Psychometrics

The MBI has been shown to have high content reliability with a Cronbach’s coefficient alpha of internal consistency of α = 0.90 recorded at the commencement of rehabilitation (Shah, et al. 1989). Kucukdeveci et al. (2000) found that the internal consistency of the MBI in spinal cord injury patients was good (α = 0.88 to 0.90) and inter-rater reliability sufficient at the item level (kappa 0.50 to 0.78) as well as good for the overall inter-rater agreement as the intra-class correlation coefficient was 0.77. The same study found correlations between the MBI and American Spinal Injury Association motor scores ranged from r = 0.58 to 0.82 excluding personal hygiene at 0.25 and feeding at 0.27, respectively. Results of the Chinese version MBI-C suggests cross-cultural stability according to data from two sets of inter-rater reliability indices that were obtained by conducting the MBI-C and the MBI concurrently on a total of 15 people with stroke. This was evidenced by the mean total scores on the MBI-C and the MBI which were similar at 57.2 (SD = 20.4) and 60.5 (SD = 22.3) (Leung, Chan, & Shah, 2007). That study also found the most inter-rater consistent item for the MBI was personal hygiene (k = 0.85) while the least consistent item was toileting (k = 0.63), whereas the most inter-rater consistent item among the MBI-C items was transfer (k = 1.00), and the least consistent item was stair climbing (k = 0.81) (Leung et al., 2007).

Advantages

Disadvantages

Possible limitations of the measure point to the fact that although the MBI builds on a significant amount of previous work relative to the original, there is limited direct evidenced-based research for a number of conditions such as neurocognitive disorders of the Alzheimer’s and related types suggesting more research is needed to definitively apply the MBI to other populations in a similar fashion as the original.

Administration

The MBI is an activity-based observational tool in which the clinician observes and grades client performance across 10 specific ADLs using a 5-item scale (range 0 to 15) where 6 items are scored as (0) unable to perform task, (2) substantial help required, (5) moderate help provided, (8) minimal help required, or (10) fully independent; 2 items scored as (0) unable to perform task, (1) substantial help required, (3) moderate help provided, (4) minimal help required, or (5) fully independent; and 2 items scored as (0) unable to perform task, (3) substantial help required, (8) moderate help provided, (12) minimal help required, or (15) fully independent (Shah, 1998; Shah et al., 1989). The cumulative score is calculated by summing each item score with higher scores suggesting a greater degree of independence in ADL functioning. The following scoring norms, as outlined by its developers, quantifies an individual’s level of dependency one might find for particular scoring ranges.