Table 2.

Clinical outcome data of the included studies.

Author (Year)	Intervention(s)	Control	Primary Outcome Measure	Secondary Outcome Measure	Follow-Up	Results
Arnold et al. (2010) [42]	Aquatic and education: aquatic exercise twice a week with once-a-week group education for 11 weeks; Aquatic: two weeks aquatic exercise for 11 weeks	no intervention	Berg Balance Scale, 6-MWT, Timed Up and Go Test	PASE score, AIMS-2 score	11 weeks	Significant improvement in fall risk factors (p = 0.038) with the combination of aquatic exercise and education.
Bearne et al. (2011) [43]	Ten 75-min group exercise and self-management sessions (twice a week for five weeks)	no intervention	WOMAC physical function	WOMAC pain, WOMAC total score	Six weeks and six months	No between-group differences in any outcome measure.
Bezalel et al. (2010) [44]	Group education program once a week for four weeks, followed by a self-executed home-based exercise program	six 20-min sessions of short-wave diathermy	WOMAC total score	Sit to stand test, Timed up and go test	four and eight weeks	Significant improvement in the timed up and go test and WOMAC total (p < 0.01) in the exercise group.
Brismee et al. (2007) [31]	Six weeks of group Tai Chi sessions, 40 min/session, three times a week, followed by another six weeks of home-based Tai Chi training	three 40-min group sessions per week for six weeks	WOMAC pain, VAS	WOMAC stiffness and physical function	3, 6, 9, 12, 15, and 18 weeks	Less overall pain and better WOMAC physical function with Tai Chi (p = 0.0089 and 0.0157, respectively).
Casilda-López et al. ( 2017) [40]	Eight-week dance-based aquatic exercise program	global aquatic exercise program	WOMAC total score	6-MWT and VAS	Eight weeks and Three months	Postintervention differences in the WOMAC pain and aggregate (p = 0.002 and p = 0.048) in favor of the experimental group.
Cheung et al. (2014) [34]	Eight-week Hatha yoga intervention involving group and home-based exercise sessions	no intervention	WOMAC total score	SPPB, SF-12	Four weeks, eight weeks and 20 weeks	Improvement in WOMAC pain p = 0.01) and stiffness (p = 0.02) in the intervention group.
Cheung et al. (2016) [35]	Yoga: one 45-min class per week for eight weeks and additional 30 min/day, four times/week of yoga practice at home; Exercises: eight weekly group-based classes	no intervention	WOMAC total score, VAS	SPPB, SF-12	Four and eight weeks	Yoga group presented improvements in WOMAC TOTAL (p = 0.001) and VAS scores (p = 0.03) compared to exercises group.
Doi et al. (2008) [46]	Four sets of 20 repetitions of quadriceps exercise every day (knee extension movements while sitting on a chair or in a supine position)	NSAIDs	WOMAC total score and VAS	SF-36	Eight weeks	Improvements in total WOMAC, SF-36 and VAS: all p < 0.001 in the exercises group; WOMAC and VAS at p < 0.001 and SF-36 at P < 0.03 in the control group.
Foley et al. (2003) [37]	Three water based, or three gym-based exercise sessions a week for six weeks, including a short warm up period, lower limb stretches, and a set of resistance exercises	no intervention	WOMAC total score, 6-MWT	SF-12	Six weeks	Walking speed and distance increased in the hydrotherapy and gym groups (both p < 0.001). No significant changes for WOMAC function or stiffness.
Fransen et al. (2007) [30]	Aquatic or Tai Chi program (with a preliminary 10-min warm-up session): 1 h, twice a week for 12 weeks	no intervention	WOMAC pain and physical function	SF-12, DASS21	12 and 24 weeks	Improvements of 6.5 and 10.5 for pain and physical function scores with hydrotherapy and improvements of 5.2 and 9.7 with Tai Chi.
Hale et al. (2012) [44]	Water-based exercise classes twice weekly for 12 weeks	community-based computer-skills training program	PPA	Step Test, Timed Up and Go Test, WOMAC total score	12 weeks	No statistically significant between-group differences were found for any outcome measured.
