As a smoker, looking at myself shows that cigarettes cause 0% chance of death in a sample of 1 - you wouldn't neccesarily expect your 1000 person sample to show

As a smoker, looking at myself shows that cigarettes cause 0% chance of death in a sample of 1 - you wouldn't neccesarily expect your 1000 person sample to show results as accurate as a much bigger sample size (222,497). But more importantly - this research was done on people over the age of 45, your 1000 people were aged 5-24. Using your age range I imagine we could prove that smoking 60 a day and eating McDonalds for every meal isn't particularly dangerous.