meta <- metacont(N.MCI_St, Mean.MCI_St, SD.MCI_St, N.MCI_AD, Mean.MCI_AD, SD.MCI_AD,

sm=“SMD”, data=data, studlab=paste(Author, Year))

## Warning in metacont(N.MCI_St,

Mean.MCI_St, SD.MCI_St, N.MCI_AD, Mean.MCI_AD, :

## Note, studies with non-positive values for n.e and / or n.c get no weight in

## meta-analysis.

meta$label.e <- “MCI_St”

meta$label.c <- “MCI_AD”

print(meta, digits=2)

## SMD 95%-CI %W(fixed) %W(random) ## Bjerke 2009 -0.74 [-1.27; -0.21] 2.8 4.8

## Mattsson 2009 -0.95 [-1.11; -0.79] 30.3 6.6

## Hertze 2010 -1.09 [-1.46; -0.72] 5.7 5.6

## Palmqvist 2012 -1.08 [-1.48; -0.69] 5.0 5.5

## Spencer 2019 -0.32 [-0.62; -0.03] 9.0 6.0

## Hansson 2006 -1.40 [-1.81; -0.99] 4.6 5.4

## Hansson 2007 NA 0.0 0.0

## Hampel 2004 -0.29 [-0.84; 0.26] 2.6 4.7

## Herukka 2007 -1.19 [-1.67; -0.70] 3.3 5.0

## Lanari 2009 -0.62 [-1.21; -0.02] 2.2 4.4

## Papaliagkas 2009 NA 0.0 0.0

## Eckerstrom 2010 -1.14 [-1.89; -0.39] 1.4 3.7

## Kester 2011 -1.12 [-1.54; -0.69] 4.3 5.3

## Seppala 2011 -1.19 [-1.82; -0.56] 2.0 4.2

## Buchhave 2012 -1.22 [-1.64; -0.81] 4.5 5.4

## Parnetti 2012 -2.99 [-3.61; -2.37] 2.0 4.3

## Prestia 2013 NA 0.0 0.0

## Leuzy 2015 -0.40 [-1.12; 0.31] 1.5 3.8

## Baldeiras 2018 -1.07 [-1.42; -0.72] 6.4 5.8

## Khoonsari 2019 -1.85 [-2.44; -1.26] 2.2 4.4

## Santangelo 2020 -0.64 [-1.10; -0.18] 3.7 5.2

## Eckerstrom 2020 -0.64 [-1.10; -0.18] 3.7 5.2

## Brys 2009 -1.00 [-1.55; -0.46] 2.7 4.7

##

## Number of studies combined: k = 20

## Number of observations: o = 2344

##

## SMD 95%-CI z p-value

## Fixed effect model -0.97 [-1.06; -0.89] -21.56 < 0.0001

## Random effects model -1.03 [-1.24; -0.82] -9.66 < 0.0001

##

## Quantifying heterogeneity:

## tau^2 = 0.1665; tau = 0.4081; I^2 = 79.0% [68.2%; 86.1%]; H = 2.18 [1.77; 2.69]

##

## Test of heterogeneity:

## Q d.f. p-value

## 90.45 19 < 0.0001

##

## Details on meta-analytical method:

## - Inverse variance method

## - DerSimonian-Laird estimator for tau^2

## - Hedges’ g (bias corrected standardised mean difference)