Monday, June 17, 2013

A medical journal's abdication of editorial responsibility

#TPSER9 One of the most moving and effective parts of the Telluride patient safety camps for residents and medical students is the presentation of the Lewis Blackman story (see trailer here.)  This is a heart-rending story about a teenage boy who undergoes elective surgery to repair a condition called pectus excavatum (sunken chest).  He dies within a few days because of post-operative complications which remain unremediated because of a series of medical errors and poor communication among the medical staff.  The date was November 6, 2000.

Less than a year later, an article was published in the Journal of Pediatric Surgery (J Pediatr Surg. 2001 Aug;36(8):1266-8.), authored by Lewis' surgeon and others.  The authors described the application of the minimally invasive surgical technique used for Lewis.  They pointed out that in 20 cases studied:

Average length of stay was 5.5 days. There were no early complications. Mean follow-up was 12 months. . . . One patient had a prolonged hospital stay of 7 days because of postoperative pain.

How could these study results be correct when Lewis had died the previous November, just a few days after the surgery?

We have to assume that the sampling of patients used by the authors ended in a manner that excluded Lewis' case.  Without knowing more, it is hard to know the scientific reasons for this limited sample.  That itself would be an interesting line of inquiry.

However, I am informed that the editor of the journal in question was made aware of Lewis' death.  Whether s/he questioned the authors about their sampling choice is not known.  But there is no indication, in the way of editorial comment or submittal by other authors, as to the issues raised in Lewis' case.

I'm not suggesting in any way that the procedure carried out by these doctors was inappropriate, but I am suggesting that the silence by the journal on this issue raises a question of editorial ethics.  Even if the death of this child was the result of circumstances not related to the specific surgical technique, it was certainly a death related to the surgery.  As a sad case of "the procedure was a success but the patient died," it warranted attention by those in the profession.

The silence by the Journal of Pediatric Surgery in this matter appears to represent a case of abdication of editorial responsibility.  Although it is years later, they owe the public an explanation.


Anonymous said...

Having pectus excavatum myself, this story struck home. It is clear the authors found a way to convince the editorial staff the death was not related to the surgery, a prime example of the opacity and outright cover-ups that go on in our profession. Hopefully, as a result of the efforts of Lewis' family and others, they would not get away with this in today's world.

nonlocal MD

Patricia L. Hale said...

From Facebook:

Study bias in case sampling is so very common that these types of studies sadly must always be suspect. Only randomized controlled studies avoid this but even then there are many many statistical ways to manipulate the data. Sad reality unfortunately when people and reputations are involved.

Bruce Randolph Tizes said...

From Facebook:

See John Ioannidis' “Why Most Published Research Findings Are False."

e-Patient Dave said...

I would welcome a response by the authors. There may have been some specific reason for excluding that case, which we don't know about. If so, the reasoning should be published so that, like all good science, it can be verified by other eyes.

You hit such a nerve on this that I just spent two hours composing what was GOING to be a comment, but at THAT size it has to be a post of its own.

The whole idea of journal editing is that only journal editors can be trusted to curate information. If that process derails, all of science suffers. That's what Ben Goldacre has been talking and writing about for several years, and it appears it's starting to get some results.

IMO every single soul who's responsible for choosing treatments needs to be well schooled on this.