The Safe to Sleep Program, SIDS and SUID

Carol Gray Babies, Birth, Infants, Postpartum, Pregnancy, Wellness , , , , , , , , , , , , , , , , , , ,

Why are We Putting Babies to Sleep on Their Backs?

The Hong Kong Studies

In 1985 after retrospectively investigating only 15 deaths, researchers concluded that that Hong Kong had a much lower rate of unexplained infant deaths than that of Western countries. They cited a low incidence of preterm birth in Hong Kong (a known risk factor for unexplained infant death) and two significant differences in infant sleep environments — crowded living conditions and the traditional Chinese practice of putting babies to sleep on their backs. Given the absurdity of recommending crowded living conditions and the unlikelihood of reducing preterm birth, they went with the supine sleep recommendation in their conclusions.

A prospective study in Hong Kong was done in 1987. Researchers investigated the deaths of 16 infants who were determined to have died from SIDS. They compared them to 32 babies who were determined to have died from other causes. They hoped to confirm the results of the retrospective study done in 1985. They found a much higher incidence (3 SIDS deaths per 10,000 live births versus 0.4 SIDS deaths per 10,000 live births in the earlier study). Nonetheless, this number was still much lower than estimated rates of 20-40/10,000 live births in Western countries.

*Note: In the Hong Kong studies, unlike any others I found, the incidence of SIDS was higher among female infants — in a culture that prefers male children.

The New Zealand Study

Next came the New Zealand Cot Death Study conducted from 1987 to 1991. New Zealand had a very high recorded SIDS rate of 25 per 10,000 live births. Researchers discovered a variety of risk factors — three of which were potentially modifiable: maternal smoking, prone sleep position and lack of breastfeeding. Recommending supine sleep was the most likely recommendation to achieve compliance. Thus the “Back to Sleep” campaign was born. It made its way to the US in 1994.

In 1992 — prior to the American Academy of Pediatrics supine sleep recommendation — 13% of US infants slept on their backs. In 2017 77% of US parents reported that they usually, but not always, put their infants to sleep on their backs.

Sudden Unexpected Infant Death (SUID) in the US

We don’t know how many babies in the US die of unexplained causes. We collect statistics, but we don’t know how many of those deaths were properly investigated and reported.

Death Certificates

The information about the causes of infant death is captured from death certificates. The quality of this information varies from place to place and from year to year.

Monitoring

The Centers for Disease Control and Prevention (CDC) supports an SUID monitoring program at 32 sites around the US. It covers 40% of all SUID cases in the United States. That’s it. Just 40%. Only these states and jurisdictions receive the funding and support to participate in the program. The SUID Case Registry uses data about SUID trends and characteristics to inform strategies to reduce future deaths.

Incidence Reduction

In 1995, the year after the Back to Sleep Program was introduced, the incidence of Sudden Unexpected Infant Death (SUID) was 11 deaths for every 10,000 live births. Between 1996 and 2020 the incidence of SUID has held steady at about 9 or 10 deaths for every 10,000 live births.

Categories of Sudden Unexpected Infant Death (SUID) in the US

The SUID umbrella currently has three sub-categories, Sudden Infant Death (SIDS), Unknown Cause and Accidental Suffocation and Strangulation in Bed (ASSB). The criteria for inclusion in these sub-categories have changed over time. The International Classification of Disease (ICD) codes used to classify these deaths have also expanded and changed over time. Not all jurisdictions have adopted the changes. This makes it difficult or impossible to compare cause of death rates year over year — especially on the national level.

Sudden Infant Death Syndrome (SIDS) in the US

The first sub-category, Sudden Infant Death Syndrome (SIDS), is not a syndrome. It is a diagnosis of exclusion. SIDS means an infant under the age of one year dies, a thorough death scene investigation and an autopsy are conducted and no cause of death is discovered.

Autopsies

A SIDS determination must include an autopsy as part of the investigation. However, an autopsy is not always done when an infant dies of unexplained causes. Furthermore, the autopsy elements vary from place to place. One commonly missing element is genetic testing. When other causes are ruled out some parents want to do genetic testing, but insurance doesn’t pay for it if the patient is dead. It costs hundreds to thousands of dollars and the person who conducted the autopsy may not have saved a vial of blood. This is yet another example of how poverty can be a risk factor for an unexplained infant death.

