This paper is in the following e-collection/theme issue:

Web-based and Mobile Health Interventions (4103) Text-messaging (SMS, WeChat etc)-Based Interventions (760) Obesity and Nutrition as Public Health Problem (416) Formative Evaluation of Digital Health Interventions (3329) Parent and Child Education on Healthy Eating and Nutrition (221)

Published on in Vol 28 (2026)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/76204, first published .
Therapeutic Effects of a WeChat Mini-Program on Metabolic Dysfunction–Associated Fatty Liver Disease: Randomized Controlled Trial

Therapeutic Effects of a WeChat Mini-Program on Metabolic Dysfunction–Associated Fatty Liver Disease: Randomized Controlled Trial

Therapeutic Effects of a WeChat Mini-Program on Metabolic Dysfunction–Associated Fatty Liver Disease: Randomized Controlled Trial

Authors of this article:

Chao Sun1 Author Orcid Image ;   Guangyu Chen2 Author Orcid Image ;   Cuicui Shi1 Author Orcid Image ;   Haixia Cao1 Author Orcid Image ;   Ruixu Yang1 Author Orcid Image ;   Jing Zeng1 Author Orcid Image ;   Xiaoyan Duan1 Author Orcid Image ;   Xin Sun2 Author Orcid Image ;   Jian-Gao Fan1 Author Orcid Image
Article Authors Cited by Tweetations Metrics

    1Center for Fatty Liver Disease, Department of Gastroenterology, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, 1665 Kong Jiang Road, Shanghai, China

    2Clinical Research and Innovation Unit, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

    Corresponding Author:

    Jian-Gao Fan, MD, PhD


    Background: For patients with metabolic dysfunction–associated fatty liver disease (MAFLD), weight loss is advised but challenging in practice. In China, there is a pronounced shortage of tailored digital lifestyle interventions for this population.

    Objective: This study aimed to assess the effects of a WeChat mini-program-delivered lifestyle intervention on weight loss and hepatic steatosis among individuals with MAFLD who were overweight or obese.

    Methods: Adults who are overweight or obese and have clinically diagnosed MAFLD with transient elastography examination were enrolled in this prospective randomized controlled trial. Patients were randomly assigned to receive either WeChat mini-program management (intervention group) or standard care (control group) at a 1:1 ratio. The intervention was structured around the development and implementation of personalized diet and exercise plans, supplemented by guided exercise video courses and reinforced through continuous monitoring and informational support. Body weight and clinical parameters were assessed at baseline and then at 6 months.

    Results: A total of 89 patients met the inclusion criteria and were randomly assigned to the intervention group (n=45) or control group (n=44). Among the 89 patients with MAFLD, 60% (27/45) of them achieved a weight loss of ≥5%, and 24.4% (11/45) of them had a weight loss of ≥10% in the intervention group, which was greater than those in the control group (27/45 vs 7/44; relative risk [RR] 3.771, 95% CI 1.836‐7.748; P<.001; 11/45 vs 3/44, RR 3.585, 95% CI 1.072‐11.988; P=.02). Importantly, patients receiving the intervention were significantly more likely to achieve a ≥10% reduction or normalization of controlled attenuation parameter (CAP) than those in the control group (26/45 vs 14/44; RR 1.816, 95% CI 1.102‐2.992; P=.01). After adjusting for key baseline covariates, multivariate analysis confirmed the intervention’s positive effect on achieving a weight loss of ≥5% (OR [odds ratio] 8.380, 95% CI 2.886‐24.331; P<.001) of ≥10% (OR 4.612, 95% CI 1.138‐18.686; P=.03), as well as on CAP reduction of ≥10 % or normalization (OR 2.853, 95% CI 1.092‐7.456; P=.03). In parallel, the intervention group presented greater reductions in liver enzymes (alanine aminotransferase, aspartate aminotransferase, and γ-glutamyl transpeptidase) and metabolic parameters (fasting insulin, hemoglobin A1c, and triglyceride) than the control group (all P<.05). According to the fibrosis assessment, only the FibroScan-aspartate aminotransferase score decreased more in the intervention group than in the control group (median difference −0.06, 95% CI −0.13 to −0.01; P=.02), as compared to other non-invasive indicators.

    Conclusions: Readily scalable in primary care and varied-resource settings, our WeChat mini-program-based intervention extends beyond weight loss to reduce hepatic steatosis and improve metabolic parameters, thereby addressing the critical gap in targeted MAFLD management in China with a low-cost model for high-burden populations. Nevertheless, larger future studies are needed to confirm these findings with greater precision and assess long-term sustainability.

    Trial Registration: Chinese Clinical Trial Registry ChiCTR2500100197; https://www.chictr.org.cn/showprojEN.html?proj=247294

    J Med Internet Res 2026;28:e76204

    doi:10.2196/76204

    Keywords

    metabolic dysfunction-associated fatty liver disease; WeChat mini-program; lifestyle intervention; weight loss; metabolic parameters; randomized controlled trial 


    Background

    Metabolic dysfunction–associated fatty liver disease (MAFLD) represents a significant and growing public health burden worldwide, paralleling the epidemics of obesity and type 2 diabetes mellitus [1]. It is a gradually progressive disease that evolves through a spectrum that begins with simple hepatic steatosis and advances to metabolic dysfunction-associated steatohepatitis (MASH), ultimately resulting in liver fibrosis, cirrhosis, and even hepatocellular carcinoma [1,2]. Despite its severe clinical sequelae, there are no pharmacologic therapies approved in China to reverse the histological progression of MAFLD, underscoring the critical role of nonpharmacological management [3].

    Lifestyle intervention involving dietary adjustments and physical exercise is still the cornerstone of MAFLD management. Clinical guidelines confirm that a body weight reduction of 3%-5% can ameliorate hepatic steatosis, a 7%-10% loss can improve MASH, and over 10% may even lead to fibrosis regression [3]. However, the real-world implementation of these interventions faces significant challenges [4,5].

    Traditional lifestyle interventions rely on face-to-face guidance from clinical doctors in hospital settings, which often results in poor execution and low compliance, limiting the coverage and sustainability of treatment and seriously affecting patient outcomes [6]. A cross-sectional survey from China revealed suboptimal self-management of lifestyle behaviors among adults with MAFLD [7]. Lack of motivation and real-time supervision are key contributing factors to the inadequate effectiveness of lifestyle interventions. In view of this, there is an urgent need for more convenient and effective measures to enhance the implementation and adherence of lifestyle changes for patients with MAFLD, thereby improving the prognosis.

    With the popularization of the internet and the rapid development of artificial intelligence, digital therapeutics (DTx), such as mobile health apps or programs, have emerged as a promising solution to bridge this care gap [8,9]. By delivering personalized health information and professional guidance, DTx enables convenient, home-based self-management, which can correct unhealthy lifestyle habits and improve treatment adherence [10-12]. A meta-analysis of 8 studies demonstrated that DTx interventions lasting 4‐24 months achieved a clinically significant weight loss at a rate of 33%, accompanied by improvements in liver enzymes and reduced liver fat [13]. In China, the near-ubiquitous use of WeChat offers a unique platform for such interventions [14,15]. Its widespread adoption eliminates the need for additional app downloads, fosters user trust through a familiar interface, and facilitates seamless communication between health care providers and patients, creating an ideal ecosystem for closed-loop chronic disease management [16,17].

    Despite this potential, there is a notable scarcity of WeChat mini-programs specifically designed and rigorously evaluated for structured lifestyle intervention in Chinese patients with MAFLD. To address this gap, we developed an innovative interactive WeChat mini-program, named the therapeutic lifestyle changes (TLCs) program, which is designed to assist patients in implementing sustainable lifestyle changes for the management of MAFLD. The program offers a suite of features, including weight loss goal setting, diet records and recommendations, and instructional exercise videos.

    Objectives

    Therefore, the aim of this study was to evaluate the efficacy of this novel TLC program in managing MAFLD. We hypothesized that, compared to participants receiving standard care, those engaged in the TLC intervention would demonstrate significantly greater improvements in reduction of liver fat content and body weight, as well as glucose and lipid metabolic parameters.


    Study Design and Setting

    A 6-month parallel, randomized controlled trial was conducted at Xinhua Hospital, affiliated with Shanghai Jiao Tong University School of Medicine, a tertiary-care hospital in China. The study period for participant recruitment and data collection spanned from August 2022 to May 2024 at our hospital. All eligible participants were randomly assigned to either the TLC program management group (intervention group) or the standard care group (control group) at a 1:1 ratio. The protocol (which was retrospectively registered) with no deviations was approved by the Institutional Review Board (IRB) prospectively. The reporting of this trial adhered to the CONSORT (Consolidated Standards of Reporting Trials) statement [18], with the checklist presented in Checklist 1.

    Participants

    The participants with clinically diagnosed MAFLD were consecutively recruited by clinicians from our hepatology clinic. Those who were interested provided written informed consent. All participants underwent an initial, face-to-face eligibility assessment at Xinhua Hospital according to the study criteria. The inclusion criteria were as follows: (1) aged 18‐65 years, (2) BMI between 24 and 35 kg/m2, (3) clinically diagnosed with MAFLD by radiological features with ultrasound confirmation and a controlled attenuation parameter (CAP) value of ≥248 dB/m, and (4) able to learn and operate a mobile-based program. Patients were excluded if they had excessive alcohol consumption (weekly ethanol intake of ≥210 g for males and ≥140 g for females weekly), etiological evidence of an alternative liver disease, such as chronic viral hepatitis, autoimmune liver diseases, drug-induced hepatic steatosis, hepatic decompensation, severe comorbidities (severe cardiopulmonary disease, uncontrolled diabetes, chronic kidney disease, psychiatric disease, and malignant tumors), recent weight loss, or weight loss medication intake.

