Bedroom at 16–18 °C

What is behind it

Thermophysiologically, falling asleep is a process of heat loss. The blood vessels in the hands and feet dilate, warm blood flows to the periphery, the body gives off heat, and core temperature falls. This is measured as the distal-proximal skin temperature gradient.

In a constant routine protocol, precisely this gradient was the best predictor of sleep onset latency — better than core temperature, its rate of change, heart rate, melatonin onset and subjective sleepiness. Two things follow from this: the environment must be cool enough for the heat to dissipate. And warm feet are not a contradiction of the cool room but the same measure from the other side.

Why heat is the problem and not cold

The review on thermoregulation during sleep provides an asymmetry that is decisive for the assessment. Under real conditions with bedding and clothing, heat increases wake time and reduces deep sleep and REM sleep; humid heat intensifies this. Cold, by contrast, does not affect sleep stages as long as bedding and clothing support thermoregulation.

That does not make cold ineffective: it changes the cardiac autonomic response during sleep without sleep stages or subjective perception changing. In semi-nude subjects the picture is even reversed; there, cold affects sleep stages more strongly than heat. Bedding is therefore part of the intervention, not an accessory.

What was measured in real bedrooms

A longitudinal study with wearable sleep monitors and environmental sensors in participants’ homes found the most efficient and most restful sleep between 20 and 25 °C. When the temperature rose from 25 °C to 30 °C, sleep efficiency fell by a clinically relevant 5–10%. The associations were mostly nonlinear, and the differences between individuals were considerable.

At population level this fits the picture: linking survey data from 765,000 US respondents from 2002 to 2011 with nighttime temperature data showed that higher nighttime temperatures increase the number of reported nights of insufficient sleep, most strongly in summer and among older people.

What the studies show

A warm shower or bath before bedtime

Systematic review and meta-analysis. Of 5,322 screened papers, 17 met the inclusion criteria and 13 provided comparable data. Endpoints included sleep onset latency, sleep efficiency and subjective sleep quality. Result: water at 40–42.5 °C, 1–2 hours before bedtime, improved sleep quality and sleep efficiency, and even 10 minutes’ duration significantly shortened sleep onset latency. The figure of an average 10-minute reduction comes from the press release of the first authors’ university, not from the abstract.

Bedroom temperature in people’s own homes

Longitudinal study with wearable sleep measurement and environmental sensors in the homes of community-dwelling older adults; sleep duration, sleep efficiency and restlessness were recorded, controlling for confounders. Result: optimum between 20 and 25 °C, a fall in sleep efficiency of 5–10% when the temperature rose from 25 °C to 30 °C. The associations were mostly nonlinear and the differences between individuals considerable. It is an observational study without randomization.

Heat disturbs, cold not in the same way

Review on thermoregulation during sleep. Under real conditions with bedding and clothing, heat increases wake time and reduces deep sleep and REM sleep; humid heat intensifies this. Cold, by contrast, does not change sleep stages as long as bedding and clothing support thermoregulation — but it does affect the cardiac autonomic response, without this being noticeable subjectively. In semi-nude subjects it is the other way round.

The lever is the hands and feet

Constant routine protocol, modified to allow nighttime sleep. Heat loss was recorded via the distal-proximal skin temperature gradient and tested against several candidate variables. In the stepwise regression, this gradient was the best predictor of sleep onset latency — better than core temperature, heart rate, melatonin onset and subjective sleepiness.

Where the number comes from — and where it does not

There is no study for the 16–18 °C range. None of the papers reviewed identified this range as the optimum; the only paper with real bedroom data found 20–25 °C, albeit in older adults and with considerable scatter. The directional claim is not right either: that a cool room promotes deep sleep and REM is not supported — what is supported is that heat reduces both. And the frequently cited lead time of 30 minutes for the warm shower deviates; the meta-analysis works with 1–2 hours, and there is no separate efficacy figure for 30 minutes.

Added to this is the quality of the evidence. There is no randomized trial that directly tested different bedroom temperatures against each other and measured hard endpoints. The bedroom study is an observational study, and the population analysis is based on self-reported rather than measured sleep.

How to do it

What was actually done in the studies: a warm shower or bath at 40–42.5 °C, taken 1–2 hours before bedtime, with as little as 10 minutes sufficing for a significant reduction in sleep onset latency. The room temperature with the best sleep efficiency in real homes was measured at 20–25 °C; the decline began above 25 °C. The World Health Organization gives 18 °C as the lower limit for health in the cold season, more for risk groups. Bedding is part of the measure: whether cold changed sleep stages depended precisely on it.

Safety

In its 2018 Housing and health guidelines, the World Health Organization makes a strong recommendation: for countries with a temperate or colder climate, 18 °C is given as a safe indoor temperature during the cold season, with moderate certainty of evidence. For older people, children and people with chronic illnesses, in particular cardiovascular and respiratory diseases, a higher minimum temperature may be necessary according to the guideline. The lower half of the widespread range therefore lies below an official recommendation. In practice, mold on cold exterior walls is an additional issue. In Raynaud’s syndrome, hypothyroidism, peripheral arterial disease or neuropathy, more bedding is the better answer than a colder room. Water temperatures of 40–42.5 °C are not automatically suitable in cardiovascular disease, low blood pressure, a tendency to circulatory collapse, pregnancy or sensory disturbances.

Sources

• Kräuchi et al., Am J Physiol Regul Integr Comp Physiol 2000
• Kräuchi et al., Nature 1999
• Haghayegh et al., Sleep Med Rev 2019
• UT Austin press release on Haghayegh et al., July 19, 2019
• Okamoto-Mizuno and Mizuno, J Physiol Anthropol 2012
• Baniassadi et al., Sci Total Environ 2023
• Obradovich et al., Sci Adv 2017
• WHO Housing and health guidelines, Geneva 2018