Sunscreen: Is it Safe?

Sorry this post is like three months too late

As everyone knows, September is the perfect time to inform people about sunscreen. Or not. Let’s say I got distracted! But bookmark this for next year.

Part of my delay was that I had somehow missed a discussion from the spring about sunscreen safety (can’t imagine why). In case you were also focused elsewhere during this period, the concern is that non-mineral sunscreens contain chemicals which can harm your health in various ways. This concern is not new. In fact, I actually already wrote about it, way back in 2014 (blast from the past here), but it’s worth revisiting given some new data.

What is the point of sunscreen?

A big pitch for sunscreen is skin cancer protection. I’m sure you’ve heard this. However: the evidence on the link between sunscreen use and decreased skin cancer is actually not obvious in the data. The theory is clear — we know that sunburns cause skin damage, and this can promote problematic cell behavior. Beyond that, we know that skin cancer is more common in locations with more sun exposure, suggesting that sun exposure matters.

But what we lack is compelling data showing that use of sunscreen mitigates this link. A 2018 meta-analysis, for example, found no evidence of a link between sunscreen usage and decreased skin cancer rates. This is pretty consistent with the existing literature, which has struggled to find these links.

Before you become a sunscreen-denier, though, it’s worth noting the many limitations of this type of analysis: although there are a couple of randomized trials, their follow-up period is short. Maybe it takes a lifetime of sunscreen use to matter. And observational studies — comparing sunscreen users to non-users — are problematic since use of sunscreen is probably positively correlated with sun exposure.

And! There are other benefits of sunscreen. Notably, not getting sunburned. I am sure we all recall the many painful sunburns of our youth. I do not want my kids to be subjected to sun-induced, fluid-filled blisters on their faces (100% this happened to me, and yes, I do use it to scare them) even if it is not a skin cancer risk.

Are these concerns about all sunscreen?

No. The concerns that people have are about non-mineral sunscreens. These work by blocking UV light with chemicals (the ones that people are concerned about, so there is no avoiding them). They do not apply to sunscreens which use minerals for this purpose. So, zinc-based sunscreens are not tarnished by these worries.

(Chemical-based sunscreens tend to be popular because they are convenient and do not turn your face white, but there are many mineral-based options out there).

It would also be remiss not to mention that wearing a hat and more clothing is another way to avoid the sun. My youth did not feature the long-sleeve bathing suits, but my kid’s definitely does.

What’s the Concern & Evidence?

The primary concern with chemical sunscreen is that the active ingredient — most notably, oxybenzone — is dangerous for health and for the environment. I am going to leave the environmental concerns for now, not because they are unimportant but just due to space and my expertise. Health worries surround disruption to the endocrine system.

Such disruptions could impact puberty, fertility, hormone development, possibly cognitive function. Basically, the endocrine system is very important.

There is a reasonable body of evidence from mice and rats that such disruptions happen. When rats are fed this type of chemical, it seems to affect their uterus size. Evidence from mice shows that exposure of pregnant mice (again, through food) impacts endocrine markers in their offspring. Generally, our understanding of these chemicals suggests this is all plausible.

It is always challenging to port evidence from mice and rats into people. Most transparently, mice are not people. A more specific thing to note is that people are much bigger than mice, and it is difficult to figure out how doses relate. Follow-up work to the first rat study above suggests that you’d need decades of full body daily sunscreen usage to achieve the rat dosage.

In contrast, we have little direct evidence of these impacts in people. One 2017 summary study reports some mixed associations between birth weight and UV-blocking chemical exposure in utero. These birth weight differences are very small, and the authors see nothing on other hormone-related outcomes. This doesn’t mean there aren’t effects, we just do not have positive evidence that they are there.

One way to think about is is that it seems very likely based on the animal evidence (it’s not just mice; also fish) that at some level these chemicals would be harmful. The question is whether these levels are common or possible with normal levels of exposure among people.

In fact, there are really two pieces to this last point, at least as it relates to sunscreen. First, what level of these chemicals in your body is “safe” and, second, to what extent does sunscreen impact this level? This first point — we just do not really know the answer.

It is to this second point which the new evidence I alluded to at the start is finally relevant. The FDA has proposed a new rule which would involve some changes in the kind of regulatory information that would be required for sunscreens. And this new study comes in as a part of informing this proposed rule.

The study is a version of a “maximal human usage” study — basically, they tried to look at what happens at the upper end of human usage. They had people apply sunscreen every two hours over 75% of their body. They applied a lot of sunscreen — think about 2.5 tablespoons of sunscreen for an adult at each application. And then they took their blood samples and tested for hormone levels.

What they found is that the chemicals from the sunscreen were absorbed into the bloodstream as early as the first day and continuing well beyond the end of the study period. The concentrations were high relative to a threshold set by the FDA for presumed safety. That is: the FDA has indicated a level below which they would presume safety and not require further follow-up. The participants in this trial exceeded that.

The bottom line conclusion of this paper is that if the FDA were to adopt their new proposed rule, most sunscreens would need to provide additional safety data. That’s it. That’s the conclusion.

I will note the conclusion is not “Do not use sunscreen.” In fact, the last line of the abstract of this paper is: “These findings do not indicate that individuals should refrain from the use of sunscreen.”

Um…okay? I kind of lost the thread? What should I do?

  • Sunscreen prevents painful sunburns, and this is the most evidence-based reason to use it.

  • The chemicals in non-mineral sunscreen can, in some concentration, in animals, cause endocrine disruption. There is reason to think this would happen — again, with some concentrations — in people. On the other hand, there is little evidence of actual harm in people.

  • With a lot of sunscreen usage, you can see chemical levels in the bloodstream which would be high enough to want to investigate more.

  • Mineral sunscreens are a bit more annoying but not subject to these concerns.

  • Long sleeve bathing suits are an alternative.

I suspect reasonable people will differ on what to do here. I will tell you, we use a chemical sunscreen. But we also wear hats and long sleeve bathing suits and try to stay in the shade.

I’ll go with my tag line here: Try not to think about it. The evidence for risks is quite weak. The annoying-ness of mineral sunscreen is minimal. Probably after I write this someone will email to tell me Zinc is also dangerous. Could be! If you forget sunscreen once, your kid will get a sunburn but this is also not the end of the world.


Look at me! I wrote a whole post without mentioning the virus. Don’t worry, we are back to regularly scheduled programming Thursday, when I expect to have a post on school mitigation as “Safety Lasagna.”

Weigh in!

Keep the thoughts coming. I cannot write back to everyone but I do read all of your emails, I promise.

Write to Me!

Where to Find Me


My Website


Expecting Better