Preliminary assessment of balance with the Berg Balance Scale in adults who have a leg amputation and dwell in the community: Rasch rating scale analysis.

Background: Self-report measures of balance and multidimensional mobility assessments are common for people with a leg amputation, yet clinical assessment of balance ability remains less explored. The Berg Balance Scale (BBS), typically used for other populations with impaired balance, has been used for young people with a high level of functioning after traumatic amputation but rarely for older people after vascular amputation.

Objective: The study objective was to examine the psychometric properties of the BBS with Rasch rating scale analysis to determine the validity and utility of the BBS in assessing balance ability in adults who have a leg amputation and dwell in the community.

Design: Rating scale analysis was applied to BBS scores obtained from a single assessment.

Methods: Adult volunteers (men and women) who had a leg amputation (any level and etiology) and dwelled in the community were recruited from a hospital-based community support group and a prosthetic clinic. Rating scale analysis of the BBS was used to assess unidimensionality, internal validity, goodness of fit, structural integrity, and person and item analyses.

Results: The study participants were 40 people (26 men and 14 women; 57.8 [SD=9.7] years old) with leg amputations (24 transtibial, 13 transfemoral, and 3 bilateral) of mixed etiology (32 vascular and 8 nonvascular). The psychometric properties of the BBS confirmed that it measures the unidimensional construct of balance ability with adequate validity and with goodness of fit and structural integrity that meet the acceptability criteria. Person measures revealed that some participants scored near the top of the BBS, suggesting a ceiling effect; item measures revealed that participants with leg amputations had the most difficulty performing the following tasks: standing with 1 leg in front, turning 360 degrees, and placing alternate foot on a stool.

Limitations: Limitations included a convenience sample and a lack of rater reliability testing.

Conclusions: The BBS cohered with the unidimensional construct of balance ability and had strong internal validity for use in a variety of people with leg amputations.