Scene and Circumstances Investigation

A SIDS determination also requires an investigation of the scene and circumstances of the death. There is no reliable way to tell the difference between SIDS and suffocation by autopsy. Without scene investigation, history, and other non-autopsy information, autopsy alone often cannot differentiate a natural infant death from an unintentional, accidental, intentional or neglectful infant death. An investigation must include a medical history exploration of the infant and family. It must also include a doll reenactment, but authorities are often reluctant to ask grieving parents to participate.

Doll Reenactments

A doll reenactment is essential because unexplained infant deaths typically have no witnesses. They usually happen while people are sleeping. When parents/caregivers discover a non-responsive infant they pick the baby up. A traumatized parent may not recall the specifics of the death scene. If there is no death scene investigation or it doesn’t include a doll reenactment there is no reliable sleep position information. This calls into question the argument in favor of back sleeping being the safest.

Incomplete Investigations

The accuracy and completeness of death scene information is variable — even in jurisdictions within individual states in the US. We don’t have up-to-date statistics for the number of death scene investigations and how many of those have doll reenactments. The apparent drop in the SIDS rate must take into consideration that in 2024 death scene and autopsy investigations are often incomplete. They were even worse in the early 1990s. Thus, I question the validity of the apparent drop in SIDS following the introduction of the Back to Sleep program.

The SIDS Numbers

In the US SIDS numbers were dropping prior to the introduction of the Back to Sleep program. From 1980 – 1988 SIDS rates in the US declined 3.5% for white infants and 19.2% for black infants. This is presumably due to an increase in autopsy rates and early efforts at standardizing death certificate information collection.

The number of SIDS cases has decreased in the US. In 1995 about 9 SIDS deaths were recorded for every 10,000 live births. In 2020 there were about 4 per 10,000 live births. Unfortunately, these numbers have stagnated in the 4-5/10,000 range for more than 20 years. There is no ongoing progress. In 2020 SIDS cases made up about 41% of all Sudden Unexpected Infant Deaths.

Unknown Cause of Infant Death in the US

This sub-category is where we don’t have enough information to state any cause of death. It differs from SIDS in that SIDS is assumed to be determined after a proper autopsy and death scene investigation. Unknown Cause implies an incomplete investigation/autopsy or no investigation/autopsy.

The Unknown Cause Numbers

Unknown Cause numbers have been gradually increasing from 2 infant deaths per 10,000 live births in 1995 to 3 per 10,000 live births in 2020. If we were doing a better job of investigation and tracking, these numbers would be going down.

In 1995 Unknown Cause made up 20% of all Sudden Unexpected Infant Deaths. In 2020 they made up 32% of sudden unexpected infant deaths largely because those deaths were not adequately investigated and therefore categorized as Unknown Cause. There is increasing resistance among death certifiers to use the term SIDS when they haven’t done a proper autopsy and death scene investigation. This is reasonable, but it makes it impossible to compare causes of death over time.

Accidental Suffocation and Strangulation in Bed (ASSB) in the US

The third sub-category is Accidental Suffocation and Strangulation in Bed (ASSB). When I first started looking at this I had no idea why Accidental Suffocation and Strangulation in Bed would be included in the Sudden Unexpected Infant Death Statistics. At first glance I thought those deaths wold have clear causation. I couldn’t understand why they would be lumped into the larger umbrella category that includes no other types of explained accidental deaths.

However, given that it is difficult to differentiate suffocation from a truly unexplained death and the bias that some death scene investigators have about babies’ sleep environments, it may be OK to combine these numbers in the overall SUID numbers. This is another example of how bad our data collection really is. I would wager that some of the babies in this category died from suffocation and others didn’t.

The ASSB Numbers

It is interesting that the apparent incidence of ASSB has been slowly rising in the US since 1995 from about 0.4 infant deaths per 10,000 live births in 1995 to about 2.5 per 10,000 live births in 2020. These numbers can be misleading because some coroners and medical examiners use this classification if they deem the baby’s sleep environment to be unsafe regardless of whether it clearly contributed to the infant’s death.

Since 1995 the number of babies overall who die unexpectedly (including ASSB) in their first year has remained steady at about 9 or 10 babies for every 10,000 live births. In 2020 ASSB cases made up about 27% of all Sudden Unexpected Infant Deaths.

