Physical activity to improve cognition in older adults: can physical activity programs enriched with cognitive challenges enhance the effects? A systematic review and meta-analysis

Table 2 Moderator analyses of effect sizes for combined PA + CA versus control

Moderator	n	k	Hedges’g (95%CI)	p	Q	p
Mode of combination^a	1620	27			3.699	.054
sequential	367	7	0.114 [−0.102; 0.331]	.301
simultaneous	1253	20	0.385 [0.214; 0.555]	<.001
Cognitive status	1696	29			0.427	.514
healthy	947	20	0.282 [0.148; 0.416]	<.001
impaired	749	9	0.389 [0.095; 0.684]	.009
Intervention length	1696	29			2.235	.327
short (< 12 weeks)	257	8	0.335 [0.092; 0.577]	.007
medium (12–23 weeks)	725	13	0.405 [0.144; 0.666]	.002
long (≥ 24 weeks)	714	8	0.194 [0.038; 0.349]	.015
Session duration^b	1568	27			3.955	.138
short (≤ 45 min)	433	6	0.248 [0.055; 0.441]	.012
medium (> 45 to ≤60 min)	679	13	0.474 [0.231; 0.716]	<.001
long (> 60 min)	456	8	0.159 [−0.041; 0.358]	.12
Session frequency^c	1518	25			4.398	.111
low (1×/week)	236	4	0.379 [0.092; 0.667]	.010
medium (2×/week)	545	9	0.542 [0.217; 0.867]	.001
high (≥3×/week)	737	12	0.190 [0.043; 0.336]	.011
EPHPP study quality	1696	29			2.080	.353
weak	560	9	0.280 [0.106; 0.454]	.002
moderate	537	13	0.454 [0.176; 0.733]	.001
strong	599	7	0.215 [0.046; 0.383]	.013

n = combined sample size; k = number of studies; Hedges’g (random effects); CI confidence interval, Q = homogeneity statistic (mixed effects)
^aTwo studies were excluded because the mode of PA + CA combination had both sequential and simultaneous components [46, 51]
^bTwo studies were excluded because average session duration (particularly, of home-based sessions) was not clearly reported [21, 40]
^cFour studies were excluded because session frequency was not clearly reported [32, 36, 40, 46]

ISSN: 1479-5868