Sleep training is one of the most contested topics in parenting, with strong views held on every side. Parents who are sleep-deprived enough to be considering it often find themselves navigating genuinely conflicting claims — that it is cruel, that it is safe, that it damages attachment, that it has no long-term effects, that it is the only solution, that it is unnecessary. The evidence, when examined directly rather than through the lens of any particular parenting philosophy, is considerably more reassuring than the debate suggests.
This article covers what the evidence actually shows about sleep training, the main methods and how they differ, who sleep training is and is not appropriate for, and how to approach choosing a method.
Healthbooq helps parents track sleep patterns before and after implementing any changes, providing concrete data about whether the approach is working and how the baby is adjusting.
What the Evidence Shows
Multiple systematic reviews and long-term follow-up studies have examined sleep training in infants, looking at outcomes including sleep, cortisol levels, attachment security, behaviour, and parent-child relationship quality. The consistent finding is that sleep training methods — across the range from extinction to graduated extinction to fading approaches — are effective at improving sleep and are not associated with long-term harm.
A frequently cited 2012 study (Price et al.) followed up children at age five whose parents had used sleep training; there were no differences in any measured developmental, emotional, or parent-child relationship outcome between sleep-trained and non-sleep-trained children. A 2016 randomised controlled trial (Gradisar et al.) found that both graduated extinction and bedtime fading were effective, and that the cortisol patterns and attachment security of sleep-trained babies showed no adverse effects.
The evidence for harm from sleep training does not exist in the peer-reviewed literature. The evidence that is cited in arguments against sleep training — primarily Darcia Narvaez's writing on evolutionary perspectives — is theoretical and extrapolated from other contexts rather than based on studies of sleep training specifically.
This does not mean sleep training is appropriate for every family or that every method suits every baby. It means that the decision can be made based on practical considerations rather than fear of harm.
The Main Methods
Extinction (sometimes called "cry it out") involves placing the baby in their sleep space awake and not returning until a specified time (usually the morning). This is the most rapidly effective method but produces the highest distress for most parents and some babies during the initial nights. It typically produces results within three to five nights.
Graduated extinction (the Ferber method, or "check and console") involves leaving the baby to settle for a set number of minutes before entering briefly to reassure (without picking up or feeding), then leaving again for a longer interval, and repeating. The intervals increase across the night and across nights. Check-ins provide parental reassurance but can extend the process by repeatedly interrupting the settling. Typically effective within five to seven nights.
Bedtime fading involves moving the bedtime progressively later until the baby falls asleep very quickly (because they are at the peak of their sleep drive), then gradually moving bedtime earlier in small steps. This reduces the settling time by working with the sleep drive rather than against resistance. It can take two to three weeks and requires accurate identification of the baby's natural sleep window.
Chair fading (the "sleep lady shuffle") involves the parent sitting in a chair beside the cot, providing verbal reassurance but not picking up, and gradually moving the chair further from the cot over one to two weeks. This provides parental presence and tends to be better tolerated by parents who find extinction or graduated extinction difficult, but it takes longer and requires consistent implementation.
Who Sleep Training Is Appropriate For
Sleep training is typically not recommended before four to five months, when developmental readiness for independent sleep onset is generally established and night feeds are beginning to reduce. The appropriate age also depends on weight and health — babies who are still feeding for nutritional reasons overnight should not have night feeds eliminated as part of a sleep training approach.
Sleep training is most appropriate when: the baby is developmentally ready (typically five months or older), feeding and nutrition do not depend on overnight feeds, the parents are consistent enough to implement the chosen approach reliably, and inadequate sleep is significantly affecting the health and wellbeing of the family.
It is not appropriate for: babies under four to five months; babies with medical conditions affecting sleep; situations where parental consistency cannot be maintained; or families where the current arrangement (including night feeding or co-sleeping with appropriate safety measures) is working adequately for everyone.
Key Takeaways
Sleep training — methods that aim to help a baby develop the ability to fall asleep independently — has a robust evidence base showing effectiveness and no demonstrated long-term harm to babies, attachment security, or the parent-child relationship. The most studied methods include extinction (leaving the baby to cry without parental intervention), graduated extinction (timed check-ins — the 'Ferber method'), and fading approaches (gradually reducing parental presence or assistance at bedtime). There is no single universally correct method; the most effective method is the one a family can implement consistently. Sleep training is typically not appropriate before four to five months.
