Time-outs: do they work, and do they harm?

The question

Do time-outs actually change behaviour — and is there any truth to the claim that they damage your child's attachment or mental health?

Short answer

Brief, calm, boring time-outs are one of the better-supported discipline tools we have: parent-training programmes that teach time-out consistently get larger improvements in children's behaviour than programmes that don't [2]. The claim that time-outs damage attachment or cause trauma sounds serious, but it is mostly theoretical — the studies that have tested it directly found no harm signal when time-outs were done properly [3][4][7]. The crucial qualifier: done properly. Long, angry, inconsistent isolation is not what the research studied, and it is associated with worse outcomes [4].

What the strongest evidence says

They work — within programmes, at least

Nobody has run a trial of time-out alone versus doing nothing for years, so the efficacy evidence is indirect but substantial. A meta-analysis of 77 parent-training programme evaluations found that teaching parents to use time-out — alongside warm positive interaction, emotional communication, and consistency — was one of the components consistently linked to bigger improvements in parenting and in children's disruptive behaviour [2]. Time-out is a core ingredient of programmes like Triple P and the Incredible Years, which have randomised-trial evidence for reducing oppositional behaviour in young children [1][2]. A detailed 2019 review concluded the evidence for time-out's efficacy and acceptability was "overwhelming" [1].

Be honest about what that means: we know time-out works as part of a package. We are less certain how much of the package's effect is the time-out itself.

The harm claims don't have the data behind them — yet

In recent years, time-out has been criticised on attachment grounds: that removing a child breaks attachment security, floods them with stress, or re-traumatises children with difficult histories. These are serious claims. But as the 2019 review pointed out, they were argued from theory, not from studies that measured harm [1].

Three studies have now tested the claims directly, and all point the same way:

• "Generation time-out grows up" (2024): 942 young adults (18–30, two samples) reported on the time-outs they received as children. In the first sample, more appropriate implementation was associated with better adult mental health, better emotion regulation, and less avoidant attachment — over and above the effects of parenting style generally; the second sample replicated the mental-health link but not the attachment and emotion-regulation ones. Simply having received time-outs showed no harm signal, including among those who had experienced childhood adversity [3].
• A clinical trial with trauma-exposed children (2022): 205 families at a Sydney conduct-problems clinic (children 2–9, mean 5.6) received a 10-session parent-management programme that included time-out (nonrandomised, waitlist-controlled). Children with high adversity exposure showed equal — and on some measures greater — reductions in overall difficulties and internalising problems than children with low adversity exposure: the opposite of what the harm claims would predict [7].
• A nationally representative survey (2024): 474 Australian parents of 6- to 8-year-olds. Appropriately implemented time-out was associated with better child mental health and attachment; inappropriately implemented time-out — inconsistent, punitive, delivered in anger, or used when the child hadn't misbehaved — was associated with worse outcomes. Adversity exposure moderated the pattern: children with a history of adversity were most likely to experience attachment enhancement from appropriately implemented time-out [4].

The two survey studies are observational and cross-sectional, so they can't prove cause and effect; the 2022 trial was nonrandomised and tested a programme including time-out, not time-out alone. But together they are the first direct evidence on the safety question, and none of them supports the alarming claims [3][4][7].

What "done properly" actually means

The version of time-out with evidence behind it is very specific — and it's not what most critics are describing:

• Brief: about one minute per year of age, up to about five minutes maximum [6].
• Boring: a neutral, safe, non-frightening spot with nothing to do — not a bedroom full of toys, not a cupboard [6].
• Calm: delivered without shouting, lecturing, or anger. If you're furious, it's not a time-out, it's a punishment [4][6].
• Consistent: used for the behaviours you've decided on in advance, the same way every time [6].
• Ended on calm: the clock effectively starts when the child settles; then brief reconnection — a hug, a short restatement of the rule, and moving on. No long debriefing lectures [6].

What the evidence calls "inappropriate" — long isolation, doing it in anger, using it for everything, being inconsistent — is associated with worse outcomes. Note the uncomfortable implication: when time-out goes wrong, it may be telling you more about the parent's state than the child's [4].

