To the Editor We propose that the conclusions of the study by Dr Swanson and colleagues1 of neurological changes reported by US government personnel in Cuba1 reflect improper interpretation of objective cognitive test results.
There are no universal rules for the classification and interpretation of cognitive test performance. Instead, several widely accepted systems based on the properties of a normal distribution are used. In their study, Swanson and colleagues1 selected a liberal cutoff to define impairment (<40th percentile), which is significantly higher than that typically used in clinical or research settings, classifying 39% of the population as impaired. The 40th percentile is in the average range, as it corresponds to only 0.25 SDs below the normative mean. Two commonly used classification systems in the references cited by Swanson and colleagues identify mild/borderline impairment as starting at less than the 16th percentile2 and less than the ninth percentile,3 which would decrease the percentage of total impaired scores in the study from 28% to 9% and 7%, respectively, a large difference. The conclusion that all 6 patients “had significant areas of cognitive weakness and/or impairment”1 thus appears to be based on a misapplication of common interpretive classification systems.
Shura RD, Kacmarski JA, Miskey HM. Neurological Symptoms in US Government Personnel in Cuba. JAMA. 2018;320(6):603. doi:10.1001/jama.2018.8698
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: