Infant Wearables: Good or Bad Idea?

In the age of the Fitbit, it is only a matter of time before devices designed for babies appear on the market. There are a whole bunch available now. There is a pacifier that doubles as a thermometer. You can buy a device to prevent heat stroke in your baby. There is even a GPS tracking devices for babies. The particular segment of the infant wearables market that concerns us here are the devices that measure infant blood oxygen levels. The unstated reason for buying such devices is the prevention of death by SIDS, or other sleep-related infant deaths.

Infant Wearables in the Age of Back-to-Sleep

These devices are based on a well-tested technology from the world of medicine called pulse oximetry. The “pulse ox” works by shining a red light onto the skin and simultaneously measuring the wave-length of the light that bounces back. The device then calculates the amount of oxygen in the bloodstream. Such devices have been used in even tiny babies for many years. Now, several companies are marketing these devices for parents worried about SIDS.

Do they work? Do infant wearables really prevent SIDS?

These are obviously two separate questions. We’ll tackle them one at a time.

To address the question of whether home pulse oximeters work, a few studies have been published. The study that best answers the question was published the the UK in 1991.

The British investigators focused on a group of babies that recently had been discharged from hospitals with apparent life-threatening events. Translated into English, the babies had turned blue at home, either because of choking or lung disease related to prematurity. This was a good population to study. The chance that these babies would actually need the alarm was greater than the general population of babies. Most of these babies had some kind of medical problem that made them more vulnerable to dropping their oxygen levels. This means the investigators were not measuring rare events and it would be unlikely that the number of events measured would be too small to give meaningful results.

The investigators also gave monitors to parents who had already lost a baby to SIDS.

The study looked at the number of events that the wearable devices picked up. Importantly, they also looked at false alarms, and measured side-effects of the devices, such as blisters and other skin burns.

The Results

A total of 201 babies were monitored for 30 months. Only 19 families withdrew from the study before it ended.

The pulse oximeters recorded 81 events, of which 52 were true episodes of blueness. No episodes were missed. In other words, the sensitivity of the device was 100%. Any time a test shows a 100% sensitivity (no true positives missed), there is going to be a substantial false positive rate. For these parents, that means an alarm going off for no reason. Sometimes the sensor would move out of position or fall off, activating the alarm. Sometimes it was a true false alarm. The device would go off when nothing was wrong.  In the UK study, false alarms happened about once every 4 days.infant wearables 2

Very few of the babies got skin redness from the red light source and none were burned.

Four babies actually died during the study. One sibling of a SIDS victim died while the infant wearables were in calibration mode. That is to say they were not turned on correctly. The cause of death was ruled SIDS as well. The other three babies died of the severe congenital heart and lung problems that got them included in the study.

There are two important take-aways from this study. Pulse oximeters work well. They will catch every dip in a baby’s blood oxygen level. But the devices cannot prevent death in severely ill babies. As was demonstrated catastrophically in this study, the devices won’t work if they are not turned on.

Infant Wearables and SIDS

The second question we asked was whether these devices can prevent your baby from dying of SIDS. The honest answer is “nobody knows”. In the UK study, the only baby to die of SIDS was not wearing an alarm. One other sibling of a SIDS victim in the study had a substantial dip in her oxygen and was later found to have meningitis. The answer remains open. We simply do not know if infant wearables prevent SIDS.infant wearables 3

But wait! Didn’t this study show that the alarm always goes off when the baby’s oxygen dips too low? If so, won’t any baby in danger of SIDS set off the alarm?

It’s a reasonable question. The answer is “yes, you are probably right”. And yet, we simply do not know because the studies have not proven the point to a degree of scientific certainty.

The Question You Didn’t Ask

It’s clear that pulse oximeters for use at home are very sensitive. They catch every event. This also means they “catch” events that are not events. Imagine how unbelievably frightening it must be to be awakened in the middle of the night by such an alarm. You cannot be sure if you are answering a false alarm or if you will be dialing 911. The question we didn’t ask at the outset was “do you think infant wearables make you more relaxed, or more anxious?”

The American Academy of Pediatrics has not taken a position on these devices for healthy infants. The organization’s silence probably reflects the absence of scientific evidence on SIDS prevention. Pediatricians disagree as to whether or not it’s a good idea for parents to buy such a device. My recommendation would be to discuss it with your pediatrician before buying.

