Convergent Validity of the Occupational Therapy Adult Perceptual Screening Test (OT-APST) with Two other Cognitive-Perceptual Tests

Abstract

Background: It is common for individuals with a neurological impairment, such as a stroke, to present with cognitive-perceptual impairments. The early identification of such problems can assist with rehabilitation efforts. Standardised cognitive and perception tests are often used by practitioners to make such identifications; therefore, it is important to examine the measurement properties of these tests. Objective: This study investigated the convergent validity of the Occupational Therapy Adult Perceptual Screening Test (OT-APST) with two other cognitive-perceptual tests: the Neurobehavioural Cognitive Status Examination (Cognistat) and the Developmental Test of Visual Perception — Adolescent and Adult (DTVP-A). The Cognistat is primarily a test of cognition but has some subscales that evaluate perceptual skills. The convergent validity between the Cognistat and the DTVP-A was also investigated. It is imperative that tests used by therapists have documented validity to ensure that the results obtained actually measure the skills and abilities that the tests claim they measure. Method: Thirty-two adult patients, ranging in age between 46 and 91 years (mean = 73, SD = 4.6) with a primary diagnosis of stroke, completed the OT-APST, the Cognistat and the DTVP-A. Spearman's rho was used to analyse the correlations between the tests' subscales. Results: Three of the Cognistat subscales were significantly correlated with OT-APST subscales: Cognistat Constructional Ability and OT-APST Body Scheme; Cognistat Memory and OT-APST Functional Skills; and Cognistat Calculations and OT-APST Body Scheme. Only one of the OT-APST subscales, Body Scheme, had statistically significant correlations with the DTVP-A's Figure-Ground, Visual-Motor Search and Visual Closure subscales. There were significant correlations between the Cognistat's Constructional Ability and Judgment subscales and five of the six DTVP-A's subscales. Conclusion: Given that only a few of the subscales from the Cognistat and DTVP-A were significantly correlated with those of the OT-APST, this indicates that many of the constructs assessed by the three tests are not associated; hence, limited convergent validity evidence for the OT-APST was present. Further inquiry is required because it is important that such instruments are assessing what they purport to test, particularly when they are used to determine clients' strengths and weaknesses in clinical practice.