Table 4 Longitudinal mediation models across sex and age.

From: Physical activity partly mediates the association between cognitive function and depressive symptoms

 

Model 1

Cognitive function → (a) physical activity → (b) depressive symptoms

Model 2

Physical activity → (a) cognitive function → (b) depressive symptoms

Men

Women

Men

Women

<65

≥65

<65

≥65

<65

≥65

<65

≥65

n

12,109

11,083

15,335

12,664

12,109

11,083

15,335

12,664

 

Unstandardized B (95% CI)

Unstandardized B (95% CI)

a1

0.007 (−0.002, 0.016)

0.029*** (0.019, 0.039)

0.022*** (0.014, 0.030)

0.057*** (0.047, 0.066)

0.054* (0.003, 0.105)

0.120*** (0.077, 0.165)

0.113*** (0.065, 0.159)

0.176*** (0.138, 0.214)

b1

−0.097*** (−0.143, −0.052)

−0.151*** (−0.195, −0.107)

−0.176*** (−0.222, −0.129)

−0.131*** (−0.172, −0.090)

−0.028** (−0.045, −0.012)

−0.057*** (−0.076, −0.039)

−0.023** (−0.039, −0.007)

−0.067*** (−0.085, −0.050)

c’1 (Direct effect)

−0.030** (−0.049, −0.011)

−0.059*** (−0.079, −0.039)

−0.021* (−0.040, −0.002)

−0.062*** (−0.082, −0.041)

−0.063* (−0.114, −0.011)

−0.102*** (−0.152, −0.051)

−0.183*** (−0.236, −0.131)

−0.120*** (−0.167, −0.073)

Total effect 1

−0.030** (−0.049, −0.011)

−0.064*** (−0.084, −0.043)

−0.024* (−0.043, −0.006)

−0.069*** (−0.089, −0.049)

−0.064* (−0.115, −0.013)

−0.108*** (−0.158, −0.058)

−0.186*** (−0.238, −0.133)

−0.132*** (−0.179, −0.085)

Indirect effect 1

−0.001 (−0.002, 0.000)

−0.004*** (−0.007, −0.003)

−0.004*** (−0.006, −0.002)

−0.007*** (−0.010, −0.005)

−0.002 (−0.004, 0.000)

−0.007*** (−0.011, −0.004)

−0.003* (−0.005, −0.001)

−0.012*** (−0.016, −0.008)

Proportion of mediation effect 1

3%

6%

17%

10%

3%

6%

2%

9%

a2

0.013** (0.016, 0.050)

0.045*** (0.035, 0.056)

0.030*** (0.023, 0.037)

0.055*** (0.045, 0.064)

0.067** (0.020, 0.113)

0.097*** (0.057, 0.136)

0.129*** (0.089, 0.170)

0.162*** (0.127, 0.198)

b2

−0.083 (−0.169, 0.004)

−0.176*** (−0.255, −0.098)

−0.104* (−0.197, −0.011)

−0.164*** (−0.237, −0.093)

−0.011 (−0.045, 0.023)

−0.107*** (−0.141, −0.073)

−0.021 (−0.053, 0.012)

−0.074*** (−0.106, −0.041)

c’2 (Direct effect)

0.017 (−0.021, 0.054)

−0.072*** (−0.109, −0.035)

−0.032 (−0.068, 0.004)

−0.078*** (−0.114, −0.041)

−0.053 (−0.162, 0.057)

−0.096* (−0.188, −0.007)

−0.070 (−0.175, 0.032)

−0.163*** (−0.249, −0.079)

Total effect 2

0.016 (−0.022, 0.053)

−0.080*** (−0.117, −0.043)

−0.035 (−0.070, 0.001)

−0.087*** (−0.123, −0.050)

−0.054 (−0.162, 0.056)

−0.106* (−0.198, −0.017)

−0.072 (−0.177, 0.029)

−0.175*** (−0.261, −0.091)

Indirect effect 2

−0.001 (−0.003, 0.000)

−0.008*** (−0.012, −0.004)

−0.003* (−0.006, 0.000)

−0.009*** (−0.013, −0.005)

−0.001 (−0.004, 0.002)

−0.010*** (−0.016, −0.005)

−0.003 (−0.007, 0.002)

−0.012*** (−0.018, −0.006)

Proportion of mediation effect 2

NA

10%

9%

10%

2%

9%

4%

7%

AIC

339384.279

330806.444

452378.299

405070.761

339400.979

330891.292

452360.175

405187.988

X2 (df)

1779.333 (37)***

2391.406 (37)***

3074.899 (37)***

3234.749 (37)***

1796.033 (37)***

2476.254 (37)***

3056.775 (37)***

3351.976 (37)***

RMSEA (95% CI)

0.062 (0.060–0.065)

0.076 (0.073–0.078)

0.073 (0.071–0.075)

0.083 (0.080–0.085)

0.063 (0.060–0.065)

0.077 (0.075–0.080)

0.073 (0.071–0.075)

0.084 (0.082–0.087)

CFI

0.880

0.867

0.859

0.870

0.879

0.863

0.860

0.865

TLI

0.796

0.774

0.759

0.779

0.795

0.766

0.761

0.770

SRMR

0.065

0.086

0.074

0.089

0.066

0.092

0.075

0.096

  1. The analysis was performed using cross-lagged panel longitudinal mediation model. NA not applicable, the proportion of mediating effect is not calculated, because the signs of the indirect and total effects are different. The “1” describes the associations between wave 4 and wave 6, while the “2” describes the associations between wave 5 and wave 7, see Fig. 1.
  2. AIC Akaike Information Criterion, CFI comparative fit index, CI confidence interval, RMSEA root-mean square error of approximation, SRMR standardized root-mean-squared residual, TLI Tucker-Lewis index.
  3. *p < 0.05. **p < 0.01. ***p < 0.001.