Implementing advanced psychometrics to identify significant change between cognitive test scores in response to CSF drainage

Introduction: Change in cognition is a clinical hallmark of Idiopathic Normal Pressure Hydrocephalus (iNPH). Accurate measurement of outcomes is crucial to optimizing intervention during cerebrospinal fluid (CSF) drainage. However, repeated administration of cognitive tests, decreases measurement reliability. Simple change scores (SCS) do not account for test–retest error. The present study uses reliable change indices (RCI) to improve reliability and validity in cognitive measurement.

Methods: A retrospective record review of 15 patients who underwent CSF drainage for suspected iNPH between October 2021 to April 2022 was conducted. Pre-post CSF drainage data for six cognitive tests was extracted: Trail Making Task A and B, RBANS Coding, Semantic Fluency, Immediate and Delayed Story Recall. SCS and RCI scores were computed using Classical Test Theory (Duff, 2012), and categorized into overall improvement, deterioration, or no change.

Results: Fifteen patients (Female n = 10; Male n = 5; age M = 75 years), underwent successful CSF drainage. Change analysis using SCS categorised most patients as improved (n = 7), rather than deteriorated (n = 4) or no change (n = 4). Change analysis using RCI categorized most patients as no change (n = 12), rather than deteriorated (n = 2) or improved (n = 1).

Conclusions: Using simple difference in raw scores to measure change in cognition, introduces risk of type 1 error (false positives). The RCI equation is proposed as a method to improve reliability in cognitive outcome measurements following CSF drainage. Development of tests that are less susceptible to test–retest effects is encouraged.