Table 3. Association of different effect modifiers on treatment effect.

Potential effect modifier	Comparisons	ROR; 95% CI	Between-trial variability (τ2); relative reduction
No effect modifier			0.1899
Staff	Nurse as receiver vs. others	1.29; 1.05 to 1.60	0.1389; 26.9%
	Other professionals as receiver vs. others	1.62; 1.29 to 2.04
Level of prevention	Primary prevention vs. others	1.30; 0.98 to 1.71	0.1692; 10.9%
	Secondary prevention vs. others	1.31; 1.09 to 1.57
Unit of randomization	c-RCT vs. RCT	1.28; 1.03 to 1.60	0.1871; 1.5%
Outcome definition	Adherence as primary vs. secondary outcome	1.38; 1.12 to 1.70	0.1719; 9.5%
Duration of follow-up	Long (>12 months) vs. short follow-up periods	1.38; 1.04 to 1.83	0.1741; 8.3%
Risk of bias	Risk of bias (high or unclear vs. low):		0.1488; 21.6%
	Sequence generation	0.88; 0.77 to 1.27
	Allocation concealment	0.93; 0.64 to 1.33
	Blinding	1.11; 0.80 to 1.53
	Incomplete outcome data addressed	1.04; 0.77 to 1.40
	Selective outcome reporting	1.58; 0.96 to 2.60
	Other	0.94; 0.74 to 1.20

Abbreviation: CI: confidence interval; c-RCT: cluster randomized controlled trial; RCT: randomized controlled trial; ROR: ratio of odds ratios.