This study was a parallel controlled study conducted from September 2018 to October 2019 in Shanghai, China. Two hundred stable COPD patients were included, and the participants were randomly divided into the following two groups (1:1 ratio): the WeChat intervention group and routine nursing control group (Figure 1). The research plan was approved by the ethics committee of Putuo Hospital (grant number: PTEC-A-2018-25-1).

The rehabilitation therapist instructed the patients to perform Baduanjin exercise until they could accomplish proficiency according to the Baduanjin video (produced by the State Sports General Administration in 2003). All patients were asked to perform the exercise two to four times a day for no less than 5 days per week.

A WeChat platform team with six members was established. The head nurse of the ward served as the team leader and was responsible for the operation and guidance of the entire project. Two nurse supervisors were responsible for teaching the patients and their families how to use the platform, providing all educational documents to the patients at the appropriate time, maintaining contact with the patients, and urging the patients to perform Baduanjin exercise. Two physicians were responsible for evaluating the patients’ condition and devising the medical plan. One technician in the pulmonary function room was responsible for evaluating the lung function of all patients, collecting the data, and performing statistical analyses of the data. The team members cooperated very well, and the platform was operated effectively. The patients very actively participated in the communication. Responses to all questions were obtained from the medical staff in a timely manner.

All patients with stable COPD were discharged from the respiratory medicine ward of a large general hospital. All candidates first completed a brief screening questionnaire. Patients who met the inclusion criteria were invited to participate in the study and received more detailed information regarding the study. After providing written informed consent, all qualified patients were divided into the WeChat group and control group.

The inclusion criteria were as follows: (1) 50 to 80 years of age regardless of gender; (2) confirmed clinical diagnosis of stable COPD according to the standard of GOLD 2018 [20]; (3) informed consent (patients and families); (4) ability to use WeChat proficiently (patients and primary caregivers); and (5) ability to perform Baduanjin independently.

The exclusion criteria were as follows: (1) severe heart, liver, and kidney diseases, tumors, or other conditions that may affect the observation; (2) life expectancy less than 1 year; and (3) history of conducting physical exercise for a long time (≥3 times/week, ≥20 minutes/time, persisting for more than 12 months) [21].

Pulmonary function (percentage forced expiratory volume in 1 second [FEV1%]) was the outcome index. Based on previous studies showing FEV1% improvement in clinical trials [22], the mean and SD of FEV1% in the control group were 57.09 and 22.53, respectively, while the mean and SD of FEV1% in the experimental group were 60.24 and 20.15, respectively. The alpha level was set to .05, the power was 80%, and the participant dropout rate was 20%. A sample size of 192 patients (96 per group) was required for the primary analysis. Thus, we recruited 200 patients in the trial (100 in the WeChat group and 100 in the control group).

Using the method of block randomization, 200 research patients were randomly assigned to the WeChat and control groups. The block size was defined as four, and there were six sequential arrangements and combinations. Excel (Microsoft Corp) randomly generated 50 (1‒50) numbers that did not repeat. Then, the numbers were divided by six to obtain the remainder, and six combinations were matched according to the remainder. A random block group table was then completed. After the research patients were included, the random block table was assigned to a group. By the nature of the trial design, neither the research staff nor the participants were blinded to the intervention.

A designated attending physician supervised the project every month and evaluated the progress.

The baseline data were based on continuous measurement data and categorized count data. The measurement data are expressed as mean (SD) or median (IQR). The count data are characterized as frequencies. Two-tailed t tests (mean and SD) or chi-squared tests were used to analyze the normally distributed data. A nonparametric test was used to analyze the nonnormally distributed data (median and IQR).

To analyze the results, repeated measures analysis of variance, a simple effect model, and two-sample t tests were used. To determine the within-group differences and between-group differences, the 95% CIs were calculated for the continuous measures. The statistical analysis was performed using SPSS 22.0 software (IBM Corp), and P<.05 was considered statistically significant.

