With the clocks gone back and darkness falling before 5pm, many people across the UK are asking how much daylight they need. Most of us spend the short days indoors, and a 2026 study suggests that daytime light, measured at the wrist, may be linked to dementia risk.
Light is measured in lux. According to the study’s authors, ordinary indoor lighting usually gives about 300 to 500 lux, while bright light therapy typically uses more than 3,000 lux. The NHS describes 10,000 lux as the standard for a SAD lamp. So for most people, a daytime average above 1,000 lux, or 42 minutes at 5,000 lux or more, usually means getting outside or sitting very close to a bright source.
The study, published in 2026 in General Psychiatry, used data from 87,577 adults in the UK Biobank, a large health project that recruited people at centres across England, Scotland and Wales. Participants had an average age of 62 and had no dementia at the start. Each wore a wrist device with a built-in light sensor for seven days, which recorded how bright their surroundings were by day and by night.
Researchers from Guangzhou Medical University, the Chinese University of Hong Kong and other institutions then followed the group for a median of 8.1 years. During that time, 741 people developed dementia, identified from GP records, hospital admissions and death records.
People whose average daytime light was above 1,000 lux had a 16% lower risk of dementia than those below that level. Those who spent at least 0.70 hours a day, or about 42 minutes, in light of 5,000 lux or more had a 17% lower risk. The results took account of age, sex, education, deprivation, smoking, alcohol, physical activity, diet, chronotype and several health conditions.
In a separate ranking of 19 predictors, less than 42 minutes a day of light at 5,000 lux or more beat six of 15 known risk factors, including alcohol intake, obesity, air pollution and hearing loss.
The links were stronger in three groups: people exposed to more light at night, evening types or night owls, and carriers of the APOE e4 gene variant, which raises the risk of Alzheimer’s disease. In these groups the cut in risk was 30 to 38% for people with more light at night, 31 to 41% for evening types, and 19 to 27% for APOE e4 carriers. Light at night, on its own, showed no clear link with dementia.
The findings come with clear limits. This is an observational study, so it shows an association and cannot prove that daylight prevents dementia. People who get more light may be healthier or more active in ways the researchers could not fully measure. UK Biobank volunteers are also healthier and better off than the wider population.
The sensors were worn on the wrist rather than near the eyes, and recorded light for only one week. The light data were collected more than a decade ago, before LED lighting and evening screen use became so common.
The authors say the thresholds need testing in other groups before they can guide advice. For now, NHS advice for people with low mood in winter is to get as much natural daylight as possible and go outside every day. Anyone worried about memory problems should speak to their GP.
Psychreg has also looked at how sleep lost during daylight saving can affect brain health.
