Bad Science:The curious case of Tina Kimmel and the CPSC

***Update to this post on June 30th. ***

I have my own ideas about cosleeping, which will not make APers very happy*. I will say that while I believe cosleeping can be a safe option, it probably rarely is in
real life, and certainly the benefits of cosleeping have been completely overblown by advocates.

That’s not what this post is really about, though: it’s to demonstrate how certain AP/NP advocates, even when confronted by evidence their practices can be harmful, are quick to dismiss and obfuscate the results. I hope you keep this in mind when reading “scientific” articles in places like Mothering Magazine, or even in the mainstream media.

In 1999, Dorothy Drago and Andrew Dannenberg of the Center for Injury Research and Policy at Johns Hopkins University School of Hygiene and Public Health, published a study based upon the 2178 cases of infant suffocation listed in the Consumer Product Safety Commission’s (CPSC) casefiles from 1980-1997. While they could not calculate the exact risk ratio, given that the percent of cosleeping babies in the population is unknown, they did note a very high rate of suffocation/SIDS of babies in adult beds as compared to cribs. They concluded:

Bed-sharing and use of adult beds for infants should be discouraged.

Naturally, this raised a howl from APers. Peggy O’Mara, the editor of Mothering Magazine, stated “It’s none of their business!” (nor is it Peggy’s, not that it ever stopped her, did it?). Later on, Dr. Sears as good as alleged the study’s authors were monetarily influenced by crib manufacturers. He also states (emphasis mine):

In fact, one independent researcher examined the CPSC’s data and came to the opposite conclusion than did the CPSC – this data supports the conclusion that sleeping with your baby is actually SAFER than not sleeping with your baby (see Mothering Magazine Sept/Oct 2002).

He’s referring to this article, in which a “researcher”, Tina Kimmel, attempted a mathematical 3-card monty to “prove” that using the data in Draco and Dannenberg, the death rate of infants in a crib is twice the death rate of those in a bed.

Tina Kimmel, MSW, MPH (USC Berkeley), is hardly a stranger to medical research. She’s been a close associate and research assistant to Dr. Michel Odent, maverick surgeon whose baseless, biased statements have been mentioned on this blog before. Kimmel and Odent co-authored a study which claimed children vaccinated with pertussis (whooping cough) vaccine were more susceptible to developing asthma (a claim which has not been supported by the subsequent literature on the subject). She’s also written and researched other AP-related subjects, apparently, though, her research was of insufficient quality to be accepted by any meical journal. According to this 2003 bio, she has “studied, worked, presented, and published on topics of primal therapy, birth, attachment, circumcision, breastfeeding, vaccination, adoption, and co-sleeping.”. Hmmmm…an “independent researcher” with a huge axe to grind, it seems.

Tina Kimmel’s analysis focuses on 2 relevant points: the first is to quantify how many deaths in the database occurred in an adult bed vs. deaths in a crib. This information can be gleaned in the original article from Table 2, and is available in Tina Kimmel’s Mothering article in Table 1. This provides us with the numerator for our relative risk assessment of beds vs. cribs.

***At this point, I strongly suggest you print out both tables and get out a calulator if you really want to follow my argument closely, because flipping between windows can be very confusing.***

The second argument she makes regards the denominator – what % of infants actually cosleep? This is also necessary in order to try and figure out which sleeping place is actually riskier. Kimmel bases her denominator on information gleaned from 5 American states and extrapolates this to all the US, and claims that the prevalence of cosleeping infants on any given night is 44% or whereabouts. I don’t think she can make such an appraisal based upon 5 states only, but let’s accept this figure, if only for lack of a better one.

If you compare the two tables I mentioned above, you’ll see Kimmel grouped hers differently than Drago & Dannenberg. I don’t think this was coincidental.

