Does outdoor lighting affect human health?

Image credit: Pexels (CC0)
1932 words / 8-minute read

Summary: ​Natural cycles of light and dark are needed for ideal wellbeing. Artificial light at night disrupts this rhythm, but does it lead to ill health? This post explores the surprisingly complicated relationship between light at night and human health, and what we still need to learn to find the optimal balance between them.

Light at night (and light pollution) are in the news these days for a reason that may surprise some readers. Consider some recent headlines:

"Bright outdoor light at night may increase stroke risk"
"Does More Outdoor Light at Night Help Cause Macular Degeneration?"
"The light you see may be as important to your health as the food you eat"
"Getting Too Much Bright Light at Night May Increase Your Cancer Risks"

The popular media are notorious for running stories about human health that seem contradictory. What's good for people one week seems harmful to them the next. Some of this results from the ways the media report news about science in general. But more often it reveals something about the often messy process by which we collect and judge evidence.
​
The same is true of questions about light and health. What do we really know about this subject? While some conclusions drawn by researchers are concerning, we're still a long way from definitive answers. In this post, we'll dig into the evidence and learn whether there's a real 'dark side' to light at night.

What you see is what you measure

Scientists study artificial light at night (ALAN) and its effects on human health along three lines of enquiry:

• Laboratory studies in controlled conditions
• "Real-world" cohort studies
• Correlation studies that match remote measurements of light with data on disease incidence

These are standard research approaches and none of them is especially problematic. Of course, there are always details of study design that are better or worse in different regards. But these basic methods are well understood, as are the limitations of their results. To get the "big picture" of what ALAN may do to our health and wellbeing requires using all the tools we have.

Laboratory studies are usually small, involving a few volunteers exposed to ALAN. Researchers measure their responses to light with various tests. While the controlled exposures mean that the "dose" of light is definite, the sample sizes are small. Information they yield can be compared against results from cohort studies. These look at effects in larger groups of people ("cohorts") in situations where the exposure isn't controlled as well. The idea there is that the larger sizes of cohorts make it easier to find the individuals who are not very representative of the group. If the presumed health effect in a cohort study is real and prevalent, it should stand out in the results.

Both laboratory and cohort studies of ALAN and human health tend to consider only indoor light exposures. And those studies show that ALAN affects people in ways that harm them. (For a comprehensive review of what we know about ALAN and health, see DarkSky International's "Artificial Light at Night: State of the Science 2023" report.) Effects reported in scientific papers include:

• Sleep problems
• Increased inflammation throughout the body
• Temporary changes at the genetic level
• Metabolic disorders (obesity, diabetes, etc.)
• Higher risk for cardiovascular disease and certain kinds of cancer
• Problems with mental health including mood disorders

​These observed changes largely stem from the tendency of ALAN to disrupt the body's "circadian rhythm". This is a roughly 24-hour cycle of biological activities tied to the rising and setting of the Sun. Over billions of years, living things became sensitive to the reliable sequence of light-dark-light-dark in nature. ALAN mimics certain qualities of sunlight, which confuses organisms. And because ALAN is so new in the environment, they have not yet evolved adaptations to it.

The view from above

All this marks ALAN as an emerging "lifestyle risk" for people. We use light at night for many reasons associated with modernity and progress. Science is telling us that there can be too much of a good thing in that regard. But so far we have only considered evidence linked specifically to indoor light at night exposure. Should outdoor exposures also concern us?

That's where the picture gets murky.

The third kind of study mentioned above attempts to answer the question by looking down from overhead. The world's cities shine at night with an intensity that makes them visible to astronauts aboard the International Space Station. Since the 1970s, satellites orbiting the Earth have measured "nighttime lights" (NTL) on our planet.

The signals they sense represent only a fraction of total light emissions. While some light emitted upward escapes the atmosphere completely, other light scatters back down to the ground. Simple models of the situation imply some relationship between light levels on the ground and the intensity of NTL seen from orbit. Reality is more complicated, and the actual relationship can be quite complex.
Potential trouble results when researchers use NTL measurements in human health studies. The idea seems reasonable enough: match up NTL data with information about the spatial distribution of various maladies. Of course, there are many other social and environmental factors that influence human diseases. Once we account for and remove those factors in the data, only the effects of ALAN should remain. But do they?

On the surface, the answer appears to be "yes". Studies find correlations — that is, mutual relationships or connections between two or more variables. People who live in areas with higher NTL levels do seem to suffer more from various illnesses than people who live in darker places. That conclusion persists even after eliminating other possible explanations for the results.

Let the data interpreter beware

It is said, famously, that "correlation is not causation". While correlation is the appearance of a relationship between variables, it says nothing about why they are related. Causation is the idea that cause precedes effect, and that for every effect there is a cause. Nature and society are rife with spurious correlations, in which there is no plausible cause-and-effect relationship.

