Table 3.

Subgroup analysis of the association between BCG administration and incident dementia.

Subgroups		Studies (n)	Hazard ratio (95% CI)	p-value(overaleffect)	Heterogeneity		Effects model	p-value(meta − regression)
					I 2	p
Region	Israel	3	0.36 (0.13–0.96)	0.041	91.7%	< 0.001	Random	0.337
	USA	3	0.73 (0.63–0.85)	<0.001	47.3%	0.150	Random
Age	<75 years	2	0.80 (0.61–1.04)	0.114	51.5%	0.151	Random	0.289
	≥75 years	2	0.68 (0.62–0.75)	<0.001	0.0%	0.904	Fixed
Gender	Male	2	0.58 (0.25–1.39)	0.223	74.7%	0.047	Random	0.955
	Female	2	0.62 (0.36–1.08)	0.092	0.0%	0.706	Fixed
Race/ethnicity	Non-Hispanic black	2	1.67 (0.83–3.36)	0.151	85.7%	0.008	Random	0.112
	Hispanic white	2	1.13 (0.92–1.39)	0.244	25.9%	0.245	Fixed
	Other	2	0.84 (0.68–1.04)	0.112	0.0%	0.567	Fixed
Dosage	≤6 doses	2	0.84 (0.75–0.94)	0.002	43.7%	0.182	Fixed	0.461
	>6 doses	3	0.61 (0.46–0.82)	0.001	80.2%	0.006	Random
Follow-up period	≤5 years	3	0.46 (0.23–0.89)	0.022	91.0%	<0.001	Random	0.463
	>5 years	3	0.55 (0.32–0.94)	0.03	80.3%	0.006	Random
Outcome	AD	4	0.47 (0.30–0.75)	0.001	89.3%	<0.001	Random	0.607
	Dementia	2	0.62 (0.33–1.17)	0.142	72.0%	0.059	Random
Total		6	0.55 (0.42–0.73)	<0.001	84.5%	<0.001	Random