Rosin's epiphany began, interestingly enough, in her pediatrician's office where she discovered an article in a 2001 issue of the Journal of the American Medical Association which dealt with the relationship between breastfeeding and childhood obesity. The researchers concluded that breastfeeding didn't make as much of a difference as family history did. This discovery led Rosin to embark on her own study of the medical literature about breastfeeding. Her conclusion: "Breast-feeding is probably, maybe, a little better; but it is far from the stampede of evidence that (Dr. William) Sears describes. More like tiny, unsure baby steps: two forward, two back, with much meandering and bumping into walls."
What about all that medical literature? Is it really that ambiguous? In a fascinating juxtapostion of timing, researchers at the Australian Center for Economic Research on Health just published a working paper in December, 2008, a version of which has been accepted for publication in the Journal of Human Lactation. This paper is entitled "'Voldemort' and Health Professional Knowledge of Breastfeeding - Do Journal Titles and Abstracts Accurately Convey Findings on Differential Health Outcomes for Formula Fed Infants?" According to authors Julie P Smith, Mark D Dunstone and Megan E Elliott-Rudder, "This study aimed to assess whether health professionals have access to unbiased information on the health implications of formula feeding through perusing the titles and abstracts of a sample of published scientific studies of health impacts of breastfeeding versus formula feeding."
For their review they chose the 78 articles used by the American Academy of Pediatrics in the formulation of their 2005 Policy Statement on Breastfeeding and the Use of Human Milk. Each of these articles showed clear, measurable advantages for babies who were breastfed or received breastmilk. The concern was that many studies which demonstrate a clear difference in outcome between breastfed and formula-fed infants fail to cast formula feeding in a negative light in the title or abstract. In fact, the authors compare infant formula to Voldemort of Harry Potter fame as "He who must not be named." Their methodology was fairly simple, but quite comprehensive. Each of the 3 researchers reviewed all 78 articles and placed the title in one of these 3 categories:
- Misleading - Title associates breastmilk or breastfeeding with an illness (ex: "Breastfeeding and Childhood Obesity")
- Neutral or Silent - Title is neutral or silent on results (ex: "Brainstem Malnutrition in Premature Infants") or the title includes positive statement about breastmilk or breastfeeding but does not mention infant formula (ex: "Longer breastfeeding and protection against childhood leukemia and lymphomas")
- Names Voldemort - Title includes a reference to artificial infant feeding or formula (ex: "Cow's milk exposure and type I diabetes mellitus")
- No mention of formula - No mention of formula or does not compare formula feeding to
breastfeeding except in describing method - Neutral/Breastfeeding Better than Formula - Breastfeeding better compared to formula feeding: This category comprises those studies that compared breastfeeding to formula feeding. Mostly conclusions couched in terms of breastfeeding advantages/or benefits over formula, not elevated risks from formula feeding.
- Increased risk of ill health associated with formula feeding 'Names Voldemort' - Formula feeding associated with increased risk of ill health: This category comprises those abstracts that conveyed that formula fed infants had higher risks of ill health, or it was clear the formula feeding was not the norm or control
Why is this significant? If the very researchers who are evaluating the beneficial effects of breastmilk - and by default the negative effects of formula - don't make that distinction clear in the abstract and/or title, how can a journalist like Ms. Rosen who conducts a serious review of the literature be expected to come to any conclusion but the one at which she arrived? The authors of the Australian study are quick to point out that mothers should NOT be put on a guilt trip for not breastfeeding. Rather, researchers must be more forthright in accurately stating their conclusions when titling their studies and writing abstracts. They do a real disservice to both medical professionals and mothers alike when they fail to highlight the increased risks to infants who receive formula.
Another glaring omission in the Atlantic Monthly article is the complete lack of any reference to the many proven health benefits the breastfeeding mother receives. To her credit, however, Ms. Rosen continues to nurse her third child. She concludes her article with the following statement which highlights yet another incredible benefit of breastfeeding: "Breast-feeding does not belong in the realm of facts and hard numbers; it is much too intimate and elemental. It contains all of my awe about motherhood, and also my ambivalence. Right now, even part-time, it’s a strain. But I also know that this is probably my last chance to feel warm baby skin up against mine, and one day I will miss it."
Thank you for this examination, it's very important that researchers title their abstracts and studies more carefully. I didn't think of it that way, but you're right, what other conclusion should Ms Rosen come to? - Sue Rickman
ReplyDeletetopic is interesting and all, but what is really cool is that Harry Potter references have seeped into scientific literature. frickin hilarious!
ReplyDeleteI can understand why scientists label studies the way they do, and I feel it is unfortunate that any reference to formula not being equal to breastmilk or breastfeeding is often dismissed as emotional and disregarded.
ReplyDeleteI have moderated online breastfeeding forums, and I found that any well-done study I shared was subject to being torn apart for a comment a researcher made or article title that hit a nerve with someone or made someone feel attacked. I felt frustrated and sad that a topic that can be so empowering to women has become a source of pain and resentment to some, and I am happy to see your post here that presents accurate information and support.
The science of an article is often totally lost in emotional reactions, and since scientists ideally present what they find and aren't there to make judgements from the results, it is really common to end up with ambiguous titles and perhaps even tiptoes around clear results in order to get important information into a format that people might more readily accept. You are right that it often shrouds the results and clouds the issue when presented this way. I hope this changes in the future, and I thank for presenting the information in such a clear and interesting manner.
Ruth Anne
What a great well-researched post. I am always appreciative of bloggers who do their homework. Makes me stay motivated to stay on top of mine too! Thanks for this!
ReplyDelete