    Randomization

    After the completion of baseline assessments, eligible participants were registered in the study. They were then automatically and randomly allocated to each study group at a 1:1 ratio by a computer-generated random system. The allocation was concealed until the moment of assignment, as the outcome was only revealed by the system after the participant’s formal registration. Given the nature of the digital intervention, participants and health care providers were not blinded to group assignment, but outcome assessors were blinded during data analysis.

    Lifestyle Intervention

    “Design of the TLC program”

    The TLC program-based lifestyle intervention was collaboratively designed by physicians from various departments, including Gastroenterology, Sport Rehabilitation, and Clinical Nutrition. Individuals in the control group were provided with standard care in the clinic according to the established treatment guidelines [19]. The TLC program intervention spanned 24 weeks and was divided into 4 distinct modules, which are described below.

    Development of Diet and Exercise Plans

    Individualized weight loss and exercise goals were set based on the baseline body weight, existing metabolic disorders, and severity of fatty liver disease. The target for weight reduction was established as 5%-10% of the initial body weight.

    Dietary Intervention Plan

    Based on preset weight loss goals, the program calculated daily dietary energy requirements, providing a recommended meal plan with the corresponding energy intake. If patients did not prefer the recipes suggested by the program, they could use the food exchange list to select alternative food with equivalent energy values. The program required patients to log and upload photos of their daily food intake, after which a human nutritionist analyzed the diet log and provided feedback via WeChat.

    Exercise Video Courses

    Participants were provided with a heart rate armband (HW702, YESOUL) to track heart rate during exercise. During each exercise session, data regarding heart rate were automatically uploaded by the armband. Each course lasted 30 minutes, and patients were to choose the training course at least 5 times per week. Additionally, patients could select other forms of exercise and record the duration of exercise, such as fast walking, running, cycling, and swimming.

    Monitoring and Information Support

    Patients were expected to complete their diet and exercise tasks through a check-in system every day. The research team sent tailored feedback messages weekly according to the status of task completion and offered regular suggestions through WeChat. Moreover, the system intelligently provided relevant educational articles, which provided professional information and support tailored to the patient’s interests and concerns.

    Measurements

    Clinical Characteristics

    Clinical characteristics regarding anthropometric and laboratory data were obtained at our hospital using a standard protocol at baseline and 6 months. Body composition was assessed via a bioimpedance analyzer (MC-980MA, TANITA). CAP and liver stiffness measurement (LSM) were performed using the FibroScan-502 device (Echosens) with an M or XL probe, following the manufacturer’s guidelines. A reliable examination was defined as at least 10 valid measurements and an IQR to median ratio of LSM below 30%. Based on an individual patient data meta-analysis, CAP thresholds of 268 and 280 dB/m were identified as optimal cutoffs for diagnosing moderate and severe steatosis, respectively [20].

    Dietary Nutrition and Physical Activity Assessment

    Dietary nutrition was assessed through a face-to-face interview with a questionnaire developed by the Chinese Society of Health Management and the Chinese Nutrition Society, which consists of 24 questions with a total score of 100 [21]. A score below 60 indicates a risk of dietary nutrition issues; a score between 60 and 75 suggests a potential risk; and a score above 75 indicates no dietary nutritional risk. Physical activity levels were evaluated via the International Physical Activity Questionnaire-Short Form (IPAQ-SF), which captures data on activity intensity, frequency, and daily duration in a typical week. Energy expenditure was quantified by the metabolic equivalent of task (MET) [22]. The weekly physical activity level for a specific intensity was calculated as the MET value of the activity× weekly frequency (d/wk)× daily duration (min/d).

    Diagnostic Criteria

    Obesity was considered a BMI≥28 kg/m2 [3]. The definition of MetS adhered to previously published criteria [23]. The homeostasis model assessment-insulin resistance (HOMA-IR) score was calculated as fasting insulin (mU/L)× fasting plasma glucose (FPG, mmol/L)/22.5. Insulin resistance was characterized by a HOMA-IR score of 2.5 or higher [1]. The fatty liver index was computed to indirectly assess the degree of hepatic steatosis using an established formula [24]. The FibroScan-aspartate aminotransferase (FAST) score was used to identify patients with MASH with significant fibrosis [25], whereas the aspartate aminotransferase-to-platelet ratio index (APRI), fibrosis-4 (FIB-4) index, nonalcoholic fatty liver disease fibrosis score (NFS), and Agile 3+ score were used to assess the risk of advanced fibrosis [26-29].

    Outcome Measurements

    The analyses included all participants who were randomized. The intention-to-treat (ITT) population for efficacy evaluation in this trial consisted of all cases that were randomized and had efficacy baseline records. The per-protocol (PP) population included participants who completed the entire planned observation period and had no major protocol deviations. The primary efficacy endpoint was the reduction of body weight by at least 5%. The secondary endpoints included a CAP value of ≥10% reduction or normalization and changes in the levels of liver enzymes and lipid and glucose parameters.

    Sample Size Calculation

    The sample size calculation was based on the primary study outcome, which is the proportion of patients achieving a weight loss of ≥5% from baseline. It was postulated that the intervention would produce a 5-fold increase in the success rate from a baseline of 8% in the control group [30]. To detect this effect with 80% power at a 5% significance level and accounting for a 20% dropout rate, a total sample size of 84 participants was required.

    Data Analysis

    Statistical analyses were performed using SPSS 23.0 software (IBM Corp). Missing data were due to patient dropouts and imputed by the last observation carried forward. Regarding the primary endpoint, this approach was chosen as it provides a conservative estimate of the treatment effect by assuming no further improvement in their weight status due to the discontinuation of active intervention, thereby avoiding overoptimistic extrapolation [6,31,32]. Descriptive statistics were conducted for all variables, with means (95% CIs) or medians (IQRs) reported for continuous variables and percentages for categorical variables. Categorical variables were analyzed using the chi-square test or Fisher exact test between groups (reporting relative risk and 95% CI) and the marginal homogeneity test within each group. Continuous variables were assessed using 2-tailed paired t tests or Wilcoxon signed-rank tests within each group, whereas unpaired t tests (reporting mean difference and 95% CI) or Mann-Whitney U tests (reporting Hodges-Lehmann median difference and 95% CI) were used to compare the changes from the baseline between groups. Binary logistic regression was used to evaluate the effect of the intervention on weight and CAP changes. The results are presented as odds ratios (ORs) with 95% CIs. A 2-tailed P value of .05 was considered significant.

    Ethical Considerations

    This research was approved by the IRB of Xinhua Hospital (XHEC-C-2021-076-2) and was retrospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500100197). This trial was registered retrospectively owing to disruptions in the research team’s workflow and uncertainties about the study’s continuity during the COVID-19 pandemic in China. We confirm that all data collection for this study was initiated subsequent to receiving approval from the IRB of Xinhua Hospital, with the IRB approval document provided in Multimedia Appendix 1. The study start was delayed due to COVID-19, and there were no other deviations or changes from the protocol after IRB approval and trial commencement. All eligible participants provided written informed consent. Participants were informed of their voluntary participation and their right to withdraw from the study at any time. It was clarified that all study procedures were offered at no cost, but no financial compensation would be made upon study completion. All personally identifiable information was coded with a unique study ID, and the anonymized data were securely stored in a password-protected cabinet.


    Participants and Baseline Characteristics

    Figure 1 presents the study flowchart in accordance with the CONSORT guidelines. A total of 89 patients met the inclusion criteria and were randomly assigned to the intervention group (n=45) or control group (n=44); these patients comprised the ITT population. At the 6-month follow-up point, 8 participants were lost to follow-up in the intervention group and 8 in the control group. A total of 73/89 (82%) patients successfully completed the study and comprised the PP population, while the remaining 16/89 (18%) patients dropped out (Figure 1). The baseline characteristics of the patients are shown in Table 1. The mean age of the whole study population was 40.0 (SD 9.8) years. Around 61.8% (55/89) were males and 55.1% (49/89) were obese (BMI≥28 kg/m2). Overall, 34.8% (31/89) had hypertension, 70.8% (63/89) had dyslipidemia, and 6.7% (6/89) had type 2 diabetes mellitus. There were no significant differences between the 2 groups in terms of age, sex, weight, presence of comorbidities, biochemical variables, diet, or activity-related parameters (P>.05). Additionally, no difference was found in the CAP or LSM between the groups at baseline (P>.05).