It's fair to ask whether there's an underlying cause for the observed correlation between NTL intensity and health problems. Do the laboratory studies sufficiently establish a causal link? We don't know. They reveal a clear connection between ALAN exposure and ill health in both individuals and groups. But they don't (yet) tell us what the threshold of harm is.

Researchers refer to the "dose-response" relationship between cause and effect. So far, we don't have a strong handle on that even under controlled conditions. In the case of outdoor lighting, we know even less about individual exposures. And there's reason to believe that the majority of light exposure most people get is from distinctly indoor sources. (Think about watching television or staring at the screen of your phone late at night, long past your usual bedtime.)
​
The bottom line is that the effect of outdoor lighting isn't zero, but the amount of indoor light exposure most people get probably dwarfs it. Most studies claiming one health effect or another from ALAN rely on NTL data as an estimate of actual exposures. But therein lies the problem: we don't know what those exposures really are. So we can't take extreme positions like "streetlights give people cancer". The evidence just doesn't support that.

(Even limited) knowledge is power

So what can we conclude from all this? Is it hopelessly confused?

For the benefit of preserving health and wellbeing we should try to limit all ALAN exposure, whether from indoor or outdoor sources. There is little scientific doubt that the observed connection between light and health is real. The hypothesis about why, with its strong biological underpinnings, is sound.
Some of it has to do with timing. During the daytime, we need exposure to bright light, especially that rich in short wavelengths (i.e., blue). This helps calibrate the internal 'clock' that maintains the circadian rhythm. Inadequate daytime light exposure is almost as harmful as too much ALAN exposure. While humans have some resilience to seasonal changes to the length of the day, we have no natural defense against the influence of ALAN.

What can we do with the information we have in hand, knowing it is incomplete? There is a case for taking steps to limit ALAN exposure, especially in sleeping rooms at night. Those spaces should be completely dark to avoid disrupting the circadian rhythm. Sleeping with lights on at night is not advised. Good "sleep hygiene" includes avoiding bright light exposure in the hours before going to bed at night.

Heavy window coverings may be needed to prevent the intrusion of outdoor lighting into sleeping rooms. While we know that exposure to bright indoor sources of light at night is bad for people, we know much less about how very dim outdoor lighting coming in through windows affects us. As one prominent light/health researcher once put it, "Photons are photons, and the body and brain don't care where they come from."
​
The owners of outdoor lighting can take care to ensure light is not shining into people's homes. Public policies governing lighting can assist by making this "light trespass" against the law. They can also assist by limiting the intensity, color, and hours of operation of outdoor lighting. That approach also benefits nocturnal wildlife, because plants and animals can't go indoors and draw the blinds to avoid ALAN exposure.

Where lighting-and-health research goes from here

When practiced well, science is a remarkable self-correcting process. New information suggests new explanations that replace old ones. That is clearly true for health research in general, and more so for efforts to understand how ALAN affects our wellbeing. Although we have learned much, there are still significant unanswered questions. These include:

• Does measured exposure to ALAN in outdoor space affected human health?
• Does outdoor light at night entering indoor spaces affect sleep and health?
• Are the observed correlations between outdoor ALAN and health problems the result of cause and effect?

These questions suggest directions for future research. We need more compact and reliable light sensors small enough that study participants can wear them. Such "wearable dosimeters", especially those that provide color information in addition to light intensity, are a potential game-changer. They may be able to better discriminate the indoor and outdoor components of ALAN exposure to disentangle their effects. That would, in turn, make light and health research less dependent on NTL measurements as proxies for real exposure data.

Of course, that is not to say that we should stop relying on satellite measurements altogether. But the available nighttime satellite data isn't optimized to address health research questions. It often doesn't allow for detailed identification of outdoor light sources in relation to where people may be seeing them. Satellites may pass over a given location at night at irregular intervals or at times of night when little human light exposure is expected. And some don't detect all the lights we know exist on the ground.

Researchers are supplementing existing NTL data sources with new ones. Some have experimented with platforms closer to the ground, such as high-altitude balloons and unmanned aerial vehicle (also known as 'drones'). These get much closer to light sources and can measure them in more detail. They are coming up with new ways to ground-truth the remote measurements by collecting information about lighting on the ground. As scientists pair these new data about outdoor lighting with better measurements of the 'dose' it yields, the quality of health research results will improve.

​When electric lighting was introduced to the world in the late nineteenth century, it was celebrated for its myriad benefits to society. Almost 150 years on, we now understand it to represent both promise and peril. As our knowledge increases, we may yet find ways to balance risk and reward that cater to social needs for ALAN while avoiding poor health outcomes.