. Author manuscript; available in PMC: 2019 Apr 7.

Published in final edited form as: Clin Psychol Rev. 2018 Apr 7;61:24–37. doi: 10.1016/j.cpr.2018.04.001

Table 3.

Moderator Analyses

Association	Moderator	k	B	SE	95% CI	z	I²
Self-Reported—Clinician-rated	Mean age	33	−.006	.006	[−.018, .006]	−.98	68.70
	Mean illness length	17	−.014	.007	[−.029, .000]^a	−2.01^*	59.26
	Mean CPZ equivalent doses	13	.000	.000	[000, .001]	.19	50.16
	% female	33	.007	.004	[−.002, .016]	1.56	66.86
	% schizophrenia diagnosis	27	−.001	.003	[−.006, .004]	−.27	57.23

AM Self-Reported—AM Clinician-rated	Mean age	23	.005	.009	[−.013, .022]	.50	76.50
	Mean illness length	10	−.008	.016	[−.038, .023]	−.50	66.74
	Mean CPZ equivalent doses	7	.000	.000	[−.001, .000]	−.90	63.06
	% female	23	.007	.005	[−.003, .017]	1.41	70.72
	% schizophrenia diagnosis	17	.000	.003	[−.005, .006]	.15	52.05

Note. k = number of studies reporting the moderator and included in the meta-regression; B = regression coefficient;

SE = standard error; 95% CI = 95% confidence interval for regression coefficient; z = z-test value for statistical significance of regression coefficient; I² = indicates the extent of between-study variability that is unexplained after accounting for the moderator; CPZ = chlorpromazine equivalent doses; AM = Amotivation.

Full CI: [−.0286, −.0003]

p< .05