    ‎
    Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flow diagram.
    Table 1. Baseline characteristics of the total study population and the randomized allocation groups in intention-to-treat population.
    CharacteristicTotal
    (N=89)    		Control
    (n=44)    		Intervention
    (n=45)    		P valuea
    ale/female), n(%)55 (61.8)/ 34(38.2)31 (70.5)/ 13(29.5)24 (53.3)/ 21(46.7).10
    Age (years), mean (SD; 95% CI)40.0 (9.8; 37.9-42.1)40.7 (10.5; 37.5-43.9)39.3 (9.1; 36.6-42.1).84
    Weight (kg), mean (SD; 95% CI)82.5 (11.6; 80.1-85.0)82.3 (11.2; 78.9-85.7)82.7 (12.1; 79.1-86.4.92
    BMI (kg/m2), mean (SD; 95% CI)28.6 (3.0; 27.9-29.228.2 (3.0; 27.2-29.1)29.0 (2.9; 28.1-29.8).25
    BMI≥28 (kg/m2), n (%)49 (55.1)21 (47.7)28 (50.9).75
    WCb (cm), mean (SD; 95% CI)92.8 (8.9; 90.9-94.7)92.6 (8.9; 89.9-95.3)93.0 (9.0; 90.3-95.7).97
    Waist to hip ratio, mean (SD; 95% CI)0.90 (0.06; 0.89-0.92)0.91 (0.06; 0.89-0.93)0.90 (0.07; 0.88-0.92).38
    Waist to height ratio, mean (SD; 95% CI)0.55 (0.05; 0.54-0.56)0.54 (0.05; 0.53-0.56)0.55 (0.04; 0.54-0.56).45
    Systolic blood pressure (mmHg), mean (SD; 95% CI)123.8 (6.8; 122.4-125.3)124.4 (6.7; 122.4-126.5)123.3 (7.0; 121.2-125.4).24
    Diastolic blood pressure (mmHg), mean (SD; 95% CI)79.3 (5.7; 78.1-80.5)79.6 (5.9; 77.9-81.4)79.0 (5.6; 77.3-80.7).35
    Hypertension, n (%)31 (34.8)17 (38.6)14 (31.1).46
    Type 2 diabetes mellitus, n (%)6 (6.7)4 (9.1)2 (4.4).43
    Dyslipidemia, n (%)63 (70.8)33 (75)30 (66.7).39
    Hyperuricemia, n (%)34 (38.2)18 (40.9)16 (35.6).60
    Metabolic syndrome, n (%)37 (41.6)19 (43.2)18 (40).76
    Platelet count (109/L), mean (SD; 95% CI)256.6 (51.7; 245.8-267.5)254.5 (51.9; 238.7-270.3)258.8 (51.9; 243.2-274.4).58
    ALTc (U/L), median (IQR)36.0 (26.0-59.0)38.0 (24.5-57.3)34.0 (26.5-61.0).90
    ASTd (U/L), median (IQR)26.0 (21.0-36.0)26.0 (21.3-35.5)25.0 (19.5-36.0).68
    GGTe (U/L), median (IQR)35.0 (23.0-53.5)39.0 (25.3-59.8)34.0 (22.0-50.0).16
    Albumin (g/L), mean (SD; 95% CI)46.3 (2.4; 45.8-46.8)46.6 (2.0; 46.0-47.2)45.9 (2.6; 45.1-46.7).20
    Urea (mmol/L), mean (SD; 95% CI)4.9 (1.1; 4.7-5.1)5.0 (1.1; 4.6-5.3)4.9 (1.0; 4.6-5.2).75
    Serum creatinine (μmol/L), mean (SD; 95% CI)66.8 (14.4; 63.8-70.0)68.9 (14.0; 64.6-73.1)64.9 (14.6; 60.5-69.3).18
    Uric acid (μmol/L), mean (SD; 95% CI)410.7 (109.9; 387.5-433.8)416.1 (112.5; 381.9-450.2)405.4 (108.3; 372.9-437.9).62
    Fasting plasma glucose (mmol/L), mean (SD; 95% CI)5.3 (0.6; 5.2-5.5)5.4 (0.6; 5.2-5.5)5.3 (0.6; 5.1-5.5).54
    Fasting insulin (pmol/L), mean (SD; 95% CI)78.0 (39.3; 69.7-86.2)80.1 (45.6; 66.2-94.0)76.0 (32.4; 66.2-85.7).93
    HOMA-IRf, median (IQR)2.4 (1.7-3.3)2.4 (1.6-3.4)2.4 (1.7-3.2).96
    HOMA-IR≥2.5, n (%)40 (44.9)19 (43.2)21 (46.7).74
    Hemoglobin A1c (%), median (IQR)5.6 (5.4-5.8)5.6 (5.2-5.8)5.6 (5.4-5.9).36
    TGg (mmol/L), median (IQR)1.6 (1.0-2.0)1.6 (1.0-2.2)1.5 (1.0-1.8).43
    TCh (mmol/L), median (IQR)4.9 (4.2-5.3)4.9 (4.2-5.4)5.0 (4.3-5.4).78
    HDL-Ci (mmol/L), median (IQR)1.2 (1.0-1.3)1.1 (1.0-1.3)1.2 (1.0-1.4).26
    LDL-Cj (mmol/L), median (IQR)3.2 (2.6-3.7)3.3 (2.6-3.8)3.2 (2.5-3.6).68
    CAPk (dB/m), median (IQR)334 (295-365)336 (288-362)334 (297-369).25
    Liver stiffness measurement (kPa), median (IQR)5.1 (4.4-6.5)5.2 (4.4-6.4)4.9 (4.4-6.7).56
    Diet quality score, median (IQR)57 (50-66)56 (49-66)58 (51-66).64
    Total METl-minutes, median (IQR)1116 (544-2057)1118 (620-2068)1102 (465-2088).59

    aContinuous variables were assessed using unpaired t tests (normally distributed data) and Mann-Whitney U tests (non-normally distributed data) between groups. Categorical variables were analyzed using the chi-square test or Fisher exact test between groups.

    bWC: waist circumference.

    cALT: alanine aminotransferase.

    dAST: aspartate aminotransferase.

    eGGT: γ-glutamyl transpeptidase.

    fHOMA-IR: homeostasis model assessment-insulin resistance.

    gTG: triglyceride.

    hTC: total cholesterol.

    iHDL-C: high-density lipoprotein-cholesterol.

    jLDL-C: low-density lipoprotein-cholesterol.

    kCAP: controlled attenuation parameter.

    lMET: metabolic equivalent of task.

    Changes in Clinical Parameters

    Table 2 and Figure 2 present the clinical parameters at baseline and 24 weeks and their changes between groups in intention-to-treat population. There was a significant reduction in weight, waist circumference, waist-to-height ratio, and BMI from baseline within each group (P<.05). The reductions in these factors were also significantly greater in the intervention group than in the control group (P<.05). Patients in the TLC program intervention group had lower alanine aminotransferase (ALT), aspartate aminotransferase (AST), and γ-glutamyl transpeptidase (GGT) levels at 6 months than at baseline, and these liver enzyme levels decreased more in the intervention group than in the control group (P<.05). At 6 months, reductions in the levels of glucose metabolism–related factors, including fasting insulin, hemoglobin A1c (HbA1c), and HOMA-IR, were observed only within the intervention group (P<.05), and the levels of these factors showed significantly greater reductions in the intervention group than in the control group (P<.05). However, there was no between-group difference in FPG (P=.22). Patients in the intervention group had greater reductions in triglyceride (median difference −0.5, 95% CI −0.7 to −0.2; P=.001) and increases in high-density lipoprotein-cholesterol (HDL-C) levels than did those in the control group (median difference 0.1, 95% CI 0‐0.1; P=.02). The diet quality score improved more in the TLC program group than in the control group (median difference 6, 95% CI 0‐10; P=.04), whereas MET did not significantly differ between the groups (P=.92).

