NocturneGet started

Evidence base

This page is public and always will be. If you are going to take scheduling advice about your baby from software, you should be able to check where it came from without paying first.

Every source, and what it is used for

These are the papers behind the numbers. Where the app extrapolates past what a source actually reports, or where something is standard practice rather than a finding, it is labelled that way rather than dressed up.

Guideline2016

Recommended amount of sleep for pediatric populations: a consensus statement of the American Academy of Sleep Medicine

Paruthi S, Brooks LJ, D'Ambrosio C, Hall WA, Kotagal S, Lloyd RM, Malow BA, Maski K, Nichols C, Quan SF, Rosen CL, Troester MM, Wise MS · Journal of Clinical Sleep Medicine 12(6):785–786 · doi:10.5664/jcsm.5866

Used for: The target range for total sleep per 24 hours at each age (12–16 h for 4–12 months, 11–14 h for 1–2 years, 10–13 h for 3–5 years). Also the reason this app makes no duration recommendation under 4 months: the AASM panel explicitly declined to issue one.

Read the source →
Guideline2016

Consensus statement of the American Academy of Sleep Medicine on the recommended amount of sleep for healthy children: methodology and discussion

Paruthi S, Brooks LJ, D'Ambrosio C, Hall WA, Kotagal S, Lloyd RM, Malow BA, Maski K, Nichols C, Quan SF, Rosen CL, Troester MM, Wise MS · Journal of Clinical Sleep Medicine 12(11):1549–1561 · doi:10.5664/jcsm.6288

Used for: The outcomes wording used in the app: regularly sleeping the recommended hours is associated with improved attention, behavior, learning, memory, emotional regulation and mental/physical health; regularly sleeping less is associated with attention, behavior and learning problems.

Read the source →
Guideline2022

Sleep-related infant deaths: updated 2022 recommendations for reducing infant deaths in the sleep environment

Moon RY, Carlin RF, Hand I; AAP Task Force on Sudden Infant Death Syndrome · Pediatrics 150(1):e2022057990 · doi:10.1542/peds.2022-057990

Used for: The safe-sleep rules the app never lets a schedule override: supine for every sleep until 1 year, a firm flat non-inclined surface free of soft goods, room-sharing without bed-sharing ideally for at least 6 months, no soft bedding, avoid overheating, offer a pacifier once breastfeeding is established.

Read the source →
Guideline2022

Evidence base for 2022 updated recommendations for a safe infant sleeping environment to reduce the risk of sleep-related infant deaths

Moon RY, Carlin RF, Hand I; AAP Task Force on Sudden Infant Death Syndrome · Pediatrics 150(1):e2022057991 · doi:10.1542/peds.2022-057991

Used for: The technical report behind the safe-sleep rules above.

Read the source →
Meta-analysis2006

Behavioral treatment of bedtime problems and night wakings in infants and young children

Mindell JA, Kuhn B, Lewin DS, Meltzer LJ, Sadeh A; American Academy of Sleep Medicine · Sleep 29(10):1263–1276 · PMID 17068979

Used for: The evidence that behavioral sleep interventions work at all: across 52 treatment studies, 94% reported efficacy and over 80% of treated children showed clinically significant improvement maintained for 3–6 months. This review is also the source of the method taxonomy the app implements (extinction, graduated extinction, positive routines/bedtime fading, scheduled awakenings, parent education).

Read the source →
Meta-analysis2012

Normal sleep patterns in infants and children: a systematic review of observational studies

Galland BC, Taylor BJ, Elder DE, Herbison P · Sleep Medicine Reviews 16(3):213–222 · doi:10.1016/j.smrv.2011.06.001

Used for: The normal-range bands the app plots behind your data: pooled means and ±1.96 SD limits for 24-hour sleep duration, night wakings, sleep onset latency, longest sleep period and number of daytime naps, by age band (34 studies, 18 countries).

Read the source →
Randomized trial2012

Five-year follow-up of harms and benefits of behavioral infant sleep intervention: randomized trial

Price AMH, Wake M, Ukoumunne OC, Hiscock H · Pediatrics 130(4):643–651 · doi:10.1542/peds.2011-3467

Used for: The long-term safety statement the app shows before you start any program. 326 children with parent-reported sleep problems at 7 months, followed to age 6: behavioral sleep techniques had no marked lasting effects in either direction, so parents can use them to reduce short- and medium-term burden.

Read the source →
Randomized trial2016

Behavioral interventions for infant sleep problems: a randomized controlled trial

Gradisar M, Jackson K, Spurrier NJ, Gibson J, Whitham J, Williams AS, Dolby R, Kennaway DJ · Pediatrics 137(6):e20151486 · doi:10.1542/peds.2015-1486

Used for: The two programs this app will actually coach: graduated extinction and bedtime fading. 43 infants aged 6–16 months randomized to graduated extinction, bedtime fading, or sleep-education control. Both interventions produced large decreases in sleep latency. Salivary cortisol showed small-to-moderate declines versus controls, and at 12-month follow-up there were no differences in emotional/behavioral problems or in secure–insecure attachment.

Read the source →
Cohort study2003

Sleep duration from infancy to adolescence: reference values and generational trends

Iglowstein I, Jenni OG, Molinari L, Largo RH · Pediatrics 111(2):302–307 · doi:10.1542/peds.111.2.302 · PMID 12563055

Used for: The split between night sleep and day sleep at each age, which is what makes a nap schedule possible at all. 493 children in the Zurich Longitudinal Studies, followed from 1 month to 16 years.

