Table 19: Summary of GRADE Quality Assessment for Exercise Interventions: Stratified by Target Population^*.

Intervention	No. of Studies	Quality Assessment				Summary of Findings
							No. of Patients		Effect (RR [95% CI])
		Design	Quality	Consistency	Directness	Other	Interv	Control		Quality
General population: Untargeted	12	RCT	Serious limitations†	Consistent	Direct	None	1250	1234	0.78 [0.66–0.91]	Moderate
		High	Moderate	Moderate	Moderate	Moderate
General population: Targeted	3	RCT	Serious limitations‡	Consistent	Some uncertainty about directness§	None	282	284	0.81 [0.66–0.98]	Low
		High	Moderate	Moderate	Low	Low
High-risk population: Untargeted	6	RCT	Serious limitations║	Some inconsistency	Some uncertainty about directness¶	None	372	270	0.89 [0.72–1.10]	Very low
		High	Moderate	Low	Very low	Very low
High-risk population: Targeted	2	RCT	No serious limitations	Consistent	Direct	None	329	330	0.93 [0.82–1.06]	High
		High	High	High	High	High
High-risk population: Combined	2	RCT	No serious limitations	Some inconsistency	Direct	None	77	61	0.87 [0.57–1.34]	Moderate
		High	High	Moderate	Moderate	Moderate

^*RR refers to relative risk; CI, confidence interval; Interv, intervention; RCT, randomized controlled trial;

^†Three studies (102) (100;113) did not describe randomization process, an omission which could conceal biases in study allocation; Exercise programs differed.

^‡One study (114) only 19% randomized; Exercise programs differed.

^§One study only on older (80+) women (115)

^║One study (116) did not describe randomization process; One study (111) not completely randomized; Three studies (111;117;118) not adequately blinded

^¶One study in women only, (119) and one study in men only. (120)