    Table 2. Clinical parameters between the two study groups at baseline and 24 weeks in the intention-to-treat population.
    ParameterControl (n=44)Intervention (n=45)Mean or Hodges-Lehmann median difference (95% CI)aP valuea
    Baseline24 weeksP valuebBaseline24 weeksP valueb
    Weight (kg), mean (SD; 95% CI)82.3 (11.2; 78.9-85.7)81.4 (12.2 ;77.7-85.0).0282.7 (12.1; 79.1-86.4)77.2 (11.7; 73.7-80.8)<.001–3.7 (–5.5 to –1.7)<.001
    BMI (kg/m2), mean (SD; 95% CI)28.2 (3.0; 27.2-29.1)27.5 (3.4; 26.5-28.6).00429.0 (2.9; 28.1-29.8)26.9 (2.9; 26.0-27.8)<.001–1.4 (–2.1 to –0.7)<.001
    WCc(cm), mean (SD; 95% CI)92.6 (8.9; 89.9-95.3)90.6 (8.6; 88.0-93.2).00393.0 (9.0; 90.3-95.7)85.3 (9.5; 82.7-88.4)<.001–5.0 (–7.0 to –2.0)<.001
    Waist to hip ratio, mean (SD; 95% CI)0.91 (0.06; 0.89-0.93)0.90 (0.06; 0.89-0.92).210.90 (0.07; 0.88-0.92)0.87 (0.07; 0.85-0.89)<.001–0.02 (–0.04 to 0.01).06
    Waist to height ratio, mean (SD; 95% CI)0.54 (0.05; 0.53-0.56)0.53 (0.05; 0.52-0.55).0030.55 (0.04; 0.54-0.56)0.50 (0.05; 0.49-0.52)<.001–0.03 (–0.04 to –0.02)<.001
    Systolic blood pressure (mmHg), mean (SD; 95% CI)124.4 (6.7; 122.4-126.5)125.6 (9.9; 122.6-128.6).98123.3 (7.0; 121.2-125.4)123.4 (11.3; 120.0-126.8).17–1.0 (–6.0 to 4.0).87
    Diastolic blood pressure (mmHg), mean (SD; 95% CI)79.6 (5.9; 77.9-81.4)81.1 (5.9; 79.3-82.9).1879.0 (5.6; 77.3-80.7)79.6 (5.8; 77.8-81.3).47–0.9 (–3.7 to 1.9).62
    Platelet count (109/L), mean (SD; 95% CI)254.5 (51.9; 238.7-270.3)256.0 (66.5; 235.7-276.2).92258.8 (51.9; 243.2-274.4)250.0 (44.3; 237.0-262.0).03–12.8 (–26.7 to 1.1).07
    ALTd (U/L), median (IQR)38.0 (24.5-57.3)30.0 (22.3-44.3).0134.0 (26.5-61.0)23.0 (17.0-33.5)<.001–8.0 (–15.0 to –1.0).02
    ASTe (U/L), median (IQR)26.0 (21.3-35.5)22.0 (19.0-27.8).00325.0 (19.5-36.0)19.0 (15.0-23.0)<.001–3.0 (–6.0 to 0).04
    GGTf (U/L), median (IQR)39.0 (25.3-59.8)41.0 (25.3-57.5).7634.0 (22.0-50.0)23.0 (16.5-34.5)<.001–8.0 (–14.0 to –3.0)<.001
    Albumin (g/L), mean (SD; 95% CI)46.6 (2.0; 46.0-47.2)45.7 (2.1; 44.8-46.0).00145.9 (2.6; 45.1-46.7)45.1 (4.0; 43.9-46.3).150.4 (–0.8 to 1.6).06
    Urea (mmol/L), mean (SD; 95% CI)5.0 (1.1; 4.6-5.3)5.0 (1.2; 4.6-5.4).404.9 (1.0; 4.6-5.2)5.0 (1.1; 4.7-5.3).610.1 (–0.3 to 0.4).95
    Serum creatinine (μmol/L), mean (SD; 95% CI)68.9 (14.0; 64.6-73.1)68.7 (13.8; 64.5-72.9).8564.9 (14.6; 60.5-69.3)64.5 (14.0; 60.3-68.7).78–0.2 (–2.2 to 1.8).79
    Uric acid (μmol/L), mean (SD; 95% CI)416.1 (112.5; 381.9-450.2)404.1 (75.2; 380.4-427.2).57405.4 (108.3; 372.9-437.9)386.8 (105.6; 355.0-418.5).008–6.6 (–36.1 to 22.8).16
    Fasting plasma glucose (mmol/L), mean (SD; 95% CI)5.4 (0.6; 5.2-5.5)5.5 (0.9; 5.3-5.8).395.3 (0.6; 5.1-5.5)5.3 (0.5; 5.1-5.4).24–0.1 (–0.5 to 0.3).22
    Fasting insulin (pmol/L), mean (SD; 95% CI)80.1 (45.6; 66.2-94.0)81.0 (40.8; 68.6-93.4).3576.0 (32.4; 66.2-85.7)60.1 (34.8; 49.6-70.6).001–16.8 (–30.0 to –3.5).002
    HOMA-IRg, median (IQR)2.4 (1.6-3.4)2.7 (1.9-3.7).252.4 (1.7-3.2)1.7 (1.1-2.8).001–0.7 (–1.1 to –0.2).002
    HbA1ch (%), median (IQR)5.6 (5.2-5.8)5.5 (5.3-5.7).585.6 (5.4-5.9)5.4 (5.3-5.7).003–0.1 (–0.2 to 0).047
    TGi (mmol/L), median (IQR)1.6 (1.0-2.2)1.9 (1.2-2.4).091.5 (1.0-1.8)1.0 (0.8-1.5).001–0.5 (–0.7 to –0.2).001
    TCj (mmol/L), median (IQR)4.9 (4.2-5.4)4.6 (4.0-5.2).155.0 (4.3-5.4)4.5 (3.9-5.1).009–0.1 (–0.4 to 0.1).32
    HDL-Ck (mmol/L), median (IQR)1.1 (1.0-1.3)1.1 (0.9-1.3).211.2 (1.0-1.4)1.3 (1.0-1.5).070.1 (0-0.1).02
    LDL-Cl (mmol/L), median (IQR)3.3 (2.6-3.8)3.1 (2.6-3.6).063.2 (2.5-3.6)3.1 (2.3-3.5).04–0.1 (–0.2 to 0.1).75
    CAPm (dB/m), median (IQR)336 (288-362)310 (288-342).04334 (297-369)291 (239-317)<.001–38 (–62 to –16).001
    CAP<268, n (%)4 (9.1)6 (13.6).551 (2.2)16 (35.6)<.001—n—
    268≤CAP<280, n (%)4 (9.1)3 (6.8)—1 (2.2)4 (8.9)———
    CAP≥280, n (%)36 (81.8)35 (79.5)—43 (95.6)25 (55.6)———
    Fatty liver index, mean (SD; 95% CI)58.4 (22.3; 51.6-65.2)57.1 (22.2; 50.3-63.8).3856.7 (21.8; 50.2-63.3)34.2 (22.3; 27.5-40.9)<.001–21.1 (–28.4 to –13.9)<.001
    Liver stiffness measurement (kPa), median (IQR)5.2 (4.4-6.4)5.1 (4.1-6.8).074.9 (4.4-6.7)5.1 (4.0-6.0).16–0.1 (–0.6 to 0.5).94
    AST-to-platelet ratio index, median (IQR)0.28 (0.20-0.37)0.24 (0.18-0.30).0080.27 (0.19-0.35)0.21 (0.16-0.27)<.001–0.03 (–0.06 to 0.01).13
    Fibrosis-4 index, median (IQR)0.68 (0.50-1.00)0.66 (0.46-0.99).200.64 (0.49-0.90)0.62 (0.49-0.76).15–0.05 (–0.12 to 0.03).85
    NAFLDo fibrosis score, mean (SD; 95% CI)–2.9 (1.2; –3.2 to –2.5)–2.6 (1.2; –3.0 to –2.3).09–2.9 (1.0; –3.2 to –2.6)–2.7 (1.0; –3.0 to –2.4).15–0.1 (–0.4 to 0.3).96
    FASTp, median (IQR)0.20 (0.11-0.42)0.11 (0.07-0.22).0010.22 (0.09-0.43)0.06 (0.03-0.15)<.001–0.06 (–0.13 to –0.01).02
    Agile3+, median (IQR)0.06 (0.02-0.15)0.05 (0.02-0.15).960.05 (0.02-0.12)0.05 (0.03-0.12).630.01 (–0.02 to 0.03).68
    Diet quality score, median (IQR)56 (49-66)64 (55-71).0158 (51-66)66 (61-76)<.0016 (0-10).04
    Total METq-minutes, median (IQR)1118 (620-2068)1442 (620-2608).291102 (465-2088)1102 (647-1725).44–10 (–204 to 192).92

    aContinuous variables were assessed using 2-tailed unpaired t tests (reporting mean difference and 95% CI) for normally distributed data and Mann-Whitney U tests (reporting Hodges-Lehmann median difference and 95% CI) for non-normally distributed data to compare the changes between the two groups.

    bContinuous variables were assessed using paired t tests (normally distributed data) or Wilcoxon signed-rank tests (non-normally distributed data) within each group. Categorical variables were analyzed using marginal homogeneity tests within each group.

    cWC: waist circumference.

    dALT: alanine aminotransferase.

    eAST: aspartate aminotransferase.

    fGGT: γ-glutamyl transpeptidase.

    gHOMA-IR: homeostasis model assessment-insulin resistance.

    hHbA1c: hemoglobin A1c.

    iTG: triglyceride.

    jTC: total cholesterol.

    kHDL-C: high-density lipoprotein-cholesterol.

    lLDL-C: low-density lipoprotein-cholesterol.

    mCAP: controlled attenuation parameter.

    nNot applicable.

    oNAFLD: nonalcoholic fatty liver disease.

    pFAST: FibroScan-aspartate aminotransferase.

    qMET: metabolic equivalent of task.

    ‎
    Figure 2. Change in clinical parameters from baseline to 24 weeks between control group and intervention group. (A) Median change in weight between groups (–0.3 vs –4.4; median difference –3.7, 95% CI –5.5 to –1.7; P<.001). (B) Median change in waist circumference between groups (–1.5 vs –7.0; median difference –5.0, 95% CI –7.0 to –2.0; P<.001). (C) Median change in alanine aminotransferase values between groups (–2.5 vs –12.0; median difference –8.0, 95% CI –15.0 to –1.0; P=.02). (D) Median change in aspartate aminotransferase values between groups (–2.0 vs –6.0; median difference –3.0, 95% CI –6.0 to 0; P=.04). (E) Median change in hemoglobin A1c between groups (0 vs –0.1; median difference –0.1, 95% CI –0.2 to 0; P=.047). (F) Median change in homeostasis model assessment-insulin resistance between groups (0 vs –0.6; median difference –0.7, 95% CI –1.1 to –0.2; P=.002). (G) Median change in triglyceride values between groups (0 vs –0.3; median difference –0.5, 95% CI –0.7 to –0.2; P=.001). (H) Median change in high-density lipoprotein-cholesterol values between groups (0 vs 0.1; median difference 0.1, 95% CI 0-0.1; P =.02). (I) Median change in controlled attenuation parameter values between groups (–4.0 vs –56.0; median difference –38, 95% CI –62 to –16; P=.001). (J) Median change in FibroScan-aspartate aminotransferase scores between groups (–0.04 vs –0.11; median difference –0.06, 95% CI –0.13 to –0.01; P =.02). ALT: alanine aminotransferase; AST: aspartate aminotransferase; CAP: controlled attenuation parameter; FAST: FibroScan–aspartate aminotransferase; HbA1c: hemoglobin A1c; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment–insulin resistance; TG: triglyceride; WC: waist circumference.

