Fans in the Nursery and SIDS: What the Research Actually Found

If you’ve spent any time on parenting forums, you’ve probably seen the line that running a fan in the nursery cuts the risk of sudden infant death syndrome (SIDS) by 72%. It’s a number that sticks, and it’s easy to see why a worried parent would want to act on it. The number is real, in the sense that it appears in a published study. But what it means is a lot narrower than the headline suggests, and the people who write the official safe-sleep guidance in the US have looked at it closely and decided it isn’t strong enough to build a recommendation on.

This guide walks through what that study actually did, why its design limits what it can tell us, what the American Academy of Pediatrics (AAP) says about fans and overheating, and what the evidence clearly supports instead.

Where the 72% figure comes from

The study is Coleman-Phox, Odouli and Li, “Use of a fan during sleep and the risk of sudden infant death syndrome,” published in Archives of Pediatrics and Adolescent Medicine in October 2008. The researchers, based at Kaiser Permanente in California, ran what’s called a population-based case-control study across eleven California counties, conducted between May 1997 and April 2000.

Here’s how it worked:

• Cases: mothers of 185 infants whose deaths had a confirmed SIDS diagnosis.
• Controls: 312 randomly selected living infants, matched to the cases on county of residence, maternal race or ethnicity, and age.
• Data collection: interviewers trained in SIDS grief counseling asked mothers about the sleep environment during the baby’s last sleep (for cases, the sleep on the day of death). The median gap between the death and the interview was 3.8 months, with a range from under one month to 20 months.
• Exposures: whether a fan was on in the room, and whether a window was open.

The headline result was an adjusted odds ratio of 0.28 for fan use (95% confidence interval 0.10 to 0.77), which the authors described as “a 72% reduction in SIDS risk.” An open window pointed in the same direction (0.64) but the confidence interval ran from 0.33 to 1.21, which means the study couldn’t rule out no effect at all.

One detail matters a lot here. Very few babies in either group had a fan running. According to the paper’s Table 2, a fan was on during the last sleep for 6 of the SIDS cases (3.6%) and 36 of the controls (11.7%). The whole finding rests on those small numbers.

The subgroup results

The authors also looked at whether the apparent fan effect differed by sleep environment. The pattern they reported was that the association looked stronger in riskier settings:

Read the “back” row carefully, because it’s the one that applies to a baby who is already sleeping the way safe-sleep guidance recommends. For infants placed on their backs, the study found no clear difference with a fan (0.84, with an interval that spans well above and below 1). The authors themselves wrote that “some differences in the effect of fan use on SIDS risk did not reach statistical significance,” and their conclusion was carefully worded: fan use “may be an effective intervention for further decreasing SIDS risk in infants in adverse sleep environments.” In other words, the study’s own framing was about babies in already risky setups, not about adding a fan to an otherwise safe nursery.

Their proposed explanation was rebreathing. They pointed to a mechanical model in which air turbulence prevented carbon dioxide from building up, and suggested a fan might disperse the exhaled air that collects around a baby’s nose and mouth. That’s a plausible idea, but it was a hypothesis, not something the study measured.

What a case-control study can and can’t tell you

A case-control study starts at the end. Researchers take a group of people who had the outcome (here, babies who died of SIDS) and a comparison group who didn’t, then look backward to see which exposures were more common in each group. That design is practical for rare outcomes like SIDS, where waiting for enough cases to happen in a forward-looking study would take enormous numbers of families. But it comes with built-in weak spots.

Correlation is not causation

The study found that fans were less common in the rooms of babies who died. That’s a correlation. It doesn’t show that the fan itself protected anyone. Families who run a fan may differ in other ways that matter: the kind of home they live in, how warm the room was, how the baby was dressed, how many blankets were in the crib, and so on. The authors adjusted for a number of factors, but no adjustment can capture everything, and an unmeasured difference between fan households and non-fan households could produce the same pattern.

Memory and grief

Every piece of exposure data came from interviews with mothers, often months later. A mother who has lost a baby and a mother whose baby is alive and well may remember and report the details of one ordinary night very differently. This is called recall bias, and the authors named it directly, along with selection bias, as a possible limitation. They also reported that participation was “relatively low, with 50% of eligible cases and 41% of eligible controls completing the study interview.” When half the eligible families don’t take part, the ones who do may not represent everyone.

Small numbers and wide intervals

With only six case babies exposed to a fan, a small change in who was counted could shift the result considerably. You can see this in the confidence intervals. The overall estimate is consistent with anything from a 23% to a 90% reduction, and some subgroup intervals stretch over a range of more than a hundredfold. That’s what uncertainty looks like on paper.

What the American Academy of Pediatrics says

The AAP’s current safe-sleep guidance is its 2022 policy statement, “Sleep-Related Infant Deaths: Updated 2022 Recommendations for Reducing Infant Deaths in the Sleep Environment” (Moon, Carlin and Hand, Pediatrics 150(1)), along with a companion technical report that lays out the evidence. On fans, the policy statement is direct:

“There is currently insufficient evidence to recommend the use of a fan as a SIDS risk-reduction strategy.”

The technical report explains why. It cites the Coleman-Phox study and its odds ratio, then says: “because of the possibility of recall bias, the small sample size of controls using fans (n = 36), a lack of detail about the location and types of fans used, and the weak link to a mechanism, this study should be interpreted with caution. Based on available data, the AAP cannot make a recommendation on the use of a fan as a SIDS risk-reduction strategy.”

