Undisclosed conflicts of interest in German-language textbooks of anesthesiology, critical care, and emergency medicine.

BACKGROUND: Medical textbooks are important throughout postgraduate internship, residency and fellowship programs but authors' conflicts of interest (COI) are rarely disclosed. In anesthesiology, two high-profile cases of research fraud/unethical publishing have previously been identified. This study evaluated whether anesthesiology textbooks in German include controversial recommendations on hydroxyethyl starch (HES) and whether authors had undisclosed COI.

METHODS: Forty-five anesthesiology, critical care, and emergency medicine electronic textbooks (January 2015-August 2017) were identified on the websites of German, Austrian, and Swiss publishers. Six textbooks were excluded (irrelevant content; compilation of peer-reviewed articles). Content on HES ('Kolloid', 'Hydroxyäthylstärke', 'Hydroxyethylstärke', 'HES', 'HAES') and COI ('Interessenkonflikt') was electronically searched; subject lists were manually searched. Recommendations on HES were analyzed in the context of indications/warnings/international guidelines. Potential COI for authors of controversial recommendations were searched on PubMed/EMBASE (January 2016-May 2018).

RESULTS: COI statements were absent in 38/39 textbooks (present for 1/553 authors; 0.2 %). Twelve textbooks did not discuss HES for volume resuscitation. HES was recommended in 13 textbooks; many recommendations were controversial (10/39 textbooks, 26 %; 25/553 authors, 5 %). Journal COI statements were available for 18/25 authors; 7 authors had financial/other COI, including 4/18 (22 %) with ties to HES manufacturers.

CONCLUSIONS: Recommendations on volume resuscitation with HES are highly controversial in German-language electronic textbooks; about one-quarter of relevant textbooks are inconsistent with regulatory authority or international guideline criteria. Undisclosed potential COI of textbook authors making controversial recommendations of HES were frequent, linking roughly one in five authors to the HES-producing industry. Against this background, COI statements must become part of standard textbook publication ethics.