    Changes in Noninvasive Factors of Hepatic Steatosis and Liver Fibrosis

    At 6 months, both groups presented a reduction in the CAP value. The intervention group experienced a decrease to 291(239-317) dB/m, whereas the control group had a median CAP of 310 (288-342)dB/m (Table 2, P=.001). CAP values were significantly lower in the intervention group than in the control group (Figure 2, median difference −38, 95% CI −62 to −16; P=.001). In the subgroup analysis, the percentage of patients with severe steatosis decreased significantly in the intervention group from baseline at 6 months (P<.001). However, the LSM values were not significantly different between groups or within each group (P>.05). Based on the noninvasive assessment, patients in the TLC program intervention group had lower FAST scores than did those in the control group (median difference −0.06, 95% CI −0.13 to −0.01; P=.02). In the comparison of these 2 groups, no significant differences were found in noninvasive scores, including the APRI, FIB-4, NFS, and Agile 3+ score (P>.05).

    Changes in Body Composition

    According to the body composition analysis (Table 3), each group presented a decrease in total body fat mass, total body fat percentage, and visceral fat level, and these variables decreased more in the intervention group than in the control group across the 6-month period (P<.05). The basal metabolic rate underwent a greater reduction in the intervention group than in the control group (mean difference −42, 95% CI −62 to −22; P<.001). The intervention group also presented more pronounced decreases in fat-free mass, total lean body mass, and appendicular skeletal muscle mass (ASM; P<.05). The ratio of ASM to weight increased in the intervention group, but no significant difference was observed in this ratio between the groups (P>.05).

    Table 3. Body composition analysis between the two study groups at baseline and 24 weeks in the intention-to-treat population.
    Body compositionControl
    n=44    		Intervention
    n=45    		Mean or Hodges-Lehmann median difference (95% CI)aP valuea
    Baseline24weeksP valuebBaseline24 weeksP valueb
    Total body fat mass (kg), mean (SD; 95% CI)25.3 (6.8; 23.3-27.4)24.3 (7.0; 22.2-26.4).00628.3 (7.3; 26.2-30.6)24.6 (7.3; 22.4-26.7)<.001–2.8 (–4.4 to –1.2)<.001
    Total body fat percentage (%), mean (SD; 95% CI)30.7 (6.9; 28.6-32.8)29.9 (6.8; 27.9-32.0).00234.3 (7.8; 32.0-36.7)30.9 (8.5; 28.3-33.5)<.001–2.6 (–4.1 to –1.2)<.001
    Visceral fat level, median (IQR)13 (11-16)12 (10-14).00112 (10-15)10 (8-13)<.001–1 (–2 to 0)<.001
    Fat-free mass (kg), mean (SD; 95% CI)57.0 (9.1; 54.2-59.7)57.1 (9.7; 54.1-60.0).4454.3 (10.6; 51.1-57.5)52.7 (10.1; 49.6-55.7)<.001–1.1 (–1.9 to –0.2).01
    Total lean body mass (kg), mean (SD; 95% CI)53.4 (9.2; 50.6-56.1)52.7 (9.3; 49.9-55.5).2951.0 (10.0; 48.0-54.0)49.9 (9.9; 47.0-52.9)<.001–0.7 (–1.2 to –0.2).002
    ASMc (kg), mean (SD; 95% CI)25.5 (5.4; 23.8-27.1)25.0 (5.8; 23.2-26.8).7024.8 (5.9; 23.0-26.6)23.9 (5.7; 22.2-25.6)<.001–0.6 (–1.0 to –0.3).001
    ASM to weight ratio (%), mean (SD; 95% CI)30.7 (3.8; 29.6-31.9)30.8 (5.3; 29.1-32.3).00929.8 (4.4; 28.5-31.1)30.8 (4.5; 29.5-32.1)<.0010.5 (–0.1 to 1.1).08
    Trunk muscle mass (kg), mean (SD; 95% CI)27.9 (4.1; 26.6-29.1)27.7 (4.1; 26.5-29.0).3126.2 (4.5; 24.9-27.6)26.0 (4.4; 24.7-27.3).06–0.1 (–0.5 to 0.1).27
    Total body water (kg), mean (SD; 95% CI)38.1 (5.3; 36.5-39.7)38.2 (5.4; 36.6-39.8).5437.7 (5.6; 36.1-39.5)36.9 (5.6; 35.2-38.6).001–0.9 (–1.5 to –0.4).003
    Total body water percentage (%), mean (SD; 95% CI)47.0 (3.9; 45.8-48.1)47.1 (5.1; 45.5-48.6).0746.0 (4.0; 44.9-47.3)47.8 (5.2; 46.3-49.4)<.0011.6 (0.2-3.0).001
    Extracellular water percentage (%), mean (SD; 95% CI)41.2 (2.1; 40.5-41.8)40.9 (2.0; 40.3-41.5).00341.4 (2.2; 40.7-42.0)40.8 (2.0; 40.2-41.4).001–0.2 (–0.6 to 0.1).10
    BMCd (kcal), mean (SD; 95% CI)1631 (252; 1555-1708)1621 (251; 1545-1698).271597 (273; 1515-1679)1545 (274; 1463-1627)<.001–42 (–62 to –22)<.001

    aContinuous variables were assessed using unpaired t tests (reporting mean difference and 95% CI) for normally distributed data and Mann-Whitney U tests (reporting Hodges-Lehmann median difference and 95% CI) for non-normally distributed data to compare the changes between the two groups.

    bContinuous variables were assessed using paired t tests (normally distributed data) or Wilcoxon signed-rank tests (non-normally distributed data) within each group.

    cASM: appendicular skeletal muscle mass.

    dBMC: basal metabolic rate.

    Analysis of Primary and Secondary Efficacy Outcomes

    As shown in Figure 3, the primary and secondary efficacy outcomes were compared between the control and intervention groups. According to the ITT analysis Figure 3, 60% (27/45) of patients achieved weight loss of ≥5% (27/45 vs 7/44; RR 3.771, 95% CI 1.836‐7.748; P<.001), and 24.4 % (11/45) of patients attained weight loss of ≥10% among patients in the TLC program-delivered intervention group, which was greater than those in the control group (11/45 vs 3/44; RR 3.585, 95% CI 1.072‐11.988; P=.02). A CAP value of ≥10% reduction or normalization was achieved in more patients who received the intervention than in those who received standard care (26/45 vs 14/44; RR 1.816, 95% CI 1.102‐2.992; P=.01). Furthermore, patients in the intervention group were more likely than those in the control group to achieve an ALT reduction of ≥50% or normalization (30/45 vs 16/44; RR 1.833, 95% CI 1.178‐2.853; P=.004), as well as triglyceride normalization (37/45 vs 20/44; RR 1.809, 95% CI 1.273‐2.570; P<.001). Similarly, these results from the PP analysis were in line with those from the ITT analysis (Figure 3).

    The effects of the TLC program on weight loss and the CAP value were also analyzed by multivariate regression (Table 4). Among the ITT population or PP population, the intervention had a positive effect on weight loss of ≥5% or ≥10% after adjusting for age, sex, and baseline weight (P<.05). Additionally, patients in the intervention group were more likely to achieve a CAP reduction of ≥10% or normalization after adjusting for age, sex, and baseline CAP in both the ITT population (OR 2.853, 95% CI 1.092‐7.456; P=.03) and the PP population (OR 3.319, 95% CI 1.117‐9.860; P=.03).

