This was a prospective, multicenter, and two-arm (allocation ratio was 1:1) trial. It was a nonblinded RCT registered at the Chinese Clinical Trial Registry with the study number ChiCTR2100046775. The design of the current RCT is shown in Figure 1. Participants were recruited from Sir Run Run Shaw Hospital Affiliated to Zhejiang University, Hangzhou, Zhejiang, China; Ningbo Ninth Hospital, Ningbo, China; Zhejiang Xinhua Hospital, Hangzhou, Zhejiang, China; Lin An People Hospital, Hangzhou, Zhejiang, China; and YuHang Integrative Medicine Hospital, Hangzhou, Zhejiang, China through doctors’ face-to-face assessments from June 2021 until December 2022. For more information about centers, see Table S1 in Multimedia Appendix 1.

The participants were recruited from among consecutive patients diagnosed with ASCVD. The inclusion criteria were as follows: (1) age from 18 to 80 years; (2) meeting ASCVD diagnostic criteria, including a history of acute coronary syndrome (ACS), myocardial infarction, stable or unstable angina, transient ischemic attack, surgical history of coronary or peripheral vascular reconstructions, etc; (3) currently taking statins by prescription; (4) no difficulty in using smartphones, essential Chinese reading and writing skills and basic calculation skills; and (5) acceptance of the terms and conditions of the study and signature of the informed consent form.

The exclusion criteria were as follows: (1) meeting super high-risk ASCVD diagnostic criteria (diagnostic criteria are determined according to the 2020 Chinese expert consensus on lipid management of very high-risk patients with ASCVD; according to the diagnostic criteria, diagnosed ASCVD patients with one of the following conditions can be diagnosed as super high-risk ASCVD: ASCVD event recurrence; multivessel disease of coronary artery; recent ACS; atherosclerotic vascular disease of heart, brain, or peripheral multivascular bed; LDL-C≥4.9 mmol/L [190 mg/dl]; and diabetes); (2) one or more of the following complications: heart failure, ventricular tachycardia, second- to third-degree atrioventricular block, uncontrolled atrial tachycardia (including atrial flutter, atrial fibrillation, etc), mural thrombus, cardiac aneurysm, dysfunction or rupture of papillary muscle, severe sinus bradycardia (heart rate <50 beats/min), sinus arrest, etc; (3) history of cardiac arrest; (4) acute noncardiogenic complications, such as infections, kidney failure, hyperthyroidism, etc; (5) severe cardiovascular conditions: moderate to severe valvular stenosis, hypertrophic cardiomyopathy, other forms of ventricular outflow tract stenosis, acute myocarditis, pericarditis, active endocarditis, suspected or known aneurysm rupture, acute pulmonary embolism, or pulmonary infarction; and (6) vulnerable populations, including patients with mental disorders, critical illness, age <18 years, pregnancy, students or subordinate of principal investigators, employees of the research institute, etc.

A power calculation based on an absolute reduction of 20% [24,25] in LDL-C from the baseline expected during 6 months of follow-up indicated that 300 cases were determined as the target sample size for 80% statistical power.

Before conducting the study, written informed consent was obtained from all participants included in the study. The study was carried out according to the Declaration of Helsinki and was approved by the Ethics Committee of Sir Run Run Shaw Hospital (NO.20210508-30). The CONSORT-EHEALTH (Consolidated Standards for Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth) checklist was adhered to report the current study [26].

The subjects were randomly assigned to the intervention group and the control group in a ratio of 1:1 through the electronic data capture (EDC) Trial Data system [27]. The intervention group accessed the WeChat MiniApp by scanning the specified QR code and bound it to WeChat. The background of the applet will review this information to ensure that the patient is a participant in the study. After the audit is passed, the patient will have access to the relevant information in the applet (clock in, video, image, text, etc).

The intervention group received CBT for ASCVD lifestyle interventions delivered by WeChat MiniApp: “CBT ASCVD” (developed by Hangzhou Kang Ming Health Consulting Co., Ltd). The MiniApp user interface (Figure S1 in Multimedia Appendix 1) and the user manual (Figure S2 in Multimedia Appendix 1) are provided. The CBT intervention consists of a continuous cycle of educational topics. A cycle of topics includes 3 sections (Section A: Overview of CBT and Lipids; Section B: Strategies for Lipid Management; and Section C: Case Study; Table S2 in Multimedia Appendix 1). Sections A and C each have 1 video session. Section B contains 4 video sessions on lipids knowledge, lipids with diets, lipids with exercise, and lipids with emotion, respectively. A total of 6 topics constitute a cycle. The corresponding videos are produced on the basis of these topics and arranged to a 6-month intervention period according to the CBT schedule (Figure S3 in Multimedia Appendix 1). The CBT schedule was divided into an intense intervention period (1-2 months) and a regular intervention period (3-6 months). Compared to the regular intervention period, the intense intervention period had more frequent CBT interaction. The contents were designed and reviewed by cardiologists and psychiatrists from Sir Run Run Shaw Hospital, Zhejiang University, Hangzhou, Zhejiang, China. This WeChat MiniApp does not limit the number of login times of patients in the intervention group, but patients need to complete the corresponding courses every week according to the learning plan; otherwise, they will receive SMS text message reminders.

