Well, this is the most commonly used trick in comparing data from different clinical trials: different patient population. Some do it deliberately, some do it due to ignorance, and some by honest mistake.