    ‎
    Figure 3. Analysis of primary and secondary efficacy outcome between control group and intervention group. (A) The percentage of patients with weight loss ≥5% between groups in the intention-to-treat population (27/45 vs 7/44, RR 3.771, 95% CI 1.836-7.748; P<.001 ) and the per-protocol population (26/37 vs 5/36, RR 5.059, 95% CI 2.184-11.719; P <.001) ; (B) The percentage of patients with weight loss ≥10% between groups in the intention-to-treat population (11/45 vs 3/44, RR 3.585, 95% CI 1.072-11.988; P=.02) and the per-protocol population (10/37 vs 3/36, RR 3.243, 95% CI 1.066-10.360; P=.04) ; (C) The percentage of patients with controlled attenuation parameter ≥10% reduction or normalization between groups in the intention-to-treat population (26/45 vs 14/44, RR 1.816, 95% CI 1.102-2.992; P=.01 ) and the per-protocol population (25/37 vs 13/36, RR 1.871, 95% CI: 1.148-3.050; P=.007; (D) The percentage of patients with alanine aminotransferase ≥50% reduction or normalization between groups in the intention-to-treat population (30/45 vs 16/44, RR 1.833, 95% CI 1.178-2.853; P=.004 ) and the per-protocol population (25/37 vs 16/36, RR 1.520, 95% CI 1.006-2.391; P=.047; (E) The percentage of patients with triglyceride normalization between groups in the intention-to-treat population (37/45 vs 20/44, RR 1.809, 95% CI 1.273-2.570; P<.001 ) and the per-protocol population (32/37 vs 17/36, RR 1.831, 95% CI 1.267-2.646; P<.001). ALT: aspartate aminotransferase; CAP: controlled attenuation parameter; ITT: intention-to-treat; PP: per-protocol; TG: triglyceride.
    Table 4. Multivariate logistic analysis model of reduction in weight and controlled attenuation parameter value in intention-to-treat and per-protocol population.
    ParameterUnadjustedModel 1aModel 2bUnadjustedModel 1Model 2
    ITTc population (N=89)PPd population (N=73)
    Weight loss≥5%
    Controlreferencereferencereferencereferencereferencereference
    Intervention, OR (95% CI)7.929
    (2.905-21.641)    		8.537
    (2.947-24.730)    		8.380
    (2.886-24.331)    		14.655
    (4.509-47.626)    		14.534
    (4.267-49.499)    		15.231
    (4.380-52.969)
    P value＜.001＜.001＜.001＜.001＜.001＜.001
    Weight loss ≥10%
    Controlreferencereferencereferencereferencereferencereference
    Intervention, ORe (95% CI)4.422
    (1.140‐17.144)    		4.677
    (1.160‐18.855)    		4.612
    (1.138‐18.686)    		4.074
    (1.018‐16.305)    		4.604
    (1.093‐19.386)    		4.571
    (1.078‐19.387)
    P value.03.03.03.047.04.04
    CAPf≥10% reduction or normalization
    referencereferencereferencereferencereferencereference
    Intervention, OR (95% CI)2.932
    (1.232-6.981)    		2.843
    (1.161-6.965)    		2.853
    (1.092-7.456)    		3.686
    (1.401-9.700)    		3.399
    (1.259-9.175)    		3.319
    (1.117-9.860)
    P value.02.02.03.008.02.03

    aModel 1: Adjusted for age, sex.

    bModel 2: Adjusted for model 1 and baseline weight.

    cITT: intention-to-treat.

    dPP: per-protocol population.

    eOR: odds ratio.

    fModel 2：Adjusted for model 1 and baseline controlled attenuation parameter value; CAP: controlled attenuation parameter.

    Safety

    During the study period, the intervention was safe in this cohort of patients with MAFLD, with no adverse events reported during the study period. The heart rate monitoring during the video exercises effectively protected participants from the risk of overexertion.


    Principal Findings

    MAFLD is a public health concern and is associated with unhealthy lifestyles. Effective weight control is a crucial measure for reversing MASH and attenuating metabolic dysfunction [11,33]. While digital health tools offer a scalable solution, there is a notable scarcity of interventions specifically designed for the population of individuals with MAFLD in China. Therefore, we conducted a longitudinal study to evaluate an innovative WeChat mini-program-based lifestyle modification in patients with MAFLD. Our study revealed that the TLC program intervention was more effective for weight loss among patients with MAFLD who were overweight or obese than standard care was. Importantly, CAP values decreased significantly more in the intervention group than in the control group, suggesting that the TLC program intervention ameliorated hepatic steatosis. Along with weight reduction, the intervention group presented significant decreases in liver enzymes, HbA1c, and triglyceride levels, indicating improved liver function and metabolic-related indicators. To investigate the changes in body composition, we found that the intervention group had a lower total body fat percentage and visceral fat level than did the control group, with parallel improvement in hepatic steatosis. In terms of liver fibrosis assessment, only the FAST score showed improvement in MASH with significant fibrosis after the intervention, whereas other fibrosis indicators, such as LSM and noninvasive scores (APRI, FIB-4, NFS, Agile 3+) showed no treatment effect.

    Based on established evidence, DTx lifestyle interventions lead to significant weight loss in patients with MAFLD [6,30]. According to a meta-analysis of 8 cohort studies on DTx, 33% of patients with MAFLD experienced significant weight loss of ≥5% [13]. In our study, more than half (27/45, 60%) of the patients with MAFLD achieved a weight loss of ≥5% , demonstrating the efficacy of weight management over 6 months. The significant improvement in diet quality scores observed in the intervention group implies a potential key role of dietary modification within the TLC program. With the reduction in body weight, there was a significant improvement in liver chemistry variables, including ALT, AST, and GGT, which was consistent with published data [30,34]. In particular, approximately two-thirds of the patients with TLC intervention experienced an ALT reduction of more than 50% or normalization, suggesting that this digital intervention improved liver function remarkably.

    Importantly, 57.8% (26/45) of the patients in the intervention group presented a CAP value of more than a 10% reduction or normalization, indicating that the severity of hepatic steatosis was highly reduced among individuals after intervention. A similar study investigated a mobile technology–based intervention program for patients with MAFLD and demonstrated that 42.4% (14/33) of participants exhibited a reduction in CAP values, with 21.2% (7/33) achieving a significant decrease of ≥10% in the CAP value [35]. Based on our study, the reduction in body weight and visceral fat led to a marked decline in hepatic fat content and subsequent improvement in liver function.

    Among patients with MAFLD, the severity of fibrosis is a major disease modifier and affects hepatic outcomes [36,37]. Gradual weight loss of more than 10% may ameliorate fibrosis in patients with MAFLD [38]. In our study, although the proportion of patients with weight loss greater than 10% reached 24.4% after intervention, the LSM value was not significantly different between the groups. Similarly, no significant difference was found in noninvasive scores, such as the APRI, FIB-4 score, NFS, and Agile 3+ score, except for the FAST score. With respect to noninvasive assessment of liver fibrosis, current evidence from different studies has shown inconsistencies. A German web-based exercise program intervention for patients with MAFLD reported improved APRI, FIB-4 score, and LSM, whereas a US mobile-based program intervention for patients with biopsy-confirmed MAFLD showed no changes in NFS or FIB-4 score [39,40]. The observed discrepancies across studies may stem from variations in intervention protocols, differential durations of implementation, and heterogeneity in patient adherence levels. Based on our data, the overall baseline LSM values of the patients enrolled seem to be remarkably low, with a median value of only 5.1 (IQR 4.4- 6.5) kPa. Furthermore, the duration of the intervention was only 6 months, which is too short for observing changes in fibrosis.

    FAST scores reflect MASH with significant fibrosis and are calculated via the CAP, LSM, and AST. In our study, the observed reduction in the FAST score in the intervention group may be associated with the significant decreases in CAP and AST levels, suggesting potential alleviation of MASH to some extent by using the TLC program. Undoubtedly, the definitive evidence of MASH alleviation is contingent upon further study with patients who have undergone liver biopsies.

    MAFLD results in impaired metabolic profiles in multiple organ systems, which not only promotes the progression of liver disease but also increases the risk of extrahepatic complications [41,42]. In the current investigation, we assessed key metrics of glucose metabolism, observing substantially greater reductions in fasting insulin, HbA1c, and HOMA-IR levels among intervention group participants, which aligns with established findings in the literature [35,39]. Nevertheless, there was no significant difference in the FPG level between these two groups. It is important to emphasize that our cohort consisted predominantly of nondiabetic individuals, with an overall low baseline glucose (5.3 mmol/L), which limited the potential for a large absolute reduction. Interestingly, some studies concerning DTx-mediated lifestyle modifications in MAFLD management have reported similar results [34,40]. Indeed, insulin resistance is the key pathogenic link between obesity and associated metabolic disorders. Moreover, it is the primary driver in the development of MAFLD and cardiovascular disease. A significant reduction in body weight has favorable effects on increasing insulin sensitivity and improving insulin resistance, leading to a reduction in fasting insulin and glycemic fluctuations, substantially reducing cardiovascular events in patients with MAFLD [43]. A reduction in the HbA1c level, reflecting good glycemic control over the past 2‐3 months, may lead to decreased hepatic fat accumulation, reduced liver fat synthesis, and increased fat breakdown [44].

    For the lipid profile, decreased TG or elevated HDL-C levels were found in the intervention group, which was in line with previously published literature [35,45,46]. Favorable modifications in TG or HDL profiles have been correlated with the amelioration of insulin resistance, which not only upregulates lipoprotein lipase expression and accelerates TG hydrolysis but also downregulates hepatic lipase activity, reduces HDL-C degradation, and promotes HDL-C synthesis [41,47]. Overall, TLC digital intervention markedly improved metabolic indices, such as glucose and lipid levels, which are pivotal determinants of cardiovascular morbidity and mortality in patients with MAFLD.

    To evaluate the quality of weight loss, we explored parameters involved in body composition. Compared with the control group, the intervention group presented greater reductions in total body fat percentage and visceral fat, as well as the alleviation of hepatic steatosis. Although the intervention group demonstrated significant reductions in total lean body mass and absolute ASM, the ratio of ASM to weight increased within the intervention group, suggesting that weight loss was driven primarily by fat reduction. As is well known, healthy weight loss involves fat loss with muscle gain. Furthermore, no significant difference in MET levels was detected between the groups in our study. To build muscle and improve body composition, future improvements to the TLC program should integrate progressive strength training, such as bodyweight exercises, resistance bands, and dumbbells. Monitoring changes in muscle mass using body fat scales can further support the personalization and optimization of these interventions.

