Is Light Pollution Giving Everyone Alzheimer's?

Despite the recent headlines, there's no good evidence that light pollution is destroying our brains

Pictured: A whole lot of lights. Photo by michael schaffler on Unsplash

Light pollution is a perennial issue. It’s deeply irritating to pretty much everyone at one point or another - we all need to sleep sometime - but it’s also unavoidable as long as we want to have human activity at night. If you want stuff like road safety and 24hr drive-through MacDonald’s you have to accept a certain quantity of mild irritation as well.

But according to recent headlines, light pollution isn’t just mildly annoying. It’s killing us. A recent piece of research, it seems, has shown that light pollution is potentially causing Alzheimer’s disease across the United States of America.

If true, this would be rather terrifying. The only real solution to light pollution is to get away from people, and that’s not an option for the majority of Americans who live in cities. Certainly not easily.

But fortunately for all of us, the data underlying these headlines is extremely bad. There’s no need to worry about light pollution killing you just yet.

The Study

The study in question is a new paper published in the journal Frontiers in Neuroscience that looked at an ecological comparison of state-level light pollution and Alzheimer’s rates between 2012 and 2018. The authors found that there was a significant positive correlation between average light pollution at night, measured by publicly-available measures of total light intensity at a state level, and rates of Alzheimer’s disease.

This is, for those in the know, hilariously useless. Terribly weak methodology. I can hear the epidemiologists, statisticians, and other experts who read my pieces laughing themselves silly as I’m writing.

Stock photos for “laughing” are brilliant. Photo by Darius Bashar on Unsplash

For everyone else, let’s look at what that paragraph means. The word ecological is simple - rather than looking at individual rates of Alzheimer’s disease and light pollution, the authors aggregated together entire states and looked at the association at this very high level. Think Texas vs California, or Wyoming vs New York (they did exclude Alaska and Hawaii, although they gave no good reason for this).

This immediately reduces our confidence in the study. I’ve written about the issues with this sort of ecological analysis before. It’s entirely possible that people with Alzheimer’s disease have lower exposure to light pollution. Statewide averages are very useful for some things, but they’re not great at giving us an idea of whether people who have degenerative neurological disease are actually exposed to the thing that these authors are studying.

Next is the issue with how the authors calculated light pollution. They took data from a publicly-available repository of nighttime brightness derived from satellite estimates between 2012-2018 and calculated an average across each state in the USA. This is, in a word, ridiculous. Different states have different population densities, which makes it a weird waste of time to calculate unweighted nighttime light intensities of entire states. You can see the issue very clearly by posting the author’s map of light intensity next to the map from Wikipedia of US population density - the authors appear to have basically just measured how densely-populated each state is with their strategy.

Left, image from Wikipedia on population density. Right, image from study in question.

This immediately makes all of the study results a waste of time. Without correcting for population density, all the nighttime light intensity data tells us is how many people each state has per square mile.

The authors did also run their analyses at the county level, but they used the same basic methodology. While counties are much smaller than states, they are still sufficiently large that the analysis makes very little sense.

The other major problem here is how the authors defined Alzheimer’s disease. They used Medicare records at a state level to identify what percentage of the state/county had a diagnosis. But the problem with such aggregate data is that it only shows that people have had a coded diagnosis of the disease. For people to get that coding, they have to first go and see a doctor and get diagnosed. It’s entirely possible that this methodology just shows which states have better access to Alzheimer’s diagnosis, rather than showing light causing an increase in the disease itself.

In addition, as the authors noted, they measured light in a somewhat useless way. Looking at the average amount of light that satellites pick up from space at night doesn’t really tell you how much light people are exposed to. Some people who live in very bright areas will have blackout blinds and not see much light after dark - others who live in generally dark places might have a streetlamp right outside their window and get a lot. Some people spend most of their nights inside in a brightly-lit room. Measuring outdoor light pollution tells us very little about the actual amount of nighttime light that people see.

There’s also the hilarious fact that the authors kind of disproved their association. The main findings that are reported are the uncorrected statistical model which doesn’t control for confounding, but in a model that controlled for potentially confounding variables the authors did not see an association between light pollution and Alzheimer’s. Once they had accounted for the risks imparted by stroke, obesity, and other risk factors for Alzheimer’s, there was no longer any relationship at all between the disease and light pollution.

Bottom Line

This study tells us very little about whether light pollution could be causing Alzheimer’s disease. Even if we ignore the issues with how the authors defined light pollution, at best we could say that there is a high-level ecological association between a fairly useless measure of light exposure and the county or state level rates of Alzheimer’s diagnoses.

There are so many things that could be causing that association that there’s very little you can take away from the publication.

The theory behind light pollution causing Alzheimer’s is also very thin. There are a handful of studies in insect models of Alzheimer’s indicating that there may be some impact of sleep disruption on the development of the disease, but it’s hard to go from studies in flies to impacts in humans. The basic theory is that nighttime light causes sleep disruption, and sleep is related to brain degeneration, but we still don’t even have a solid grasp on how sleep patterns impact Alzheimer’s never mind how that’s mediated by light.

Pictured: Despite what Kafka might’ve told you, this is not in fact a human being in great distress. Photo by MOHD AZRIEN AWANG BESAR on Unsplash

It’s possible that light pollution is causing brain degeneration. But there’s currently no strong reason to worry, at least based on the evidence we’ve got so far.

There’s definitely data showing that having a dark bedroom can help you sleep, so if you’re worried about light pollution I’d say grab some blackout blinds. It’s also not great to stay up all night playing video games, although I doubt the people doing that are overly interested in the possible health impacts of a 24-hour Elden Ring binge. Otherwise, I’d try not to stress about the headlines, they’ll only make it harder to sleep.