Professor Ed Mitchell
SUDI Academic Expert
Auckland University

Ed Mitchell qualified at St George’s Hospital Medical School in London and has worked in the UK, Zambia and New Zealand.

He was the Cure Kids Professor of Child Health Research at the University of Auckland from 2001 to 2015 and is now a Professorial Research Fellow. He has published over 400 original papers, particularly on the epidemiology of sudden infant death syndrome (SIDS). He was awarded a Doctor of Science for his work on “The Epidemiology and Prevention of SIDS” by the University of London. He has received several awards for his landmark studies of SIDS and in 2009 was made a fellow of the Royal Society of New Zealand.

Professor Ed Mitchell answers questions from the community.

You have advised that parents do not sleep with their baby. Why is that?

Many studies have shown that sleeping with baby on the same sleeping surface, usually a mattress, increases the risk of sudden unexpected death in infancy (SUDI). Indeed over 50% of SUDI cases now occur in a bed sharing situation. There are certain factors that increase the risk further. These include the first 3-4 months of life, whether or not the mother smoked in pregnancy, whether she took drugs or alcohol before sleeping with the baby or was excessively tired and whether the baby was vulnerable, such as low birthweight or preterm. Even without these additional risks bed sharing increases the risk 3-fold, so our advice is “For the first six months, the safest place for baby to sleep is in a cot in the parent’s bedroom.”

  • bed sharing
What is about bed sharing that causes SUDI?

We believe the mechanism is accidental suffocation. The nose of the baby is soft and can easily be compressed. Also the jaw can be pushed back and occlude the airway. A healthy baby will arouse (wake up) and wriggle, but if mother has taken drugs or alcohol she won’t be responsive to this. Also if the baby has been exposed to smoking when in the womb, the arousal mechanisms are blunted, and the baby may not wake up.

  • smoking
  • bed sharing
  • cause
Is it safe to leave babies to sleep in car seats/capsules for extended periods of time as sometimes parents don't want to wake baby after traveling?

Babies will often fall asleep in a car seat when travelling. Not surprisingly it is tempting to leave the baby asleep in the capsule when they arrive home. But is it safe? Studies of babies in car seats show that their head may flop forward and cause partial airway obstruction and hypoxia (low oxygen levels). This will briefly wake baby up (micro-arousals). Despite these concerns, SUDI in a car seat is rare. In the 3-year Nationwide SUDI Study there was only 1 death (out of 137) that occurred in a car seat. In comparison there were 7 control (living) babies (out of 249) in a car seat during the comparison sleep. This suggests it isn’t dangerous in healthy babies, but I would be concerned about repeated hypoxia in low tone babies, such as those with Down syndrome.

  • car seat
  • car
  • sleep
Should we be putting a hat on baby to sleep at night now its’ got cold?

Questions relating to type and amount of baby clothing are frequent. In the original New Zealand Cot Death Study (1987-1990) baby hats (e.g. beanies and bonnets) were used in 8.3% of cases and 5.2% of controls. In the Nationwide SUDI Study hats were used by 4.8% of cases and 3.9% of controls. These differences are small and not statistically significant.

In infants the surface area of the head and neck is 20% of their total body area (compared with just 9% in adults). Heat loss can be considerable, so in cold conditions a hat will reduce this. This may prevent the baby getting cold and uncomfortable. But it doesn’t affect the risk of SUDI.

Head covering increases the risk of SUDI, but this does not refer to hats; it refers to baby’s head being covered with blankets. A meta-analysis of 10 studies found the prevalence in cases was 24.6% and 3.2% among controls. This suggests head covering by blankets is a modifiable risk factor associated with SIDS deaths and has led to the UK recommendation to place baby at the foot of the bed (“Feet to foot”). Theoretically this might prevent babies kicking the blankets over their heads, but there isn’t much evidence to support that this happens, which is why we do not emphasise this in New Zealand.

  • hat
  • cold
  • baby clothing
What is the difference between SUDI and SIDs and why the terms have changed from "cot death" as it was previously known?

Sudden infant death syndrome (SIDS) has been present since antiquity. Indeed it is described in the Old Testament of the Bible: 1 Kings 3:19 ‘‘And this woman’s child died in the night.’’ “Cot death” was well known to pathologists, but not to paediatricians as they never saw them, as the deaths occurred in the home. These deaths were often labelled as pneumonitis (inflammation of the lung). It was not until 1965 that a specific International Classification of Diseases (ICD) code was allocated for SIDS. SIDS is the scientific term for a sudden unexplained infant death. These deaths generally occur during a sleep, and no cause is found despite a detailed examination of the death scene, medical history and thorough autopsy. Nowadays 50+% of deaths occur in a bed sharing context, so using the term “cot death” is inappropriate. The cause of deaths associated with bed sharing may be labelled as accidental asphyxia (i.e. suffocation), SIDS (that is unexplained) or unascertained (often used when the pathologist cannot distinguish between suffocation and SIDS). The trouble is that one pathologist or coroner might call the death SIDS and another accidental asphyxia. Sudden unexpected death in infancy (SUDI) captures all three causes, including unexplained and explained.

