Gas Stoves, Finally

I did it! I finally wrote this! I meant to do it in November. And it’s only a day late in posting, too.

Gas stoves: The next terror?

Sometime this fall, as if we didn’t have enough stuff to worry about, a new panic ensued about gas stoves. This very thorough article in Slate exemplifies much of the debate and concern. Basically: the worry is that the byproducts of natural gas (particulate matter, nitrogen dioxide, carbon monoxide and others) cause health problems in kids in particular. They might exacerbate asthma or other respiratory problems, an issue which has added valence in the era of COVID-19.

Part of the source of the panic is a lot of people (myself included) have gas stoves and like them. Relative to electric stovetops, gas is easier to control in cooking (electric coils are often harder to adjust) and, of course, you can’t blister peppers on an electric stove. (What’s that? You’re not frequently blistering your own peppers? Hmmm…) Perhaps more importantly: switching out stove types (let alone natural gas boilers, etc, etc) may be really, really expensive.

Given the “switching costs” (as we economists would call them), I think it’s important to separate the concern into several possible levels.

This is not a concern, do nothing
This is a small concern so if you were choosing from scratch, it would be better to do electric or (this is a newer, more expensive option) induction cooktop
This is a really serious concern and you should invest significant funds in switching from gas right now.

Which is it?

Background and data on indoor air pollution

Indoor air pollution is a big issue for human health, especially in the developing world. The WHO cites it as the most important environmental health risk in the developing world. Most exposure in the developing world comes from indoor cookstoves, and higher exposure levels are associated with worse health and increased mortality. Improving cooking technologies is an important policy issue, and many NGOs are focused on technologies like solar cookstoves which avoid open indoor flames.

There is a big step, though, between indoor cookfires in small dwellings and natural gas stoves. The level of indoor air pollution produced by gas stoves is significantly lower than open cook fires. The question is: at the levels you might experience, is there evidence of harm?

Academic studies on this take a familiar shape. This 2013 meta-analysis combines 41 of them and argues for a small increase in asthma symptoms from indoor gas cooking. It’s often useful to dive into the individual papers in these summary analyses to see what’s under the hood.

Here is one example: a 2004 paper which found a large increase in risk of asthma from gas cooking. The study covers almost 6,000 children in Russia. Their parents are surveyed about living conditions and a number of respiratory health measures for the children. The authors find that asthma risk is elevated for kids who live in a house with gas cooking. This particular study has a lot of different analyses, and some of the result are mixed (for example, gas stoves do not increase the risk of asthma symptoms or any other respiratory risks, only “Doctor-Diagnosed Asthma”).

Among the largest papers in the meta-analysis is a publication from 1991 which analyzed the determinants of asthma and wheezing in almost 18,000 young Canadian children. The authors find a significant link between gas cooking and asthma (but not wheezing). They caution, though, that gas cooking is rare in their data and even with a large sample size they worry that because this group is uncommon, there could be other things driving the results.

In general, these studies all suffer from similar problems. They are all observational (i.e. exposure to gas stoves isn’t randomized). They are all pretty imprecise (i.e. they are statistically consistent with a wide range of values). They are all testing a wide variety of hypotheses (many treatments, many outcomes) which contributes to worries about statistical power. In a number of cases the populations chosen are explicitly high risk, and many of these are older studies run in less-developed countries. Their general-population relevance for the current period in the US isn’t obviously nil, but there is some stretch.

When the meta-analysis puts everything together, they come to the conclusion that gas cooking produces a small increase in asthma risk for children, perhaps a 1.5 percentage point increase in risk.

It is worth saying that many things increase risk of asthma, including other types of pollution (i.e. living near a highway) and parental smoking. Data just out this year shows a (correlational) link between asthma and bedroom carpeting.

By far the most important factor in asthma is genetics. The impacts of gas stoves here (or any other environmental causes) pale in comparison to the predictive value of a family history of asthma. Of course, you cannot control your genes, but it may be useful to simply note that other factors matter more. Gas stoves are not some kind of unique risk.

What’s new in the indoor air pollution data?

This meta-analysis was published in 2013. Why all the new attention now? One reason is COVID-19 — we are thinking more about respiratory health, and conditions like asthma do raise the risks from COVID-19.

There are also a couple of new summary reports out (one from RMI and one from UCLA) which talk through these issues, although they do not contain new data analysis.

So, the short answer, I think is that there isn’t anything especially new, just new awareness.

What does it all mean?

First of all: there are other considerations around gas stoves which deserve mentioning. For example, NEJM published an opinion piece in 2018 arguing against the use of natural gas for climate reasons.

Second: how you think about this is going to depend on the health situation with you and your kids. I noted above that family history is among the most significant predictors of asthma. Other data suggests that the indoor air environment can exacerbate asthma symptoms in kids who already have it. This means that some of these considerations may differ if you have a child with asthma or a family history.

Third: The issue isn’t gas stoves per se it’s the byproducts of cooking on them. If you have an efficient range hood, this may remove some of these products. If your cooking space is larger and more open, that will help. This study shows that in terms of nitrogen dixoide concentrations, putting a hepa filter in your kitchen achieves about half of the effect of fully replacing your gas stove. This is perhaps the most positive concrete news: hepa filters are widely available and not very expensive.

Putting this all together…here is my take.

Gas stoves seem like they might slightly exacerbate asthma in children who already have it, or who are prone to it.
I am less convinced by the data that these are a significant factor in driving risk for kids in general.
If you are starting completely from scratch, it might be worth considering a non-gas-stove if you are otherwise neutral. For most people who cook a lot, gas remains the mechanism of choice, so this could push the other way. Induction cooktops are expensive so, again, that might push the other way.
If you already have a gas stove, and you are very worried about this, get a hepa filter.

UPDATE: March 22, 2021

Readers on gas stoves

A number of you wrote me after the gas stove posts, and your thoughts were so helpful I wanted to share. They came in three forms.

A number of people pointed out a key non-asthma reasons to avoid gas stoves (and gas heat): the contribution of natural gas to climate change. This is among the central arguments of that NEJM article I cited at the end of the piece. This is absolutely correct and worth factoring in, especially when thinking about new construction.
Several people pointed out that if you are worried about air quality, you can get an indoor air monitor, which can give you a sense of how your indoor air quality is, and how it varies with cooking. If you do choose to get a portable hepa filter, you can also then monitor how it is working.
Finally, a number of you pointed out that induction cooking has gotten much more accessible and affordable over time. Someone sent this link to an portable induction cooktop from Amazon (reminder: you also need magnetic-bottom cookware to use this). They said they don’t always use it, but for many kinds of cooking it’s better than gas (see some discussion of the tradeoffs here). This suggests another intermediate approach along with the Hepa filter.

Hope this helps and thank you to all of you for input!