(Hauser) Ambulation Index

Multiple Sclerosis:

(National MS Society webpage http://www.csp.org.uk/outcome-measures/hauser-ambulation-index)

• Test-retest reliability reported to be good

(Amato et al, 1999)

• AI might have questionable test-retest reliability, especially for patients with EDSS scores between 2.0 – 3.0 and 4.0 – 5.0, as scores in these ranges may change by one point in a short time period, but no data provided

(Syndulko et al, 1996)

• Excellent test-retest reliability (ICC = 0.91)

Multiple Sclerosis:

(National MS Society webpage)

• Interrater is good but no references cited

(Amato et al, 1999)

• Reported that AI is more precise and has better interrater reliability as compared to the EDSS, but no data provided

(Sharrack et al, 1999; n = 64 individuals with MS; mean EDSS = 4.5; range of EDSS = 0.0 – 7.5)

• Excellent intrarater reliability (ICC = 0.93; K = 0.59; repeatability coefficient = 1.5 points
• Intrarater agreement = 66, 94, 97 and 100% when agreement was defined as no difference, <1 point, <2 points and <3 points
• Excellent interrater reliability (ICC = 0.96; K = 0.73; repeatability coefficient = 1 point)
• Interrater agreement = 77 and 100% when agreement was defined as no difference and <1 point

Multiple Sclerosis:

(Cattaneo et al, 2006; n = 63 individuals with MS; able to stand independently for >3 seconds and walk 6 meters with or without an assistive device)

• Excellent correlation between AI and Berg’s Balance Scale (rho = 0.74)
• Excellent correlation between AI and Dynamic Gait Index (rho = 0.80)
• Excellent concurrent correlation between AI and Timed Up and Go (rho = 0.74)
• Adequate correlation between AI and Activities-specific Balance Confidence Scale (rho = 0.45)
• Adequate correlation between AI and Dizziness Handicap Inventory (rho = 0.32)

(Provinciali et al, 1999)

• AI is unable to predict handicap as measured by London Handicap Scale and quality of life impairment as measured by Functional Assessment of MS

(Sharrack et al, 1999; mean EDSS = 4.5; range of EDSS = 0.0 – 7.5; AI correlated significantly with EDSS = 0.68; all p < 0.001 – 0.008)

• Excellent correlation between AI and the Scripps Neurological Rating Scale (r = 0.67)
• Excellent correlation with Functional Independence Measure (r = 0.73)
• Adequate correlation with Cambridge MS Basic Score disability (r = 0.54) and handicap (r = 0.55)
• Excellent correlation with Barthel Index and London Handicap Scale (r = 0.72)
• Excellent correlation with the EuroQoLVAS (r = 0.73)
• Excellent correlation with the SF-36 physical functioning (r = 0.87)
• Adequate correlation with the physical role limitation (r = 0.52), social functioning (r = 0.52), vitality (r = 0.39) and general health perception (r = 0.38)
• Did not correlate significantly to SF-36-emotional role limitation, and social functioning (mental health bodily pain and health change)
• Adequate correlation to patients’ ability to work (r = 0.59); do housework (r = 0.55
• Excellent correlation with the disability rank (r = 0.88) at p < 0.001
• Adequate correlation for patients to look after themselves (r = 0.35) at p < 0.01 – 0.02

(Vaney et al, 1996; mean EDSS = 6.6(1.7) for inpatients with MS compared to groups of MS subjects with various walking capabilities of normal, slow and unable)

• Adequate correlation of the Hauser’s AI for inpatients with MS to the Rivermead Mobility Index with multiple groups of MS (rho = 0.45, p < 0.01 for the normal walk group; rho = 0.96, p = 0.001 for all groups

Multiple Sclerosis:

(Cattaneo et al, 2006; n = 63 individuals with MS and able to stand independently for > 3 seconds and walk 6 meters with or without as assistive device)

• AI unable to discriminate between non-fallers and fallers

(Schwartz et al, 1999; individuals with MS)

• Excellent correlation for EDSS levels of 1 – 2.5 (mean = 0.8 (0.7))
• Excellent correlation for EDSS levels of 3.0 – 6.0 (mean = 3.1 (1.3))
• Excellent correlation for EDSS levels above 6.0 (mean = 7.0 (1.5))

(Vaney et al, 1996; discriminate among inpatients with MS who have normal walking capability vs. slow walking vs. unable to walk)

• Mean AI scores for 3 groups = 2.2(0.9), 5.1(1.0), and 8.5( 0.8), p < 0.001

Multiple Sclerosis:

(Cattaneo et al, 2006; n = 63 individuals with MS and able to stand independently for > 3 seconds and walk 6 meters with or without as assistive device)

• Adequate ceiling effect with 7.8% of subjects reaching maximum AI score

(Vaney et al, 1996; subjects with MS had mean EDSS = 6.6( 1.7))

• Poor ceiling effect for 28% of subjects reaching maximum score of 9 on the AI
• Excellent floor effect with 0% of subjects reaching minimum score

Multiple Sclerosis:

(Sharrack et al, 1999; individuals with MS; mean EDSS = 4.5, range = 0.0 – 7.5, p = 0.039)

• Moderate effect size in individuals with MS indicating limited responsiveness to change

(Syndulko et al, 1996)

• Using a signal-to-noise ratio, determined that the AI has responsiveness values (R1 = 2.37) for all patients, 2.65 for patients with EDSS < 5.5, and 2.14 for patients with EDSS > 5.5; indicating better sensitivity to change as compared to the EDSS and two components of the Incapacity Status Scale composites, but not as responsive as neuro-performance composites (global and lower and upper extremity)

(Vaney et al, 1996)

• The AI is reported to be more able to detect change as compared to 10 meter walk test and EDSS but less responsive than the Rivermead Mobiliity Index; the AI was able to detect changes in 18.5% of patients with MS (RMI was able to detect changes in 39%, 10 meter walk test 16.5% and EDSS 7.5%