Table 1

Results of moderation analyses for categorical moderators

Moderator	Summary effect						n_study	Test of moderation
Moderator	g	LL	UL	t	df	P value	n_study	Statistic	df	P value	I²
Treatment level
Type of intervention								HTZ=1.86	13.8	0.183	69.64
Aerobic training	0.35	0.02	0.69	2.35	10.06	0.041	10
Resistance training	–	–	–	–	<4	–	1
Multicomponent	0.16	−0.08	0.39	1.46	12.37	0.169	10
Alternative forms	0.02	−0.08	0.13	0.58	4.99	0.585	4
Intervention length								HTZ=0.44	11.7	0.657	69.90
Short	0.14	−0.03	0.31	1.72	13.33	0.108	10
Medium	0.21	−0.11	0.54	1.44	11.20	0.177	10
Long	–	–	–	–	<4	–	5
Exercise training intensity								HTZ=0.74	7.12	0.511	69.47
Moderate	0.13	−0.24	0.50	0.93	4.50	0.399	5
High	–	–	–	–	<4	–	3
Mixed	0.24	−0.02	0.51	2.16	7.83	0.064	6
Unspecified	0.10	−0.12	0.32	1.00	12.20	0.333	11
Study level
Type of dementia								HTZ=3.06	3.63	0.166	69.46
Alzheimer’s disease	0.56	0.03	1.09	2.49	7.35	0.040	7
Vascular dementia	–	–	–	–	<4	–	2
Multiple types	0.10	−0.08	0.27	1.22	11.37	0.248	10
Unspecified	0.09	−0.09	0.27	1.12	7.95	0.294	6
Overall risk of bias								HTZ=0.69	18.1	0.718	70.68
Low	0.14	−0.003	0.28	2.32	6.82	0.054	7
Some concerns	0.17	−0.08	0.41	1.45	12.07	0.174	10
High	0.28	−0.07	0.63	1.79	9.59	0.156	8
Comparison group								HTZ=1.85	6.73	0.230	70.34
Usual care	0.16	−0.16	0.48	1.16	7.38	0.281	8
Other activities	0.31	0.06	0.56	2.66	14.00	0.019	11
Physically active	–	–	–	–	<4	–	4
Cognitive training	–	–	–	–	<4	–	2
Individual level
Percentage female								HTZ=0.88	27.8	0.355	65.52
Low proportion	0.22	−0.01	0.45	2.05	13.9	0.060	13
High proportion	0.103	−0.04	0.24	1.56	14.5	0.140	11
Unspecified	–	–	–	–	<4	–	1
Outcome level
Cognitive domain								HTZ=0.69	14.9	0.574	69.31
Global cognition	0.31	0.05	0.57	2.49	16.76	0.024	23
Memory	0.09	−0.08	0.26	1.13	9.56	0.285	15
Executive function	0.10	−0.06	0.26	1.26	18.10	0.223	9
Processing speed	0.14	−0.18	0.47	1.01	8.01	0.344	8

df, Degrees of freedom; LL, Lower limit of the 95% confidence interval; UL, Upper limit of the 95% confidence interval.