The study looks reasonably well done to me. If anything, it underestimates the numbers. The effects would be even larger if it included kids - with the overweight/obesity rates among kids also being the highest in the usual suspect nations. One other not mentioned limitation is that there should be ethnicity-specific BMI limits for obesity. Research indicates that, for example, some Asian groups show disease rates at BMI>23 that you find in the US at BMI>25. If you included that into this study, it would also lead to higher overall bio mass due to overweight/obesity but the percentages would shift slightly.
As for overweight/obesity in general, that could certainly fill looong threads of discussion. I would like to just point one thing out that is a major concern in public health and nowhere as misunderstood as in the US: In most cases, food choice is NOT strictly an expression of free will. Americans seem obsessed with free choice and government interference with it - but they have little problems with the food industry not only selling them cheap crap that is based on maximized profit but also, most importantly, advertized to them with budgets in the tens of billions of $$ every year. Advertizing is there to create 'needs' that are independent of things like health or well-being. Fast food places are not randomly placed in cities, they are strategically placed to cater to specific target groups. Between price structures, advertizing influences, access/availability, economically caused society changes (e.g. both parents need to work), and lack of education (which is a topic in itself...), the range of free choice is minimal. The majority of the population can not shop for healthy stuff at Wholefoods.
Stefan