Background: Motivational interviewing (MI) has been suggested as a useful tool in the management 37 of overweight and obesity, but training staff and increased delivery time adds cost to the intervention, and its independent effectiveness is unclear. Purpose: To assess whether the independent contribution of MI, as part of a behavioral weight management program (BWMP), is effective in improving weight control and psychological well41 being. Data Sources: Six electronic databases and two trial registries were searched from 7 February 2020 through 24 September 2021. Study Selection: Randomized controlled trials in adults or adolescents aiming to support weight loss or maintenance with programs including MI, compared to similar interventions without MI, or to no/minimal interventions. Data Extraction: Two independent reviewers screened studies, extracted data, and assessed risk of bias. Weight, anxiety, depression, quality of life, and other aspects of psychological well-being, were extracted. Data Synthesis: Pooled mean differences or standardized mean differences were obtained from random and fixed effects meta-analyses depending on the number of studies being pooled. Subgroup analyses explored differences in effectiveness. Results: Forty-six studies reporting on 11,077 participants, predominantly with obesity, were included. At 6 months follow-up, BWMPs using MI were more effective than no/minimal 55 intervention (-0.88kg; 95%CI -1.27 to -0.48; I2=0%), but they were not statistically significantly more 56 effective than lower-intensity (-0.88kg; 95%CI -2.39 to 0.62; I2=55.8) or similar-intensity BWMPs (- 57 1.36kg; 95%CI -2.80 to 0.07; I2=18.8%). At one year, data were too sparse to pool for comparators of 58 no/minimal intervention, and MI did not produce statistically significantly greater weight change 4 compared to lower or similar-intensity 59 BWMPs without MI (-1.16kg; 95%CI -2.49 to 0.17; I2=88.7%, 60 and -0.18kg; 95%CI -2.40 to 2.04; I2=72.7% respectively). At 18 months follow-up, MI did not produce 61 statistically significant benefit in any of the comparator categories, with -0.69kg (95%CI -1.54 to 0.17; 62 I2=0.0%) compared to minimal no/minimal intervention, -0.57kg (95%CI -1.20 to 0.06; I2=62.9%) compared to lower-intensity, and -0.36kg (95%CI -1.06 to 0.34; I2=66.4%) compared to similar64 intensity programs, but analyses were limited by sparsity of data. There was no evidence of subgroup differences based on study, participant or intervention characteristics, though limited by sparse data. Too few studies assessed effects on psychological well-being to pool, but data did not suggest that MI changed these outcomes. Limitations: There was high statistical heterogeneity between the studies, which was largely unexplained even after sensitivity and subgroup analyses. Categorization by comparator and follow up resulted in pooling few studies. Conclusions: There is no evidence that MI adds to the effectiveness of BWMPs or improves weight loss in people being treated for weight-related non-communicable diseases. Given the intensive training required for its delivery, this evidence suggests MI may not be a worthwhile addition to BWMPs.
Annals of Internal Medicine
American College of Physicians