Originally published in MedPage Today
by Todd Neale, MedPage Today Staff Writer
Just over half — 54% — of sudden infant death syndrome (SIDS) cases in southwest England occurred when the baby was co-sleeping in the same bed as a parent, a case-control study showed.
That compares with 20% of infants among randomly selected families and controls deemed to be at high risk for SIDS, Peter Fleming, MBChB, PhD, of the University of Bristol in England, and colleagues reported online in BMJ.
In a multivariable analysis, sleeping on the same bed or couch as a parent was associated with a 21.77-fold (95% CI 3.79 to 125.00) increased risk of SIDS.
Much of the elevated risk appeared to be attributable to drug and alcohol use among the parents. In multivariable analyses, there were significant interactions between bed-sharing and substance use compared with both random controls (P=0.002) and high-risk controls (P=0.02).
But rather than advising against co-sleeping outright, they said “a better approach may be to warn parents of the specific circumstances that put infants at risk. Parents need to be advised never to put themselves in a situation where they might fall asleep with a young infant on a sofa.”
Parents should also be advised to never co-sleep with an infant in any environment after consuming alcohol or drugs, they said.
The American Academy of Pediatrics recommends against co-sleeping. The 2005, recently affirmed policy statement advises parents to place their infants in “a separate but proximate sleeping environment” in the parents’ room.
The rate of SIDS dropped in the early 1990s after the AAP and other child health organizations began the “Back to Sleep” campaign advocating that parents put their babies on their backs.
Early recommendations also included reducing the amount of bedding and telling mothers to stop smoking during pregnancy.
More recent guidance has included recommendations against co-sleeping, placing infants on their sides to sleep, the use of duvets and pillows, and postnatal exposure to smoke.
It’s unclear, however, how well all of this advice has been heeded and whether there are new risk factors for SIDS emerging, Fleming and his colleagues said.
So they examined all unexpected deaths of infants up to 2 years old during a four-year period in six counties in southwest England.
They gathered two control groups, one comprising 82 randomly selected families and one comprising 87 families at high risk for SIDS because of young maternal age, low socioeconomic status, and current smoking.
There were 90 SIDS deaths identified during the study period, for a rate of 0.49 per 1,000 live births.
In addition to co-sleeping, there were several factors associated with an increased risk for SIDS:
* Child sleeping alone in a room: OR 21.34, 95% CI 2.99 to 152.56
* Using a pillow: OR 10.59, 95% CI 1.43 to 78.39
* Swaddling: OR 31.06, 95% CI 4.21 to 228.94
* Maternal smoking during pregnancy: OR 13.36, 95% CI 3.07 to 58.83
* Preterm birth: OR 11.52, 95% CI 1.64 to 80.82
* Parental report of child being in fair or poor health for the last sleep: OR 8.06, 95% CI 1.11 to 58.42
* Sleeping on stomach: OR 6.61, 95% CI 1.57 to 27.88
These risk factors were statistically significant regardless of the control group used as a comparison.
“This is important because it indicates that risk factors for SIDS apply to all sections of the community and are not just a consequence of social deprivation, because SIDS now occurs predominantly in disadvantaged communities,” wrote Edwin Mitchell, DSc, of the University of Auckland in New Zealand, in an accompanying editorial.
Other risk factors have become less significant over time, according to Fleming’s group.
The presence of bedding over the head or face of the baby, a side sleep position, excessive bedding and clothing, and postnatal exposure to tobacco smoke were not associated with SIDS in this study. They had been related in a previous study conducted 10 years earlier.
In his editorial, Mitchell said that very little could be gained from further epidemiological studies on the risk factors for SIDS, which are largely preventable.
The current challenge, he said, is finding out how to reduce these risk factors, although he did not propose any specific solutions.
Fleming and his colleagues acknowledged that the relatively small number of deaths from SIDS during the study may limit the interpretation and significance of some factors.
The study was also limited, they said, by the difficulty in recruiting control families and by the failure to interview control families on weekends, when drug and alcohol use might be more common.