Rater Reliability and Responsiveness of GOT–Cog™, a Novel Occupational Therapy Cognitive Screen

IMPORTANCE: The Gaylord Occupational Therapy Cognitive screen (GOT-Cog™) is a novel cognitive screening tool developed to support inpatient occupational therapy treatment planning. It has demonstrated excellent content, construct, and criterion validity, as well as good internal consistency. OBJECTIVE: To assess GOT-Cog rater reliability and responsiveness. DESIGN: Repeated-measures design. GOT-Cog was administered to participants twice at admission and once at discharge. When different investigators collected the first two assessments, interrater reliability was tested; when collected by the same investigator, intrarater reliability was tested. The first and final GOT-Cog scores were used to assess responsiveness. SETTING: Single long-term acute care hospital (LTACH). PARTICIPANTS: 192 participants, recruited from inpatients admitted with an order for occupational therapy services. OUTCOMES AND MEASURES: GOT-Cog total score was the primary outcome measure. Rater reliability was evaluated using interclass correlation coefficients (ICCs); responsiveness was calculated using effect size and score change. RESULTS: Scale-level interrater reliability of GOT-Cog was moderate to good, ICC(2, 1) = .80 (95% confidence interval [CI] [.71, .87], r = .81, n = 87); intrarater reliability was good to excellent, ICC(2, 1) = .84 (95% CI [.76, .90], r = .86, n = 89). Domain-level rater reliability ranged from poor to excellent. A small effect size was demonstrated from admission to discharge (d = .37, n = 149) over a mean length of stay of 24.06 d (95% CI [20.70, 27.41]; n = 192). CONCLUSIONS AND RELEVANCE: Previous findings are supported, which indicates that GOT-Cog has the validity and reliability needed to support its use in the LTACH setting with a mixed medically complex inpatient population. Plain-Language Summary: Conducted in the long-term care setting, this study explains how consistent the Gaylord Occupational Therapy Cognitive screen (GOT-Cog™) is when given to the same person, multiple times, and by the same or different people, as well as how sensitive it is to change over time. To test this, GOT-Cog was given to participants twice at inpatient admission and once at discharge. GOT-Cog was found to have moderate to good consistency when delivered by different people and good to excellent consistency when used by the same person. Over an average of 24 days, GOT-Cog showed a small level of change in participants who were tested admission and again at discharge.