← Back
Research and new methods

The limits of a small study

A result in a few, or a few dozen, patients can be encouraging, but chance, selection and the lack of a comparison can exaggerate it.

Last updated: ~1 min read
A small group of people sitting in a circle and talking
Figure 1.A small group

01Short answer

In small studies, the outcome for one or two patients changes the percentages sharply. Without a comparison group it is hard to know whether improvement came from the medicine or from something else. If the fittest patients were selected, results look better. That is why early-phase results must be confirmed in larger randomised trials. One patient's story, however striking, is not proof.

02Simple explanation

Few participants.

Chance plays a role.

A comparison is needed.

Wait for confirmation.

03Scientific explanation

NCI: randomisation makes groups comparable and reduces bias.

NCI: new treatments are tested step by step, from the laboratory to clinical trials; only those proven safe and beneficial become standard care.

04What happens in the body?

Why be cautious

Percentages are unstable in small numbers.

Scales
Figure 2.Scales — The limits of a small study

What to look for

Number of participants, comparison and length of follow-up.

A narrow window
Figure 3.A narrow window — The limits of a small study

How to read the news

Research news brings hope, but not every advance suits everyone. Look at who it was tested in, at what phase and in how many people.

Talk to your doctor

Show your doctor what you have read; they will explain how it relates to your diagnosis and treatment. Do not change your treatment on your own.

05From research news to a conversation: four steps

Read, check, ask, decide together.

Read — From a trusted source.1ReadFrom a trusted source.
Check — Phase, participants, result.2CheckPhase, participants, result.
Ask — Show your doctor.3AskShow your doctor.
Decide — Together, without rushing.4DecideTogether, without rushing.
SourceA trusted site.
DoctorA personal explanation.
SafetyNot changing treatment yourself.

06Key points at a glance

TopicIn brief
Why be cautiousPercentages are unstable in small numbers.
What to look forNumber of participants, comparison and length of follow-up.
How to read the newsResearch news brings hope, but not every advance suits everyone. Look at who it was tested in, at what phase and in how many people.
Talk to your doctorShow your doctor what you have read; they will explain how it relates to your diagnosis and treatment. Do not change your treatment on your own.

07Key terms

Selection bias
Non-random choice of participants.
Randomisation
Assigning groups by chance.
Clinical trial
Testing a new treatment with volunteers.
Evidence
A result confirmed by reliable research.

08Key points

  • Learn.
  • Ask your doctor.
  • Write it down.
  • Check the source.
  • Talk to your doctor.

09Frequently asked questions

8 out of 10 responded: is that a lot?

Encouraging, but it may change in a larger trial.

Is a small study useless?

No, it is important for the next step.

Is this advance right for me?

Only your treating doctor can say, based on your diagnosis, stage and condition.

10Questions for your doctor

Take these questions to your appointment. The ones you save are collected on the “My questions” page and stay on this device only.

  1. 8 out of 10 responded: is that a lot?

  2. Is a small study useless?

  3. Is this advance right for me?

My questions

12Sources

  1. National Cancer Institute. Randomization in Clinical Trials
  2. National Cancer Institute. Phases of Clinical Trials
  3. MedlinePlus. Evaluating Health Information
  4. National Cancer Institute. What Are Clinical Trials?
  5. National Cancer Institute. Using Trusted Resources
  6. National Cancer Institute. NCI Dictionary of Cancer Terms: tyrosine kinase inhibitor
  7. NMDP (Be The Match). About transplant
  8. National Cancer Institute. NCI Dictionary of Cancer Terms: cytokine release syndrome
  9. National Cancer Institute. NCI Dictionary of Cancer Terms: mutation
  10. National Cancer Institute (NCI). NCI Dictionary of Cancer Terms: flow cytometry
Medical disclaimerThis information is for educational purposes only and does not replace medical advice.
Last updated: ← Back to the list

Open the interactive site →