrogate with the target outcome

Distinguish association, predictive validation, causal mediation, qualification and decisions by particular regulators or health systems. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The report says the surrogate is supported in this context, not universally accepted for every intervention. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

Use in one approval or guideline does not automatically validate the endpoint elsewhere. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “Avoid Causal and Regulatory Overclaim”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

CHAPTER 13 OF 13

13. State the Decision the Surrogate Can and Cannot Support

Back to contents

Why this move matters

Connect the result to the next trial, development or evidence decision and identify what direct follow-up remains necessary.

Pair the surrogate with the target outcome

Connect the result to the next trial, development or evidence decision and identify what direct follow-up remains necessary. Name what was measured early and what patient-relevant outcome it is intended to substitute for. Never let “endpoint” conceal which one is direct and which one is a proxy.

Open the validation claim

Worked example: The marker result supports continuing development with mandatory two-year mobility follow-up. Identify the population, intervention class, level of evidence and uncertainty supporting the surrogate–target relationship.

Read benefit and harm on separate tracks

A favourable surrogate result may not establish a favourable target outcome, and it may reveal little about harms. Report direct clinical outcomes and adverse effects whenever available.

Interpret without shortcut language

An efficient early decision is not the same as a final benefit–harm conclusion. A surrogate can make a trial faster or more feasible; it cannot erase uncertainty about whether the treatment helps people feel, function or survive.

Student, parent and career transfer

Make one evidence card for “State the Decision the Surrogate Can and Cannot Support”: source location, plain-language meaning, limit and next question. A parent or teacher can ask for one everyday explanation and one specialist explanation; both must preserve the same conditions. This habit supports clinical research, drug and device development, biostatistics, health technology assessment, evidence appraisal, regulation and science communication without promising a grade, admission, funding decision or job.

PRACTICE LAB

Build a surrogate-to-target inference decision trail

Student route

1. Choose one original report inside this guide’s scope.
2. Copy its objective and underline actor, action, context, evidence and intended decision.
3. Draw the specific reasoning chain used in this article.
4. Mark each step reported, inferred, contested or unknown.
5. Rewrite the conclusion so it preserves the narrowest important limitation.

Parent or teacher route

Ask the learner to explain CONSORT-Surrogate’s central reporting method twice: once to a classmate and once to a method specialist. Praise a precise correction and a well-stated unknown; both show disciplined attention to evidence.

Early-career route

Create a one-page CONSORT-Surrogate audit with source link, date checked, scope, method logic, decisive wording, contrary evidence, limit and next action. A colleague should be able to continue without guessing.

Continue with eduKate’s guides to reading course and career information and giving useful feedback in group projects.

USEFUL NEXT READING

Connect this narrow owner to its neighbour

Read the surrogate-endpoint design-theory guide for a related but different reader job. Return here when the question is specifically how to report a surrogate endpoint without letting a convenient proxy become a silent promise about patient benefit. This boundary keeps one canonical owner per intent.

For English practice across research, school and career communication, visit the English at work and in public life guide.

FREQUENTLY ASKED QUESTIONS

Questions students and parents often ask

What is a surrogate endpoint?

A measure used in place of a target outcome of direct relevance to participants or other stakeholders.

What is the target outcome?

The outcome the surrogate is intended to predict or substitute for, such as symptoms, function, survival or quality of life.

How many CONSORT-Surrogate items are there?

Nine modified CONSORT items and two new items.

Does correlation validate a surrogate?

Not by itself. Validation may require evidence that treatment effects on the surrogate predict treatment effects on the target in the relevant context.

Why report harms separately?

A benefit surrogate may not capture adverse effects or treatment burden.

Can a favourable surrogate result prove patient benefit?

No. The strength of that inference depends on validation evidence and remaining uncertainty.