Huang et al. (2017) [45]	Quadriceps isometric contraction exercise (two sets of exercises in the morning and evening)	local physiotherapy and oral NSAIDs	WOMAC total score, VAS	/	One and three months	Significant improvement of WOMAC and VAS score in the experimental group (p < 0.05).
Hurley et al. (2007) [48]	12 supervised sessions that combined discussion on specific topics regarding self-management and coping, with an individualized, progressive exercise regimen	no intervention	WOMAC physical function	WOMAC pain, WOMAC total score	Six weeks and six months	Individual and group rehabilitated participants had better WOMAC score (p = 0.01) than control group.
Lee et al. (2009) [32]	Eight weeks of group Tai Chi Qigong sessions, with 60 min per session twice a week	no intervention	SF-36	WOMAC total score, 6-MWT	Eight weeks	Improvements in SF-36 (p = 0.010) and 6- MWT (p = 0.005) in the Tai Chi group.
Lund et al. (2008) [38]	Aquatic and land-based exercise programs for eight weeks with 2 sessions per week	no intervention	KOOS function and pain, VAS	Muscle Strength	Eight weeks and three months	Only in the land-based exercise group a decrease of pain was detected (p = 0.039). There were no significant differences between groups for KOOS.
Marconcin et al. (2017) [47]	PLE2NO program: 90-min intervention twice a week for 12 weeks	no intervention	KOOS pain	KOOS function and total score, 6-MWT	12 weeks	Significant clinical improvement was found for all KOOS (larger than 10 points) and in the 6 MWT (p = 0.035) in the exercise group.
Taglietti et al. (2018) [41]	Aquatic program twice a week for eight weeks	educational program: once a week for eight weeks	WOMAC total score, VAS	SF-36, Timed up and go test, Yesavage Geriatric Depression Scale	Three months	WOMAC pain reduced in favour of the aquatic exercise group (p = 0.021). No differences for the outcome’s functional mobility or depression.
Takacs et al. (2017) [49]	Ten weeks of exercises targeting dynamic balance and strength performed four times per week	no intervention	CB&M, WOMAC physical function	Muscle Strength	10 weeks	Improvements in self-reported pain (p = 0.005), physical function (p = 0.002), and fear of movement (p = 0.01) in the training group.
Tsai et al. (2013) [33]	Three sessions a week of Tai Chi exercise (12-form Sun Tai Chi) for 20 weeks	no intervention	WOMAC pain	WOMAC physical function and stiffness, timed up and go test and Sit to stand test	21 weeks	WOMAC pain (p < 0.001), physical function (p = 0.001) and stiffness scores (p = 0.001) improved in the Tai Chi group.
Wang et al. (2011) [36]	Aquatic/land-based exercise protocol with a 60-min flexibility and aerobic training class, three times a week for 12 weeks	no intervention	KOOS total score, 6-MWT	knee ROMs	Six and 12 weeks	Aquatic and land group presented less pain than control group (respectively p < 0.001 and p = 0.002).
An et al. (2013) [50]	Short-term Baduanjin exercise: 30-min sessions five times a week for one year	/	WOMAC total score, SF-36	6-MWT	one year	WOMAC pain, stiffness and physical function subscales, SF-36 body pain and 6-MWT were significantly improved.
Bove et al. (2017) [29]	16 sessions of task-specific training at a frequency of two visits per week	/	KOOS total score	30-Second Chair Rise, Timed Stair Climb Test, Floor Transfer Test	Four, six, and eight weeks	Improvements in patient-rated and performance-based outcomes.

PASE: Physical Activity Scale for the Elderly; AIMS-2: Arthritis Impact Measurement Scales 2; WOMAC: Western Ontario and McMaster Universities Arthritis Index; VAS: Visual Analog Scale; 6-MWT: 6 min walk test; SPPB: Short Physical Performance Battery SF-12: Short Form-12; NSAIDs: Nonsteroidal anti-inflammatory drugs; DASS21: Depression, Anxiety and Stress Scale; PPA: Physiological Profile Assessment; KOOS: Knee injury and Osteoarthritis Outcome Score; CB&M: Community Balance and Mobility Scale; ROM: Range of Motion.