A prospective randomized controlled trial was conducted in Harbin Chest Hospital, China from May to November 2020. Convenience sampling was performed to recruit 114 patients with pulmonary TB. They were assigned into the intervention group and control group in a 1:1 ratio by using a computer-based random number generator, and all received a 3-month intervention. WeChat groups were created by a pharmacist to provide health education for the participants. Patients in the intervention group and the control group received routine medical and nursing care in the TB clinic, including the supervision of their medications. In addition, pharmacist-assisted mHealth intervention based on the ITHBC theory about TB management was provided to the intervention group. This trial is reported in accordance with CONSORT-eHEALTH (see Multimedia Appendix 1 for the completed CONSORT-eHEALTH form V 1.6.1).

Based on the ITHBC theory, we summarized the epidemical characteristics of TB and the self-management and psychology of patients with TB through a literature review. The research team created a multidisciplinary panel comprising of a TB pharmacist, 2 nursing experts, 2 TB physicians, and a public health expert to discuss and develop the intervention program. A statistician was also included to be responsible for the random allocation of the participants and data analysis. Semistructured interviews were conducted with 4 patients randomly selected in the finally included participants to know their mastery of TB knowledge, diets, exercise habits, risk factors and awareness, personal needs and wishes, and concerns and inertia. After communication, we conducted individualized assessments to refine the intervention program. Finally, 20 participants who would not be enrolled in the final trials were randomly selected to complete the pretest to refine self-designed scales. The face and content validity of the self-designed questionnaire were determined by pretest and experts, respectively.

The formula for sample size calculation was as follows:

We assumed σ=1.6 and δ=1.02 with reference to a preliminary study [19] when α=1.6, 1-β=0.80, and then we got n=40 for 1 group. Considering dropouts (20%) over the course of the study, 100 patients were included (50 for control group and 50 for intervention group).

The participants in this study were patients who were admitted consecutively to the TB clinic of Harbin Chest Hospital from May to August 2020, and the study was conducted between May and November 2020. Each participant’s intervention started as long as they were enrolled offline at the hospital by face-to-face communication. Therefore, all the patients were not recruited or started with the intervention at the same time, but all completed the 3-month intervention.

The eligibility criteria were (1) 18 years old or older, (2) experiencing newly diagnosed active pulmonary TB according to the classification of TB (WS 196-2017) issued by the People’s Republic of China State Health and Family Planning Commission [20], (3) literate and capable of using WeChat, and (4) hospitalized when enrolled in the study. The exclusion criteria were as follows: (1) experiencing drug-resistant pulmonary TB or treated according to the treatment plan of drug-resistant TB and (2) complicated with serious diseases such as AIDS, malignant tumor, and severe diseases that were newly diagnosed in the heart, brain, liver, and kidney. All patients in both the groups received the therapy in the hospital for nearly 1 month since the intervention started and then received TB outpatient treatment for 2 months. The median hospitalization time for the intervention group was 25 (IQR 18-32) days, while that for the control group was 24 (IQR 16-31) days (P=.35). All patients gave web-based informed consent to participate in this study (see Multimedia Appendix 2 for the informed consent). Patients whose treatment regimens were changed or lost to follow-up were terminated from this study.

Intervention for the intervention group on the day of hospitalization included health education in WeChat groups, web-based health education lectures, and receiving health education plans as well as routine medical and nursing care such as supervision of their medications, which continued for 1 month. Then they received therapy at outpatient clinics and were provided with the same intervention as provided in the hospital for 2 months (see Multimedia Appendix 3 for the screenshots of the health education in the WeChat group). Patients in the control group received routine medical and nursing care in the TB clinic, including supervision of their medications. Meanwhile, the pharmacist also created a WeChat group for the control group to allow patients to communicate with each other but without any intervention. The content of the intervention was designed based on the ITHBC theory, and the flow chart is presented in Figure 2.

