Development of a Short Form of the Severe Impairment Battery

Abstract

Objective The authors sought to develop a short form of the Severe Impairment Battery (SIB). Methods Authors describe the development of an empirically-derived short form of the SIB (SIB–S) by use of data from 191 subjects with severe dementia in the United States and France. Results Mean (standard deviation) Mini-Mental State Exam scores for the American and French samples were 7.7 (4.8) and 5.7 (3.4), respectively, and original SIB scores were 71.87 (18.34) and 58.38 (26.86), respectively. Exploratory factor analyses were conducted separately and in combination for the two samples, to determine the number of clinically meaningful factors. An eight-factor model, explaining 60.2% of the common variance, was selected. The eight constructs were described as: expressive language, memory (verbal and nonverbal), social interaction, color-naming, praxis, reading and writing, fluency, and attention. Derived SIB–S scores were 38.41 (9.12) and 29.79 (13.17) for the American and French samples, respectively. Conclusions The original SIB is a valid and reliable research tool developed to enable reliable assessment of patients with severe dementia; it takes approximately 30 minutes to administer. The SIB–S takes only 10–15 minutes to administer, making it more appropriate for use in patients with very severe dementia, while it maintains the attributes of the original SIB. The authors sought to develop a short form of the Severe Impairment Battery (SIB). Authors describe the development of an empirically-derived short form of the SIB (SIB–S) by use of data from 191 subjects with severe dementia in the United States and France. Mean (standard deviation) Mini-Mental State Exam scores for the American and French samples were 7.7 (4.8) and 5.7 (3.4), respectively, and original SIB scores were 71.87 (18.34) and 58.38 (26.86), respectively. Exploratory factor analyses were conducted separately and in combination for the two samples, to determine the number of clinically meaningful factors. An eight-factor model, explaining 60.2% of the common variance, was selected. The eight constructs were described as: expressive language, memory (verbal and nonverbal), social interaction, color-naming, praxis, reading and writing, fluency, and attention. Derived SIB–S scores were 38.41 (9.12) and 29.79 (13.17) for the American and French samples, respectively. The original SIB is a valid and reliable research tool developed to enable reliable assessment of patients with severe dementia; it takes approximately 30 minutes to administer. The SIB–S takes only 10–15 minutes to administer, making it more appropriate for use in patients with very severe dementia, while it maintains the attributes of the original SIB.