Oklahoma State University-led research has come to an interesting research conclusion: that more than half of clinical trial abstracts published in elite psychology and psychiatry academic journals are based on spin—the exaggeration of the clinical significance of a specific or particular therapy or treatment without the scientific statistics to back up the claims. This is a troubling finding.
One of the points of taking the time to read a lengthy article in an academic publication is the confidence professionals have in the substance of the content. Professionals expect that when the findings of a clinical trial are reported on, that the claims are backed by solid statistics: metrics and various quantitative and qualitative attributes.
Oklahoma State University Study
It turns out that this may be more of a belief than a reality. Researchers sought to uncover how much spin in fact existed in clinical trial abstracts published in academic journals. The team reviewed PubMed for randomized clinical trials focusing on psychiatric and behavioral health treatments between 2012 and 2017 in six elite psychology and psychiatry journals.
The focused on trials whose results were not statistically significant and used a previously published definition of “spin” to help, as objectively as possible, assess the quantity of article that went through the “spin” process.
Surprisingly, the Oklahoma team found a lot more spin than they had anticipated. With a total of 65 trials, evidence of spin in abstracts was at 56%–spin in conclusion sections equaled 49%. In 17 trials, they found evidence of spin in both results and conclusion 15% of the time.
Interestingly, spin become more common in clinical trials that compared specific drugs or behavioral methods with placebo intervention or usual care.
The authors couldn’t correlate industry funding with spin which was a relief for some.
Samuel Jellison, Oklahoma State University Center for Health Sciences, Tulsa, Oklahoma, USA
Will Roberts, Oklahoma State University Center for Health Sciences, Tulsa, Oklahoma, USA