Information and desire to change are the prerequisites that patients engage in recommended behaviors. In our study, health education can be divided into knowledge education and belief education. Knowledge was defined as information of pulmonary TB, and belief was defined as personal perceptions about patients’ health condition or health behaviors [14]. Knowledge education contained basic knowledge, diet management, hygiene routines, and medication management. Belief education was conducted in weekly meetings to increase the consciousness of crisis for disease and patient confidence for disease management. The pharmacist gave individualized health education based on baseline information. Education forms included WeChat groups and online lectures. The topics of the health education are presented in Multimedia Appendix 4, which were sent to the WeChat group in the form of articles, pictures, or videos. Patients were guided every Monday for 10 minutes in WeChat. Web-based lectures were conducted every Thursday for an hour through WeChat.

Patients with the desire to change health behaviors should learn self-regulation skills and the ability to apply TB knowledge to their lives. To improve self-regulation skills and ability, the pharmacist summarized patients’ current behavior and issues measured at the baseline to make an individualized health education plan in a month and a short-term aim in a week during hospitalization. The plans in the latter 2 months would be made according to performance of the patients in the previous month. Patients were asked to self-monitor and record daily health behaviors and emotions. A meeting was held by the pharmacist in the WeChat group every Saturday, where patients reported and self-evaluated their own behaviors and emotions; moreover, they had exchange of views with wardmates. The pharmacist would give some timely guidance to help the patients make the next aim themselves.

Excellent social facilitation positively influences and supports patients’ engagement in health behaviors. According to the ITHBC theory, social facilitation includes social influence and social support. Patients are more likely to engage in health behaviors when authorities sway their thinking and motivation, which is called social influence. In our study, social influence came from health care providers, especially pharmacists who offered the intervention, family, and significant others (eg, wardmates, friends, neighbors, colleagues, social media). Social support includes emotional, instrumental, and informational support, which were offered by health care providers, family, friends, and others. The patient’s family received health education in rounds with patients, attended web-based lectures, and supervised the patients. They were encouraged to take the lead in adopting health behaviors and provide comfort and company for the patients in their family. The pharmacist provided instrumental and informational support by establishing a WeChat group and health education. Wardmates communicated with and encouraged each other in wards and in the WeChat group. During the intervention, when the country was in lockdown during the COVID-19 epidemic, patients also received much tutorial advice on preventive behaviors, as the coronavirus transmission is similar to that of pulmonary TB virus.

All the patients filled out a web-based questionnaire of demographic information at baseline (see Multimedia Appendix 5 for the questionnaire 1). The primary outcome was self-management behavior. Secondary outcomes were TB knowledge awareness, self-efficacy, social support, and degree of satisfaction with health education. The outcome data were collected by scales at baseline and at the end of the study when patients were hospitalized and at home, respectively. Clinical pharmacists used the web-based platform of Wen Juan Xing [21] to distribute the questionnaire (see Multimedia Appendix 5 for the questionnaire 2), and patients filled it while the pharmacist guided them face-to-face (in hospitalization) or telephone (after discharge). The demographic information of the patients was collected by the medical record system and the self-designed questionnaire.

Self-management behavior was measured using a self-designed structured scale with item 4 in questionnaire 2. General self-management behaviors consisted of COVID-19-related behaviors, medication behaviors, and lifestyle habits with 13 items and a 5-level Likert scoring method. Each item was scored as 1-4 points, with 1=no intention to act, 2=intention to act, 3=having acted but stopped, 4=starting to act, and 5=having acted and considering keeping on. Higher scores indicated greater self-management behaviors. Cronbach α was .966, suggesting good internal consistency of the scale.

In order to ensure the quantity and homogeneity of the participants, the participants were selected in strict accordance with the criteria of inclusion and exclusion, which were checked by TB physicians. We chose newly diagnosed patients with active pulmonary TB to make sure of a consistent treatment plan and to reduce bias caused by different medication therapies. A study [25] has shown that the generation of a random sequence should be done by some independent personnel, usually a statistician, who is not going to be involved in the conduct of the randomized controlled trial. To ensure the scientific nature of the random allocation, this study had a statistician responsible for the random allocation of the participants and the data analysis. The researchers discussed the intervention measures of this study with an interdisciplinary panel in the field of TB and conducted structured interviews with the participants in order to further improve the intervention program. In addition, a pretest was conducted before formal intervention to find the problems in the design of the intervention and to optimize the intervention program. The clinical pharmacist explained the purpose and significance of this study to the participants and obtained informed consent. In the process of filling out the questionnaire, the clinical pharmacist helped to explain the questionnaire to the patients to ensure that the questionnaire was completed efficiently, but it was not instructive. Moreover, when using the data collection tool, known as Wen Juan Xing, patients could not submit the questionnaire unless they completed every question, which avoided missing data.

