IRT Analysis of the Frontal Systems Behavior Scale: Identifying the Best Items for Use with Geriatric Patients in Primary Care

ABSTRACT

Objective: The Frontal Systems Behavior Scale (FrSBe) (Grace & Malloy, 2001) assesses behavioral dysfunction associated with frontal-subcortical damage; it is often used to measure these indicators of executive dysfunction in older adults with possible dementia. Although prior research supports the FrSBe’s clinical utility and factorial validity, little attempt has been made to examine which items are most useful for geriatric cases. The goal of the present study is to identify these items.

Method: Data from 304 older patients referred for neuropsychological assessment were used to examine the FrSBe’s three subscales: Apathy (A; 14 items), Executive Dysfunction (E; 17 items), and Disinhibition (D; 15 items). Item properties were investigated using the Graded Response Model, a two-parameter polytomous item response theory model.

Results: Difficulty parameters, discrimination parameters, and information curves identified 18 items that effectively discriminate (a ≥ 1.70) between levels of behavioral dysfunction and measure a range of dysfunction (b_A: -1.23 – 2.22; b_D: -.29 – 2.14; b_E: -1.81 – 1.77).

Conclusions: Most FrSBe items were effective at discriminating various levels of behavioral dysfunction, though weaker items were identified.

Clinical Implications: The findings suggest the FrSBe is a useful clinical tool when working with a geriatric population, though some items provide more information than others.

KEYWORDS:

• Assessment tools
• behavioral assessment
• IRT item analysis