Skip to main content

Table 3 Estimated 10-year cognitive decline in each gait speed and handgrip strength quintile by linear mixed effects models

From: Role of gait speed and grip strength in predicting 10-year cognitive decline among community-dwelling older people

Model terms

DSST

MMSE

Estimate (SE)

P-value

Estimate (SE)

P-value

【Gait speed】

 Intercept

40.88 (0.62)

< 0.001

27.96 (0.10)

< 0.001

 Gait speed Q1a

Reference

   

 Gait speed Q2

2.41 (0.86)

0.005

0.16 (0.14)

0.261

 Gait speed Q3

2.21 (0.85)

0.010

0.14 (0.14)

0.305

 Gait speed Q4

3.48 (0.91)

< 0.001

0.08 (0.15)

0.598

 Gait speed Q5

3.96 (0.94)

< 0.001

0.25 (0.15)

0.097

 Timeb

−0.05 (0.09)

0.606

−0.16 (0.02)

< 0.001

 Time-squaredb

−0.05 (0.01)

< 0.001

–

 

 Gait speed Q1 x time

Reference

   

 Gait speed Q2 x time

0.14 (0.09)

0.140

0.02 (0.03)

0.421

 Gait speed Q3 x time

0.15 (0.09)

0.101

0.05 (0.03)

0.062

 Gait speed Q4 x time

0.13 (0.09)

0.146

0.02 (0.03)

0.590

 Gait speed Q5 x time

0.28 (0.09)

0.003

0.05 (0.03)

0.078

【Handgrip Strength】

 Intercept

41.57 (0.63)

< 0.001

27.93 (0.10)

< 0.001

 Handgrip strength Q1c

Reference

   

 Handgrip strength Q2

0.49 (0.86)

0.567

0.07 (0.14)

0.614

 Handgrip strength Q3

1.64 (0.88)

0.063

0.07 (0.14)

0.628

 Handgrip strength Q4

2.15 (0.90)

0.017

0.38 (0.14)

0.008

 Handgrip strength Q5

3.42 (0.95)

< 0.001

0.23 (0.15)

0.127

 Timeb

−0.04 (0.09)

0.710

−0.17 (0.02)

< 0.001

 Time-squaredb

− 0.05 (0.01)

< 0.001

–

 

 Handgrip strength Q1 x time

Reference

   

 Handgrip strength Q2 x time

0.07 (0.10)

0.439

0.04 (0.03)

0.189

 Handgrip strength Q3 x time

0.11 (0.09)

0.259

0.05 (0.03)

0.060

 Handgrip strength Q4 x time

0.20 (0.09)

0.033

0.05 (0.03)

0.082

 Handgrip strength Q5 x time

0.20 (0.10)

0.040

0.06 (0.03)

0.039

  1. All models were adjusted by covariates (age, gender, education years, marital status, smoking status, body mass index, Tokyo Metropolitan Institute of Gerontology Index of Competence, depressive symptoms and Charlson Comorbidity Index) and their interactions with time. The covariates were grand-mean centered. However, these estimates were not presented, and only the estimates of interest (intercept, time, and their interactions with physical function) are shown
  2. In males: Q1 as the lowest quintile and Q5 as the highest quintile; ≦1.12; 1.13–1.25; 1.26–1.37; 1.38–1.47; and > 1.48, m/s
  3. In females: Q1 as the lowest quintile and Q5 as the highest quintile; ≦1.13; 1.14–1.22; 1.23–1.32; 1.33–1.42; and > 1.43, m/s
  4. In males: Q1 as the lowest quintile and Q5 as the highest quintile: ≦30.70; 30.71–33.90; 33.91–37.40; 37.41–41.25; and > 41.25 kg
  5. In females: Q1 as the lowest quintile and Q5 as the highest quintile: ≦18.50; 18.51–20.90; 20.91–23.10; 23.11–25.70; and > 25.70 kg
  6. SE standard error, MMSE Mini-Mental State Examination, DSST Digit Symbol Substitution Test
  7. Among a sample of 1096 adults, those with missing values in physical and cognitive functions and covariates at baseline were excluded from the analyses. Thus, the sample sizes varied according to the models: in the gait speed models, N = 1006 for DSST, and N = 1007 for MMSE; in the grip strength models, N = 1022 for DSST, and N = 1024 for MMSE
  8. aGait speed in each group
  9. bYears since baseline. Time-squared was included in modeling non-linear change of DSST
  10. cHandgrip strength in each group
Back to article page

BMC Geriatrics

ISSN: 1471-2318

Contact us