When I was first diagnosed with AF in 1988 at 30 years of age there was utter consternation as I was extremely fit. AF had then only been seen in unfit people (smokers, overweight etc.). Unsurprisingly these people die early because of a range of factors and they also have a higher incidence of stroke. So the question is does AF actually increase risk once you remove all these other factors. The technical jargon is "co-morbidity". I am aware of one study that was done which actually stripped out all the other factors so see if AF (single variable) affected life-span.
Here is the conclusion
"Overall survival of the 76 patients with lone atrial fibrillation was 92% and 68% at 15 and 30 years, respectively, similar to 86% and 57% survival for the age- and sex-matched Minnesota population. Observed survival free of heart failure was slightly worse than expected (P=0.051). Risk for stroke or transient ischemic attack was similar to the expected population risk during the initial 25 years of follow-up but increased thereafter (P=0.004), although CIs were wide. All patients who had a cerebrovascular event had developed > or = 1 risk factor for thromboembolism."
So you'll see NO increase in stroke risk for 25 year and longer life (actually statistically not significant, but hey let's celebrate)
The study is here ncbi.nlm.nih.gov/pubmed/175... or try searching for the title "Long-term progression and outcomes with aging in patients with lone atrial fibrillation: a 30-year follow-up study"
It was done in 2007 so there may be more up to date research but I haven't seen it. Have you?