There's a well known pattern that studies conducted by a drug company are more likely to report positive results than studies conducted by a neutral party.