    Strengths and Limitations

    This study has several strengths. Its innovation lies in being the first tailored digital tool to address the specific self-management challenges of this population in China, leveraging a widely accessible platform. Additionally, this mobile-based program offers customized dietary plans based on each patient’s health status. Notably, this smartphone-delivered lifestyle program aids in weight control and liver fat reduction by facilitating healthier behaviors, which could be implemented in home settings. This high accessibility removes common barriers to care, such as transportation and time constraints, while its timely, personalized feedback promotes greater adherence and sustained habit formation. Together, these advantages make it a scalable and cost-effective strategy for the long-term management of MAFLD.

    This study has several limitations. First, all the patients enrolled were from a single tertiary hepatology clinic, and more motivated individuals tended to self-select into the program. This selection bias was inevitable. Second, hepatic steatosis and fibrosis were evaluated by FibroScan or noninvasive tests rather than liver biopsy. On the one hand, FibroScan is a widely used and noninvasive tool to detect hepatic steatosis and screen for fibrosis. On the other hand, it is extremely challenging to obtain liver biopsy specimens in clinical practice. Third, no postintervention follow-up was conducted to assess the maintenance of weight loss. The lack of long-term follow-up data limits the conclusions regarding the sustainability of the WeChat-based intervention effects. Future research will include a follow-up phase where the research team will formulate personalized weight-management plans and encourage the involvement of family or friends for support and supervision to facilitate long-term weight maintenance for participants. Additionally, the sample size herein was limited, and the dropout rate was relatively high (18%). The application of last observation carried forward for handling missing data may have introduced bias into the treatment effect estimate. These factors, collectively, may have contributed to the imprecision in our effect estimates, as evidenced by the wide CIs. Subsequent study should incorporate larger, multicenter cohorts and advanced modeling approaches. Furthermore, the mini-program itself will be further developed to integrate incentives, gamification, and social support. This comprehensive strategy aims to enhance user adherence, boost engagement, and reduce dropout rates.

    Conclusions

    Our study introduces a WeChat mini-program that delivers a tailored lifestyle intervention designed for patients with MAFLD in China, addressing a gap in scalable management. Our intervention is characterized by its ability to not only reduce weight but also to significantly improve hepatic steatosis and metabolic measures. These findings contribute to the field by establishing a feasible, low-cost model for managing MAFLD in high-burden populations, with strong potential for implementation in primary care and resource-limited settings. Given the imprecision of effect estimates, further larger-scale studies are essential to confirm these preliminary findings, obtain more precise estimates, and evaluate sustained long-term clinical outcomes with long-term follow-up.

    Acknowledgments

    This study was supported by grants from Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0508704), the Construction Project of the “Discipline Peak-Climbing Plan” from Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (XKPF2024B400). The authors thank all the participants and clinicians who participated in our study. We would like to thank Hangzhou Jianhai Technology Company for providing supportive digital technology.

    Artificial intelligence (AI) was not used in any portion of the manuscript writing.

    Funding

    The funder had no involvement in the study design, data collection, analysis, interpretation, or the writing of the manuscript.

    Data Availability

    The datasets generated and analyzed in the current study are available from the corresponding author upon reasonable request.

    Authors' Contributions

    SC and FJG contributed to the conceptualization of the study. SC, SCC, CHX, YRX, ZJ, and DXY were responsible for the investigation and data curation. CGY and SX guided the methodology. FJG and SX provided resources and secured funding. SC, SX, and FJG were responsible for the program development, project administration, and supervision. SC and FJG contributed to formal analysis and writing-original draft. SC, CGY, SCC, CHX, YRX, ZJ, DXY SX, and FJG carried out validation, writing-review, and editing of the manuscript.

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    Institutional Review Board approval document.

    PDF File, 2513 KB

    Checklist 1

    CONSORT eHealth checklist (V1.6.1).

