Sugar-sweetened beverages (SSBs) significantly contribute to weight gain and increased risk of type 2 diabetes, cardiovascular disease, and certain cancers. Dietary guidelines often recommend lower-calorie alternatives like water or NNS beverages to reduce sugar intake. However, the long-term effects of NNS beverages on weight management remain controversial. While some observational studies link NNS beverage consumption to weight gain, meta-analyses of shorter-term randomized controlled trials (RCTs) have shown reduced energy intake and modest weight loss with NNS beverages compared to SSBs. A critical gap in the literature is the lack of long-term RCTs directly comparing NNS beverages to water for weight maintenance after an initial weight loss phase. This study aimed to address this gap by conducting a 52-week RCT comparing the effects of NNS beverages and water on weight change in adults participating in a behavioral weight management program. Previous research, such as the Colorado/Temple trial, has shown promising results for NNS beverages in weight loss and maintenance, but further long-term studies are needed to solidify these findings and inform dietary guidelines.

The existing literature on the effects of NNS beverages on weight management is mixed. Some observational studies have suggested a positive association between NNS beverage consumption and weight gain or increased BMI. However, meta-analyses and systematic reviews of RCTs, primarily comparing NNS beverages to SSBs, often report reduced energy intake and modest weight loss with NNS beverages. A key limitation of many previous studies is their short- or medium-term duration, highlighting the need for long-term RCTs directly comparing NNS beverages with water. The lack of data from longer-term trials comparing the effects of NNS beverages to water on weight maintenance following weight loss limits the evidence base for policy recommendations.

This parallel-group, open-label, controlled equivalence trial, known as the SWITCH trial, enrolled healthy adults (18-65 years) with a BMI of 27–35 kg/m² who regularly consumed >3 cold beverages per week. Participants were randomly assigned (1:1) to consume either water or NNS beverages (at least two 330 ml servings daily) as part of a 52-week behavioral weight management program. The program consisted of a 12-week weight-loss phase (weekly meetings) followed by a 40-week weight-maintenance phase (monthly meetings). Randomization was stratified by sex, age, BMI, and NNS naïveté. The primary outcome was weight change at week 52. Secondary outcomes included changes in anthropometrics, cardiometabolic risk factors, appetite, and activity levels. Data collection involved monthly weigh-ins, anthropometric measurements, blood tests (subset of participants), and activity tracking (Fitbit). Adherence to beverage consumption was assessed via daily online logs, returned packaging, and food frequency questionnaires. Data analysis used analysis of covariance (ANCOVA), with baseline weight as a covariate. Multiple imputation and last observation carried forward analyses addressed missing data. The trial was conducted at the University of Liverpool, England, and received ethical approval. The study aimed to achieve 90% power to detect a ± 1.5-kg weight change difference between groups at week 52, accounting for an expected 27% attrition rate.

Of 493 randomized participants, 262 completed the 52-week program. At week 52, the water group showed a mean weight loss of 6.1 kg, while the NNS beverage group lost 7.5 kg. This difference (1.4 kg) was statistically significant (P < 0.05) in the complete cases analysis. However, this difference did not reach the pre-defined threshold for clinical significance (1.5 kg). In analyses using multiple imputation and last observation carried forward methods, the difference between groups was not statistically significant. Significant reductions from baseline were observed in both groups for waist and hip circumference. The reduction in hip circumference was significantly greater in the NNS beverage group compared to the water group. Most biomarkers showed improvements from baseline in both groups, with a significant difference observed in high-density lipoprotein cholesterol (driven by a modest increase in the NNS beverage group). Sweetener consumption was significantly reduced in the water group but not in the NNS beverage group. Activity levels (steps/day) increased in the NNS beverage group but decreased in the water group, although the between-group difference was not significant. A subset of participants (n=57) underwent DXA scans which revealed significant reductions in fat mass and fat-free mass in both groups, with no significant difference between groups.

This study provides long-term data on the comparative effects of NNS beverages and water on weight management within a behavioral weight-loss program. The results partially support previous findings from shorter-term trials and the Colorado/Temple study, indicating that both water and NNS beverages can support weight loss and maintenance. The statistically significant but not clinically significant greater weight loss observed with NNS beverages may be attributable to factors such as increased satiety or reduced caloric intake from replacing other beverages. The inconsistent findings across different data imputation methods highlight the influence of missing data on the results. The study's findings contribute to the ongoing debate surrounding the role of NNS beverages in weight management, suggesting that while they may offer a modest advantage over water in this specific context, the difference is not substantial enough to be considered clinically significant.

The 52-week SWITCH trial demonstrates that both water and NNS beverages can support weight loss and maintenance within a structured behavioral weight management program. Although the NNS beverage group demonstrated statistically greater weight loss, this difference was not clinically meaningful. Future research should explore the long-term effects of NNS beverages beyond the assisted weight maintenance phase and investigate potential mechanisms underlying the observed differences. The impact of individual factors (e.g., baseline habits, genetics, metabolism) on response to NNS beverages also needs further investigation.

The main limitations include the relatively low completion rate (53%) at week 52, potentially influenced by the COVID-19 pandemic and the study's duration. The single-site location in England limits the generalizability of the findings to other populations or ethnic groups. The lack of racial or ethnicity data also limits the interpretation of potential effects based on these factors. The use of self-reported data during some assessments due to COVID-19 restrictions also introduces a potential bias into the collected data.