To evaluate the potential impact of sugar-sweetened beverage (SSB) taxes on overweight and obesity prevalence in countries of different income classifications.

Systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (PROSPERO number CRD42020161612). Five databases (Cochrane Library, Embase, LILACS (via Virtual Health Library) and MEDLINE (via PubMed), and Web of Science were searched, from January 2009 to December 2019. Articles that reported changes in purchases, sales, intake, body weight, BMI, overweight and/or obesity prevalence due to a tax on or price change in SSB were included.

Studies conducted in countries of different income classifications.

The search yielded 8349 articles of which 21 met inclusion criteria.

Among the sixteen studies selected, only two did not show that consumption, sales and purchase decreased as the price of SSB increased. In eight of the thirteen studies selected, a positive effect of an SSB tax on decreasing overweight and obesity prevalence was expected. It is estimated that a 20 % taxation on SSB would result in a greater decrease in the prevalence of overweight and obesity compared to a 10 % rate. Studies with no significant effect of taxing on sales, purchases, consumption and prevalence of obesity were from high-income countries, while significant effects of taxing on reducing purchase, consumption and/or obesity prevalence were found in studies from upper-middle- and middle-income countries.

A high SSB tax might be an effective fiscal policy to decrease purchase and consumption of SSB and reduce overweight/obesity prevalence, especially if the tax were specific for beverage volume.

To explore if the secular trend in height is contributing to delay overweight rise among Brazilian adolescents.

Changes in BMI mean over time were fitted using linear regression including as independent variables survey year, height, survey-specific income quintiles, age and an interaction term of height × survey year. Overweight was defined as BMI≥25·0 kg/m2. Changes in overweight prevalence over time were fitted using Poisson regression.

Four national household surveys: 1974/5, 1989, 2002/3 and 2008/9.

Brazilian adolescents.

Mean values of height and BMI increased over the period, for both sexes and in all age ranges, except for girls aged 14–19 years from 1989 to 2002/3. The highest average increment and mean rate of height were between 1989 and 2002/3 and in 10–15-year-olds. The annual increment of height decreased from 2002/3 to 2008/9 in parallel with the increment in BMI rate. After fitting the regression model, the height × survey year interaction and per capita income were strong vectors to increase BMI mean. Changes in increment rate of height played a protective role against overweight in the last two periods for both sexes, mainly for girls. The period from 1989 to 2002/3 was the strongest vector associated with overweight in boys and the association decreased to the next period, from 2002/3 to 2008/9.

BMI and height of adolescents have increased in a wavering and alternate way throughout four decades in Brazil. The rate of height increment has played a protective role against overweight in adolescents.