Read the source →
Cohort study2004

A brief screening questionnaire for infant sleep problems: validation and findings for an Internet sample

Sadeh A · Pediatrics 113(6):e570–e577 · doi:10.1542/peds.113.6.e570

Used for: The three referral thresholds the app screens against every week (the BISQ clinical cutoffs): more than 3 night wakings, more than 1 hour of nocturnal wakefulness, or less than 9 hours of sleep per 24 hours. In validation, night-waking count and nocturnal sleep duration classified clinical vs control cases at 85% accuracy.

Read the source →
Theoretical model2006

Understanding sleep–wake behavior and sleep disorders in children: the value of a model

Jenni OG, LeBourgeois MK · Current Opinion in Psychiatry 19(3):282–287 · doi:10.1097/01.yco.0000218599.32969.03

Used for: Justification for applying the two-process model to infants and children specifically, including the developmental change in how fast sleep pressure accumulates — which is why wake windows lengthen with age.

Read the source →
Theoretical model2016

The two-process model of sleep regulation: a reappraisal

Borbély AA, Daan S, Wirz-Justice A, Deboer T · Journal of Sleep Research 25(2):131–143 · doi:10.1111/jsr.12371

Used for: The homeostatic sleep-pressure curve drawn on the Today timeline. Process S rises exponentially during wake and decays during sleep; the app uses it to explain why a given wake window is being suggested. It is an illustrative model, not a measurement of your baby.

Read the source →

What comes from a trial, and what comes from us

“Evidence-based” often means a real finding wrapped around a lot of invented detail. For each program the app coaches, here is that line drawn explicitly.

Routine & schedule

No crying protocol

Parent education and consistent bedtime routines are one of the intervention categories reviewed by the AASM task force, which found behavioural interventions efficacious in 94% of 52 treatment studies.

From the evidence
That routine and schedule interventions belong to the efficacious category.
Our implementation choice
The specific routine length and the schedule itself are generated from your own logged wake windows and the age reference values.

Bedtime fading

No crying protocol

One of two active arms in a randomized controlled trial of 43 infants aged 6–16 months. Produced large decreases in sleep latency versus a sleep-education control, with small-to-moderate declines in salivary cortisol and no differences in attachment or in emotional and behavioural problems at 12-month follow-up.

From the evidence
The method itself and its efficacy: setting bedtime near actual sleep onset, then advancing it, reduces sleep latency.
Our implementation choice
The 15-minute step size, the two-consecutive-good-nights rule for advancing, and the automatic hold when settling slows are our implementation, chosen to match standard clinical protocol. The trial did not prescribe them.

Graduated extinction

Significant crying

The other active arm of the same randomized trial, and the most studied method in the AASM review. Large decreases in sleep latency; salivary cortisol declined rather than rose relative to controls; at 12 months there were no differences in emotional or behavioural problems or in secure–insecure attachment. A separate randomized trial followed 326 children to age 6 and found no lasting effects in either direction.

From the evidence
The method and its efficacy and safety profile. In the trial, parents chose their own initial checking interval and increased it progressively.
Our implementation choice
The specific ladder of intervals is generated by us from your chosen starting interval. Nothing about the exact minute values comes from the trial.

Gradual withdrawal

Some crying

Parental-presence-fading approaches are among the intervention categories covered by the AASM task force review of 52 studies, which found 94% reported efficacy with over 80% of children clinically improved and maintained at 3–6 months.

From the evidence
That gradual-withdrawal approaches belong to the efficacious category.
Our implementation choice
The five-position ladder and its pacing are our implementation of a commonly described protocol, not a transcription of a specific trial arm.

Safe sleep

These are from the American Academy of Pediatrics 2022 policy statement. No schedule or program in this app will ever suggest anything that conflicts with them, and they are never behind a subscription.

  • On the back, for every sleep

    Supine for every sleep by every caregiver until 1 year — naps included, and by grandparents and childcare too, not just at home.

  • Firm, flat, non-inclined surface

    A firm flat sleep surface free of soft goods. Inclined sleepers and sitting devices are not for sleep.

  • Same room, separate surface

    Room-share without bed-sharing, ideally for at least the first 6 months. The baby sleeps close to your bed, but on their own surface.

  • Nothing else in the sleep space

    No soft bedding, pillows, bumpers, or loose blankets.

  • Avoid overheating

    Dress for the room; avoid head covering and over-bundling.

  • Offer a pacifier for sleep

    Once breastfeeding is well established, offering a pacifier at sleep onset is protective.

Where this app is weakest

  • Wake windows are derived, not measured. No study reports wake windows directly. Ours are computed from published sleep durations, day/night splits and nap counts. The arithmetic is sound and the answers land where clinical practice sits, but it is a derivation and you should treat the first week’s numbers as a starting point rather than a fact.
  • The shape of the day is a heuristic. That the first wake window is shortest and the pre-bedtime one longest is widely practised and poorly evidenced. It is deliberately a small effect here, and your own data overrides it quickly.
  • Parent-reported data has known limits. Almost all the reference values come from diaries and questionnaires, which systematically miss brief wakings that a parent sleeps through. Your data has the same bias, which is partly why comparing like with like still works.
  • The reference ranges are international, not local. Galland pooled 18 countries and found children in predominantly Asian countries slept about 59 minutes less across the 0–12 year range. A single international norm will be slightly off for any given population.
  • Sleep pressure is drawn, not measured. The curve on the timeline is the two-process model applied illustratively. It is not a measurement of your baby, and no consumer device measures it either.