Study is loaded with confounding factors, and problems relating it as general advice vs cohort: here's a bunch quickly of the top of head, simply for discussion

Study is loaded with confounding factors, and problems relating it as general advice vs cohort: here's a bunch quickly of the top of head, simply for discussion's sake -- not as a direct criticism of the authors / study. > Questionnaire-based diet and lifestyle assessments were administered in 1997 and 2009 These kinds of studies are not good because of a number of factors -- Morality of foods (people often answer questionaires with ideals or what they "ought" rather than fact), People are very bad at estimating their intakes, the data is ~15 yrs ago meaning lots of things can change in the interim > One aspect to take into consideration is however that there is a social tradition of “fika” in Sweden, where people get together with friends, relatives, or coworkers for coffee and pastries (41). Thus, one could hypothesize that the intake of treats is part of many people's everyday lives without necessarily being related with overall poor dietary or lifestyle patterns, and that it might be a marker of social life. Consumption of sweetened beverages, one the other hand, has been linked to lower overall dietary quality (42). Note that the positive associations were correlated with a _social_ form of eating. We already know folks with strong social connections, and distracting ways to slow life and eating, have better health outcomes. Also consider the zero sum form of eating whereby 100 calories of sugary beverage crowds out 100 calories of nutrient dense food (eg: broccoli). My own personal bias is that this is why vegetarians enjoy some health benefits -- they're simply eating a lot more vegetables (and not because meats are themselves deleterious). > No mention of adiposity (body fat percentage) A person with low and stable fat% is likely to have a very different response to sugars than someone with a high fat% (insulin resistance) and instability (yoyo-ing seems to matter albeit unclear how) > Cohort Unless you're a Swede of the average genetic profile ~97-09 then there's a gradient of issues related to your differences.