Randomized clinical trial: what this kind of study can — and can't — tell parents

Families are randomly assigned to an intervention or a comparison group, so the groups start out alike and any later difference can be credited to the intervention itself.

What it can show

It can support causal language — “made”, “improved” — for the outcomes it actually measured, in the ages it actually studied.

What it can't

It says nothing about outcomes it didn't measure, and its results apply first to the age group and setting it tested.

1 episode in the launch series uses this design:

Episode 1⏱ 3:40 video · 📖 1 min read

Randomized clinical trial Video coming soon

No Screens in the Last Hour Made Toddlers Sleep Better

Pickard et al., JAMA Pediatrics, 2024

Try tonight: Build a bedtime box — books, puzzles, soft toys — and swap the last hour before bed from screens to the box, every night for a week.

Questions about this design

What is a randomized clinical trial?

Families are randomly assigned to an intervention or a comparison group, so the groups start out alike and any later difference can be credited to the intervention itself.

Can a randomized clinical trial prove cause and effect?

It can support causal language — “made”, “improved” — for the outcomes it actually measured, in the ages it actually studied. It says nothing about outcomes it didn't measure, and its results apply first to the age group and setting it tested.

Which Parenting by Science episodes use this design?

Episode 1 — “No Screens in the Last Hour Made Toddlers Sleep Better” (Pickard et al., JAMA Pediatrics, 2024).

Meta-analysis (2)Systematic review & meta-analysis (2)Observational survey (1)Longitudinal cohort study (1)Survey + classic experiments (1)Cross-sectional survey (1)Observational cohort study (1)Cross-sectional study (1)Within-person observational study (1)