So the honest summary is not “fans are dangerous” and not “fans prevent SIDS.” It’s that the evidence isn’t there either way, and the AAP won’t tell parents to buy one for this purpose.

Has the finding been repeated?

Not that we could find. The AAP’s 2022 technical report, which reviewed the literature for its recommendations, still rests its fan discussion on this one study. The report does mention two older pieces of evidence about room ventilation more broadly: a study finding that bedroom heating, compared with no bedroom heating, was associated with higher SIDS risk (odds ratio 4.5), and another finding a lower risk in a well-ventilated bedroom with windows and doors open (odds ratio 0.4). Those point in a consistent direction, but they’re also observational and don’t test fans specifically. If a newer, larger study on fans does appear, we’ll update this page.

Overheating: the part the evidence does support

Behind the fan debate is a real and better-established concern: babies who get too hot. The AAP lists “avoid overheating and head covering in infants” among its top-level (Level A) recommendations. The technical report notes that excessive clothing or blankets, and the room temperature, “are associated with an increased SIDS risk,” and that babies who sleep on their stomachs have a higher risk of overheating than those on their backs.

What the AAP does not give is a target room temperature, and it explains why: “the definition of overheating in these studies varies. Therefore, it is difficult to provide specific room temperature guidelines to avoid overheating.” If you see a precise “ideal nursery temperature” quoted as official AAP guidance, it isn’t from the 2022 policy. Instead, the AAP asks parents to dress the baby for the room:

• Dress the baby with “no greater than one layer more than an adult would wear to be comfortable in that environment.”
• Check for signs of overheating “such as sweating, flushed skin, or the infant’s chest feeling hot to the touch.”
• Avoid overbundling and covering the face and head.
• Don’t put a hat on the baby indoors, “except in the first hours of life or in the NICU.”

The research on hot weather is less settled than you might expect. The technical report describes a Montreal study that linked very hot days (above 29°C, or 84.2°F) to almost three times the odds of SIDS, a Vienna study that used a similar approach and could not replicate that, and a study of 210 US cities that found a modest increase in SIDS risk with higher summer temperatures. The AAP notes these environmental studies have “significant limitations,” including no data on clothing or air conditioning at the time of death. It also says it is unclear whether overheating is an independent factor or partly a reflection of the extra risk from blankets and soft objects in the sleep space.

What this means at home

If you want to lower your baby’s risk of sleep-related death, the strongest evidence points to the basics in the AAP policy and the NICHD Safe to Sleep campaign. None of them involve a fan:

• Back to sleep for every sleep. Safe to Sleep says to place all babies on their backs “until they are 1 year old,” including babies born preterm and babies with reflux.
• A firm, flat, level surface. A crib, bassinet or play yard mattress “covered only with a fitted sheet.”
• A bare sleep space. No pillows, quilts, comforters, soft toys or loose blankets, and no crib bumpers, which Safe to Sleep says research links to serious injuries and deaths. It puts it simply: “Remove everything from baby’s sleep area, except a fitted sheet covering the mattress.”
• Room sharing without bed sharing. The AAP recommends the baby sleep in the parents’ room, on a separate surface designed for infants, “ideally for at least the first 6 mo.”
• Never on a sofa or armchair. The AAP calls couches and armchairs “extremely dangerous places for infants.”
• Dress for the room, and watch for overheating. One layer more than you’d wear, no hat indoors, and check for sweating or a hot chest.
• Offer a pacifier at naps and bedtime. This is also a Level A AAP recommendation.

If your house is warm and you like running a fan for your own comfort, the AAP guidance doesn’t forbid it. It simply doesn’t count it as SIDS protection. The checks that matter are the ones above. A fan is not a substitute for any of them, and it doesn’t make a stomach-sleeping position or a crib full of blankets safe.

For a single printable list, see our white noise machine safety checklist.

What about the “Fan” sound on hushnap?

hushnap has a sound called Fan. It’s a sound and nothing more. Like every sound in the player, it’s synthesized live in your browser with the Web Audio API to resemble the steady whoosh of a fan. It doesn’t move any air, it doesn’t cool the room, and it has nothing to do with ventilation or carbon dioxide. Whatever a real fan might or might not do in a nursery, a fan sound coming out of a phone or speaker can’t do it.

Some parents like fan noise simply because it’s a steady, broadband sound that masks household noise. If you use it for that, the same rules apply as for any sound machine: keep the volume moderate and the speaker away from the crib. Our guides on how loud a baby sound machine should be and how far it should be from the crib go through the details.

What we don’t know yet

• Whether fans have any effect on SIDS risk at all, once the other things that differ between fan and non-fan homes are accounted for. One study with six exposed cases can’t settle that.
• Whether the rebreathing explanation holds up in real nurseries, not only in a mechanical model.
• What room temperature range is best for sleeping babies. The AAP says the research doesn’t define overheating consistently enough to set one.
• How much warmer weather itself adds to risk, separate from clothing, bedding and sleep position. The studies so far disagree.

What isn’t uncertain is the core of safe sleep. Back to sleep, a bare and flat surface, the baby close to you but not in your bed, and not too warm. Those are where your attention will do the most good.