And if you have any issues or (non-medical) concerns about your baby’s sleep, I’m a sleep consultant.


Swaddling and SIDS

As if parents don’t have enough to worry about these days: Now they are worried about swaddling and SIDS. I am sure that parents all across the world saw the news come across their Facebook feed and were sent into a panic.

Does swaddling your baby increase the rate of Sudden Infant Death Syndrome? You would think so, to read media accounts of a paper published recently in the journal Pediatrics. Here’s a link, and the reference for those who want to read it.

Pease AS, Fleming PJ, Hauck FR, et al. Swaddling and the Risk of Sudden Infant Death Syndrome: A Meta-analysis. Pediatrics. 2016;137(6):e20153275

Here are some headlines from some of our most respected media outlets. Almost all of them misunderstood the paper’s findings:

New York Times: “Swaddling May Increase the Risk of SIDS”

Washington Post: “Swaddling babies is tied to heightened risk of sudden infant death syndrome”

Yahoo! News: “New Research Suggests Swaddling Could Increase Risk Of SIDS”

Thank God for the Atlantic Monthly and to science writer Adrienne LaFrance.  LaFrance is the only medical and science writer I’ve read so far who detected the flaws in the paper. Here’s the headline:

“About That Scary Swaddling Study: A new meta-analysis seems to link infant swaddling with a higher risk of SIDS. But there’s more to the data than that”

Indeed there is more to the story. Let’s get to the bottom of that study on swaddling and SIDS.

Why Even Study Swaddling and SIDS?

In the early 2000s, there was just as much uncertainty as there is today about the causes of SIDS.  The famous Back-to-Sleep campaign in the US was already succeeding in dropping SIDS rates. Our success was similar to results seen in other countries.  Researchers were stumped. Was it really as simple as all that? How could a silly little change like having the baby sleep on her back reduce the risk of SIDS? Many studies were performed looking at body functions like heart rate and breathing in babies. It was known that swaddled babies were generally calmer and less sensitive to waking up suddenly. The question was asked whether being quiet and less sensitive put a baby at risk for SIDS.

A Look at the Studies

Many studies were done, but most of them were lousy. Pease, et al., who wrote the paper that got all the headlines, decided to perform a “meta-analysis” of all the studies they could find on the subject of swaddling and SIDS. A meta-analysis is considered to be the highest form of scientific study. It looks at all scientific studies of a particular question and sort of pools the results. The idea is that many different groups looking at the same problem collectively get to the truth better than any one study does. It’s never a good idea to rely on only one study to answer a question. The more investigators who come up with the same result, the closer to the truth you are probably getting.swaddling and sids 3

They looked at a lot of studies. Since the 1950’s there have been almost 400 studies asking the question: Is there a link between swaddling and SIDS? Of these studies, Dr. Pease and her colleagues could find only 4 that met a standard rigorous to be considered worthy. And one of these had never been published the data on swaddling. When they were done with their analysis the investigators discovered that they could not easily compare the results of all four studies. So to present their final results they needed to eliminate one study, leaving them with three. Out of 400.

What Did They Find?

After pooling all these data, the investigators found that swaddling increased the risk of SIDS very slightly if the baby was put down on her stomach or side. In other words, it was riskier to violate the “back-to-sleep” rule. It was also riskier to swaddle a baby older than 6 months of age. The most confusing part of the study showed that there was a slightly increased risk to swaddling if you lay the baby on her back. It’s confusing because even so, many more SIDS deaths occurred in un-swaddled babies than in swaddled babies. The so-called “increased risk” was only compared to the comparison (“control”) group.


More Problems

There were other problems with the meta-analysis. The investigators could not be sure that all the studies used the same definition of “swaddling”. Swaddling means different things in different places. Some of the studies lumped together swaddled and “wrapped” babies. It isn’t entirely clear what was being compared. Perhaps more damaging, the pooled studies were so different from one another that it was impossible to eliminate all the features that could confound the results. In other words, they couldn’t really be sure that swaddling was the thing that increased the risk of SIDS!swaddling and sids 2

The Bottom Line on Swaddling and SIDS

  • It’s okay to swaddle, but if you do, lay the baby on her back. Swaddling is only risky if the baby is face down or on her side.
  • Older babies probably shouldn’t be swaddled, but that’s okay, since it’s really difficult to swaddle a six-month old baby anyway. They fight out of the swaddle too easily.
  • Learn to swaddle a baby correctly. There are lots of terrific videos on YouTube that show you how. Here’s one:

Back to Sleep Saves Lives

What position is the best for putting my baby to sleep? We know now that putting the baby on her back, on a firm surface, is the safest? But how do we know this? The famous “Back to Sleep” campaign did not appear out of thin air. It was the result of many years of careful study in the field of epidemiology. One tiny little change in our behavior, putting the baby down on her back instead of her belly, has saved millions of lives in the 20th century.


