People who got more bright daylight had a lower dementia risk in a large UK study

Our body clock, the internal timer that tells us when to be alert and when to sleep, is set mainly by light. Bright light during the day and darkness at night keep it on time. People with dementia, the loss of memory and thinking that interferes with daily life, often have a scrambled daily rhythm, and treatment with very bright lamps can ease some of their symptoms.

Most of us, though, spend our days indoors. Ordinary indoor lighting is roughly 300 to 500 lux (a unit of brightness), the authors note, while daylight outside is many times brighter. A team led by researchers at the Affiliated Brain Hospital of Guangzhou Medical University in China asked whether how much light people actually get during a normal day is linked to their later risk of dementia.

What they did

The researchers used data from the UK Biobank, a long-running study of about half a million British volunteers. Between 2014 and 2018, about 100,000 of them wore a wrist device for a week that measured movement and, through a small sensor, the brightness of the light around them. After removing people who already had dementia and recordings that looked faulty (for example, a sensor that was covered most of the time), 87,577 people remained. They were 62 years old on average, and 57% were women.

From each recording the team worked out:

• the average light level during the day (7:30 to 20:30) and at night (0:30 to 6:00);
• how long each person spent in bright light, at three levels: at least 3,000, 5,000 and 7,000 lux.

They then followed everyone through GP, hospital and death records for a median of 8.1 years. In that time, 741 people (under 1%) were diagnosed with dementia. The analysis took into account a long list of other factors, including age, sex, education, income area, the season the device was worn, air pollution, diet, smoking, alcohol, physical activity, whether people were "morning" or "evening" types, social isolation, and conditions such as diabetes, high blood pressure and hearing loss.

What they found

• People whose average daytime light was above 1,000 lux had a 16% lower risk of a dementia diagnosis than those below it (hazard ratio 0.84).
• More time in bright light went with a similar lower risk: more than 1.4 hours a day at 3,000 lux or more (18% lower), more than 0.7 hours (about 42 minutes) at 5,000 lux or more (17% lower), or more than 0.45 hours at 7,000 lux or more (17% lower).
• Light at night showed no clear link with dementia risk.
• The link with daylight was stronger in three groups: people who also had more light at night (30–38% lower risk), evening types (31–41% lower) and carriers of APOE4, the best-known gene variant that raises Alzheimer's risk (19–27% lower).
• In a computer model that ranked 19 factors by how well they predicted dementia, getting less than about 42 minutes of light at 5,000 lux or more came 10th, between low education and alcohol use, and ahead of factors such as obesity and hearing loss.
• In extra analyses, part of the link appeared to run through steadier daily activity rhythms and through the size of some brain areas measured on MRI scans (in about 16,900 people). Only one brain area, the fusiform cortex, held up after the statistical correction for testing many things at once. Blood vitamin D levels did not explain the link.

What to keep in mind

This is an observational study: it shows that brighter days went together with lower dementia risk, not that light lowers the risk. People who get more daylight may differ in other ways that the team could not fully measure; adjusting for many factors reduces this problem but cannot rule it out.

The authors list several limitations of their own. Light was measured for only one week, which may not reflect long-term habits. It was measured at the wrist, not at the eye, so it is only a rough estimate of the light that actually reaches the brain's clock, and the sensor cannot tell darkness from a sleeve covering it. UK Biobank volunteers are healthier and better off than the general population, so the exact thresholds (such as 1,000 lux) may not carry over to other groups. The data were collected before LED lighting and late-night phone use became as common as they are now. Several of the "mediation" results were no longer significant after correcting for multiple tests and should be seen as exploratory.

It is also worth knowing that the cut-offs (1,000 lux, 42 minutes and so on) were chosen from the shape of the data in this same study, so they need to be confirmed in other groups of people. And the extra risk reductions in subgroups come from smaller numbers and are less certain than the main result.

Why it matters

Daytime light is easy to measure with a cheap wrist device, and the link held after accounting for most well-known dementia risk factors. The authors suggest it could help identify people at higher risk and that trials should now test whether more daytime light actually lowers dementia risk. Until such trials are done, the study adds to the case that the body clock, and the light that sets it, is part of the story of how the brain ages.