Reference Values Derived From The 90+ Study: The Five Times Sit to Stand Test.

Background and purpose: Physical performance reference values play an important role in older adult care, yet data are extremely limited in individuals 90 years and older, the "oldest old." The Five Times Sit to Stand Test (5XSST) is a frequently used method of quantifying functional lower extremity strength. To improve the classification and interpretation of 5XSST scores, we aim to develop 5XSST reference values in individuals 90+ years.

Methods: Participants are members of The 90+ Study, a longitudinal oldest-old cohort study. The current study is cross-sectional, using data from the first visit on which each participant completed the 5XSST. Participants performed the 5XSST with arms folded (traditional test) or using their arms to push from the chair (modified test). We calculated means, standard deviations, and percentiles (5th, 10th, 25th, 50th, 75th, 90th, 95th) by test type and age category. We used linear regression to compare mean scores by sex (men, women), age category (90-91, 92-94, 95+), test type (traditional, modified), living situation (home alone, home with another person, nursing home/assisted living), and falls in the past year (0, 1+).

Results and discussion: The 972 participants had a mean age of 93.0 years (range = 90.0-103.1, SD = 2.5). Of these, 64.8% performed the traditional test and 36.2% the modified test. Mean time for the traditional test was 16.2 seconds (SD = 6.3) and, for the modified test, 22.6 seconds (SD = 9.9). Scores were significantly slower in participants in the oldest age category, who performed the modified test, lived in a facility, or fell in the past year. No significant differences were found according to sex. We present 5XSST reference values in men and women by test type and age category. Previous studies in younger groups have reported faster 5XSST times than those from our 90+ cohort, which suggests using reference values established in younger groups to categorize the performance of oldest-old individuals is not optimal for accurate categorization of scores.

Conclusions: The reference values we present will allow providers to correctly classify and interpret 5XSST scores in the rapidly growing group of individuals 90 years and older.