Reliability and validity of the Norwegian version of the Severe Impairment Battery (SIB)

Int J Geriatr Psychiatry. 2008 Sep;23(9):896-902. doi: 10.1002/gps.2001.

Abstract

Objective: The Severe Impairment Battery (SIB) is developed to test cognitive function in patients with dementia of moderate to severe degree. We have conducted a study to assess the inter-rater reliability and the validity of the Norwegian version of SIB.

Methods: The reliability study comprised 30 patients, and the validity study 59 patients in nursing homes. We assessed Cronbach's alpha coefficient of the scale and the inter-rater reliability for the total SIB score and its nine sub scores between two testers by means of the Spearman's correlation coefficients. In the validity study we compared the SIB scores with the scores on the Clinical Dementia Rating Scale.

Results: The mean SIB score was 72.10 (SD 25.37). The Cronbach's alpha was 0.97, and the inter-rater reliability of total SIB score was Spearman's rho 0.85, and ranged from 0.46 to 0.76 for the nine sub-scores. The mean SIB score for patients with a CDR score <2 was 84.2(13.4), whereas total scores for patients with CDR 2 and 3 were 74 (18.9) and 48.4 (33.3), respectively. A cut-off point of 80.5 points gave the highest accuracy in discriminating between patients with CDR 2 and CDR 3, while a cut-off point of 87.5 best discriminated between CDR<2 and CDR 3.

Conclusion: The study indicates that the Norwegian version of SIB is a reliable and valid test with which to evaluate the cognition in patients with dementia of moderate to severe degree.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Cognition Disorders / diagnosis
  • Dementia / diagnosis*
  • Dementia / psychology
  • Female
  • Humans
  • Male
  • Neuropsychological Tests* / statistics & numerical data
  • Norway
  • Observer Variation
  • Reproducibility of Results
  • Severity of Illness Index