Statistical analysis was implemented using SPSS Statistics 26.0 software (IBM Corp). Normality tests were applied to assess the distribution of the continuous data, and nonnormally distributed data were presented as median (IQR). The Mann-Whitney U test was used to compare the changes in each scale score between the 2 groups at the baseline and the end point, while the Wilcoxon test (for paired samples) was applied to compare the difference between the baseline and the end point of each scale score in each group. The demographic characteristics of the 2 groups were compared using the chi-square test. P<.05 (2-tailed test) was considered as statistically significant.

This study was approved by the ethics committee of Harbin Chest Hospital, China (2020-10).

A total of 114 patients participated in this study, and they were randomly assigned to the intervention group (n=59) and the control group (n=55). During the study period, 2 patients dropped out (2/114, 1.8%). Therefore, 112 patients were included in the statistical analysis finally. Of the 112 patients, 64 (57.1%) were males. All patients were aged 18 years and older and were mainly aged between 18 and 30 years (39/112, 34.8%), and only 7 (6.3%) were older than 60 years. Approximately half of the patients had a college degree or higher (51/112, 45.5%). Most patients (101/112, 90.2%) lived with their families or friends, and 71 (63.4%) were married. The demographic information of the patients is shown in Table 1. At baseline, approximately all patients’ characteristics were similarly distributed between the intervention and the control groups.

The total scores of all the 13 self-management behaviors were calculated. Both groups scored greater than that at baseline (P<.001 vs P=.04, respectively), but the total score of the intervention group was higher than that of the control group (P<.001) (Figure 3). Table 2 shows that there was no significant difference in the scores of all self-management behaviors between the intervention group and the control group at baseline.

After the intervention, there were significant differences in the scores of the above behaviors between the 2 groups except for “covering your mouth and nose when coughing or sneezing” (P=.23) and “wash hands properly” (P=.60) (Table 3).

The scores of self-management of health behaviors in the intervention group increased significantly after the intervention (P<.001). The 4 health behaviors with significant increases in the scores in the control group were “stay in a good mood” (P=.002), “maintain a balanced diet and ensure the intake of high protein, vitamins, and minerals” (P=.002), “cover your nose and mouth when coughing or sneezing” (P<.001), and “wash hands properly” (P<.001). Although the score of the item “self-isolate from families in home-based treatment” was lower than that at the baseline (P=.01), there was no significant change in the scores of other behaviors in the control group (Table 4).

This study had 2 strengths. First, we applied the ITHBC theory to study the self-management of patients with TB for the first time. Second, mHealth interventions did not face any restrictions during the COVID-19 pandemic and met the needs of the patients to the greatest extent. Moreover, healthy people are exposed to less risks owing to the application of mHealth [43,44].

This study had the following limitations. First, the intervention was only conducted in 1 hospital, which may restrict the general applicability of our results, and future studies should be conducted with larger sample sizes to confirm our results. Second, scores of TB knowledge awareness and friend support in both groups had significant differences at the baseline (P=.01), which may be due to the small sample size. Third, the short duration of observation (3 months) cannot prove the long-term efficacy. We plan to conduct studies with a long-term intervention (6 or 12 months) to focus on assessing the patient’s clinical efficacy and quality of life. Fourth, the nonrandom sampling method could result in some selection bias.

In conclusion, ITHBC-based mHealth intervention may be a new promising therapeutic strategy for the management of TB for improving patients’ subjective initiative and self-management behaviors, which are beneficial for promoting compliance and the quality of prevention and control for pulmonary TB. In addition, mHealth provides an effective solution for outpatients without nursing care.