The design of this content is intended to improve the cognition and behavior of patients on blood lipid management through CBT and improve the self-management level (self-efficacy). Long-term accumulation will help to change the outcome of blood lipid management so that the blood lipid level of patients can not only meet the treatment goal but also be stable for a long time. The “CBT ASCVD” can systematically interfere with patients’ cognition, emotion, and behavior through mobile terminals. Participants in the intervention group have been told how to access the program after registration and will use it during the study.

In addition, routine care was delivered to both control group and CBT group. Routine care was designed as a brief oral health education by healthcare providers during each follow-up, involving disease-related knowledge, lifestyle guidance, complication prevention, rational drug use, and other contents. Besides, all participants received standard pharmacological treatments.

The EDC system was used for data collection and management. Data collection occurred at 4 points: baseline, follow-up in the first, third, and sixth months. Baseline data were collected when patients were enrolled, including height, weight, disease history, vital signs, laboratory results, electrocardiogram, medications, General Self-Efficacy Scale (GSE), and quality of life index (QL-index). Follow-up data were collected at the follow-up in the first, third, and sixth months, including weight, vital signs, laboratory results, electrocardiogram, medications, GSE, and QL-index (Figure 1).

All laboratory test samples were obtained after an overnight fast. Triglyceride (TG) level was measured by the standard enzymatic method. The Friedewald formula calculated the LDL-C level [28]. The definition of variability is as follows: (1) SD method SD is used to describe the variability of the univariate during the follow-up period; (2) coefficient of variation (CV) method: CV = (SD / mean) × 100(%) [29]. The LDL-C standard was defined as LDL-C below 1.8 mmol/L [30].

The GSE is a psychometric scale that assesses self-efficacy [31]. It consists of 10 items that represent attitudes toward obstacles rated on a scale from 0 (not optimal) to 4 (very optimal). The QL-index is a tool to evaluate the QL [32], consisting of activities of daily living, principal activities, health, outlook, and support.

Given routine clinical practice and the time required to achieve maximal lipid reduction, the follow-up period was designed to be 6 months from the start of lipid-lowering medication, which was concurrent with the CBT intervention. The study’s primary outcome was the change in the LDL-C level from enrollment to each follow-up node. The secondary outcomes were TG, C-reactive protein (CRP), the score of GSE, and the QL-index.

The normally distributed continuous variable was presented as mean (SD) with comparisons by independent samples t tests. The nonnormally distributed continuous variable was presented as median (IQR) with comparisons by the Kruskal-Wallis test. The categorical variable was presented as a count (percentage) with comparisons by the chi-square test.

The distribution of indicators (LDL-C, TG, QL-index, and GSE) between the control and the CBT group in the sixth month was visualized by the violin plot and compared by the Kruskal-Wallis test. The trajectory of indicators during the follow-up period was visualized according to the control and CBT groups. The linear regression model was used to assess the association of CBT with the change in LDL-C and the variation of LDL-C, TG, CRP, and the score of QL-index and GSE. The logistic regression model was used to assess the association between CBT and LDL-C during the follow-up period. We additionally adjusted this rate at baseline in the adjusted model. The subgroup analyses were conducted for the change of LDL-C, QL-index, and GSE at the sixth month stratified by the baseline LDL-C (<1.8 or ≥1.8 mmol/L), gender (male or female), age (<65 or ≥65 years), hypertension (no or yes), and diabetes (no or yes).

A total of 300 patients were enrolled and randomly allocated to the control group (n=150) and the CBT group (n=150). Overall, 296 patients completed assessments during the 6-month follow-up. In the control group, 148 patients completed the assesments, and 2 patients dropped out due to contact loss (n=1) and health problems (n=1). In the CBT group, 148 patients completed, and 2 patients dropped out due to contact loss (n=1) and unwillingness to accept follow-up (n=1). Figure 2 presents the flowchart of patient recruitment.

