Armed with three years of grocery shopping data, researchers found that total sugar sales are down by almost 20%, driven largely by falling soda purchases.

Ever feel like it’s impossible to make sense of whether soda taxes actually work to reduce sugar consumption? We don’t blame you. For years, proposals to levy fees on sugary drinks have stoked impassioned debate and generated thousands of pages of competing research over their effectiveness. Just as quickly as some jurisdictions have implemented such policies, others have responded by banning them, creating a confusing patchwork of soda legislation across the country.

For readers who aren’t in the weeds of public health, the topline findings of soda tax studies may even seem contradictory. Just take this sample of headlines, for example, all published in the past few years: “Soda taxes really do work,” reads one article in TIME. “Philly’s soda tax didn’t lead to people drinking less soda, study says,” reads another from The Philadelphia Inquirer. And then there’s this post from NPR: “U.S. Soda Taxes Work, Studies Suggest — But Maybe Not As Well As Hoped.”

It’s enough to make you throw your hands up and think about simpler things instead. But stick with us as we parse out a new, first-of-its-kind study — one that may tell us once and for all just how much soda taxes cut back sugar consumption in a major city that adopted them. Spoiler: The reductions are big.

In a recent analysis, researchers used a large set of grocery shopping data to track food and drink sales before and after the city of Seattle implemented a sugary drinks tax in January of 2018. The data set itself—gathered by marketing insight firm Nielsen — was huge, representing 45% of all food store sales in the city for 2017, 2018, and 2019. To create a control group, researchers also obtained the same data for the nearby city of Portland, Oregon, which doesn’t have a sugary drinks tax in place.

To account for different sugar levels in various beverages, a team led by Lisa Powell, health policy professor at the University of Illinois at Chicago, coded each type of drink by its exact sugar content. In doing so, they found that the total amount of sugar sold through taxed drinks dropped by 23% in Seattle compared to sales of the same products in Portland, one year after the implementation of the soda tax. That decline held for the following year as well, suggesting that the dip was not just a fluke.

“These results suggest that sugar-sweetened beverage taxes may effectively yield permanent reductions in added sugars sold from sugar-sweetened beverages in food stores,” the authors wrote in a paper published in Journal of the American Medical Association.

But the researchers didn’t stop there. They also wanted to know if shoppers might be getting sugar from other foods instead—a possibility that soda tax opponents have argued would become commonplace. Were Seattle residents simply swapping out Mountain Dew for candy bars? To find out, the researchers also analyzed sales data for untaxed drinks like flavored milk, sweets — which the team defined to include candies, desserts, and baked goods — as well as loose sugar. Over the course of months, Powell’s team painstakingly coded each product sold by its sugar content and then calculated just how much sales of these products changed after the soda tax went in place.

They found a slight increase in sugar consumed through untaxed drinks in 2018, which then dissipated in 2019. They also noticed a small, sustained increase in sugar consumed through sweets. In both cases, however, those upticks were not large enough to overcome the significant reduction in sugar sold through taxed drinks.

“After accounting for substitutions, there was a net reduction of 19% in grams of sugar sold from taxed beverages,” Powell said. “So there was some offset for sure, but there’s still at the end of the day, a substantial reduction in grams of sugar sold.”

The findings bolster the city of Seattle’s own early findings about the effectiveness of the soda tax. Last year, researchers at the University of Washington found that low-income families in particular saw a significant decrease in soda consumption following the implementation of the tax. The findings were based on a survey of residents, meaning that it relied on self-reporting, which is less accurate than sales data. However, self-reported numbers play a crucial role in capturing soda sales outside of grocery stores, such as at restaurants and bars, which were not included in Powell’s study.

“Ideally, results from both types of studies can help researchers triangulate the likely efficacy of the tax,” wrote Jesse Jones-Smith, a lead researcher at the University of Washington who has been involved in research about the tax’s impacts, in an email.

The results may have promising public health implications. Almost 60% of Americans exceed the Dietary Guidelines’ suggested limit on added sugar consumption. This can contribute to health issues including diabetes and heart disease, making eaters more vulnerable to diseases and more likely to incur higher healthcare costs over the course of their lives.

Whether these findings might compel other cities to adopt similar policies is still an open question. In response to growing support for soda taxes, food and drink lobby groups have pushed back forcefully to curtail their implementation through a political strategy known as “preemption.” Pioneered by the tobacco industry in the 1970s, this involves lobbying for state-level lawmaking that prevents local jurisdictions from adopting their own public health ordinances.

And that’s exactly what happened in Washington shortly after Seattle passed its soda tax: In November of that same year, soda companies campaigned successfully for the passage of a ballot initiative banning all future fees on sugary drinks across the state. If Seattle’s neighbors want to pass their own soda taxes, they’ll have to wait for its repeal.

This story originally appeared in The Counter and is reprinted here as part of the SoJo Exchange from the Solutions Journalism Network, a nonprofit organization dedicated to rigorous reporting about responses to social problems.