Introduction
Multisite clinical trials increasingly use standardized cognitive assessment batteries, such as the NIH Toolbox. While computerized cognitive batteries streamline the administration process, operator error exists, and the need to train Outcomes Assessors (OAs) remains a critical component of standardized multi-site data collection. We describe the training and evaluation process used to ensure the standard administration of the cognitive portion of the NIH Toolbox (NIHTB-CB) across 69 sites in ICECAP, a trial designed to identify the optimal duration of induced hypothermia for neuroprotection in comatose survivors of cardiac arrest.
Methods
OAs were trained to administer the NIHTB-CB as a subset of a larger test battery including paper and pencil neuropsychological tests. All OAs completed a standardized training protocol that included: 1. review of online NIHTB training modules; 2. meeting with a member of ICECAP’s Quality Assurance Team (QAT) to discuss common administration errors; 3. completion of 5 practice administrations; and 4. submission of a certification video. The QAT used a detailed scoring “rubric” to evaluate errors in administration. The QAT provided detailed feedback to the OA, who was either immediately certified to administer the NIHTB-CB or asked to re-train and resubmit an additional certification video. In addition, each OA video recorded and submitted their first assessment with an ICECAP participant. A paired samples t-test was used to examine differences in the number of administration errors made on certification administrations versus first-trial participant administrations.
Results
To date, 44 OAs have been certified to administer the NIHTB-CB, and 34 completed their first study participant. 20 OAs made ≥1 administration error on the practice submission and 5 OAs made ≥1 administration error during their first participant assessment. There were 74 total administration errors on the NIHTB-CB at certification and 7 on first participant administration. OAs demonstrated significantly fewer errors from initial certification to actual participant testing, t(33)=3.07, p<.01.
Conclusions
Findings revealed that the implementation of a comprehensive training process improved the standard administration of the NIHTB-CB for the ICECAP trial, thus ensuring the rigor of the assessments that are administered as a part of this trial.