The study, Accelerometer-measured physical activity is not associated with two-year weight change in African-origin adults from five diverse populations, followed 1,944 young adults of predominantly African descent in 5 different countries of varying degrees of socio-economic development (Ghana, South Africa, Jamaica, Seychelles and the USA). Researchers were interested in the relationship between the participants' weights and their baseline physical activity levels (as measured by 8 days worth of objective accelerometer data) over the 3 years of study.
The participants were categorized as either meeting the USA Surgeon General's recommendations for weekly physical activity (PA) of 30mins/d most days of the week or not.
The researchers were surprised to learn that the total weight gain at every site was greatest among those meeting the PA guidelines, though to be fair, the differences were minimal. For instance in the USA those meeting the PA guidelines were found to be gaining an average of roughly half a pound per year, while those not meeting them were found to be losing half a pound per year. What was clear though was that the measured PA did not appear protective vs. weight gain, and clearly not a driver of weight loss.
Researchers also explored sedentary behaviour as a separate entity and found that it too failed to correlate with weight change over time (in either direction).
All this to say, this is another paper for the pile suggesting that forks tend not to be outrun.