Table 2.

Effect of eHealth interventions by outcome and comparator.

Outcome by comparator		Impact	Number of participants (studies)	Certainty of evidence (GRADEa)
Anxiety
	Inactive control	Four studies had large to small effects, and 1 study had a trivial effect	446 (5 RCTsb)	Very lowc,d,e
	Active control	Two studies had large to small effects, and 1 study had a trivial effect; 1 study reported inferior results, but effect sizes could not be calculated	380 (4 RCTs)	Very lowc,d,e,f
Bereavement
	Inactive control	No studies	0 RCTs	N/Ag
	Active control	No studies	0 RCTs	N/A
Coping
	Inactive control	One study had a small effect; and 1 study showed a trivial effect; 1 study had small to trivial effects favoring the control group	433 (3 RCTs)	Very lowc,d,e
	Active control	One study had a large effect	63 (1 RCTs)	Very lowd,e,f,h
Depression
	Inactive control	Nine studies had large to small effects, and 2 studies had trivial effects; 1 study comparing 2 interventions found small effects in one and trivial effects in the other. The intervention was inferior to control in 1 study (small effect)	1488 (13 RCTs)	Lowc
	Active control	Three studies had medium to small effects, and 2 studies had trivial effects	500 (5 RCTs)	Very lowc,d,e
Mental distress
	Inactive control	Four studies had large to medium effects, and 1 study had a trivial effect	268 (5 RCTs)	Very lowc,f,h,i
	Active control	The intervention was inferior to control in 1 study (large effect)	100 (1 RCTs)	Very lowd,e,f
Quality of life
	Inactive control	One study had a small effect, and 2 studies had a trivial effect. The intervention was inferior to control in 1 study (small effect)	268 (4 RCTs)	Very lowc,d,f,j
	Active control	Two studies had large to small effects, and 2 studies had a trivial effect	421 (4 RCTs)	Very lowc,e,i,j
Substance use
	Inactive control	No studies	0 RCTs	N/A
	Active control	No studies	0 RCTs	N/A

^aGRADE: Grading of Recommendations Assessment, Development, and Evaluation.

^bRCT: randomized controlled trial.

^cMost articles were rated high using the Cochrane Risk-of-Bias Tool.

^dMagnitude and direction of effect varied across studies.

^eVariability in how the outcome is measured and the types of interventions.

^fThe total number of participants across studies was small (400 or fewer), and some studies had small improvements, whereas others had nonsignificant results likely because of a small sample size (borderline imprecision).

^gN/A: not applicable.

^hAt least one study was rated high, and multiple studies rated some concerns overall with the Cochrane Risk-of-Bias Tool.

ⁱPopulations are limited to a few specific conditions and disorders or by sex, which limits generalizability.

^jThe total number of participants across studies was >400, but some studies found no effect (borderline imprecision).