Cogstate for grant submission

NEW analyses 12/02

CogState scores by group: Linear models (2 groups w/premorbid IQ & remote/non-remote)

2 groups: Non-PTSD (low-exposed controls & resilient); PTSD

Reference group = Non-PTSD

NO participants in the PTSD group completed the CogState remotely so cannot look at interaction

We see effects of PTSD vs non-PTSD on GML, GMR, ISRL (but only when standarized IQ is in model)

Groton maze learning test (Executive function)

Call:
lm(formula = score ~ group_2grp * wtar_standard + cogState_remote, 
    data = gml)

Residuals:
    Min      1Q  Median      3Q     Max 
-23.188 -11.051  -2.282   8.155  44.707 

Coefficients:
                             Estimate Std. Error t value Pr(>|t|)   
(Intercept)                  104.9436    32.0552   3.274  0.00213 **
group_2grpPTSD               -95.1573    52.6810  -1.806  0.07804 . 
wtar_standard                 -0.4688     0.2865  -1.637  0.10921   
cogState_remote1               5.8503     6.1573   0.950  0.34748   
group_2grpPTSD:wtar_standard   0.9285     0.4962   1.871  0.06828 . 
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 15.15 on 42 degrees of freedom
  (6 observations deleted due to missingness)
Multiple R-squared:  0.112, Adjusted R-squared:  0.02746 
F-statistic: 1.325 on 4 and 42 DF,  p-value: 0.2766

`geom_smooth()` using formula 'y ~ x'

(In the above plot, TRUE = standardized WTAR is above the sample mean)

Groton maze learning test - delayed recall (Memory)

Call:
lm(formula = score ~ group_2grp * wtar_standard + cogState_remote, 
    data = gmr)

Residuals:
    Min      1Q  Median      3Q     Max 
-5.3581 -2.9329  0.1011  2.2804  7.9636 

Coefficients:
                              Estimate Std. Error t value Pr(>|t|)   
(Intercept)                   25.78259    7.56086   3.410   0.0015 **
group_2grpPTSD               -29.20027   12.25624  -2.382   0.0220 * 
wtar_standard                 -0.15747    0.06758  -2.330   0.0249 * 
cogState_remote1              -2.82138    1.59295  -1.771   0.0842 . 
group_2grpPTSD:wtar_standard   0.26307    0.11534   2.281   0.0280 * 
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 3.496 on 40 degrees of freedom
  (8 observations deleted due to missingness)
Multiple R-squared:  0.2076,    Adjusted R-squared:  0.1284 
F-statistic: 2.621 on 4 and 40 DF,  p-value: 0.04903

`geom_smooth()` using formula 'y ~ x'

(In the above plot, TRUE = standardized WTAR is above the sample mean)

International shopping list - delayed recall (Verbal memory)

Call:
lm(formula = score ~ group_2grp * wtar_standard + cogState_remote, 
    data = isrl)

Residuals:
    Min      1Q  Median      3Q     Max 
-4.2657 -0.9958  0.2083  1.0347  2.8018 

Coefficients:
                              Estimate Std. Error t value Pr(>|t|)    
(Intercept)                   13.21375    3.39601   3.891 0.000369 ***
group_2grpPTSD               -12.91514    5.52466  -2.338 0.024491 *  
wtar_standard                 -0.03374    0.03038  -1.111 0.273267    
cogState_remote1               0.42271    0.67874   0.623 0.536960    
group_2grpPTSD:wtar_standard   0.12402    0.05202   2.384 0.021945 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.579 on 40 degrees of freedom
  (8 observations deleted due to missingness)
Multiple R-squared:  0.1333,    Adjusted R-squared:  0.04664 
F-statistic: 1.538 on 4 and 40 DF,  p-value: 0.2096

`geom_smooth()` using formula 'y ~ x'

(In the above plot, TRUE = standardized WTAR is above the sample mean)

 

 

 

 

 

 

 

  # 11/30/21

Number of exposures and CogState

New analyses 11/28/21

NEW: Premorbid IQ (WTAR)

3 groups

2 groups

NEW: CogState scores by group: Linear models (2 groups w/premorbid IQ & REMOTE)

2 groups: Non-PTSD (low-exposed controls & resilient); PTSD

Reference group = Non-PTSD

NO participants in the PTSD group completed the CogState remotely so cannot look at interaction

Detection test (Psychomotor function)

(not administered remotely)

Groton maze learning test (Executive function)

Groton maze learning - delayed recall test (Memory)

Identification test (Attention)

(not administered remotely)

International shopping list test (Verbal learning)

International shopping list - delayed recall test (Memory)

One-card learning test (Visual memory)

One-back test (Working memory)

(not administered remotely)

Composite score

Previous

CogState Mean/SD by group

3 groups

2 groups

Non-PTSD = Low-exposed controls & Resilient

Plots for CogState scores by group

Plot all scores on same plot: 3 groups

Plot all scores on same plot: 2 groups (including Low-exposed)

Plot all scores on same plot: 2 groups (NOT including Low-exposed)

Detection

Groton maze learning

Groton maze learning - delayed recall

Identification

International shopping list

International shopping list - delayed recall

One-card learning

One-back test

Composite score

CogState scores by group: Linear models (3 groups)

3 groups: Low-exposed controls; Resilient; PTSD Reference group = PTSD

Detection test (Psychomotor function)

Groton maze learning test (Executive function)

Groton maze learning - delayed recall test (Memory)

Identification test (Attention)

International shopping list test (Verbal learning)

International shopping list - delayed recall test (Memory)

One-card learning test (Visual memory)

One-back test (Working memory)

Composite score

CogState scores by group: T-tests (2 groups)

2 groups: Non-PTSD (i.e, Low-exposed controls & Resilient); PTSD

Detection test (Psychomotor function)

Groton maze learning test (Executive function)

Groton maze learning - delayed recall test (Memory)

Identification test (Attention)

International shopping list test (Verbal learning)

International shopping list - delayed recall test (Memory)

One-card learning test (Visual memory)

One-back test (Working memory)

Composite score

MOCA

MOCA score (3 groups)

MOCA score (2 groups)

Non-PTSD = resilient & low-exposed controls.