  • cot death
  • SUDI
  • SIDS
When is it OK for our baby to sleep with his older sibling?

Only a few studies have examined this. One of the best is the Chicago Infant Mortality Study. This case-control study found that bed sharing with siblings (with or without the parents) raised the risk fivefold.

Although observational studies do not tell us the mechanism, all parents would have observed how deeply preschool children sleep, and it would not be surprising if the sibling did not respond to the struggles (arousal) of the baby if baby was overlaid.  
However, by 6 months of age 85% of SUDI cases have occurred and after 12 months SUDI is very rare. So for the safety of your baby do not let baby sleep with an older sibling until at least 12 months of age.

What about co-bedding twins? Certainly there are SUDI deaths that have occurred in a twin when sleeping together (co-bedding), however, this practice is quite common in New Zealand, so it is difficult to estimate whether co-bedding is a risk or not, but given the increased risk of SUDI associated with sleeping with older siblings, I would recommend caution.

  • bed sharing
  • siblings
What temp is best for our baby’s room?

This simple question is actually quite complex! There is little information on the room temperature when the death occurred. Researchers have tried to estimate the room temperature from the environmental (outside) temperature, but this is difficult, especially if room heating and insulation need to be taken into account. So one has to rely on case reports. I’ve reviewed cases where the room temperature was very high, but more frequently the rooms are cold. Is this a risk from the cold or is this a marker of poor, disadvantaged families living in poorly heated and uninsulated homes? I don’t know.
Recommendations vary: 16-20°C and 20-22°C have both been recommended. Not everyone has a thermometer, so I’d recommend a temperature that is comfortable for you with a light jersey. Plunket has sensible advice on their website. They recommend:

  • the [baby’s] room is well aired with the door open, especially if you use a heater
  • the temperature should feel comfortable for a lightly-clothed adult - the temperature around the cot will be more even if it’s away from windows
  • the room is not too hot - using an electric heater with a thermostat is best (fan heaters may overheat the room and gas heaters can give off dangerous fumes)
  • the cot is away from windows, curtains, blind cords, power points and heaters.
  • temperature
  • heater
What is the difference between SIDS and SUDI? Much of the available information on Safe Sleep is geared towards preventing suffocation.

Terminology has changed over time. It was once called ‘cot death’ despite many of the deaths not occurring in the cot. It was then changed to SIDS which is unexplained infant death. SUDI (Sudden Unexpected Infant Death) is a broader term than SIDS. It includes unexplained deaths (that is SIDS) and sleep related deaths from asphyxia or suffocation, such as may occur while bed sharing. In part this change has been driven by changes in diagnostic fashion. One pathologist might call the death SIDS, another suffocation in bed while bed sharing and another unascertained. SUDI captures all the deaths that were once labelled SIDS or cot death. SIDS is by definition unexplained, and therefore cannot be predicted or prevented. However, we can predict what increases the risk and we know that the ‘Back to Sleep’ campaign and the Safe Sleep programme has and is reducing mortality. The good thing about describing the mechanism as due to suffocation is that it suffocation is obviously preventable.

  • asphyxia
  • suffocation
I am so worried about my baby dying of SUDI that I get up several times a night to check my baby. My husband tells me I shouldn’t worry.

I was sorry to hear that about your worry over SUDI. Infants of families that follow the recommended infant care practices (non-smoker, room sharing but not bed sharing, back sleeping and breastfeeding) are at very low risk of SUDI. In our recent study we calculated there would be only 6-7 deaths per year in New Zealand if these recommended advice was followed by all. This is a dramatic reduction from the 250 deaths a year that occurred in the late 1980s. There are of course some things which one cannot alter after the baby is born, such as low birthweight. We hope to implement a SUDI risk calculator which will give the absolute risk of death and identify which families need additional support.

  • risk
  • recommended
Has the ‘toxic gas theory’ been proven or disproved?

This interesting theory goes back to the 1980s. It postulated that a specific fungus grew on the plastic covering of damp cot mattress. The fungus would metabolise the fire retardants in the plastic covering and would release nerves gases which could kill the baby. Proponents of the theory recommended covering the mattress with a polythene covering to stop the gases reaching the baby. To me it never seemed a likely explanation, as 50+% of deaths occurred in the parental bed, which was unlikely to be wet and contaminated with the fungus unlike the cot. The theory stimulated considerable research which was not able to replicate the gases in anything closely resembling a cot environment. In my opinion the theory was clearly wrong when it was found that the postulated nerve gases wasn’t toxic!

