Table 3.

Reported clinical trials of Qi Gong for patients with Parkinson disease

Study	Size	Design	Duration	Intervention	Controls	Outcomes	AEs
Wassom et al. (2015) [58]
USA	7	Open-label	6 weeks	Weekly group “six-healing sounds” Qi Gong with twice daily at-home exercises	–	Motor: significant reduction in stride time (5.27%), gait velocity (8.73%), and double support time (9.01%); non-significant improvement in stride length (4.15%). Non-motor: improvement in some aspects of sleep quality (decreased motor symptoms at night, less disturbed sleep); there was no improvement in fatigue from baseline	Did not report
Schmitz-Hubsch et al. (2006) [59]
Germany	56	Randomized, controlled	2 8-week sessions	Weekly 60-min Qi Gong sessions (n = 32)	Usual care (n = 24)	Motor: The proportion of patients with improvement in UPDRS was significantly greater at 3 months (p = 0.008), almost significant at 6 months (p = 0.0503), but not 12 months (p = 0.635) Non-motor: decreased constipation, decreased pain, reduced daytime sleepiness relative to controls; both groups reported decreased depression	Did not report
Liu et al. (2016) [60]
Germany	56	Randomized, controlled	10 weeks	5 60-min Health Qi Gong sessions per week (n = 28)	Usual care (n = 26)	Motor: Muscle hardness of pronator teres using myometry was reduced compared to controls (p < 0.01); significant improvement in TUG test from baseline; improved hand–eye coordination and balance on various assessments	Did not report
Lee et al. (2018) [61]
Korea	56	Randomized, controlled, wait-list, single-blind	8 weeks	Twice weekly 60-min Qi dance (Turo) sessions per week (n = 25)	Usual care (n = 16)	Motor: mUPDRS and total UPDRS were both significantly improved in the Turo group compared to controls (p = 0.004 and p = 0.001, respectively). There was a tendency toward improvement on the BBS (p = 0.051). Non-motor: PDQL improved compared to controls (p = 0.049), particularly social functioning and systemic symptoms	Did not report
Burini et al. (2006) [62]
Italy	26	Randomized, controlled, cross-over	7 weeks of assigned treatment, followed by an 8-week “washout” without treatment, then 7 weeks of the other treatments	3 sessions of Qi Gong per week, followed by 3 sessions of aerobic exercise per week (n = 13)	3 sessions of aerobic exercise per week, followed by 3 sessions of Qi Gong per week (n = 13)	Motor: There was no change in mUPDRS in either group from baseline. The aerobic exercise group significantly improved their 6MWT and modified Borg scale (breathlessness) compared to baseline, but the Qi Gong group did not. Non-motor: no change in PDQ-39, BDI, Brown’s disability scale in either group	Did not report
Xiao and Zhuang (2016) [63]
China	89	Randomized, controlled, single-blind	6 months	4 45-min sessions of Baduanjin Qi Gong (BDQ) per week plus 30 min of daily walking (n = 45)	30 min of daily walking only (n = 44)	Motor: The BDQ group had significant improvement in mUPDRS (p = 0.038), 6MWT (p = 0.045), TUG test (p = 0.028), gait speed (p = 0.021), and BBS (p = 0.037). Non-motor: The PDSS-2 total score was improved in the BDQ group (0.045) relative to controls, including the following subgroups: motor symptoms at night (p = 0.037) and disturbed sleep (p = 0.045)	Did not report
Xiao et al. (2016) [64]
China	98	Randomized, controlled	6 months	4 60-min sessions of Baduanjin Qi Gong per week (n = 49)	4 sessions of conventional physical therapy per week (n = 49)	Motor: The Qi Gong group had significant improvement in TUG test, BBS, ABC, and 6MWT compared to baseline, but not when compared to controls. They also had fewer falls compared to baseline, which approached significance between groups	Did not report
Moon et al. (2017) [65]
USA	10	Randomized, controlled	6 weeks	Twice daily at-home sessions of “six-healing sounds” Qi Gong plus a weekly group exercise class (n = 5)	Sham Qi Gong without breathing exercises or meditation plus a weekly group exercise class (n = 5)	Non-motor: significant reduction in TNF-α (an inflammatory marker) in the Qi Gong group, but not in controls. There were also significant improvements in the PDSS-2 total score (p < 0.0005) and PD symptoms at night subscale (p < 0.05), which correlated to a decrease in TNF-α	Did not report

6MWT = 6-min walk test; ABC = Activity-Specific Balance Confidence; AEs = adverse events; BBS = Berg Balance Scale; BDI = Beck Depression Inventory; BDQ = Baduanjin Qi Gong; mUPDRS = Unified Parkinson’s Disease Rating Scale, motor subscale; PD = Parkinson Disease; PDQ-39 = 39-item Parkinson’s Disease Questionnaire; PDQL = Parkinson’s Disease Quality of Life Questionnaire; PDSS-2 = Parkinson’s Disease Sleep Scale 2; TNF-α = tumor necrosis factor-α