Does thoracic kyphosis severity predict response to physiotherapy rehabilitation in patients with osteoporotic vertebral fracture? A secondary analysis of the PROVE RCT

To estimate the magnitude of small meaningful and substantial individual change in physical performance measures and evaluate their responsiveness. Secondary data analyses using distribution- and anchor-based methods to determine meaningful change. Secondary analysis of data from an observational study and clinical trials of community-dwelling older people and subacute stroke survivors. Older adults with mobility disabilities in a strength training trial (n=100), subacute stroke survivors in an intervention trial (n=100), and a prospective cohort of community-dwelling older people (n=492). Gait speed, Short Physical Performance Battery (SPPB), 6-minute-walk distance (6MWD), and self-reported mobility. Most small meaningful change estimates ranged from 0.04 to 0.06 m/s for gait speed, 0.27 to 0.55 points for SPPB, and 19 to 22 m for 6MWD. Most substantial change estimates ranged from 0.08 to 0.14 m/s for gait speed, 0.99 to 1.34 points for SPPB, and 47 to 49 m for 6MWD. Based on responsiveness indices, per-group sample sizes for clinical trials ranged from 13 to 42 for substantial change and 71 to 161 for small meaningful change. Best initial estimates of small meaningful change are near 0.05 m/s for gait speed, 0.5 points for SPPB, and 20 m for 6MWD and of substantial change are near 0.10 m/s for gait speed, 1.0 point for SPPB, and 50 m for 6MWD. For clinical use, substantial change in these measures and small change in gait speed and 6MWD, but not SPPB, are detectable. For research use, these measures yield feasible sample sizes for detecting meaningful change.

Background The Short Physical Performance Battery (SPPB) is a well-established tool to assess lower extremity physical performance status. Its predictive ability for all-cause mortality has been sparsely reported, but with conflicting results in different subsets of participants. The aim of this study was to perform a meta-analysis investigating the relationship between SPPB score and all-cause mortality. Methods Articles were searched in MEDLINE, the Cochrane Library, Google Scholar, and BioMed Central between July and September 2015 and updated in January 2016. Inclusion criteria were observational studies; >50 participants; stratification of population according to SPPB value; data on all-cause mortality; English language publications. Twenty-four articles were selected from available evidence. Data of interest (i.e., clinical characteristics, information after stratification of the sample into four SPPB groups [0–3, 4–6, 7–9, 10–12]) were retrieved from the articles and/or obtained by the study authors. The odds ratio (OR) and/or hazard ratio (HR) was obtained for all-cause mortality according to SPPB category (with SPPB scores 10–12 considered as reference) with adjustment for age, sex, and body mass index. Results Standardized data were obtained for 17 studies (n = 16,534, mean age 76 ± 3 years). As compared to SPPB scores 10–12, values of 0–3 (OR 3.25, 95%CI 2.86–3.79), 4–6 (OR 2.14, 95%CI 1.92–2.39), and 7–9 (OR 1.50, 95%CI 1.32–1.71) were each associated with an increased risk of all-cause mortality. The association between poor performance on SPPB and all-cause mortality remained highly consistent independent of follow-up length, subsets of participants, geographic area, and age of the population. Random effects meta-regression showed that OR for all-cause mortality with SPPB values 7–9 was higher in the younger population, diabetics, and men. Conclusions An SPPB score lower than 10 is predictive of all-cause mortality. The systematic implementation of the SPPB in clinical practice settings may provide useful prognostic information about the risk of all-cause mortality. Moreover, the SPPB could be used as a surrogate endpoint of all-cause mortality in trials needing to quantify benefit and health improvements of specific treatments or rehabilitation programs. The study protocol was published on PROSPERO (CRD42015024916). Electronic supplementary material The online version of this article (doi:10.1186/s12916-016-0763-7) contains supplementary material, which is available to authorized users.

The interpretation of patient scores on clinical tests of physical mobility is limited by a lack of data describing the range of performance among people without disabilities. The purpose of this study was to provide data for 4 common clinical tests in a sample of community-dwelling older adults. Ninety-six community-dwelling elderly people (61-89 years of age) with independent functioning performed 4 clinical tests. Data were collected on the Six-Minute Walk Test (6MW), Berg Balance Scale (BBS), and Timed Up & Go Test (TUG) and during comfortable- and fast-speed walking (CGS and FGS). Intraclass correlation coefficients (ICCs) were used to determine the test-retest reliability for the 6MW, TUG, CGS, and FGS measurements. Data were analyzed by gender and age (60-69, 70-79, and 80-89 years) cohorts, similar to previous studies. Means, standard deviations, and 95% confidence intervals for each measurement were calculated for each cohort. The 6MW, TUG, CGS, and FGS measurements showed high test-retest reliability (ICC [2,1]=.95-.97). Mean test scores showed a trend of age-related declines for the 6MW, BBS, TUG, CGS, and FGS for both male and female subjects. Preliminary descriptive data suggest that physical therapists should use age-related data when interpreting patient data obtained for the 6MW, BBS, TUG, CGS and FGS. Further data on these clinical tests with larger sample sizes are needed to serve as a reference for patient comparisons.