“Back to Sleep” campaigns came about as the result of the search for the causes of Sudden Infant Death Syndrome (SIDS). SIDS has also been called “cot death” or “crib death”.

In the US, unexplained sudden infant death is the #3 cause of infant mortality, after congenital malformations and extreme prematurity. But between the ages of 1 month and 1 year, SIDS is the #1 cause of death. The most common age for SIDS deaths is between 2-4 months of age.



By definition, SIDS has no cause. For example, if an infant suffocates on a blanket in her crib, the cause is suffocation, not SIDS.  We cannot make a diagnosis of SIDS unless a thorough forensic investigation and an autopsy has been completed. Only then, when no other medical or criminal cause has been found, can SIDS be given as the reason for death.

Because the causes of SIDS remain mysterious, researchers had to find other ways to figure out how to prevent it. The search began with a number of observations starting in the middle of the 20th century.

The Origins of Back to Sleep

In 1944, in an article entitled “Accidental mechanical suffocation in infants”, Dr. Harold Abramson suggested that someone ought to do a study to ask whether sleeping position had anything to do with infant deaths. It was not until the mid 1960’s and early 1970’s that these studies were actually done. Both showed that sleeping on the belly was associated with infants dying of SIDS. But it was not until 1986 that any researcher even suggested that infants sleep on their backs to reduce the incidence of SIDS. This study was done in Australia. The suggestion was taken up and SIDS rates began to go down. The United Kingdom and Denmark soon followed with their own campaigns, with positive results.

back to sleep
She’s on her back

Several other interesting observations were made. The rate of SIDS is extremely low in Hong Kong, compared to the US. A well-done study of the the territory in 1985 showed that the most Hong Kong babies go to sleep on their backs.

Another carefully-performed study in Sweden showed that rates of SIDS went up and down as recommendations about infant sleeping position changed over time.

The US climbs on board

Finally, public health officials in the US began to pay attention to the mounting evidence. As you can see from the graph, the American Academy of Pediatrics started its back to sleep campaign in 1992. From that point on, an interesting thing began to happen: Rates of back sleeping began to increase (green dotted line), and rates of SIDS began to fall (red bars). This was a similar result to what had happened in other countries.

Causation is not correlation, but…

You always have to be careful when you look at studies like the ones mentioned here. It is always possible that the cause (putting baby on her back) and the effect (lowered SIDS rate) are not related to each other. This could be a coincidence. This is especially so because we still don’t know what causes SIDS. There’s a lot of speculation, but we just don’t know. What we can say is that all over the world, the same result has happened. Back to sleep campaigns get started and SIDS rates go down. If this is the case, we should be less concerned about what the true causes of SIDS are, because we appear to have found a way to prevent many cases of SIDS.back to sleep

Keeping in mind that we don’t know what actually causes SIDS, there are a number of other risk factors that have emerged from these studies:

  • Tobacco smoke: SIDS rates are higher for mothers who smoke during pregnancy
  • Sleeping with a pacifier: This appears to lower the risk of SIDS
  • Breastfeeding: Infants who breastfeed are at slightly lower risk of SIDS
  • Genetics: Boys are more likely than girls to die of SIDS
  • Vaccinations: These appear to be protective. Again, no one knows why

What to do about “flat head”?

There’s no question that it’s a very good thing that SIDS is so rare now, thanks to what we’ve learned. But there is one unintended consequence of all the back to sleep campaigns: flat head. Called “positional plagiocephaly“, it’s the result of, well, laying on the back while your baby head is growing rapidly. It’s occurring now in about 10-50% of all babies that sleep on their backs. Most babies who have a mild case of flat head will grow up to have a normally-shaped skull. The best way to reduce the amount of head flattening your baby has is to do observed “tummy time” about 20 minutes per day. It’s also recommended that you position the baby’s head alternatively to the right and to the left when she sleeps. This will help keep her most symmetrical. Do not use positioners. These are not recommended.

