According to "Poe’s bitter enemy Dr. Thomas Dunn English," Poe did not have a drug addiction.
“Had Poe the opium habit when I knew him, I should, both as a physician and a man of observation, have discovered it in his frequent visits to my rooms, my visits to his house, and our meetings elsewhere. I saw no signs of it."
The Edgar Allan Poe Society of Baltimore authoritatively states that Poe was not a drug user and surely not a drug addict (though opium was prescribed medicinally during that era). They state that the notion of Poe's opium addiction came from (1) his narratorial mode, which was often first-person (thus a confusion amongst readers between first-person and the person); (2) the fact that some of Poe's first-person narrators describe themselves as opium users; (3) a critical opinion in The Daily Cincinnati Gazette, upon the 1845 publication of Poe's Tales, that dismissed the tales as "the strange outpourings of an opium eater.”
In short, the author became confused with his narrators and works. This said, there are numerous reports, including his own, of having at times--such as while at the University of Virginia and later at West Point--been first a heavy episodic drinker, then a constant drinker or an alcoholic. Yet, in one of his later 1841 letters to Dr. J. Evans Snodgrass, he confesses to having been "temperate" for four years.
"I am temperate even to rigor. . . . it is now quite four years since I have abandoned every kind of alcoholic drink."
Nor was Poe mad. He is said, as a result of forensic psychological analysis based heavily upon his letters, to have had bipolar disorder, which earlier was called manic-depressive illness. Yet, if "madness" means a loss of the ability to reason coherently about a rationally perceived reality, then bipolar disorder is most assuredly not "madness." This disorder is an imbalance in the brain's neuro-regulators that causes swings in mood, energy and activity levels such that the affected person alternates between hyper-active mood elation and inactive mood depression.