Since I’ve taken the night shift of baby-watching, I’ve developed a distressing habit of checking to see whether my daughter is still alive.
After trying a variety of different methods, I finally found one that offers reassurance without disturbing her. I stand near her head at a 45 degree angle to her crib and lean sideways towards her face so that my ear is two inches from her nose. Even when she is in a deep sleep — with the accompany slow, deep breathing — I can still hear her breathing. By angling by body I avoid breathing on her face (which wakes her up) and I can easily slip the life-verification routine into normal movement around her crib area.
Despite the comfort I receive from knowing my daughter hasn’t died, I struggle with whether this is normal behavior or if I am just paranoid about SIDS. My concerns about SIDS don’t necessarily stem from any specific problems or risk factors, but rather from reading too many articles about the disturbingly high rate of infant death from unexplained causes. Sudden Infant Death Syndrome (SIDS) is the leading cause of death for babies between one month and one year [see What is SIDS?] and since our daughter was lucky enough to be born healthy and at full-term, SIDS is now the primary risk factor for death.
A strict query of academic research shows nearly 4000 articles on SIDS and there are entire organizations dedicated to the prevention of SIDS across the country. Most references cite 2007 Centers For Disease Control data which links birth and death records to categorize infant mortality in the United States. The CDC Wonder database allows any concerned parent to analyze infant death rates across of host of variables including geographic, maternal, and birth characteristics.
There were 2461 deaths attributed to SIDS out of 4,316,233 live births in 2007 for a reported incidence rate of 0.57 per 1000 live births. Phrased differently, one baby died of SIDS for every 1754 live births in the country. However, the CDC data uses “live births” as a denominator and this figure includes severely preterm infants and other live births with congenital defects. For example, 13,286 of the 33,391 infants born before 28 weeks died during 2007.
Rather than look at my daughter’s SIDS risk as simply the nationwide average, I spent the entire night playing with the database to learn more about our specific risk profile. Whereas her risk of death from SIDS is 1:1754 when seen as just an average baby, the conclusion is far more reassuring when adjusted for her specific profile.
The CDC database includes four distinct time periods [1995 to 1998; 1999 to 2002; 2003 to 2006; and 2007]. Because of the data presentation and a change in the International Classification of Disease coding, it is impossible to aggregate the time periods into a single query. In addition, the American Academy of Pediatrics initiated their “Back to Sleep” campaign in 1992 and the trend of SIDS deaths follows a pattern consistent with the gradual diffusion of the public health message into mainstream consciousness. I decided to focus on most recent two time periods [2003 to 2006 and 2007] to try to avoid the inter-period confusion likely generated by the huge success of the AAP campaign.
My initial query using (what I considered to be) broad information about our case was too strict to generate useful information. Focusing on the relevant eight major risk factors for SIDS, I created a second query with
broad information about our family: Gender = Female / Age of Infant = 28 to 364 days / Race = Asian or Pacific Islander or White / Plurality = Single / Maternal Age = 20 to 39 / Prenatal Care = 1st month to 6th month.
The results of the second query show 173 deaths from SIDS out of 469,934 live births for an incident rate of 0.37 per 1000.
However, a quick glance at the untouched variables show (what I feel are) two glaring omissions: maternal education and gestational age. I added Maternal Education = Bachelor’s Degree or higher / Gestational Age = 37 weeks to 42 weeks to the second query to create a third query.
The results of the third query show 10 deaths from SIDS out of 128,409 live births for an incident rate of .08 per 1000 (note that incident rates are “suppressed” in the CDC data when there are fewer than 20 deaths given a query due to questionable sample sizes below that level).
A similar query for the 2003 to 2006 dataset shows 16 deaths from SIDS out of 219,847 live births for an incident rate of 0.07 per 1000 (again “suppressed” in the CDC data).
While my adventure in the CDC data should not be confused for scientific research, it has made me a little less concerned about my daughter dying from SIDS. As even my third query excludes some negative factors (i.e. our risk should be even lower than that of the subjects in the query since nobody in the house smokes, nobody uses drugs, etc.), I suspect that if there were 100,000 babies in an identical situation to my daughter’s, fewer than 10 would die of SIDS over the course of their first year of life.
Since our family is lucky enough to possess a strong profile for avoiding SIDS, I am now less concerned about checking every 30 minutes to see if my daughter is still breathing. Unfortunately for her, I am now tempted to kiss her forehead (thus disturbing her) instead.