If you’re concerned about any of this, SIDS and/or flat head, please consult your pediatrician!

Baby Box: Good Idea, But Not That Good

The Finnish “Baby Box” is enjoying something of a moment. Since the appearance of an article on the subject in BBC Magazine, the concept has “gone viral”. Entrepreneurs and public health experts on every continent have jumped on that crazy Finnish baby box bandwagon. It’s a neat idea, and it’s nice for parents. But claims are being made for the “box” that are not supported by the evidence.

What’s a “Baby Box”?

In 1938, the government of Finland began mailing packages to low-income pregnant women. The packages contained diapers, bedding, gauze towels and other child-care products. But the most notable feature of the package was the box itself: there was a mattress at the bottom, that the baby was meant to sleep in. The precise reason for the baby box is unclear, but one suggestion is that many poor Finnish women could not offer a clean place for their babies to sleep.

Beginning in 1949, the program was expanded to include all mothers, which is to say, the “means testing” mechanism was removed. But there was a catch: in order to get the box, mothers had to begin attending a prenatal clinic before the fourth month of pregnancy. This was done to ensure that mother enjoyed a healthy pregnancy, giving her baby the best chance at being born healthy.


The contents of the box has changed with time, depending on contemporary trends or the privations of war time. Seventy-eight years later, Finnish moms continue to look forward to receiving their baby box no matter how many children they have.

The Box Goes Viral

After the appearance of the original article in BBC magazine, the Finnish government sent a baby box to the Duke and Duchess of Cambridge when they were expecting their first baby. Expectant parents all over the world wrote to the Kela, the Finnish social security agency, to request their own baby boxes. Entrepreneurs jumped on board and began selling baby boxes world-wide, from Finland and the United box

The concept also captured the attention of public health experts throughout the world. The reason for the attention is very likely the inclusion of a single graph in the original BBC article, showing the decline of infant mortality in Finland throughout the 20th century.

The implication was that the box itself contributed to the decline in the Finnish infant mortality rate.

How likely is this to be true?

In the early part of that century, throughout the world, infant mortality was high by modern standards. But infant mortality rates were doing something very significant since the late 19th century: they have been falling rapidly.

The Defeat of Infant Mortality

We would like to say we know the reasons why infant mortality has been plummeting throughout the developed world. The truth is that we don’t really know why. There are many factors involved. It’s almost certainly the case that several factors were important. According to experts, in the early 20th century the most important factor was the decline in deaths from infectious disease. This was due mostly to improvements in nutrition and public health. By mid-century, the development of antibiotics contributed to further improvement in life expectancy. Today, the largest contributors to infant mortality are congenital abnormalities and the consequences of extreme premature birth.

Further down the list of causes are SIDS and injuries. The latter group includes death by suffocation, usually in the crib. These are the causes of death most likely directly related to the place where the baby sleeps, as in, say, a baby box. The claim, hinted at in several articles, is that the box has been responsible directly for a reduction in SIDS and suffocation deaths.

baby box
Big Drop, No Box

Not so fast.

It’s important to note that infant mortality has been plummeting in every other developed country that did not distribute baby boxes. There’s no evidence to suggest that Finland’s very low rate (around 3 per 1000 live births) has anything to do with the box. It’s more likely that other improvements in public health and medical innovations had more to do with it.

But wait. Could the box have nothing to do at all with the low Finnish infant mortality rate?

It’s probably the case that the box was part of a multi-level public health effort, the cumulative effects of which was better overall health of new mothers and their babies. Other features of the effort included better prenatal care and education for mothers and fathers.

Similar efforts in India include mosquito netting over the box, to help stem the spread of malaria.

So what about cities in the US with relatively high infant mortality rates?

An initiative in Fort Worth, TX, hopes that a baby box distribution program will help lower the rate of 7.9/1000 closer to the national average of 6.1/1000. It’s unclear how much of the excess infant mortality is due to sleep-related deaths. If the program is to succeed, and I wish them well, it will likely succeed because of the increased attention that pregnant woman pay to their own health. The effect on rates of premature birth may be substantial.

So the baby box itself may not be the answer to the infant mortality problem. But if the box helps persuade mothers to get prenatal care, the program may be worth the investment. Also there are all those cool boxes.