Subscribe to eTOC

For Parkinson Disease Patients, Tai Chi Can Improve Balance and Reduce Falls


patients (shown here) taking a tai chi class at a senior center.

Patients with Parkinson disease (PD) can improve their balance and reduce their likelihood of falling with twice-weekly tai chi, according to a large, randomized trial that compared tai chi with resistance training and stretching. The trial, reported in the Feb. 9 New England Journal of Medicine, adds to a growing list of behavior-based interventions shown to improve balance in PD patients, a symptom of the disease that is largely resistant to medication and surgical interventions.

“Tai chi is a set of slow, continuous, and perhaps most importantly, self-initiated voluntary movements,” explained lead author Fuzhong Li, PhD, senior research scientist at the Oregon Research Institute in Eugene, OR. “That's why we postulated it might help patients with motor difficulties.” Previous studies have shown its ability to reduce the incidence of falls in older adults, but little research has been published on its potential to provide the same benefits to PD patients.

The practice of tai chi involves a connected series of movements that emphasize control, coordination, balance, and posture. For working with PD patients, Dr. Li said, “We take the tai chi framework and then we modify it to fit the impairments people with Parkinson's disease face. We might call it ‘therapeutic tai chi.’ We push the patient toward the periphery of the base of support, challenging the proprioception system.”

A series of movements or postures make up a “form,” which in traditional tai chi may include as few as four postures to more than 100. In Dr. Li's trial, he combined six movements specifically chosen to maximize the involvement of balance and gait. Movements included weight shifting, ankle sway, stepping, and shifting the center of balance.


One hundred seventy-six patients were enrolled in the trial, and were randomized to either tai chi, resistance training, or stretching for one hour twice weekly for 24 weeks. Resistance training exercises focused on strengthening the muscles responsible for posture, balance, and gait. Stretching served as a control. All patients performed their routines in a group setting, and were taught breathing exercises as part of the program. Patients were primarily in Hoehn & Yahr stages 1 through 3 (mild to moderate PD), although several patients were in stage 4.

The primary outcome measures were “maximum excursion” and “directional control” during a challenge to their postural stability. For these tests, patients stood in place and leaned forward, backward, or to the side while remaining erect. When leaning forward, for instance, the theoretical maximum excursion (100 percent) is to move so that the center of gravity is directly over the tips of the toes; any further, and physics dictates you will fall. To measure directional control, motion-capture sensors record the path length of the movement, and compare it to the straight-line path for the same stopping point.

Secondary outcome measures included a variety of gait measures, as well as the motor section of the Unified Parkinson's Disease Rating Scale (UPDRS) and a functional reaching measure. Assessments were performed at baseline, three months, six months, and again three months after the end of the intervention, all in the “on” state. Falls were monitored by a daily fall diary.


At six months, patients in the tai chi group did significantly better than those in either of the other groups on both primary outcome measures. At baseline, maximum excursion was 64 percent of the theoretical maximum in each group. At six months, those in the tai chi group had improved to 74 percent and those in the resistance-training group to 68 percent, while those in the stretching group had declined to 62 percent. Tai chi patients improved their directional control by about 18 percent, while those in the two other groups worsened slightly.

On all secondary outcomes, patients doing tai chi also did better than those who did stretches, and better than those doing resistance training on two measures: stride length and functional reach. UPDRS improvement from baseline was 6.4 points for tai chi, 5.1 points for resistance training, and 1.4 points for stretching.

The number of falls per participant per month was 0.22 for tai chi, 0.51 for resistance training, and 0.62 for stretching; the difference was not significant between tai chi and resistance (p=0.05) but was between tai chi and stretching (p=0.005). Three months after the end of the intervention, tai chi patients had significantly fewer falls than either of the other groups.

“Our study shows that six months twice-weekly tai chi could help people with Parkinson's disease, especially with postural instability, and it may help people with their functional performance as well,” Dr. Li said. Importantly, these results apply to those with mild-to-moderate disease, he said, for whom the balance challenge is relatively safe. “I would recommend to clinicians that tai chi be considered as an add-on to existing physical therapy, or as a rehabilitation strategy for balance therapy.” The benefits of tai chi, he noted, include that it is relatively easy to learn, it doesn't require any equipment, and can be done indoors at home.


Oksana Suchowersky, MD, professor of clinical neurosciences at the University of Calgary, Alberta, commented that the trial was valuable for its comparison of exercise regimens, a rarity in the literature, she said. “It helps us recommend what types of exercise patients should do. That tai chi helped balance and reduced falls is particularly clinically relevant, since falls are a major reason for nursing home placement in PD.”

Chris Hass, PhD, associate professor of applied physiology and kinesiology at the University of Florida Center for Movement Disorders and Neurorestoration in Gainesville, said: “I think the improvements Dr. Li is reporting after tai chi are very promising. My concern, however, is that the take-home message is that tai chi is much better than resistance training, and we don't know that.”

The form of resistance training used in this trial was much lower in intensity than “what traditionally would be considered resistance training,” he said. Dr. Hass's most recent work shows that the effect of resistance training on gait in PD is at least as large as that of tai chi. He also cautioned that any study of tai chi must be replicated, because not only are PD patients heterogeneous and have different needs, but tai chi itself comes in many different styles, and may be taught by instructors of very different skill.

Merrill Landers, DPT, associate professor and chair of the department of physical therapy at University of Nevada at Las Vegas, was impressed by the improvements from tai chi, but commented that it was not surprising it performed better at this than resistance training. “It's apples and oranges to compare resistance training to balance training. Pushing the limits of stability is exactly what tai chi does, so it's no surprise it was more successful.” A more useful comparison, he said, would be with other balance training programs. “The big issue we have right now is which balance program is the best. We just don't have good comparisons.”

For Dr. Li, the question is not which type of exercise is better, but what each can offer. “I encourage patients to do all kinds of exercise. Each exercise modality is unique, with resistance training focusing on strength and tai chi focusing on control and initiation. Stretching is also important for rigidity. They're all good.”