Evening Wind-Down

What is behind it

Two pathways run in parallel. The circadian one: light in the evening inhibits melatonin release via melanopsin-containing retinal cells. Compared with dim light below 3 lux, room light below 200 lux delayed melatonin onset in 99.0% of people and shortened melatonin duration by about 90 minutes.

The thermoregulatory pathway: a warm shower increases blood flow to the palms and soles, enlarges the distal-to-proximal temperature gradient and thus accelerates the drop in core body temperature — the signal to fall asleep. Added to this is the autonomic component: slow breathing lowers heart rate and increases heart rate variability. The term “activating the parasympathetic nervous system” is backed by these markers — but it is supported for the breathing exercise, not for dimming and not for the shower.

What the individual components deliver

Light: for the evening there is an international consensus recommendation with concrete numbers. From at least 3 hours before bedtime, at most 10 lux melanopic EDI at the eye, and at most 1 lux in the sleep environment. Warm shower: water at 40 to 42.5 °C improved sleep quality and sleep efficiency; 1 to 2 hours before bedtime, even 10 minutes significantly shortened sleep onset latency.

Breathing exercise: in a crossover study with N = 42 students, 5 minutes each of slow breathing at 40%, 60% and 80% of the spontaneous breathing rate, lower rates reduced arousal — lower heart rate, higher heart rate variability, higher peripheral skin temperature.

What the routine has to compete with

The benchmark is full cognitive behavioral therapy for insomnia. A meta-analysis of 30 RCTs with non-active control groups shows Hedges’ g for the Insomnia Severity Index of 0.64 after 3 months, 0.40 after 6 months and 0.25 after 12 months. Against this benchmark, an isolated evening routine loses — but remains effective even in losing.

What the studies show

Sleep hygiene as the sole measure (Ruan et al. 2025)

Systematic review with meta-analysis, seven databases up to September 30, 2024, RCTs only, risk assessment with RoB 2. Scope: 42 RCTs, 4,245 adults, 65.5% women. Sleep hygiene education alone improved the Insomnia Severity Index by MD 3.4 (95% CI 2.08–4.64), but was inferior to full CBT-I (MD 3.8; 95% CI 2.92–4.76), partial CBT-I (MD 4.5; 3.33–5.60), exercise (MD 2.9; 0.81–5.04) and acupressure (MD 1.9; 0.82–3.00).

Room light before bedtime (Gooley et al. 2011)

Controlled laboratory analysis, n = 116 healthy volunteers between 18 and 30 years, at least five days in the research center. Compared with dim light below 3 lux, room light below 200 lux delayed melatonin onset in 99.0% of people. Light during the hours of sleep suppressed melatonin by more than 50% in 85% of the runs.

The warm shower in meta-analysis (Haghayegh et al. 2019)

Of 5,322 hits, 17 papers met the criteria, and 13 provided comparable quantitative data. Passive body heating with water at 40 to 42.5 °C improved sleep quality and sleep efficiency; 1 to 2 hours before bedtime, even a duration of 10 minutes significantly shortened sleep onset latency.

Blue blockers in the Cochrane review (Singh et al. 2023)

17 RCTs with samples of 5 to 156 participants. No study had a low risk of bias in all seven Cochrane RoB 1 domains, and only 35% were preregistered. No meta-analysis could be calculated for any endpoint. On sleep quality: very low certainty of evidence, six RCTs with a total of 148 participants — three reported an improvement, three no difference.

Where the routine reaches its limit

The evening routine has been tested as an isolated measure and found inferior: compared with full CBT-I, sleep hygiene trails by 3.8 ISI points — the difference from the full version is larger than the effect of the routine itself (MD 3.4). For the blue-blocker component the picture is even thinner: according to the Cochrane review, it cannot be said whether blue-light-filtering lenses are superior to comparison glasses for sleep quality — very low certainty of evidence across six RCTs with a total of 148 participants. For daytime sleepiness it was likewise very low, across 2 RCTs with 42 participants.

The time frame “no screen 1 h before” is not supported by studies. What is supported is the light pathway: room light in the 8 hours before bedtime delays melatonin onset, and the consensus recommendation sets the 10-lux limit from 3 hours before. Screen use in any case does not act only via light, but via content, arousal and later bedtimes. That “always the same order” is necessary has not been tested either. And the breathing component has been measured acutely: relaxation, heart rate, HRV and skin temperature directly after 5 minutes of breathing, not sleep onset latency the same night.

How to do it

For three components it is possible to state concretely what the studies did. Light: at most 10 lux melanopic EDI at the eye from at least 3 hours before bedtime, at most 1 lux in the bedroom. Even room light below 200 lux over the preceding 8 hours shortened melatonin duration by about 90 minutes.

Warm shower or warm bath: 40 to 42.5 °C, planned 1 to 2 hours before bedtime, with significantly shortened sleep onset latency from a duration of 10 minutes. Not immediately before lying down — the interval is part of the mechanism. Breathing exercise: 5 minutes at a reduced breathing rate, the lower the rate the stronger the drop in arousal. And for context: in the 42 RCTs, sleep hygiene was delivered as structured education, not as a list to follow on your own.

Safety

This is the safest tip in this group: dimming the lights, tea, a book and a warm shower carry no relevant risks. There are nonetheless four caveats. Blue-blocker glasses: in nine RCTs with 333 participants adverse events were reported, and in three they occurred; those mentioned are increased depressive symptoms, headaches, comfort problems with wearing them and lower mood. Warm bathing: in cardiovascular disease, low blood pressure or pregnancy, the decision belongs in a doctor’s hands. Breathing exercises are not harmless for everyone — people with panic attacks, with heart or lung disease and pregnant women are mentioned. And fourth: persistent insomnia is a case for CBT-I, not for more routine.

Sources

• Ruan et al., Sleep Medicine Reviews 2025
• van der Zweerde et al., Sleep Medicine Reviews 2019
• Gooley et al., J Clin Endocrinol Metab 2011
• Brown et al., PLoS Biology 2022
• Singh et al., Cochrane Database of Systematic Reviews 2023
• Haghayegh et al., Sleep Medicine Reviews 2019
• Moebus et al., iScience 2026
• Sharma et al., Frontiers in Psychiatry 2026