Based on the inclusion and exclusion criteria, we screened 200 patients and completed baseline assessments before randomization to the trial of 100 patients in each group. Of the 200 patients, 67 were female and 133 were male. The ratio of male and female patients assigned to the two groups was not different. The mean age was 68.16 years (WeChat group, 69.43 years; control group, 68.87 years), and the mean course of COPD was 4.46 years (WeChat group, 4.50 years; control group, 4.31 years). Most patients had severe COPD, accounting for 73% (146/200) of cases (WeChat group, 72% [72/100]; control group, 74% [74/100]). Patients with moderate and very severe COPD accounted for 11.5% (23/200) of cases (WeChat group, 12% [12/100]; control group, 11% [11/100]) and 15.5% (31/200) of cases (WeChat group, 16% [16/100]; control group, 15% [15/100]), respectively. We also assessed the number of acute exacerbations and current smokers in both groups. Patients who had acute exacerbation at least once in the past year accounted for 31% (31/100) of cases in the control group and 35% (35/100) of cases in the WeChat group. The proportion of current smokers in the control group (34% [34/100]) was comparable to that in the WeChat group (37% [37/100]). According to statistical analysis, the baseline characteristics were balanced in both groups (Table 1).

A repeated measures analysis of variance was performed to analyze the data from the two groups. The data did not conform to the sphericity test (P<.001). The analysis of the multivariate test revealed that in the within-group comparisons in both the control and WeChat groups, the Baduanjin exercise frequency significantly differed across various time points (F=214.87, df=11, P<.001). This finding indicates that in both the control and WeChat groups, Baduanjin management was effective. Additionally, in the between-group comparisons, the different groups (control group vs WeChat group) showed a significant difference in the time points (F=33.82, df=1, P<.001) (Multimedia Appendix 1, Figure 2).

Further analysis by a simple effects model showed a difference within and between the groups. The Baduanjin exercise frequencies in the control and WeChat groups (1 week to 4 weeks) increased from week 1 to week 4 and then maintained a plateau until the end of the project. In the within-group comparisons, there were statistically significant differences at each time point in weeks 1 to 3 compared with week 4, but no difference was observed between weeks 5 and 12 and week 4. According to the between-group comparison results, significant differences were found between the WeChat and control groups at the same time points (all P<.001). We found that WeChat management was more effective for performing Baduanjin exercise (Multimedia Appendix 1, Figure 2).

The FEV1% predicted value significantly differed before and after the trial in both the control group (Z=−3.686, P<.001) and WeChat group (Z=−6.985, P<.001), and the FEV1% predicted value in both groups improved after Baduanjin exercise. A significant difference was found in the FEV1% predicted values after Baduanjin exercise between the two groups (Z=−3.679, P<.001), and no difference was observed before the trial (Multimedia Appendix 2, Figure 3). The results demonstrate that Baduanjin exercise was more effective in the WeChat group.

All patients completed the questionnaire independently. We found that the CAT score significantly differed before and after the exercise in the control group (Z=−4.937, P<.001) and the WeChat group (Z=−5.246, P<.001). The life quality of the patients in both groups improved after Baduanjin exercise (Figure 4). No statistically significant difference was found in the CAT score between the two groups before the study (Z=−1.407, P=.30), and a significant difference was found after the exercise (Z=−5.246, P<.001). These results demonstrate greater improvement in the life quality of patients in the WeChat group after the exercise (Multimedia Appendix 3, Figure 4).

Communication in the WeChat group was mainly conducted via text/graphics, with 13,911 items, followed by voice messages, with a total of 1317 items. Text/graphics and voice accounted for the main methods of message exchange. Pictures and videos were also used; however, the frequency was relatively low (Table 2).

All messages could be divided into the following three groups: information provided by the medical team, medical staff-patient interaction communication, and patient-patient interaction communication. The frequency of the provided information was relatively fixed. Medical staff-patient interaction was relatively frequent during the first 2 weeks but was less frequent thereafter. However, patient-patient interaction gradually increased, indicating that a COPD family was established. Interestingly, we found that the patients interacted and encouraged each other. The platform became a common home for all WeChat participants.

We divided the WeChat group into active patients and nonactive patients based on the number of messages exchanged. Those with more than two messages per day were considered active patients, and those with one or fewer messages were considered inactive patients. Eighty-six patients were more active in the group. The number of completed Baduanjin exercises, lung function, and CAT scores of these active patients were higher than those of the nonactive patients (Figure 5). The amount of information exchanged by the WeChat group reflected the activity of the WeChat management platform, indicating that COPD patients actively participated in the management of the disease and received better outcomes.

According to the score of the 5-point Likert scale, items that received more than 4 points were as follows: continued WeChat management, effective Baduanjin exercise management, practical information content, convenient communication and interaction, smooth platform operation, simple operation process, service attitude of the medical staff, and COPD rehabilitation. The results indicated that the COPD patients were satisfied with the management of the WeChat platform. Among these items, the highest score was for continued WeChat management (mean 4.54, SD 0.77), followed by effective management of Baduanjin exercise (mean 4.46, SD 0.87), indicating the feasibility, sustainability, and effectiveness of disease management with the WeChat platform (Table 3).

