Skip to main content
. 2014 Jul 18;13(5):918–925. doi: 10.1111/acel.12251

Table 2.

LP/Lean, short physical performance battery (SPPB), and 400-m backward elimination (BE) regression models at baseline

β ± SE P-value
LP/Lean: Adjusted R2 = 0.22, P = 0.0009
 Sex 4.2 ± 1.4 0.003
 FA Factor 9: 2-hydroxypalmitate, 2-hydroxystearate 1.6 ± 0.6 0.01
 AA Factor 6: Hydrocinnamate, cinnamoylglycine −1.4 ± 0.6 0.03
 Age −0.4 ± 0.2 0.05
 Dipeptide Factor 3: γ-glutamylamino acids −1.3 ± 0.6 0.05
 AA Factor 19: N-acetylglycine 1.2 ± 0.6 0.07
SPPB: Adjusted R2 = 0.15, P = 0.001
 AA Factor 6: Hydrocinnamate, cinnamoylglycine −0.4 ± 0.2 0.008
 AA Factor 7: N-methylproline, N,N-dimethylproline −0.4 ± 0.2 0.009
400-m: Adjusted R2 = 0.41, P < 0.0001
 Age −0.0 ± 0.0 0.0005
 AA Factor 19: N-acetylglycine 0.1 ± 0.0 0.0006
 AC Factor 4: Isobutyrylcarnitine −0.1 ± 0.0 0.004
 AA Factor 4: BCAA degradation products, indolelactate 0.0 ± 0.0 0.005
 AA Factor 12: Phenol Sulfate −0.1 ± 0.0 0.009
 Sex 0.1 ± 0.0 0.01
 AA Factor 5: BCAA −0.0 ± 0.0 0.02
 AA Factor 13: Trans-urocanate −0.0 ± 0.0 0.08

BCAA, branched-chain amino acids; LP/Lean, leg press one repetition maximum divided by total lean mass.

Sex, age, total fat (LP/Lean model) or BMI (SPPB, 400-m models), and principal components analysis factors that were associated (P ≤ 0.10 and q ≤ 0.30) with measures of physical function were considered as candidate variables for BE linear regression. Covariates significantly associated (P ≤ 0.10) with LP/Lean, SPPB, and 400-m are shown with parameter estimates and standard errors (β ± SE), in order of significance (P-value).