Table 12.

Multiple meta-regression analysis, predicting between-study variance in the level of differential attrition using the level of similarity between active and sham control groups for a number of selected features.

Descriptives
Mean (SD) similarity ratings	n of sessions	Standardisation	Fidelity monitoring	Individualisation	Application mode	Participation	Treatment environment
	1.69 (0.9)	0.97 (1.2)	0.61 (1.0)	0.5 (1.5)	0.8 (1.8)	0.9 (1.5)	1.2 (1.1)
Mean ES				R 2	k
0.3068				0.127	111

Homogeneity analysis
	Q	df	P	T 2	I2
Model	15.1	6	0.0197	0.08	60.2%
Residual	104.	104	0.48	0.00	0%
Total	119.1	110	0.26

Regression coefficients
	B	SE	−95% CI	+95% CI	Z	P	Beta
Constant	−1.13	1.5	−4.1	1.83	−0.75	0.45	0.0
Similarity: fidelity monitoring	−2.14	0.68	−3.5	−0.8	−3.13	0.0017	−0.3
Similarity: application mode	0.44	0.45	−0.45	1.32	1.0	0.34	0.12
Similarity: environment	0.87	0.7	−0.51	2.25	1.23	0.22	0.13
Similarity: N of sessions	0.62	0.86	−1.06	2.29	0.72	0.47	0.08
Similarity: standardisation	0.57	0.59	−0.59	1.72	0.96	0.34	0.1
Similarity: participation	−0.16	0.53	−1.19	0.88	−0.29	0.77	−0.03
Method of moments random effects variance component
v = 46.5

Means and variance of the similarity ratings included in the model are provided, having a possible range of −2 (different for all studies) to 2 (similar across all studies).