Uncorrected versus demographically-corrected scores on the NIH Toolbox Cognition Battery in persons with traumatic brain injury and stroke.
Kristian P Nitsch, Kaitlin B Casaletto, Noelle E Carlozzi and 3 others
PMID 29265869WHAT IT FOUND
For therapists interpreting NIH Toolbox cognition scores after TBI or stroke, corrected scores make acquired cognitive problems easier to see.
They reduce the influence of age, education, sex and race, and better separate severe TBI from milder TBI.
Key findings
01On the fluid cognition composite, people with TBI or stroke scored lower than matched controls, and the difference was larger after demographic correction (TBI d=0.83 vs d=0.66; stroke d=1.10 vs d=0.97).
02Demographic factors accounted for 29% to 38% of variance in uncorrected fluid scores, but less than 5% after correction.
03Corrected fluid scores distinguished severe TBI from complicated mild-to-moderate TBI (d=0.52), while uncorrected scores did not (d=0.12); both score types distinguished severe from mild-to-moderate stroke (d=0.45 to 0.54).
STILL TO COME
How it was doneWhat they foundWhat it means for OTsWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Stroke controls were not matched on race/ethnicity because the stroke sample had many African Americans, so some uncorrected group differences may reflect demographic differences rather than injury. All participants were community-dwelling and at least 1 year after injury, so the results may not apply to acute rehabilitation or patients with more severe disability. Stroke severity was classified with the Modified Rankin Scale, and the community-dwelling sample restricted the usable range to 0 to 4, which may miss subtle differences in severity. The study did not measure functional outcomes such as independence, return to work or daily living performance, so it cannot show that corrected scores predict recovery or participation. Injury location and neuroimaging were not consistently documented, so some unexpected crystallized score differences may reflect language or other injury-related effects rather than premorbid ability. The crystallized composite is partly language-based, and subtle language impairment may not have been detected even though frank aphasia was excluded.
The easy way to misread this
Do not conclude that demographically corrected scores are better for every rehab decision. They were more sensitive for detecting differences from demographic peers, but the study did not show they predict functional outcomes such as independence or return to work, and it notes uncorrected absolute scores may be better for some functional prediction.