The baseline characteristics are displayed in Table 1. The mean age was 63.04 (SD 9.69) years, 69.6% were male participants, and the mean LDL-C level was 2.48 (SD 0.90) mmol/L, and 21.3% up-to-standard rates for LDL-C. There were no significant differences in baseline characteristics by group condition (CBT group vs control group), including demographics (age, gender, BMI, smoking, and drinking), comorbidity (diabetes and hypertension), laboratory results (LDL-C, TG, CRP), LDL-C <1.8 mmol/L, LDL-C <1.8 mmol/L and TG <5.6 mmol/L, GSE score and QL-index (all P values >.05).

After 6 months of follow-up, the CBT intervention group (vs control group) had more LDL-C reduction, higher scores of GSE and QL-index, and no difference in TG decrease (Figure 3).

The trajectory of the indicators is visualized in Figure 4. The mean level of LDL-C and TG showed an overall downward trend during the follow-up period in both the control group and the CBT intervention group, and the mean score of QL-index and GSE showed an upward trend.

In Table 2, CBT intervention has no significant effect on the changes in the LDL-C level at the first (β=–1.723, 95% CI –9.762 to 6.315) and third (β=2.641, 95% CI –5.785 to 11.068) months of follow-up. The CBT intervention significantly reduced LDL-C levels compared to the control group at the sixth follow-up month (β=–10.026, 95% CI –18.111 to –1.940). Moreover, a linear regression analysis showed no significant effect on the variation of LDL-C calculated by the SD and the CV method was observed during follow-up (SD: β=–0.018, 95% CI –0.098 to 0.061; CV: β=0.003, 95% CI –0.028 to 0.034).

In Table 3, the CBT intervention has no significant effect on the up-to-standard rates of LDL-C at the first (odds ratio [OR] 1.177, 95% CI 0.745-1.861) and third (OR 0.873, 95% CI 0.553-1.378) months of follow-up. CBT intervention significantly increased the up-to-standard rates of LDL-C at the sixth follow-up month (CBT vs control: 45.3% vs 57.4%; OR 1.631, 95% CI 1.030-2.582). After adjusting the baseline up-to-standard rates of LDL-C, the CBT intervention remained the moderately increased up-to-standard rates (OR 1.579, 95% CI 0.994-2.508).

In Figure 5, the results of the subgroup analysis are consistent with the main finding according to baseline LDL-C (<1.8 or ≥1.8 mmol/L), gender (male or female), age (<65 or ≥65 years), hypertension (yes or no), and diabetes (yes or no). The CBT intervention created a significant reduction in the LDL-C level at the sixth follow-up month in both the baseline LDL-C ≥1.8 mmol/L subgroup (β=–11.746, 95% CI –18.645 to –4.847) and baseline LDL-C <1.8 mmol/L subgroup (β=–24.103, 95% CI –43.110 to –5.095). Hence, the benefits of the CBT intervention were universal. CBT could reduce LDL-C levels even when the baseline LDL-C meets the standard. Besides, the significant reduction in receiving the CBT intervention was also observed in older adults (age ≥65 years; β=–12.709, 95% CI –23.960 to –1.457) and the nondiabetes (β=–12.991, 95% CI –22.352 to –3.630) subgroup.

In Table 4, patients with ASCVD who received the CBT intervention showed significant improvements in the GSE and QL-index during the whole follow-up period (all P values <.05). However, the CBT intervention did not significantly affect TG and CRP levels (all P values ≥.05). Furthermore, subgroup analyses also suggest that mobile-based CBT increases GSE (Figure S4 in Multimedia Appendix 1) and QL-index (Figure S5 in Multimedia Appendix 1) at the sixth month across the stratified population.

This multicenter RCT showed that compared with the conventional interventions group, a significant reduction of LDL-C level in the CBT group was observed at the 6-month follow-up. This difference was consistently significant even when the baseline LDL-C had reached the standard. The CBT intervention could significantly improve the self-efficacy and quality of life of patients with ASCVD, which was present during the whole follow-up period.