    PDF File, 37271 KB
    1. Eslam M, Fan JG, Yu ML, et al. The Asian Pacific association for the study of the liver clinical practice guidelines for the diagnosis and management of metabolic dysfunction-associated fatty liver disease. Hepatol Int. Apr 2025;19(2):261-301. [CrossRef] [Medline]
    2. European Association for the Study of the Liver (EASL), European Association for the Study of Diabetes (EASD), European Association for the Study of Obesity (EASO). EASL-EASD-EASO Clinical Practice Guidelines on the Management of Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD). Obes Facts. 2024;17(4):374-444. [CrossRef] [Medline]
    3. Fan JG, Xu XY, Yang RX, et al. Guideline for the Prevention and Treatment of Metabolic Dysfunction-associated Fatty Liver Disease (Version 2024). J Clin Transl Hepatol. Nov 28, 2024;12(11):955-974. [CrossRef] [Medline]
    4. Seeberg KA, Borgeraas H, Hofsø D, et al. Gastric bypass versus sleeve gastrectomy in type 2 diabetes: effects on hepatic steatosis and fibrosis: a randomized controlled trial. Ann Intern Med. Jan 2022;175(1):74-83. [CrossRef] [Medline]
    5. Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD Practice Guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. May 1, 2023;77(5):1797-1835. [CrossRef] [Medline]
    6. Pfirrmann D, Huber Y, Schattenberg JM, Simon P. Web-based exercise as an effective complementary treatment for patients with nonalcoholic fatty liver disease: intervention study. J Med Internet Res. Jan 2, 2019;21(1):e11250. [CrossRef] [Medline]
    7. Zhou R, Zhang B, Zhang W, et al. Self-management behaviours in adults with non-alcoholic fatty liver disease: a cross-sectional survey from China. BMJ Open. Feb 22, 2024;14(2):e078333. [CrossRef] [Medline]
    8. Tincopa MA, Patel N, Shahab A, Asefa H, Lok AS. Implementation of a randomized mobile-technology lifestyle program in individuals with nonalcoholic fatty liver disease. Sci Rep. Mar 28, 2024;14(1):7452. [CrossRef] [Medline]
    9. Zhou R, Gu Y, Zhang B, et al. Digital therapeutics: emerging new therapy for nonalcoholic fatty liver disease. Clin Transl Gastroenterol. Apr 1, 2023;14(4):e00575. [CrossRef] [Medline]
    10. Kwon OY, Lee MK, Lee HW, Kim H, Lee JS, Jang Y. Mobile app-based lifestyle coaching intervention for patients with nonalcoholic fatty liver disease: randomized controlled trial. J Med Internet Res. Feb 15, 2024;26:e49839. [CrossRef] [Medline]
    11. Sato M, Akamatsu M, Shima T, et al. Impact of a novel digital therapeutics system on nonalcoholic steatohepatitis: the NASH App clinical trial. Am J Gastroenterol. Aug 1, 2023;118(8):1365-1372. [CrossRef] [Medline]
    12. Björnsdottir S, Ulfsdottir H, Gudmundsson EF, et al. User engagement, acceptability, and clinical markers in a digital health program for nonalcoholic fatty liver disease: prospective, single-arm feasibility study. JMIR Cardio. Feb 15, 2024;8:e52576. [CrossRef] [Medline]
    13. Albhaisi S, Tondt J, Cyrus J, Chinchilli VM, Conroy DE, Stine JG. Digital therapeutics lead to clinically significant body weight loss in patients with metabolic dysfunction-associated steatotic liver disease: a systematic review and meta-analysis. Hepatol Commun. Aug 1, 2024;8(8):e0499. [CrossRef] [Medline]
    14. Chen D, Shao J, Zhang H, et al. Development of an individualized WeChat mini program-based intervention to increase adherence to dietary recommendations applying the behaviour change wheel among individuals with metabolic syndrome. Ann Med. 2023;55(2):2267587. [CrossRef] [Medline]
    15. Duan Y, Li X, Guo L, Liang W, Shang B, Lippke S. A WeChat mini program-based intervention for physical activity, fruit and vegetable consumption among Chinese cardiovascular patients in home-based rehabilitation: a study protocol. Front Public Health. 2022;10:739100. [CrossRef] [Medline]
    16. Duan Y, Liang W, Guo L, et al. Effectiveness of a wechat mini program-based intervention on promoting multiple health behavior changes among Chinese patients with cardiovascular diseases in home-based rehabilitation: randomized controlled trial. J Med Internet Res. Jun 3, 2025;27:e66249. [CrossRef] [Medline]
    17. Pan Y, Tang J, Lu B, et al. Effects of cognitive behavioral therapy for diet on postprandial glucose and pregnancy outcomes in gestational diabetes mellitus: multicenter randomized controlled trial. J Med Internet Res. Jul 29, 2025;27:e71075. [CrossRef] [Medline]
    18. Eysenbach G, CONSORT-EHEALTH Group. CONSORT-EHEALTH: improving and standardizing evaluation reports of web-based and mobile health interventions. J Med Internet Res. Dec 31, 2011;13(4):e126. [CrossRef] [Medline]
    19. National Workshop on Fatty Liver and Alcoholic Liver Disease, Chinese Society of Hepatology, Chinese Medical Association, Fatty Liver Expert Committee, Chinese Medical Doctor Association. Guidelines of prevention and treatment for nonalcoholic fatty liver disease: a 2018 update. Zhonghua Gan Zang Bing Za Zhi. Mar 20, 2018;26(3):195-203. [CrossRef] [Medline]
    20. Karlas T, Petroff D, Sasso M, et al. Individual patient data meta-analysis of controlled attenuation parameter (CAP) technology for assessing steatosis. J Hepatol. May 2017;66(5):1022-1030. [CrossRef] [Medline]
    21. Chinese Society of Health Management, Chinese Nutrition Society, Reproductive Medicine Branch of China International Exchange and Promotion Association for Medicine and Healthcare, China Health Promotion Foundation, Zhejiang Provincial Clinical Nutrition Center. Expert consensus & standard on weight management for overweight or obese people. Chin J Health Manage. 2018;12(3):200-207. [CrossRef]
    22. Craig CL, Marshall AL, Sjöström M, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. Aug 2003;35(8):1381-1395. [CrossRef] [Medline]
    23. Sun C, Goh GBB, Chow WC, et al. Prevalence and risk factors for impaired renal function among Asian patients with nonalcoholic fatty liver disease. Hepatobiliary Pancreat Dis Int. Jun 2024;23(3):241-248. [CrossRef] [Medline]
    24. Bedogni G, Bellentani S, Miglioli L, et al. The Fatty Liver Index: a simple and accurate predictor of hepatic steatosis in the general population. BMC Gastroenterol. Nov 2, 2006;6:33. [CrossRef] [Medline]
    25. Newsome PN, Sasso M, Deeks JJ, et al. FibroScan-AST (FAST) score for the non-invasive identification of patients with non-alcoholic steatohepatitis with significant activity and fibrosis: a prospective derivation and global validation study. Lancet Gastroenterol Hepatol. Apr 2020;5(4):362-373. [CrossRef] [Medline]
    26. Kruger FC, Daniels CR, Kidd M, et al. APRI: a simple bedside marker for advanced fibrosis that can avoid liver biopsy in patients with NAFLD/NASH. S Afr Med J. Jun 27, 2011;101(7):477-480. [CrossRef] [Medline]
    27. Angulo P, Hui JM, Marchesini G, et al. The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology. Apr 2007;45(4):846-854. [CrossRef] [Medline]
    28. Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. Jun 2006;43(6):1317-1325. [CrossRef] [Medline]
    29. Sanyal AJ, Foucquier J, Younossi ZM, et al. Enhanced diagnosis of advanced fibrosis and cirrhosis in individuals with NAFLD using FibroScan-based Agile scores. J Hepatol. Feb 2023;78(2):247-259. [CrossRef] [Medline]
    30. Lim SL, Johal J, Ong KW, et al. Lifestyle intervention enabled by mobile technology on weight loss in patients with nonalcoholic fatty liver disease: randomized controlled trial. JMIR Mhealth Uhealth. Apr 13, 2020;8(4):e14802. [CrossRef] [Medline]
    31. Björnsdottir S, Ulfsdottir H, Gudmundsson EF, et al. Long-term feasibility and outcomes of a digital health program to improve liver fat and cardiometabolic markers in individuals with nonalcoholic fatty liver disease: prospective single-arm feasibility study. JMIR Cardio. Sep 12, 2025;9:e72074. [CrossRef] [Medline]
    32. Harrison SA, Ruane PJ, Freilich B, et al. A randomized, double-blind, placebo-controlled phase IIa trial of efruxifermin for patients with compensated NASH cirrhosis. JHEP Rep. Jan 2023;5(1):100563. [CrossRef] [Medline]
    33. Armandi A, Bugianesi E. Dietary and pharmacological treatment in patients with metabolic-dysfunction associated steatotic liver disease. Eur J Intern Med. Apr 2024;122:20-27. [CrossRef] [Medline]
    34. Mazzotti A, Caletti MT, Brodosi L, et al. An internet-based approach for lifestyle changes in patients with NAFLD: two-year effects on weight loss and surrogate markers. J Hepatol. Nov 2018;69(5):1155-1163. [CrossRef] [Medline]
    35. Tincopa MA, Lyden A, Wong J, Jackson EA, Richardson C, Lok AS. Impact of a pilot structured mobile technology based lifestyle intervention for patients with nonalcoholic fatty liver disease. Dig Dis Sci. Feb 2022;67(2):481-491. [CrossRef] [Medline]
    36. Lin H, Lee HW, Yip TCF, et al. Vibration-controlled transient elastography scores to predict liver-related Events in steatotic liver disease. JAMA. Apr 16, 2024;331(15):1287-1297. [CrossRef] [Medline]
    37. Lee J, Vali Y, Boursier J, et al. Prognostic accuracy of FIB-4, NAFLD fibrosis score and APRI for NAFLD-related events: a systematic review. Liver Int. Feb 2021;41(2):261-270. [CrossRef] [Medline]
    38. Romero-Gómez M, Zelber-Sagi S, Trenell M. Treatment of NAFLD with diet, physical activity and exercise. J Hepatol. Oct 2017;67(4):829-846. [CrossRef] [Medline]
    39. Huber Y, Pfirrmann D, Gebhardt I, et al. Improvement of non-invasive markers of NAFLD from an individualised, web-based exercise program. Aliment Pharmacol Ther. Oct 2019;50(8):930-939. [CrossRef] [Medline]
    40. Stine JG, Rivas G, Hummer B, et al. Mobile health lifestyle intervention program leads to clinically significant loss of body weight in patients with NASH. Hepatol Commun. Apr 1, 2023;7(4):e0052. [CrossRef] [Medline]
    41. Chew NWS, Mehta A, Goh RSJ, et al. Cardiovascular-liver-metabolic health: recommendations in screening, diagnosis, and management of metabolic dysfunction-associated steatotic liver disease in Cardiovascular disease via modified Delphi approach. Circulation. Jan 7, 2025;151(1):98-119. [CrossRef] [Medline]
    42. Chan WK, Wong VWS, Adams LA, Nguyen MH. MAFLD in adults: non-invasive tests for diagnosis and monitoring of MAFLD. Hepatol Int. Oct 2024;18(Suppl 2):909-921. [CrossRef] [Medline]
    43. Li M, Cui M, Li G, et al. The Pathophysiological associations between obesity, NAFLD, and atherosclerotic cardiovascular diseases. Horm Metab Res. Oct 2024;56(10):683-696. [CrossRef] [Medline]
    44. He S, Lu S, Yu C, et al. The newly proposed plasma-glycosylated hemoglobin A1c/High-Density lipoprotein cholesterol ratio serves as a simple and practical indicator for screening metabolic associated fatty liver disease: an observational study based on a physical examination population. BMC Gastroenterol. Aug 19, 2024;24(1):274. [CrossRef] [Medline]
    45. Axley P, Kodali S, Kuo YF, et al. Text messaging approach improves weight loss in patients with nonalcoholic fatty liver disease: a randomized study. Liver Int. May 2018;38(5):924-931. [CrossRef] [Medline]
    46. Motz V, Faust A, Dahmus J, Stern B, Soriano C, Stine JG. Utilization of a directly supervised telehealth-based exercise training program in patients with nonalcoholic steatohepatitis: feasibility study. JMIR Form Res. Aug 17, 2021;5(8):e30239. [CrossRef] [Medline]
    47. Colantoni A, Bucci T, Cocomello N, et al. Lipid-based insulin-resistance markers predict cardiovascular events in metabolic dysfunction associated steatotic liver disease. Cardiovasc Diabetol. May 20, 2024;23(1):175. [CrossRef] [Medline]

    ‎
    ALT: alanine aminotransferase
    APRI: aspartate aminotransferase-to-platelet ratio index
    ASM: appendicular skeletal muscle mass
    AST: aspartate aminotransferase
    CAP: controlled attenuation parameter
    CONSORT: Consolidated Standards of Reporting Trials
    DTx: digital therapeutics
    FAST: FibroScan-aspartate aminotransferase
    FIB-4: fibrosis-4
    FPG: fasting plasma glucose
    GGT: γ-glutamyl transpeptidase
    HbA1c: hemoglobin A1c
    HDL-C: high-density lipoprotein-cholesterol
    HOMA-IR: homeostasis model assessment-insulin resistance
    IPAQ-SF: International Physical Activity Questionnaire-Short Form
    ITT: intention-to-treat
    LSM: liver stiffness measurement
    MAFLD: metabolic dysfunction–associated fatty liver disease
    MASH: metabolic dysfunction-associated steatohepatitis
    MET: metabolic equivalent of task
    OR: odds ratio
    PP: per-protocol
    RR: relative risk
    TLC: therapeutic lifestyle changes

    Edited by Stefano Brini; submitted 18.Apr.2025; peer-reviewed by Jing Zhang, Nia Maye; final revised version received 26.Nov.2025; accepted 02.Dec.2025; published 27.Jan.2026.

    Copyright

    © Chao Sun, Guangyu Chen, Cuicui Shi, Haixia Cao, Ruixu Yang, Jing Zeng, Xiaoyan Duan, Xin Sun, Jian-Gao Fan. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 27.Jan.2026.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research (ISSN 1438-8871), is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.