The right age window

Time-out is designed for roughly 18 months to 5 or 6 years [6]. Under about 18 months, children can't reliably connect the consequence to the behaviour — for babies, the answer is distraction and redirection, not discipline. Over about 6, removal of privileges and natural consequences tend to work better, because older children understand them and find them more meaningful [6]. Using time-out outside its window is one of the quiet reasons it "doesn't work" for some families.

Time-in: the popular alternative with almost no evidence

"Time-in" — staying close and connecting through the misbehaviour instead of separating — is widely recommended by the same writers who criticise time-out. It sounds lovely, and it may well be fine. But there are essentially no outcome studies of time-in as a discipline strategy; it is recommended on theoretical grounds [1]. The evidence log marks this as a gap, not a finding.

What it means for the parents

Here's the part the parenting books sometimes skip: time-out is also a tool for the parent.

The most common reason time-outs fail isn't the child — it's the adult. Clinicians report that parents most often give up because they can't tolerate their own distress while the child pleads, bargains, or screams [6]. That's not weakness; sitting outside a door listening to your child cry goes against every instinct. Knowing that this is the expected hard part — and that giving in mid-time-out teaches the child that escalating works — genuinely helps.

There's a second parent-side benefit that is underappreciated: a time-out taken calmly is one of the few discipline tools that removes the parent from the escalation cycle too. A brief separation when you're about to shout is harm reduction for everyone in the room. No trial has measured parental stress or anger as an outcome of time-out use — that's a real gap — but the mechanism is straightforward and clinicians endorse it [1][6].

The honest trade-off: time-out done badly (angry, long, inconsistent) is associated with worse child outcomes [4], and doing it well requires parental regulation at exactly the moment regulation is hardest. If you can't do it calmly today, skip it today.

What remains uncertain

• The exact recipe. One minute per year, ages ~18 months to 5–6, boring location — these are expert consensus, not experimentally optimised parameters [5][6]. (Dadds & Tully frame time-out for ages roughly 2–8 [1]; AAFP clinical guidance targets 18 months to 5/6 years [6].)
• Long-term RCTs of time-out alone. All efficacy evidence is bundled into programmes; the standalone effect size is unknown [2].
• Attachment, measured directly. No study has run attachment assessments (like the Strange Situation) before and after a time-out programme. The reassuring evidence is questionnaire-based [3][4].
• Stress physiology. Whether time-out raises cortisol or other stress markers in young children is essentially unstudied.
• Time-in. Recommended widely, studied barely [1].

Benefits and risks in absolute terms

No trial event rates exist, so this table summarises the direction and strength of the evidence rather than per-1,000 figures.

Practical considerations

• Teach it before you need it. Explain at a calm moment what time-out is, which behaviours earn it, and where it happens [6].
• One warning, then act. For defiance: state the instruction, warn once, then time-out. For aggression (hitting, throwing), most guidance says go straight to time-out [6].
• Boring and brief. About a minute per year of age, max ~five minutes; no toys, no screens, no audience [6].
• Minimal attention during. Don't lecture, don't negotiate, don't make eye contact as a reward. If they leave, return them calmly without a speech [6].
• End warm. When the time is up and they're calm, reconnect — a hug, one short sentence about the rule, then move on completely [6].
• Mind the ratio. Time-out only makes sense against a background of generous attention for good behaviour ("time-in" all day, every day). If the only attention your child gets is corrective, no discipline strategy will work well [6].
• Reserve it. For a small number of specific behaviours (aggression, serious defiance), not for every annoyance. Overuse dilutes it [5][6].

When to talk to your doctor, midwife, or pediatrician

• If your child's aggression or defiance isn't improving after a few weeks of consistent, calm time-outs — the strategy may not fit the problem, or something else may be going on.
• If you find yourself using time-outs in anger, for long periods, or many times a day — that's a sign the approach (or the day) needs rethinking, and a professional can help.
• If your child shows extreme distress during time-outs (panic, self-harm, dissociation) — stop and seek advice rather than pushing through.
• If your own anger feels out of control during discipline — talk to your GP. That's common, treatable, and nothing to be ashamed of.