  • damp mattress
  • plastic covering
What is the definition and use of the term co-sleeping? My understanding was that co-sleeping was an infant sharing the same bed as an adult or older child, but I’ve also seen it defined as sharing the same room and therefore a safer option than bed sharing. This has confused me.

Yes it has been made confusing by an anthropologist. He postulated that co-sleeping (that is sharing the same sleeping surface) would reduce the risk of SIDS. As you know that is not the case, indeed it increases the risk. We showed that being in the same room as the parent but not in the same bed is associated with a reduced risk. He then redefined co-sleeping as including sleeping in the same room, and stated he was right all along!!

I believe we should avoid the word co-sleeping, and use bed sharing and room sharing, because we’re trying to keep the terminology succinct and clear to professionals and families when we discuss SUDI. The meaning of bed sharing and room sharing is self-evident and thus preferred.

  • bed sharing
There are many modifiable risk factors that greatly add to the risk of bed sharing.  Do you think if those risk factors were taken out of the equation, and certain safety measures, such as having parent sleep on a harder mattress, with minimal blankets, and keeping the infant on one side and not in between parents, could ever make bed-sharing safe?  I also wonder what monitoring the infant's vitals in such situations would show.

One of the aims of the Nationwide SUDI case-control study was to see if there were ways of bed sharing that were safe. The answer is yes, providing infants are older than 3 months, mothers did not smoke or smoke in pregnancy, no alcohol, no drugs, infant not born preterm or small for gestational age, not sharing with another child. However, if any of those factors are present there is an increased risk, sometimes markedly (32 fold increased risk where mother is a smoker).

For infants less than 3 months who have none of these risk factors then the risk is very small – that is the parents are doing everything right. If they also bed share the risk increases 3 fold but the risk in absolute terms remains small (see Carpenter).

There is no physiological studies on infant bed sharing when other risk factors are present. All the studies that have been done are with extremely low risk families. It is unethical to do such studies when you know the baby is at increased risk of death.

  • bed sharing
Does swaddling pose a risk of SUDI?

The jury is still out on whether or not it is related to SUDI. There is a good paper published a few years ago: Anna S. Pease, Peter J. Fleming, Fern R. Hauck, Rachel Y. Moon, Rosemary S.C. Horne, Monique P. L’Hoir, Anne-Louise Ponsonby and Peter S. Blair. Swaddling and the Risk of Sudden Infant Death Syndrome: A Meta-analysis. Pediatrics June 2016, 137 (6) e20153275; DOI: https://doi.org/10.1542/peds.2015-3275 

Basically it found that there was a non-statistically significant increased risk of SIDS from swaddling. The risk was high for babies placed prone and moderate risk for on the side compared with babies on the back. This makes sense from an airway obstruction perspective.

Infants in sleeping sacks or bags are at lower risk of SUDI. These babies are not swaddled.

  • swaddling
What do people mean by swaddling?

It varies from being very restricting of arms and legs to being lightly wrapped in a muslin cloth.

  • swaddling
What is developmentally supportive swaddling?

Developmentally Supportive swaddling (DOI: https://doi.org/10.1542/peds.2006-2083 and below) has its advocates, and I would support its use in NICU. It may have some value in managing the unsettled infant at home, but I definitely would not recommend it for normal infant care (babies born at full term).

  • swaddling
What are the advantages of swaddling?

This was discussed by Bregje E. van Sleuwen, Adèle C. Engelberts, Magda M. Boere-Boonekamp, Wietse Kuis, Tom W.J. Schulpen and Monique P. L'Hoir. Swaddling: A Systematic Review. Pediatrics October 2007, 120 (4) e1097-e1106; DOI: https://doi.org/10.1542/peds.2006-2083 “In general, swaddled infants arouse less and sleep longer [EM: Not necessarily a good thing]. Preterm infants have shown improved neuromuscular development, less physiologic distress, better motor organization, and more self-regulatory ability when they are swaddled. Excessively crying infants cried less when swaddled, and swaddling can soothe pain in infants. It is supportive in cases of neonatal abstinence syndrome and infants with neonatal cerebral lesions [EM: This justifies its use in NICU].”

  • swaddling
Are there adverse effects?

There is an increased risk of the development of hip dysplasia, which is related to swaddling with the legs in extension and adduction (straight and together). There is some evidence that there is a higher risk of respiratory infections related to the tightness of swaddling. Swaddling immediately after birth can cause delayed postnatal weight gain under certain conditions, but does not seem to influence breastfeeding parameters.

  • swaddling