The main purpose of this study was to test the feasibility and efficiency of WeChat in the management of Baduanjin exercise in COPD patients. The results of this study suggest that the management of Baduanjin exercise with the WeChat platform improves the enthusiasm and compliance of patients. The frequency of Baduanjin exercise using the WeChat-based intervention was 0 to 21 times (mean 18.84) to 0 to 13 times (mean 10.43) per person per week compared with traditional interventions at 12 weeks. The Baduanjin exercise frequency in the WeChat group was markedly higher than that in the control group. Under the management of WeChat, some patients did not perform Baduanjin exercise. However, after the follow-up on WeChat, the number of patients with a Baduanjin exercise frequency of 0 decreased (Figure 6). Therefore, Baduanjin exercise management based on the WeChat platform improved the enthusiasm and compliance of patients. These results are consistent with the beneficial effect of compliance using WeChat management in other clinical populations [26].

Based on the WeChat platform, communication between medical staff and patients increased, and the effectiveness of Baduanjin lung rehabilitation improved. Studies have shown that the frequency of the use of the platform is crucial for the effectiveness of the intervention [27]. Our results indicate that the total number of messages exchanged gradually increased and reached a peak at approximately 4 weeks; thereafter, the number maintained a plateau. Interaction with text/graphics exhibited the most obvious increase (Figure 7). We also found that active participants had better outcomes. The interaction between patients increased, and mutual exchange, sharing, and supervision promoted more enthusiasm among the patients. The role of the team was always favorable.

All participants completed the satisfaction survey, and a high score was obtained in the WeChat group. Based on the questionnaire responses, we found that continuing WeChat management and Baduanjin exercise management were the top two ranked items. The other items also achieved relatively high scores. Thus, the patients expected to continue WeChat management, and the WeChat management of Baduanjin exercise is feasible.

Early hospital follow-up, coaching, COPD action plans and management programs, telemedicine, pulmonary rehabilitation, and other interventions require more research. Improvement in management is an effective strategy for improving patient recovery [28]. A recent study also noted that reducing readmission could improve prognosis [29]. Although the management of COPD patients has been studied since the 1960s [30], the situation of management is not optimistic. Many problems are associated with COPD management, and a key problem is lung rehabilitation [31]. Currently, published data lack sufficient evidence for the identification of an effective pathway to improve the symptoms of COPD by management. Therefore, better ways to achieve management must be explored [32].

With the development of the internet, major communication platforms have begun to be applied in health care. Social media platforms, such as Twitter and Facebook, have been steadily applied in medical education [33,34]. Facebook has been used to study many aspects of health care, and it provides a convenient and easily accessible way to collect unsolicited and observational patient data [35]. An analysis of the content of COPD group chats on Facebook showed that COPD Facebook group members share specific disease-related experiences and request information regarding select self-management topics. This information can be used to improve the quality of self-management support provided to members of popular COPD Facebook groups [36]. WeChat is also gradually being used in medical teaching and the clinic, but the clinical application needs improvement [37]. Clinical studies based on the WeChat platform, such as cough variant asthma and promoting weight loss, have achieved good results [38]. The internet could also help patients with COPD understand the disease and manage the condition by themselves [39].

This study has some limitations. First, the study period was quite short (approximately 12 weeks), which may have affected the judgement of compliance, but our study is ongoing, and more data will be reported in the future. Second, the number of participants in this study was relatively small, although the sample size was calculated statistically. However, because of the complexity of clinical trials and the uncontrollability of patients, a larger number of patients could supply more intact information. Third, as the original data were obtained from patient reports, inaccurate reports could not be fully excluded. Thus, there is a problem of credibility in patient reports using WeChat management. Finally, our study only explored the effectiveness of Baduanjin rehabilitation training under WeChat platform management. More projects in self-management supported by the WeChat platform could be fully studied in the future.

In this study, a new management platform involving WeChat from the in-hospital period to the out-of-hospital period was established for patients. Pulmonary rehabilitation using Baduanjin exercise for COPD patients was applied using this platform. This online rehabilitation training model met the patients’ requirements as this platform was convenient, easy to understand, and time saving. Information is shared in a timely manner, and seamless communication is maintained with patients. Additionally, the compliance and enthusiasm of the patients were much higher. Baduanjin exercise based on WeChat platform management was effective in improving the life quality and lung function of COPD patients. Patient management based on WeChat is effective, feasible, and sustainable.