Summary of findings for the main comparison. Aerobic exercise training compared with control for fibromyalgia.

Aerobic exercise training compared with control for fibromyalgia
Patient or population: individuals with fibromyalgia Settings: group and supervised Intervention: aerobic exercise training Comparison: control comparison (treatment as usual, wait list control, continuation of daily activities including physical activity) Outcome: measured at the end of the intervention
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Aerobic exercise training
Health‐related quality of life FIQ Total. Scale from 0 to 100 (0 is best) Follow‐up: 12‐24 weeksa	Mean health‐related quality of life ranged across control groups from 54 to 63.7 units	Mean health‐related quality of life in the intervention groups was 7.89 lower (13.23 to 2.55 lower)		372 (5 studies)	⊕⊕⊕⊝ moderateb	Absolute change 8% (3% to 13%) better Relative changec 15% (5% to 24%) better NNTB 6 (4 to 16)d
Pain intensity Visual analogue scale Scale from 0 to 100 mm (0 is best) Follow‐up: 6‐24 weekse	Mean pain intensity ranged across control groups from 56 to 80.5 mm	Mean pain intensity in the intervention groups was 11.06 lower (18.34 to 3.77 lower)		351 (6 studies)	⊕⊕⊝⊝ lowb,f	Absolute change 11% (4% to 18%) better Relative changec 18% (7% to 30%) better NNTB 4 (2 to 15)d
Fatigue Visual analogue scale Scale from 0 to 100 mm (0 is best) Follow‐up: 14‐24 weeksg	Mean fatigue ranged across control groups from 62.3 to 72 mm	Mean fatigue in the intervention groups was 6.06 lower (‐12.41 lower to 0.30 higher)		246 (3 studies)	⊕⊕⊝⊝ lowb,h	Absolute difference 6% improvement (12% improvement to 0.3% worse) Relative changec 8% improvement (16% improved to 0.4% worse) NNTB n/a
Stiffness FIQ Scale from 0 to 100 mm (lower scores mean less stiffness) Follow‐up: 16 weeks	Mean stiffness in control groups was 69 mm	Mean stiffness in the intervention groups was 7.96 lower (14.95 to 0.97 lower)		143 (1 study)	⊕⊕⊝⊝ lowb,i	Absolute difference 8% (1% to 15%) improvement Relative changec 11% improvement (1% to 21% improved) NNTB 6 (3 to 218)d
Physical function FIQ and SF‐36 converted, 0 to 100 scale (0 is best) Follow‐up: 8‐24 weeksj	Mean physical function ranged across control groups from 6 to 22 units	Mean physical function in the intervention groups was 10.16 lower (15.39 to 4.94 lower)		246 (3 studies)	⊕⊕⊝⊝ lowb,h	Absolute change 10% (95% CI 15 to 5) improvement Relative changec 21.9% (95% CI 33.2 to 10.7) improvement NNTB 5 (3 to 13)d
Withdrawals All‐cause attrition Follow‐up: 6‐24 weeks	17 per 100	20 per 100 (14 to 25)	RR 1.25 (0.89 to 1.77)	456 (8 studies)	⊕⊕⊕⊝ moderateb	Absolute change: 5% more withdrawals with exercise (3% fewer to 12% more) Relative change 25% more (11% fewer to 77% more) NNTH n/a
Adverse events Descriptive information	"the present study findings confirm earlier studies that have shown aerobic exercise to reduce tender point tenderness, increase work capacity without adverse side effects..." (Wigers 1996; pages 83‐84). The following statements show some minor adverse events following aerobic exercise training: "...unable to exercise after an injury" (Sanudo 2010; pages 1840), but it is unclear whether the injury was related to intervention participation: "One participant assigned to the short bout exercise withdrew after developing a metatarsal stress fracture" (Schachter 2003; page 347)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; NNTB: number needed to treat for an additional beneficial outcome; NNTH: number needed to treat for an additional harmful outcome; RR: risk ratio
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Very low quality: We are very uncertain about the estimate

^aStudy author (intervention length in weeks): King 2002 (12), Schachter 2003 and Kayo 2011 (16), Gowans 2001 (23), Sanudo 2010 (24)
 ^bDowngraded because of lack of blinding (performance and detection bias)
 ^cRelative change calculation as per Cochrane Musculoskeletal Review Group procedures: mean difference (MD)/pooled control group baseline means
 ^dNumber needed to treat for an additional beneficial outcome (NNTB): NNTB for continuous outcomes calculated using the Wells calculator (from the CMSG Editorial office; http://musculoskeletal.cochrane.org/) for statistically significant outcomes only. Mean and standard deviation (SD) for the control group at baseline taken from Schachter 2003 (health‐related quality of life 55 (1.3), pain 61 (1.97), stifness 7 (1.9), and physical function 38 (1.86)
 ^eStudy author (intervention length in weeks): Sencan 2004 (6), Wigers 1996 (14), Kayo 2011 and Schachter 2003 (16), Mengshoel 1992 (20), Sanudo 2010 (24)
 ^fDowngraded for imprecision, wide confidence intervals, issues related to selective reporting (ie, Mengshoel 1992 and/or Kayo 2011)
 ^gStudy author (intervention length in weeks): Wigers 1996 (14), Kayo 2011 and Schachter 2003 (16), Sanudo 2010 (24)
 ^hDowngraded for imprecision
 ⁱOne study

^jStudy author (intervention length in weeks): Wigers 1996 (14), Kayo 2011 and Schachter 2003 (16), Sanudo 2010 (24)