Sudden Unexpected Infant Death Incident Reporting Form

The SUIDIRF was developed in 1996. It is regularly updated, most recently in 2021. It is a wonderful tool that helps a death certifier do a thorough investigation. Unfortunately, its use is voluntary and by no means universal.

Another Way to Look at the Overall SUID Numbers

We claim that sudden unexpected infant deaths have dropped as a result of the Back to Sleep program, but the overall numbers are relatively steady from 11 unexpected deaths per 10,000 live births in 1995 (the year after the Back to Sleep program was announced) to 9 or 10 out of 10,000 live births since then.

In this paper I looked at unexplained infant births as the number of deaths per 10,000 live births because I wanted the numbers to reflect a scale that was approachable and understandable.

Infant mortality from all causes in the US has decreased dramatically since 1980. There were 1,400 infant deaths per 100,000 live births from all causes in 1980. In 2021 there were 540/100,000.

When I look at the total number of deaths per 100,000 live births compared to the number of SUID deaths I found that the percentage change is minimal. In 1995, the year after the Back to Sleep program was introduced, SIDS and Unexplained deaths dropped from 15% to 14% of all infant deaths — not exactly earth shattering.

Since 1996, the year the Sudden Unexpected Infant Death Incident Reporting Form and international autopsy reporting protocols were introduced, the percentage of all infant deaths attributed to SUID has grown from 13% of all infant deaths to 17% of all infant deaths. If we were doing a decent job of data collection I would expect this percentage to drop significantly. It hasn’t in 28 years. In fact, it has steadily risen.

Problems With the Numbers
  • Both Hong Kong studies had a ridiculously low number of research subjects.
  • In some jurisdictions unexplained infant death numbers refer to deaths from birth to one year. In others, deaths during the first month are separately categorized as neonatal deaths.
  • ICD Code shifting — the ways to classify unexpected infant deaths have expanded and changed especially between 1999-2001.
  • We don’t know how many unexplained infant deaths are properly investigated and reported.
  • Not all jurisdictions use the Sudden Unexpected Infant Death Incident Reporting Form introduced in 1996 and revised most recently in 2021.
  • The deaths that are driving the overall numbers are not adequately investigated.
  • The quality of death certificate information varies from place to place.
  • The quality of death certificate information varies over time.
  • Despite the significant increase in infants being put to sleep on their backs (from 13% in 1992 to 77% in 2017. That’s a 492% increase.) We haven’t seen a dramatic reduction in unexplained infant deaths.
  • The use of SIDS as a cause of death has fallen out of favor. In many jurisdictions, this term has been reduced considerably or abandoned altogether. This makes it impossible to compare death statistics over time.
  • Unknown Cause deaths have been increasing while SIDS deaths have been decreasing due to the growing resistance of death certifiers to use the term SIDS when they haven’t done a proper autopsy and death scene investigation.
  • Death certifiers have varying credentials. Some are coroners who may be elected officials and have no medical, pathology or forensics training. An 18-year-old made headlines when she was elected deputy coroner in Jay County, Ind. while still in high school.
  • The CDC-supported SUID Case Registry only covers 40% of all SUID cases in the United States. This number has been gradually increasing, though.
Conclusions

I believe that back sleeping is saving lives, but due to data collection, analysis and interpretation issues we don’t know how many lives are being saved and we don’t know why. We make recommendations for best practices, but we can’t adequately evaluate their effectiveness.

Unfortunately, despite decades of effort, we are not yet fully prepared to begin collecting and interpreting accurate infant death statistics in the US. When we are ready it will take time to see long-term trends because the historical data were so poorly produced.

We have no shortage of statisticians to do the job. It’s the information being collected by death certifiers that needs to be more complete. Some of those certifiers are poorly trained. Some have biases that influence how they classify deaths. Some are bound by policies that require them to only consider certain causes of death and not others.

We have no national system for collecting infant death information. We have a patchwork of certifiers who, in their individual jurisdictions, follow a variety of ever-changing protocols. Established international autopsy protocols are not consistently followed even when autopsies are conducted. We’ve had a great tool for collecting this data (SUIDIRF) since 1996, but it has not been universally adopted.