Kimmel counts 428 cases of infant deaths occurring in cribs. She further states that she can only be sure of 139 infant death actually happening in beds (102 cases of overlaying in a bed and 37 cases of oronasal obstruction in a waterbed):

We can’t use the other 739 bed- or bedding-related cases in our analysis, because the place of death is not specific enough; these deaths may have occurred in a large adult bed, a single-size adult bed, a child’s bed, or a misused crib.

Not so. An bed is a bed, no matter what its size, nor is it likely that most parents cosleep with their children in child-sized beds, or put tiny babies to sleep with older siblings in child beds. Misused cribs are listed under death in cribs, and this is stated very clearly. If she really thought this point wasn’t clear enough, the proper thing to do would be to contact the study’s authors.
If we count all deaths in the original table that can be attributed to adult beds, we get:

Between bed/mattress and wall – 285
Between bed/mattress and frame – 149
Between bed/mattress and other – 57

Oronasal obstruction:waterbed – 37

Overlain: in a bed – 102

Entrapment with suspension:in bed – 29

Positional asphyxia:bed – 6

Unknown:bed – 44

For a total of 709 bed-related deaths.

There is another wedging category containing 102 cases, called “Other within crib/bed” which apparently includes both crib and bed deaths. We’ll get to that category in a minute.

To remind you, Kimmel states that she counted 428 crib-related deaths. However, looking at the original table in Drago & Dannenberg, we find only 326 certain crib-related cases:

Wedging: Between mattress and crib frame – 193
Entrapment with suspension:in crib/cradle – 64
Compression:collapsed crib – 13
Compression:Other crib/cradle/playpen – 5
Positional asphyxia:crib – 3
Unknown:crib/cradle/playpen – 48

Kimmel’s table enumerates the first 3 crib-related categories, listing the last 3 (totalling 56 deaths) under the category of “other”. However, that category lists 158 deaths. Where have the extra 102 deaths in cribs come from? Why, Tina’s decided to inflate the crib-related deaths by putting all the deaths in the category “Wedging:Other within crib/bed” into the crib group.

How very curious! I would guess Tina Kimmel, with her MPH and all, really should know better, but she knew quite well that none of her groupies at Mothering would understand what she’d done. I did expect a wee bit better from Doc Sears, though.

So let’s pretend the deaths in that group were 50% bed-related and 50% crib-related (being nice and all) and that Kimmel is right about the cosleeping stats – some 44% of parents share their beds with their babies on any given night. That would make 760 bed-related deaths and 377 crib-related ones. The relative risk for bedsharing in this case, using the same calculation she used, would actually be 1.66.

This, of course, is making certain assumptions (if the actual % of bedsharers is smaller, the ratio goes up, for example), so don’t quote this risk ratio as gospel. The point is, however, that bedsharing is certainly not SAFER, certainly not twice as safe, as Kimmel contends. The only thing we can be sure of is that Kimmel intentionally “cooked” the data to prove something that wasn’t true.

Another worrying thing to me is that Drago & Dannenberg clearly showed that wedging in between a bed’s mattress and the wall is one of the most common methods of infant suffocation. However, Dr. Sears in the article on “safe sleep” mentioned above, is still recommending (emphasis mine)

Yet, to be safe, place baby between mother and a guardrail or push the mattress flush against the wall and position baby between mother and the wall.

Safe sleep, indeed. It’s a good thing baby product manufacturers have taken notice, though; hence, I guess, the invention of this.


*Or to be precise, cosleeping that is defined as sharing a sleep surface with your baby. Room sharing with your baby, or sharing sleep on a separate sleep surface, is something that I, along with most medical sources, wholeheartedly endorse.

add to : Add to Blinkslist : add to furl : Digg it : add to ma.gnolia : Stumble It! : add to simpy : seed the vine : : : TailRank : post to facebook


2 Responses

  1. […] basis, rather than an objective look at the facts.  If science is used, we often see examples of poor studies that are deeply flawed.  This creates a Chicken Little situation making it hard to know when […]

  2. Extremely persuasive use of facts and still a fun read. I thank you on behalf of many others.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: