TOPLINE

Adults with a BMI of 27-35 who followed a behavioral weight loss program lost similar amounts of weight at 2 years regardless of whether they drank diet beverages or water, and both groups stayed below their starting weight.

METHODOLOGY

• The role of diet beverages in long-term weight control has been debated for years. Some observational studies have tied them to weight gain, while randomized trials have shown modest benefits compared with sugary drinks. Extended trials that compare them directly with water after initial weight loss are scarce.
• Researchers analyzed the results obtained at week 104 of SWITCH, an open-label randomized trial conducted at the University of Liverpool in healthy adults aged 18-65 years with a BMI of 27-35 who habitually drank cold beverages; 493 began their assigned regimen (246 had water, 247 had diet drinks), and about 76% were not naïve to diet drinks.
• Each participant had at least two 330-mL servings daily. The diet group chose from 20 commercial products sweetened mainly using aspartame, acesulfame potassium, and sucralose.
• All participants took part in a nutritionist-led behavioral weight management program, with weekly sessions for 12 weeks and monthly sessions through week 52; the 104-week analysis covered the voluntary, unassisted extension that followed, which had no sessions.
• The primary endpoint was the change in body weight from the start of the trial to week 104. Only 220 participants (117 had water, 103 had diet drinks) had weight data at both timepoints.

TAKEAWAY

• At week 104, the mean weight loss was 3.7 kg with water and 4.8 kg with diet drinks. The 1.1-kg between-group difference was not significant (90% CI, -0.2 to 2.4).
• Both groups showed significant reductions in waist circumference (4.5 cm with water vs 5.5 cm with diet beverages) and hip circumference (2.7 cm with water vs 3.1 cm with diet beverages) at 2 years, with no significant differences noted between groups.
• Some biomarker levels changed from baseline to week 104 in both groups; between-group differences were significant for total cholesterol (rose by 0.2 mmol/L with diet beverages vs fell by 0.3 mmol/L with water; difference, -0.5 mmol/L; 95% CI, -0.9 to -0.1), A1c (rose by 1.8 mmol/mol vs fell by 0.6 mmol/mol; difference, -2.4 mmol/mol; 95% CI, -4.5 to -0.3), and diastolic blood pressure (fell by 1.7 mm Hg with diet beverages vs fell by 4.5 mm Hg with water; difference, -2.8 mm Hg; 95% CI, -5.3 to -0.3); the authors considered these differences too small to matter clinically.
• Self-reported hunger edged up in both groups but reached significance only with diet beverages, with no significant difference observed between groups.

IN PRACTICE

"This study provides the long‑term clinical evidence that has been missing from international discussions. For people trying to manage their weight, non‑nutritive sweetened beverages offer an effective alternative to sugar‑sweetened drinks—and perform equivalently to water over two years," Professor Jason Halford, of the University of Leeds and University of Liverpool wrote in a press release.

SOURCE

The study was led by Joanne A. Harrold, PhD, Department of Psychology, University of Liverpool, England. It was published online in the British Journal of Nutrition.

LIMITATIONS

The study was conducted at one site in England and did not record participants' race or ethnicity; hence, the findings may not apply to other populations. Only 220 participants had data on body weight measurements at 2 years. The authors attributed this to the COVID pandemic, the trial's length, and the voluntary final phase.

DISCLOSURES

The study was supported by the American Beverage Association. All authors reported receiving funding from the American Beverage Association (paid to the University of Liverpool). Some authors also disclosed receiving consulting fees, honoraria, research grants, personal fees, or meeting support from various organizations, along with industry funding for a doctoral studentship.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.