Article

The objective of this study was to develop a prognostic clinical
prediction rule to identify people not achieving community walking level prosthetic use after 1 yr. DESIGN: This is a prospective longitudinal cohort
study of community-dwelling adults with lower-limb amputations recruited from
support groups and prosthetic clinics. Participants completed Activities-specific
Balance Confidence and Houghton prosthetic use for mobility self-report scales
and the Berg Balance Scale.
The clinical prediction rule was developed using
multivariate logistic regression, receiver operating curves, and probability
statistics to identify people not achieving community walking level prosthetic
use (Houghton scores <9) at 1 yr. RESULTS: Forty (74.1%) of 54 participants
provided follow-up data. Participants averaged 57.0 +/- 11.9 yrs old, and the
most recent amputation had occurred an average of 6.6 +/- 11.0 yrs ago. Seventy
percent had vascular amputations and 52.5% had transtibial amputations. The
clinical prediction rule predicted who would not reach the community prosthetic
walking level with excellent accuracy (area under the curve >0.96) using four
criteria: initial Houghton, Activities-specific Balance Confidence, and Berg
Balance Scale tasks 9 (retrieve object from floor) and 10 (look behind over
shoulders). Failure to exceed cutoff scores in two or more criteria yielded
posttest probability of not reaching community walking prosthetic use 1 yr later
for 90% of participants or higher. CONCLUSIONS: Accurate 1-yr prosthetic use for
mobility prognoses can be obtained by screening prosthetic use, balance
confidence, and balance ability to identify community-dwelling people with
lower-limb amputations unlikely to achieve community walking prosthetic use.

Langue : ANGLAIS