Outcomes of a randomized-controlled digital weight management program over 2.5-years

Coaching-led digital weight-loss interventions are successful in the short-term, but have limited evidence supporting long-term changes. This study evaluated a commercially available behavioral weight-management program compared to digital control on initial and long-term outcomes.

A total of 684 participants were screened to achieve a sample of English-speaking individuals 18-60 years old with a body mass index (BMI; kg/m²) > 27 kg/m² without contraindications to using a smartphone program. In a single-blinded, parallel-group trial, a centralized computer-generated program randomized 600 participants (n = 302 intervention; n = 298 control) to a 6-month intervention or control with a maintenance phase during months 7-30. The primary outcome was change in BMI; secondary outcomes included effects on diet, physical activity, sleep, quality of life, psychosocial functioning, and health status.

Fifty-seven participants withdrew from the study and all future assessments; 78% of the sample completed 6-months and 65% completed 30-month follow-up. All available data were used under assumptions of missing at random. Latent growth curve models showed the differences were not significant [β = 0.99, 95%CI = -1.38, 3.36, d = .17, 95%CI = -0.24, 0.58]. Over 30-months, changes for intervention and control were 0.13 kg/m² (SD=3.09) and 0.58 kg/m² (SD=3.64), respectively, a significant difference [β = -4.42, 95%CI = -8.72, -0.12, d = -0.76, 95%CI = -1.51, -0.02].

Although the acute effects on self-reported BMI were not significant, the use of coaching-led weight loss led to better two-year BMI outcomes, and were greater for those at a lower BMI. Interventions using personal support may be important for realizing longer-term benefits of digital interventions.