Validity and Reliability Testing of the Gaylord Occupational Therapy Cognitive (GOT-Cog®) Screen: A Novel Cognitive Screen for Inpatient Occupational Therapy Treatment Planning

Research Objectives: To evaluate the construct validity, interrater and intrarater reliability, and responsiveness of the Gaylord Occupational Therapy Cognitive (GOT-Cog®) screen in the long-term acute care hospital (LTACH) setting.

Design: GOT-Cog construct validity was assessed using a randomized cross-over controlled study design using the St. Louis University Mental Status (SLUMS) exam as the comparator. To minimize mental fatigue, the assessments were collected over two consecutive days. For reliability and responsiveness testing, a separate cohort was recruited and given GOT-Cog three times; twice at admission over two consecutive days and again at discharge. If the first two assessments were collected by different investigators, interrater reliability was tested; if collected by the same investigator, intrarater reliability was tested. Responsiveness was evaluated using the first and final GOT-Cog scores.

Setting: Gaylord Hospital, a 137-bed LTACH located in Wallingford, Connecticut, USA.

Participants: Inpatients with an order for occupational therapy.

Interventions: n/a

Main Outcome Measures: Construct validity was primarily evaluated using the correlation of GOT-Cog and SLUMS, Reliability was evaluated using interclass correlation (ICC), and responsiveness using Coden's d for effect size.

Results: For construct validity, total GOT-Cog and SLUMS scores positively correlated (r=0.75, p< 0.0001, n=98), and no ordering (p≥0.8081) education (p≥0.4469) or program (p≥0.1738) effects were observed. All shared domains between GOT-Cog and SLUMS showed significant correlations (r-value range = 0.24-0.69, p≤0.0155), and all unique domains showed significant correlations to both GOT-Cog and SLUMS total score (r-value range = 0.24-0.65, p≤0.0194). For reliability, GOT-Cog demonstrated a moderate-to-good interrater reliability (ICC2,1 (95% CI) = 0.74(0.63, 0.83), r=0.82, n=83), and good-to-excellent intrarater reliability (ICC3,1 (95% CI) = 0.85(0.79, 0.90), r=0.82, n=83). For responsiveness, GOT-Cog demonstrated a small effect size from admission to discharge (d=0.40, n=144) in the LTACH setting over a mean (95% CI) length of stay of 22.6(20.5, 24.7) days.

Conclusions: To our knowledge, GOT-Cog is the first cognitive screen to be developed specifically for the LTACH setting. If widely adopted as the measure of choice, we believe this screen will allow occupational therapists to better identify patients’ functional cognitive deficits that may impact their recovery. Being a more comprehensive screen, GOT-Cog may lead to timely interdisciplinary referrals at admission versus later in a patient’s stay, thus improving the care, and we suspect the rate of recovery, of patients who may not have otherwise been evaluated by these specialties.

Author(s) Disclosures: Authors have no conflicts of interest to disclose. Please note, a subset of this data was presented previously at ACRM 2023 as poster.