Table 18: Summary of GRADE Quality Assessment for Exercise Interventions: Stratified by Intervention Length^*.

Intervention	No. of Studies	Quality Assessment				Summary of Findings
							No. of Patients		Effect (RR [95% CI])
		Design	Quality	Consistency	Directness	Other	Interv	Control		Quality
Exercise (untargeted, long duration)	8	RCT	Serious limitations†	Consistent	Direct	None	462	434	0.76 [0.64–0.91]	Moderate
		High	Moderate	Moderate	Moderate	Moderate
Long duration: targeted	3	RCT	No serious limitations	Consistent	Some uncertainty about directness‡	None	454	456	0.89 [0.79–1.01]	Moderate
		High	High	High	Moderate	Moderate
Long duration: Combined	1	RCT	No serious limitations	Only 1 study	Direct	None	43	27	0.73 [0.57–0.95]	High
		High	High	High	High	High
Short duration: Untargeted	10	RCT	Serious limitations§	Slightly inconsistent	Direct	None	1160	1070	0.85 [0.71–1.01]	Low
		High	Moderate	Low	Low	Low
Short duration: Targeted	2	RCT	No serious limitations	Consistent	Direct	None	157	158	0.91 [0.73–1.13]	High
		High	High	High	High	High
Short duration: Combined	1	RCT	No serious limitations	Only 1 study	Direct	None	34	34	1.11 [0.73–1.70]	High
		High	High	High	High	High

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

^†Several studies (100-103) did not describe randomization process and by this omission might conceal biases in study allocation. Heterogeneity in exercise programs.

^‡Two studies on older individuals aged 80+ and 85+ (104;105). Study by Campbell et al. on women only. (106)

^§Five studies didn’t have adequate blinding. (107-111) Two studies were not completely randomized (111;112)