Clinical Core

Summary (Clinical Core) In speech-language or stroke rehabilitation literature, relatively little attention has been paid to threats to a study’s validity that relate to correct implementation of treatment, and also assessment, procedures. These threats can increase the chance of Type I or Type II error, or the additional error (“Type III”) of concluding significance or nonsignificance when in fact the tests or the treatments were not correctly implemented. Additionally, these threats can prevent causal inference. In the absence of fidelity monitoring, investigators are unable to confidently determine whether or not results (significant or insignificant) were caused by the targeted independent variable or due to other random factors introduced because the clinician “drifted” from the protocol, or “contaminated” the protocol by adding or omitting elements. Lack of attention to implementation, then, could be a contributor to the historically mixed results in aphasia treatment research emphasized by Project 1 and Project 2. Most of the aims proposed by Fridriksson, Hillis, Bonilha, Rorden, and Hickok require recruitment and retention of patients and collection of high quality data. High quality data necessitate that assessors use prescribed administration procedures and raters adhere to prescribed scoring procedures. Seven of twelve project aims rely upon treatment, and therapists that adhere to prescribed treatment procedures. The assumption of the investigators is that clinical staff functioning as assessors, raters, and/or therapists adhere to the study protocol. Efforts to optimize adherence to assessment and treatment procedures, referred to as assessment and treatment fidelity, are critical to the proposed projects. The Clinical Core will coordinate and activities designed to optimize fidelity, thereby guarding against threats to validity and improving the power to detect effects for nearly every aim proposed. To guard against threats to validity stemming from selection bias, attrition, or missing data points, the Clinical Core will conduct training and monitoring activities related to recruitment, retention, and data collection and management. To guard against threats to validity stemming from drift, cross-contamination, clinician-to- clinician variability, or unblinding, the Clinical Core will conduct extensive training, initial and throughout the study, and monitoring activities that will ensure that there is standardized training for all clinical staff, that assessment and treatment was delivered as intended, and that there is evidence that treatment was received as planned. As far as we can tell, these will be among the first studies in stroke or aphasia rehabilitation to: (1) incorporate multiple-tier training programs for all study member staff, (2) monitor treatment fidelity through several complementary methods, and (3) monitor assessment fidelity via adoption of established treatment fidelity guidelines. Finally, the Clinical Core will also support the Projects by scoring behavioral data, both for reliability measures and for blinded measures of outcome. Thus, the main goals of the Core are training the study team, scoring and management of behavioral data, and optimizing adherence to the protocols.