Behavioral motivation support versus personalized dietary management for metabolic and behavioral outcomes in MASLD: a network meta-analysis
Chao Wang, Ziying Huang, Pan Liu, Niuniu Sun, Lanhong Ruan, Mengyuan QiaoAbstract
Objectives
MASLD has a prevalence of almost one-third in adults worldwide and is not currently treated pharmacologically with a first-line therapy. Lifestyle change is still the foundation of management; however, the relative effectiveness of behavioral and dietary approaches remains uncertain. We compared behavioral motivation support (BMS), biomarker-guided personalized dietary management (BPDM), general dietary education (GDE), and prescribed dietary models (PDM) with usual care (UC) in adults with MASLD.
Methods
Eight databases and two trial registers were searched from inception to 31 March 2026. The review record was first submitted to PROSPERO on 17 April 2026 and formally registered on 18 April 2026 (CRD420261371973), after the search but before data extraction. Random-effects frequentist NMA was used where networks were sufficiently connected; outcomes measured with different instruments were analyzed using Hedges’ g or narratively.
Results
Fourteen RCTs (n=1,805) were included. Compared with UC, BMS was associated with lower ALT (MD −7.41 U/L, 95 % CI −11.07 to −3.76) and BMI (MD −1.34 kg/m 2 , 95 % CI −2.31 to −0.37), while BPDM was associated with lower ALT (MD −6.72 U/L, 95 % CI −11.76 to −1.68), BMI (MD −1.82 kg/m 2 , 95 % CI −3.15 to −0.50), and body weight (MD −6.66 kg, 95 % CI −8.47 to −4.86). The HbA1c network was sparse and did not include BMS. Nutrition behavior was analyzed as an exploratory two-study pairwise meta-analysis (Hedges’ g 1.64, 95 % CI −0.25 to 3.52), and self-efficacy was synthesized narratively. CINeMA confidence was very low across all 88 assessed network comparisons.
Conclusions
The findings are hypothesis-generating and do not establish superiority, complementary clinical roles, or treatment recommendations. Direct comparative trials are required.