On February 9, 2005, Health Canada (HC) announced that it was suspending sales of Adderall XR in Canada because of concerns that it might lead to sudden death and stroke. The FDA, however, has begged to differ.
Both agencies claim to have analyzed the same data, yet according to their respective health advisories, their numbers look different. HC reported 20 sudden unexplained deaths (“SUDs”) linked to Adderall from 1999-2003, while the FDA reported only 12. What gives?
We called both agencies for answers, but only the FDA responded. According to FDA spokeswoman Crystal Rice, six of HC’s deaths were in adults, which the FDA omitted from its analysis because of the high baseline rate of SUDs in adults. Of HC’s 14 remaining pediatric SUDs, the FDA claims that one was a duplicate case, and that another should not be included because it involved a child who was found with cocaine “on his lap” at death.
All right, so the number of pediatric Adderall-related deaths is actually 12. But is this greater than the rate associated with other stimulants?
Unclear. According to Dr. Robert Peterson, director general of the therapeutic products directorate at HC, the death risk from Adderall was “two or three times” higher than that of Ritalin or Concerta (NYT, Feb 11, 2005). However, in the same article, Dr. Robert Temple, director of the FDA’s Office of Medical Affairs, said that seven cases of sudden death occurred in patients taking Ritalin and Concerta vs. 12 deaths for Adderall from 1999-2003, an insignificant difference, according to Temple.
We asked the FDA for the raw data, but they told us that this data is “considered proprietary.” However, we did find some relevant information on the DEA’s (Drug Enforcement Administration) website, which reported that in 1999, 11 million methylphenidate scripts were written vs. 6 million amphetamine scripts (they reported no data beyond 1999). Since the FDA’s Health Advisory reported that 30 million Adderall prescriptions had been written over the 5 years, we assumed that the DEA’s number was accurate (6 million per year x 5 years = 30 million). We did this same math with DEA’s methylphenidate number (11 million x 5 = 55 million). (Granted, it seems unlikely that the same number of prescriptions were written in each year, but unless the FDA releases its data, this is the best we can come up with.) Dividing deaths by prescriptions, we discovered that the death rate per million methylphenidate scripts is 7/55= 0.13, while the death rate per million amphetamine scripts is 12/30=0.3. Thus, amphetamines appear to be about 2.3 times more likely than methylphenidate to lead to sudden death.
We emailed our calculations to the FDA, who pointed out that since Adderall is newer and more intensively marketed than Ritalin, the higher Adderall numbers may be due to excess reporting rather than an actual difference in death rates. Furthermore, the baseline rate of sudden unexplained death (so-called “SUD”) in children is substantial. While the Adderall death rate may be 0.3/million, a recent study of SUD in the U.K. reported a rate of about 2 deaths/million/ year in kids older than 2 years old (Heart 2000; 83:410-413), meaning that the background risk may be 10 times higher than the supposed Adderall risk.
Ultimately, TCR has to side with the FDA in this inter-agency skirmish. Any connection between Adderall and death is probably spurious. After a series of high profile embarrassments, it appears that Team FDA has finally gotten its numbers right.
TCR VERDICT: Adderall and death? No connection.