References

1. Dadds, M. R., & Tully, L. A. (2019). What is it to discipline a child: What should it be? A reanalysis of time-out from the perspective of child mental health, attachment, and trauma. American Psychologist. — [C, conceptual review] http://dx.doi.org/10.1037/amp0000449
2. Kaminski, J. W., Valle, L. A., Filene, J. H., & Boyle, C. L. (2008). A meta-analytic review of components associated with parent training program effectiveness. Journal of Abnormal Child Psychology, 36(4), 567–589. — [B] https://pubmed.ncbi.nlm.nih.gov/18205039/
3. Xu, J., Tully, L. A., & Dadds, M. R. (2024). Generation time-out grows up: Young adults' reports about childhood time-out use and their mental health, attachment, and emotion regulation. European Child & Adolescent Psychiatry, 33, 3471–3479. — [C] https://doi.org/10.1007/s00787-024-02408-8
4. Roach, A., McLean, R., Mendoza Diaz, A., Hawes, D., & Dadds, M. (2025). Time-out under scrutiny: Examining the relationships among the discipline strategy time-out, child well-being and attachment and exposure to adversity. British Journal of Psychiatry, 227(2), 538–544 (published online 2024). — [C] https://pmc.ncbi.nlm.nih.gov/articles/PMC12355463/
5. Sege, R. D., et al. (2018). Effective discipline to raise healthy children. Pediatrics, 142(6), e20183112. DOI: 10.1542/peds.2018-3112. — [guideline]
6. American Academy of Family Physicians. (2002). Childhood discipline: Challenges for clinicians and parents. American Family Physician. — [clinical guidance] https://www.aafp.org/pubs/afp/issues/2002/1015/p1447.html
7. Roach, A. C., Lechowicz, M., Yiu, Y., Mendoza Diaz, A., Hawes, D., & Dadds, M. R. (2022). Using time-out for child conduct problems in the context of trauma and adversity: A nonrandomized controlled trial. JAMA Network Open, 5(9), e2229726. — [C] https://doi.org/10.1001/jamanetworkopen.2022.29726

Changelog

• 2026-09-09: Final source-verification pass. Every empirical number checked against an opened primary source: Dadds & Tully 2019 (PubMed abstract — "overwhelming evidence" quote verified), Kaminski et al. 2008 (PubMed abstract — 77 evaluations, time-out among components with larger effects), Xu et al. 2024 (PubMed abstract — 407 + 535 samples, Study 1 vs Study 2 replication pattern confirmed; journal added: Eur Child Adolesc Psychiatry 33:3471–3479), Roach et al. 2024 full text (PMC12355463 — moderation narrowed to attachment enhancement for adversity-exposed children; citation year corrected to 2025, epub 2024), Roach et al. 2022 (PubMed abstract — 205 children, nonrandomised waitlist-controlled programme including time-out, high-adversity children showed equivalent-or-greater benefit), AAFP 2002 clinical guidance full text (verified: 18 months–5/6 window, 1 min/year max 5 min, neutral/boring/safe spot, end only when calm 15 s, parent-distress as most common failure reason, time-in background). Dadds & Tully bracket [B]→[C] (conceptual review). Removed the aggregated "1,600 children and adults" framing. Rating C confirmed.
• 2026-09-09: Author review under the v1 author-review model (REVIEW_PROCESS.md, 2026-09-09); reviewer: Guille (via delegated agent review). All open items resolved; status → approved. Added Roach et al. 2022 (JAMA Network Open) nonrandomised clinical trial on time-out-including programmes in trauma-exposed children; corrected Xu et al. 2024 characterisation (Study 1 vs Study 2 replication); Roach et al. cited to the print version (Br J Psychiatry 2025;227(2):538–544, epub 2024). Age window standardised to ~18 months–5/6 years with the noted review-level variation (2–8). Rating C confirmed.
• 2026-09-08: Topic created (draft). Evidence gathered 2026-09-08; awaiting independent reviewer. (Superseded 2026-09-09: author-review model, no external review before v1.)