Controlling lipids and managing chronic conditions represent extensive and costly challenges for patients with ASCVD, which require long-term medication adherence and lifestyle modifications. Moreover, psychological distress can be considered for patients living with a chronic disease [33,34]. Lifestyle changes, such as losing weight, choosing healthy food, exercising, and stopping smoking, are challenging. CBT is devoted to positively changing how a person thinks (“cognitive”) and what they do (“behavior”) [35]. It has been increasingly used to modify cardiovascular risk factors in psychosocial intervention programs [36,37]. When difficulties in controlling lipids occur, patients with cognitive distortions become frustrated and abandon medical interventions, which worsens their condition. CBT can increase the readiness of patients to cope with their problems and help them with their resolution [38,39]. A cardiac rehabilitation program showed that CBT significantly improved depressive symptoms and high-density lipoprotein cholesterol levels [40]. Similarly, another small-scale RCT demonstrated that CBT could reduce LDL-C, TC, and TG levels in patients with diabetes [20]. Consistent with the previous results, this study demonstrated a significant reduction in LDL-C levels in patients with ASCVD after they receive a 6-month CBT invention.

Although the traditional CBT model can provide professional medical interventions during admission, it is limited by location. It cannot achieve timely and accurate medical follow-up once patients leave the hospital. Web-based CBT is as efficacious as face-to-face CBT [41] and can be delivered at a fraction of the cost and clinical time. With the development of the mobile internet, WeChat has built a critical communication platform for social intercourse [42]. In WeChat, we also developed a public account, thus making it easier to communicate between physicians and patients. To ensure data security and privacy, MiniApp does not collect any patient data from the hospital information system, nor does it require patients to upload any medical information to WeChat. MiniApp only serves as a convenient mobile-based CBT platform that works one way from the health care provider to the patient. Despite potential information security issues, WeChat remains one of the preferred platforms for CBT due to its large number of users in mainland China.

CBT controls lipid levels for patients with ASCVD by improving drug compliance, strengthening physical exercise, a balanced diet, and a healthy lifestyle [43-45]. Moreover, CBT-based interventions can help improve lipid profiles by decreasing physiological stress caused by negative emotions and improving autonomic nervous and endocrine systems [46]. The efficacy of this treatment modality might be slow and long-lasting. Patients may gradually benefit from a sustained CBT intervention. As shown by the results of this study, the lipid-lowering effect of CBT is not immediately observable; no differences were observed at follow-up months 1 and 3, and the difference was not significant until follow-up month 6 (β=–10.026, 95% CI –18.111 to –1.940). The requirement of lipid standard for patients with ASCVD is LDL-C below 1.8 mmol/L [30]. The crude model showed that the up-to-standard rate of LDL-C in patients receiving the CBT intervention increased significantly in the sixth month, compared with the control group (OR 1.631, 95% CI 1.030-2.582). Patients may benefit further by extending the CBT intervention beyond the current 6 months. However, this benefit may change after the end of the intervention due to the potential for relapsing into bad behaviors.

The comprehensive evaluation of the efficacy of the intervention should involve objective indicators and subjective perception. Improving self-efficacy is a strategy to help patients achieve and keep treatment goals [47]. The GSE is a psychometric scale that assesses self-efficacy [31]. The QL-index is a tool to evaluate the quality of life [32], another indicator of subjective perception in this study. Gratifyingly, the CBT intervention’s improvement in self-efficacy and quality of life is immediately present and remains persistent.

Our study is the first multicenter, prospective, randomized, controlled, open-label parallel-group superiority trial conducted in a Chinese population with ASCVD to explore the efficacy of mobile-based CBT on reducing LDL-C levels. The results of this study support that CBT should act as a treatment key in the comprehensive management of ASCVD. On the one hand, CBT can effectively reduce LDL-C levels in patients with ASCVD by improving self-efficacy, and the clinical benefit is more significant with the extension of the intervention time. The CBT intervention is expected to reduce the burden of lipid-lowering drugs or even replace them. On the other hand, the quality of life of patients with ASCVD improved by the CBT intervention is instant and effective. The development of mobile technology has substantially reduced the cost of CBT and increased its convenience and accessibility. Given the large number of patients with ASCVD, future work can be devoted to quantifying the economic benefits of mobile-based CBT by conducting cost-effectiveness analyses. Mobile-based CBT remains a promising option in the comprehensive management of patients with ASCVD.

This study has some limitations worth noting. First, due to the nature of psychotherapy studies, subjects and researchers (excluding statisticians) were unblinded. Therefore, they may be subject to bias from treatment expectations. Second, a reduction of LDL-C was shown after the 6-month mobile-based CBT intervention. Extending the duration of the mobile-based CBT beyond 6 months may be helpful to determine the efficacy of CBT on other indicators. Third, the follow-up was terminated after the end of the 6-month CBT intervention. Further assessment of relapsing into bad behaviors was not conducted.

In patients with ASCVD, mobile-based CBT is effective in reducing LDL-C levels (even for those who are already meet the LDL-C standard) and can improve self-efficacy and quality of life.