Good news from the Internet of flab: Data from connected scales suggests Americans aren’t piling on lots of pounds while in coronavirus isolation.

Withings, the maker of popular Internet-connected scales and other body-measurement devices, studied what happened to the weight of some 450,000 of its American users between March 22 — when New York ordered people home — and April 18. Despite concerns about gaining a “quarantine 15,” the average user gained 0.21 pounds during that month. Some 37 percent of people gained more than a pound.

How you view those numbers, however, is a matter of perspective. In a typical year, Americans gain one to two pounds.

The analysis released Friday is the latest effort by a tech company to quantify how lives have changed during the pandemic using data from “Internet of Things” devices. And like other covid-19 studies of fevers from connected thermometers, social distancing from smartphones and activity from fitness trackers, we should also read the weight results with some caution.

Some of these studies are more marketing efforts than contributions to public health. But scientists and officials have also been turning to personal tech for serious academic studies and to even track covid-19 cases in real time.

Withings, which has contributed data to other academic research, said the company is publishing this data on its own because it kept getting asked about it. “We heard a lot of assumptions about how the lockdown would have an impact on people’s weight and activity,” said marketing head Lucie Broto.

Withings looked at last year, too, to understand how we’re coping at a time so many people are stuck at home and presumably stressed. Over the same March-April period in 2019, Withings said its American users gained slightly less weight — 0.19 pounds on average — though fewer people had the scales last year.

Withings also found similar trends in other countries. Since the onset of coronavirus quarantine orders, the average weight gain in the United Kingdom is 0.35 pounds, China is 0.55 pounds and France is 0.19 pounds.

Like other data efforts, there are also privacy questions. Withings said all the data in its study was anonymized and can’t be traced back to individuals. But contributing aggregate data is a condition included in its terms of service; its customers don’t get the option to opt out if they want to use Withings products.

Dariush Mozaffarian, dean of the School of Nutrition Science and Policy at Tufts University — who wasn’t involved with the Withings analysis — said he found the results a bit disappointing. “With the shutdown of the restaurants, I thought the numbers would have gotten better,” he said. Home-cooked meals tend to be healthier than dining out.

The population with Withings scales, which sell for $60 to $150, might be a self-selecting bunch. Withings said some 46 percent of its users are women, and while it has thousands of customers in every state, they tend to be in urban areas. The users in its sample also skew young and over-index for people in their 40s (some 24 percent of the total).

“Our users are in no way the super fit or ultra-health conscious but are generally mindful of health,” said Broto.

People who are concerned enough about their weight to buy and keep using a scale may be better — or worse — at controlling it. Withings said its users gained an average of 1.06 pounds last year, which is in line with other studies.

Other experts not involved in the Withings analysis warn average data can be misleading. “The thing we find in diet is that it is extremely variable,” said nutrition scientist and Stanford University professor Christopher Gardner. Some people, he said, may be hitting their groove during stay-at-home orders by embracing cooking and taking up jogging. But others could be using food to cope with stress and gaining large amounts of weight.

Withings didn’t have data to share about the range or variability in weight gains in its research. “That could be telling a very different narrative,” said Gardner.

Doctors say we should be thinking about what we eat now. “There’s no time like a pandemic to eat healthy,” said Eric Rimm, a public health professor at Harvard. Diet is important for immune function, and evidence indicates people with preexisting conditions — including being overweight — fare worse with covid-19, he said.

“There is an opportunity right now for people to be doing better — and some are,” said Gardner.

Research has shown that people who plan their meals in advance tend to eat healthier. Limiting grocery runs during the pandemic for safety reasons also offers people the opportunity to be more intentional about what they eat, Gardner said.

He also recommends focusing on eating meals with family and friends, even over video conference. “Connecting with family over quality food and slowing down what you’re eating would help people manage their hunger and their satiety and their health,” Gardner said.