Infant Risk Factors for Sudden Unexpected Infant Death
  • Prematurity
  • Low birth weight
  • Brain stem defect
  • Respiratory infection
  • Prone or lateral sleeping
  • Sleeping on a soft surface
  • Bed sharing
  • Overheating
  • Sex – males are more vulnerable
  • Age – between second and fourth months
  • Race – Black, Native American and Alaska Native infants are at higher risk
  • Family history – siblings died of unexplained causes
  • Secondhand smoke
Pregnancy Risk Factors for Sudden Unexpected Infant Death
  • Younger than 20
  • Smokes cigarettes
  • Uses drugs or alcohol
  • Inadequate prenatal care
Protective Strategies Against Sudden Unexpected Infant Death
  • Breastfeeding
  • Room-sharing
  • Pacifier use during sleep
  • Safe sleep surface
  • Avoid overheating
  • Supine sleep
References for The Safe to Sleep Program, SIDS and SUID:

Death Scene Investigation and Autopsy Practices in Sudden Unexpected Infant Deaths:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5063238/

Sudden Unexpected Infant Death and Sudden Infant Death Syndrome Data and Statistics:
https://www.cdc.gov/sids/data.htm

A Fresh Look at the History of SIDS: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6474527/

Understanding Sudden Infant Death Syndrome: https://www.theatlantic.com/health/archive/2016/06/understanding-sids/485147/

New Zealand’s National Cot Death Prevention Programme:
https://www.centreforpublicimpact.org/case-study/national-sids-prevention-programme

Sudden Infant Death History in Hong Kong: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1835990/pdf

Sudden Infant Death Syndrome — United States, 1980-1988: https://www.cdc.gov/mmwr/preview/mmwrhtml/00017250.htm

Co-sleeping in Infancy: the Controversies: https://www.fhs.gov.hk/english/health_professional/OMP_eNewsletter/enews_20180614.html

The Sudden Death in the Young Case Registry: Collaborating to Understand and Reduce Mortality: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5330401/

SUID and SDY Case Registry: https://www.cdc.gov/sids/case-registry.htm

Infant mortality rate in the United States from 1990 to 2021: https://www.statista.com/statistics/195950/infant-mortality-rate-in-the-united-states-since-1990/

Cot death in Hong Kong: a rare problem?: https://pubmed.ncbi.nlm.nih.gov/2866397/

Dummy use, thumb sucking, mouth breathing and cot death: https://pubmed.ncbi.nlm.nih.gov/10541944/

The Epidemiology of Sudden Infant Death Syndrome and Sudden Unexpected Infant Deaths: Diagnostic Shift and other Temporal Changes: https://www.ncbi.nlm.nih.gov/books/NBK513373/

International comparison of sudden unexpected death in infancy rates using a newly proposed set of cause-of-death codes: https://adc.bmj.com/content/100/11/1018

Infant Death Scene Investigation and the Assessment of Potential Risk Factors for Asphyxia: A Review of 209 Sudden Unexpected Infant Deaths: https://onlinelibrary.wiley.com/doi/10.1111/j.1556-4029.2007.00477.x

Breastfeeding and Reduced Risk of Sudden Infant Death Syndrome: A Meta-analysis: https://publications.aap.org/pediatrics/article-abstract/128/1/103/30379/Breastfeeding-and-Reduced-Risk-of-Sudden-Infant?redirectedFrom=fulltext

Results from the first year of the New Zealand cot death study: https://pubmed.ncbi.nlm.nih.gov/2020450/

Unexplained Pediatric Deaths: Investigation, Certification, and Family Needs: https://www.ncbi.nlm.nih.gov/books/NBK577032/

SUIDI Reporting Form: https://www.cdc.gov/sids/SUIDRF.htm

How Qualified Is Your Coroner?: http://www.pbs.org/wgbh/pages/frontline/post-mortem/things-to-know/how-qualified-is-your-coroner.html

Factors Associated With Choice of Infant Sleep Position: https://publications.aap.org/pediatrics/article-abstract/140/3/e20170596/38387/Factors-Associated-With-Choice-of-Infant-Sleep?redirectedFrom=fulltext

Most moms aren’t putting babies to sleep safely, study says: https://www.cnn.com/2017/08/21/health/baby-sleep-guidelines-sids-study/index.html

Infant mortality rate (under one year old) in the United States from 1935 to 2020* https://www.statista.com/statistics/1042370/united-states-all-time-infant-mortality-rate/