Can More Daylight Help Protect You From Dementia?

A light sensor replaced the questionnaire

Earlier UK Biobank work on daytime light and dementia relied on people's recall. A 2022 analysis of the UK Biobank, a long-running study of about half a million British volunteers, asked participants how long they spent outdoors and linked those answers to later dementia. Self-reported time outside is a rough measure, though, since it says nothing about how bright those hours were or how much light people got indoors.

The new analysis, published in June in General Psychiatry, drew on the same cohort but used sensors. It followed 87,577 dementia-free adults, average age about 62, who wore a wrist device with a built-in light meter for a week between 2014 and 2018. Over a median of about eight years, 741 were diagnosed with dementia through GP, hospital, and death records. People whose average light between 7:30 a.m. and 8:30 p.m. reached 1,000 lux or more had about 16% lower risk than those below that level. That figure comes after accounting for demographic, lifestyle, environmental, and medical differences that could help explain the gap. It also held up after a correction for testing several light measures at once, and when shift workers and people with vision problems were left out or deaths from other causes were accounted for.

A one-sixth difference in relative dementia risk could matter if light is what produces it, and this kind of study can't settle that.

Nobody in the study was assigned to brighter days, and the most obvious alternative explanation runs backward. Changes leading to dementia can begin years before diagnosis, and declining activity could keep someone indoors long before anyone names the cause. Statistical adjustment can't fully remove that, or other differences between bright-day and dim-day people that weren't measured.

The light measurement itself was thin. One week stood in for years of habits. The sensor sat on the wrist, where a sleeve can cover it, and it can't see what reaches the eyes, which is the route by which light affects the body clock. The recordings may not reflect today's lighting habits, and UK Biobank volunteers are healthier and less deprived than the British population, so the authors themselves say the lux cut-offs may need recalibrating in other groups.

The headline targets deserve the most caution. The researchers looked for cut-offs in the exposure-risk curves and then tested those cut-offs in the same data. None of these cut-offs was set in advance or checked in an independent group, and when daytime was defined by actual sunrise and sunset, some associations weakened enough to lose statistical significance. The link also looked stronger in evening types, in carriers of the APOE e4 gene variant, and in people with brighter nights; those differences held up under formal testing but haven't been replicated, so they don't justify genetic testing or a special light plan.

Night light came out close to neutral, with people in the brightest quarter of nighttime exposure having about the same estimated dementia risk as those in the darkest. The estimate still allowed for lower or higher risk, though, and a wrist under the covers can read as darkness whatever the bedroom is like, so the study doesn't establish that light at night is harmless for the brain.

What happens when light is added on purpose

If daylight protects memory, deliberately adding light should show some effect. Intervention studies exist, but none has established that extra light prevents dementia, and what they show is narrower.

A recent trial was published in July 2026. It enrolled 61 older adults with mild cognitive impairment and sleep problems, who got tailored home lighting or placebo lighting for 24 weeks. A memory subsection of a standard cognitive test moved in favor of the active lighting, but overall cognitive scores didn't differ significantly, the adjusted analysis found no significant cognitive treatment effect over time, and the cognitive comparisons weren't corrected for multiple testing. It's an encouraging hint in a small group that already had symptoms, and it tells you little about people whose memory is healthy now.

In people already living with dementia, results are similarly uneven. A placebo-controlled nursing-home trial found better caregiver-rated sleep but no clear change in sleep measured by wrist devices.

Where daylight belongs in your routine now

If you're already living with dementia and sleep poorly, daytime daylight and physical activity are part of current NHS advice, along with a night light or blackout blinds depending on your needs. That use rests on symptom management and doesn't depend on settling the prevention question.

For prevention, this evidence doesn't justify buying a light box or tracking lux on a wearable. If you'd like more daylight anyway, time outdoors costs little, and nothing in this research argues against it. Getting outside doesn't mean giving up sun protection: shade, protective clothing, sunglasses, and sunscreen with UVA and UVB protection still apply, and so does not looking directly at the sun. A therapeutic light box is a different matter, because certain eye conditions and medicines that increase light sensitivity can make one unsuitable, so that choice is worth checking with a clinician first.

What would settle the prevention question is a large, long randomized trial in dementia-free adults that sets the light exposure in advance, measures how much light reaches the eyes and how well people stick with it, and counts dementia diagnoses along with any harms. Independent cohorts with repeated light measurements, and analyses designed to separate early disease from cause, would show whether a level like 1,000 lux holds up beyond healthy British volunteers.