Note: There happened to be several articles with ethical implications recently. Enjoy JournalFeed - Ethics Week.
Written by Clay Smith
Spin is common (>40%) in emergency medicine RCTs. Watch out. Here’s how to spot spin.
Why does this matter?
Spin - “use of specific reporting strategies, from whatever motive, to highlight that the experimental treatment is beneficial, despite a statistically non-significant difference for the primary outcome, or to distract the reader from statistically non-significant results.” If we can spot spin, we can interpret the literature better and come closer to truth.
I’m not sure how to spin this…
The authors pulled emergency medicine RCTs from the top 5 EM journals and top 5 general medical journals and looked for spin. Specifically, spin was defined as above. Their strategy was to include the RCTs with non-significant primary outcomes and see if the authors tried to spin them. Out of 348 RCTs, 114 had non-significant primary outcomes. Of these, authors used spin in 50 (44.3%). This happened more often in general medicine journals than EM journals and more often in industry funded studies. The most common spin was to emphasize a statistically significant subgroup analysis or to conclude they had, “accomplished an objective that was not a prespecified endpoint.” Be on guard for spin.
How to Spot the Spin:
Optimistic language (“trend toward significance”…I feel guilty, having said this)
Focus on modified treatment, off protocol, not intention-to-treat
Focus on significant secondary endpoint or subgroup
Claim of equivalence or noninferiority with non-statistically significant results (when not designed as a noninferiority study)
Claim of accomplishing a goal that was not pre-specified
Registered and published endpoints don’t match
Evaluation of Spin in the Abstracts of Emergency Medicine Randomized Controlled Trials. Ann Emerg Med. 2019 May 14. pii: S0196-0644(19)30230-6. doi: 10.1016/j.annemergmed.2019.03.011. [Epub ahead of print]
Open in Read by QxMD
Reviewed by Thomas Davis