Ask Mallard
Study design

Randomized screening trial

4 min read

Randomize an invitation to screening, then follow everyone assigned to each group to measure the effect of offering the program.

NHMRC EVIDENCE LEVEL · 2009Level II · Screening interventionProvisional design classification. Study quality and certainty of findings require separate appraisal.
Why this level?Participants or clusters are randomly allocated to concurrent comparison groups. NHMRC 2009 source

Random allocation of an invitation with concurrent controls is an NHMRC level II screening intervention design. The label describes allocation, not whether the screening program works.

Fit

When to use it, and when not to

Use it when

  • You need to know whether offering screening improves patient outcomes.
  • You can randomize invitations before people choose whether to attend.

Not when

Before you start

What you need

Decide these first
  1. The invitation and control strategiesassignmentSpecify tests, repeat rounds, follow-up care and the screening available to controls.
  2. The eligible populationDefine who can receive an invitation before allocation.
  3. A patient outcome and follow-up periodoutcomeTypeUse disease mortality or another patient benefit. Record false positives, extra procedures and overdiagnosis.
  4. Attendance and screening outside the trialMeasure both. They affect how different the assigned strategies are in practice.
  5. A missing-data plan that starts from an assumptionDecide why deaths or follow-up may be missing before choosing a method. The assumption picks the method, not the share missing.
It rests on these
  1. Invitations are allocated by chance with concealment.How to check: Keep the next allocation hidden until eligibility is fixed.
  2. Outcome capture is comparable in both groups.How to check: Use the same follow-up and cause-of-death rules. Retain non-attenders in their assigned group.
  3. Missing deaths and follow-up follow the assumption you named.How to check: Register links fail more often for people who move or emigrate, so missing completely at random (MCAR) is rarely plausible. Missing at random (MAR) means the gaps depend only on recorded factors, such as age, area and group. Under MAR, multiple imputation can use those factors, but it is not automatically better than analyzing everyone followed. If people who leave differ in their risk of dying of the disease, even given recorded factors, missing not at random (MNAR) is plausible. Then add a tipping-point analysis, which shifts the imputed deaths until the conclusion changes.

Plain language

How it works

Fix eligibility before randomizing the invitation. Specify what the control group can receive. Follow all assigned people, whether they attend or not. This estimates the effect of offering screening at the observed uptake.

What to preserve

Disease mortality is a common endpoint; all-cause mortality and harms also matter. Earlier diagnosis alone is not evidence of benefit. Non-attendance and screening in controls can shrink the difference between assigned strategies. Plan follow-up long enough to observe the benefit and harms that matter. Use CONSORT to report allocation, participant flow, outcomes and harms.

Randomized screening trial in outline
Compare outcomes in everyone assigned, including non-attenders. Attendance is not the randomized comparison.Randomized screening trialDefine eligible populationRandomize invitationAssign control strategyFollow everyone as assignedCompare outcomes in everyone assigned, including non-attenders. Attendance is not the randomizedcomparison.
Conceptual diagram. It shows the structure of the idea, not data from any study.

Worked example

A worked example

Invented for illustration
Setting
A program randomizes 10000 eligible people to invitation and 10000 to usual care.
Question
Does offering screening reduce deaths from the disease?
Approach
Follow both assigned groups for the same period, including people who decline screening.
Result
Deaths are 120 of 10,000 invited and 150 of 10,000 controls. The risk difference is -0.3 per 100 and the risk ratio is 0.8.
What it means
The offer reduces observed risk by 3 deaths per 1000 people assigned. These invented figures do not establish a real program benefit.

In practice

Recent studies PubMed search · not vetted

Searching PubMed…

Watch out

Common mistakes

The one to remember

Compare everyone assigned an invitation with everyone assigned the control strategy. Comparing attenders with non-attenders loses randomization and can make screening appear beneficial.

  • Using longer survival after diagnosis as proof of benefit.Instead: Earlier diagnosis can lengthen measured survival without delaying death. Compare outcomes from allocation in all assigned people.
  • Counting extra diagnoses as benefit.Instead: Some disease would never cause harm. Report overdiagnosis and harms alongside patient outcomes.

Inside Mallard

How Mallard uses this

Plans do not link to this design yet. Mallard has no dedicated sizing form for randomized screening trial.

No Mallard code example points to this page.
Atlas node design.randomized_screening_trial, the id a plan cue records.

Related Learn pages

Related designs and methods

Evidence

Sources & further reading

  1. Screening and prostate-cancer mortality in a randomized European study (opens in a new tab) · N Engl J Med · doi 10.1056/NEJMoa0810084
  2. Reduced lung-cancer mortality with low-dose computed tomographic screening (opens in a new tab) · N Engl J Med · doi 10.1056/NEJMoa1102873
  3. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials (opens in a new tab) · BMJ · doi 10.1136/bmj.c332
  4. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls (opens in a new tab) · BMJ · doi 10.1136/bmj.b2393
Prepared with AI assistance; source titles, authors, years and DOIs checked against PubMed on 2026-11-06. No independent professional approval is claimed. ERSPC supports the invitation-versus-control design and population mortality endpoint. NLST supports randomized comparison of screening strategies; its comparator was radiography, not no invitation.
Last source review: 2026-11-06
AI-generated educational content
Mallard uses reputable methodological and reporting sources, but this material may contain errors or omit important context. Verify important methodological decisions with the cited sources and, when appropriate, a qualified research methodologist or statistician.