It sounds like they didn't cover lifestyle factors, but I'm wondering if you even could. Over 7 years, does 170,000 participants means people lying or not remembering things properly cancel each other out? or would the data be too muddied to tell.
I'm a habitual coffee drinker and I can tell you with a high degree of certainty that I almost never drink non-diet soda, and haven't for at least 7 years. It's pretty hard to just forget behaviors that are habitual for you.
While true for habits, I eat cake like maybe 3 times a month four months out of the year? I never drink soda, but I couldn’t tell you what my consistent sugar/fat/cholesterol intake is, it varies with the seasons/stress, etc. is it better than average? Worse than average? I dunno. Same with exercise.
