If no significant heterogeneity was detected, a fixed-effect model Small molecule library was used. Statistical significance was set at p < 0.05. Database searching using the method described led to the retrieval of 570 articles. After the screening of titles and abstracts, nine articles appeared to be eligible

(Singh et al 1997, King et al 1997, Tworoger et al 2003, Li et al 2004, Elavsky and McAuley 2007, King et al 2008, Irwin et al 2008, Altena et al 2008, Reid et al 2010). Three articles were subsequently excluded, two because their control groups had engaged in some form of exercise (Tworoger et al 2003, Li et al 2004) and one because the experimental group had engaged in additional therapies that did not meet the inclusion criteria (Altena et al 2008) (Figure 1). No additional articles were identified by the scanning of reference lists. Therefore six trials were included in the analysis. The six included trials involved 305 participants. The quality of the included trials is presented in Table 1 and a summary of the trials is presented in Table 2. Quality: The quality of the included trials ranged from 5 to 8 on the PEDro scale ( Table 1). No trials blinded participants or therapists, while two trials blinded

assessors. All trials had retention rates of 85% or greater and all reported between-group differences with point estimates and measures of variability. Participants: Most of the included trials recruited both men and women participants with sleep problems. The mean age of the participants ranged from 48 to 72 years. However, the 305 participants were predominantly Vemurafenib ic50 Isotretinoin female because one trial recruited only postmenopausal women ( Elavsky and McAuley 2007). Interventions: Five trials examined aerobic exercise (endurance training, walking, or

Tai Chi) and one trial examined a resistance exercise program. The duration of most of the trials was between 10 and 16 weeks, with one study continuing for 12 months. The control groups in all the trials received either no treatment or health education for 90–120 minutes per week. All the aerobic exercise programs examined were of moderate intensity, instructing the participants to reach 60–70% of their heart rate reserve or 60–85% of their peak heart rate for 40 to 60 minutes. Self-reported sleep quality: The effect of exercise training on sleep quality as indicated by the global Pittsburgh Sleep Quality Index score was examined by pooling data from 288 participants across five trials. Participation in exercise training improved sleep quality, with an SMD of 0.47 (95% CI 0.08 to 0.86) ( Figure 2, see also Figure 3 on the eAddenda for a detailed forest plot.) The effect of exercise training on the ‘subjective sleep quality’ subscale of the Pittsburgh Sleep Quality Index was examined by pooling data from 239 participants across five trials.