Aim and design
We aimed to compare the effects of Tai Chi and aerobic exercise on weight, body mass index, blood pressure and glycosylated hemoglobin (HbA1c) level in older persons with mild to moderate-severe depressive symptoms. A randomized controlled trial was conducted.
Participants
Cluster sampling method was used in this trial to avoid contamination. Six communities were randomly selected from 24 communities in Ya’ an City, then the selected communities were assigned to either the Tai Chi group or the aerobic exercise group according to random numbers generated by Microsoft Excel software version 2016. Older persons who met the inclusion criteria were selected as the study subjects in each community. Either a 12-week Yang’s Tai Chi or aerobic exercise intervention was offered to older persons with depressive symptoms from these six communities depending on their allocation. Older persons were eligible if they: had mild to moderate-severe depressive symptoms; aged 60 years old and above; did not have any other mental illness; had lived in the community for at least 1 year. Older persons were excluded if they: were not suitable for Tai Chi and aerobic exercise, such as severe vision or hearing impairment; had a history of cardiovascular disease or stroke in the past 6 months; were accepting antidepressants; exercised regularly at a frequency of equal to and more than 3 times per week.
Depressive symptoms were assessed by the Patient Health Questionnaire 9-Item (PHQ-9), which was initially developed by Kroenke [20]. It was a self-rating scale for the depression-related symptoms experienced in the past 2 weeks, scoring each item on a four-point Likert scale of 0–3, with 0 indicating “not at all” and 3 indicating “almost every day”. PHQ-9 is widely used in both clinic and community settings. It was introduced to China in 2007 [21]. The reliability and validity of PHQ-9 were demonstrated in the elderly populations in mainland China [22]. The cut points of 5, 10, 15, and 20 were interpreted as representing mild, moderate, moderate-severe, and severe levels of depression on the PHQ-9 [20].
Intervention
Tai Chi and aerobic exercise interventions were carried out simultaneously to eliminate potential confounding from seasonal influences on symptom severity. Tai Chi intervention was offered 3 times per week for 12 consecutive weeks, and each session lasted 60 min, which was demonstrated to be the most popular approach in previous studies [23]. Likewise, aerobic exercise intervention was offered 3 times per week for 12 consecutive weeks, each session lasted 60 min. Attendance forms and sign-in sheets were used to monitor attendance of each participant at intervention lessons. Study subjects were allowed to take one day off a month, and participants who missed two or more times a month were excluded. Participants in both Tai Chi and aerobic exercise groups received a closely supervised, group-format intervention sessions in their community's outdoor plaza. Two to four research assistants who were responsible for supervising and observing participants to avoid accidents, were scheduled for each intervention session.
Tai Chi
24-Style Yang’s Tai Chi was adopted in the study. The 24-Style Tai Chi is simplified from classical 108-movement Yang style Tai Chi, which was more suitable for older persons to learn and practice within 12 weeks and was also the most common style adopted in published literature. The key points of Tai Chi were regular motion of the whole body, breathing regulation and mental concentration during the practice. Tai Chi intervention session was leaded by 3 instructors who had extensive experience of 20 years and over. Moreover, all 3 instructors had been trained for study subject protection at the beginning of the intervention classes. Safety precautions were informed to all participants. The theory and procedures of Tai Chi were explained by the instructor at the first session to ensure that participants grasp the essentials of Tai Chi exercise. Learning and reviewing occurred in the consequent sessions. The total duration of Tai Chi exercise was about 60 min.
Aerobic exercise
Participants allocated to aerobic exercise group received a 20-movement low impact aerobics. The 20-movement exercise consisted of several components: shoulder movement, arm movement, chest expand, waist movement, and leg movement. These 20 movements were low-intensity and were mainly dynamic stretching. Each movement took 1 min. The 20-movement aerobic exercise cycled for 3 times during each session. The total duration of Tai Chi exercise was about 60 min. Three nursing students conducted aerobic exercise intervention as instructors. Before the intervention began, the three nursing students were trained about the 20 movements and the subject protection procedure. Moreover, safety precautions were told to all participants. The aerobic exercise sessions were strictly monitored by research assistants to ensure participants’ comfort and safety and to minimize adverse events.
Outcomes measures and follow up
Participants’ sociodemographic data was collected at baseline using a self-designed form, including age, marital status, living status, education years, living aera, and number of diseases. The outcomes of the study were weight, BMI, BP and HbA1c level, which were assessed at baseline and after completion of the interventions (week 12). Height and weight were self-reported by participants. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. BP was assessed by mercurial sphygmomanometer (YUWELL, 20,152,200,947). BP of each participant was measured three times after a 30-min rest, and the average of the three measurements was taken as the final value. Hypertension was defined as a systolic blood pressure (SBP) of 140 mmHg or greater or a diastolic blood pressure (DBP) of 90 mmHg or greater. Peripheral blood was collected by EDTA tube before intervention and HbA1c was assessed after 12-week intervention.
Sample size
Sample size was calculated based on the results of one previous meta-analysis [13]. The hypothesized was that Tai Chi was more effective than aerobic exercise. A two-sided hypothesis test at 0.05 significance level with an allocation ration of 1:1 was used, for a 90% statistical power to detect a significant effect of Tai Chi over aerobic exercise, a 20% attrition rate was estimated. A sample size of 240 (120 in the Tai Chi group, 120 in the aerobic exercise group) was finally determined.
Randomization
Six communities were selected from 24 communities and assigned to the Tai Chi group and aerobic exercise by random number table method. Microsoft Excel (version 2016) was used to generate table of random numbers with “1” and “2”. The number “1” was set to appear six times and the number “2” was set to appear 18 times. Six communities corresponding to the number “1” were included, while 18 communities corresponding to the number “2” were excluded. Then, 3 communities corresponding to the number “1” were assigned to the Tai Chi group and 3 communities corresponding to the number “2” were assigned to the aerobic exercise group by using the same procedure.
Qualified older persons in the six communities were considered as candidate participants of our study. After screening for eligibility, candidates were recruited by the same random number table method mentioned above, according to the proportion determined by the sample size and qualified population base of each community. Microsoft Excel (version 2016) was used to generate table of random numbers with “1” and “2”. Participants who corresponded to number “1” were included, while participants corresponding to number “2” were excluded.
Blinding
The data collectors, physical indicators assessors, and laboratory tester who conducted the baseline and follow-up assessments were blinded to treatment allocation.
Data analysis
The qualitative data was described by frequency and proportion, and the data were statistically inferred by chi-square test. The quantitative data was statistically described by mean and standard deviation (SD), and the data were statistically inferred by independent t-test. A two side P value < 0.05 was considered to indicate significant differences. All analyses were performed in Stata version 16.0.