The evidence base for shaken baby syndrome

Editor—Reece et al's response to the editorial of Geddes and Plunkett claims to be a response of “106 doctors.”¹,² What, precisely, do the 106 signatures attached to this letter signify? That all had reviewed the letter and were in full agreement with the entire content? That they agreed in general with the thrust of the letter? Or was this more a show of solidarity on the part of doctors who care deeply about the risks of shaking on children?

This needs clarification if the signatures are to carry any weight whatsoever. Science—even medical science—is not a popularity contest. The meaning of a signature must be made explicit for it to add weight to a document.

Each signature carries with it responsibilities of authorship. Reece's letter declared no competing interests, but all signatories would need to comply for this to be true.

Six of the signatories (Levin, Chadwick, Alexander, Barr, Jenny, and Reece) are medical practitioners on the International Advisory Board of the National Center on Shaken Baby Syndrome (www.dontshake.com). They participate in this group's conferences and are presumably compensated or reimbursed for this work—information requiring disclosure under BMJ guidelines.

Figure 1.

The letter of Reece exemplifies a problem identified in my own paper³—that the literature on shaken baby syndrome is polarised and based more on strong beliefs and opinions than strong data. Ten thousand signatures cannot change this.⁴