Study of Pot Smokers' Brains Shows That MRIs Cause Bad Science Reporting
This week a study of cannabis consumers published by The Journal of Neuroscience provided powerful evidence that MRI scans cause shoddy science reporting. Researchers at Northwestern University and Massachusetts General Hospital used MRIs to compare the brains of 20 young adults who reported smoking pot at least once a week and 20 controls who had used marijuana no more than five times in their lives and had not consumed it at all in the previous year. The pot smokers had higher gray-matter densities in the left nucleus accumbens, and there were "significant shape differences" between subjects and controls in that area and in the right amygdala. The differences were more pronounced in subjects who reported smoking marijuana more frequently. "Because this is a cross-sectional study," the authors noted, "causation cannot be determined." In other words, it is not clear whether the brain differences were caused by marijuana. It also is not clear how long the differences last or whether they have any functional significance.
Those nuances generally were lost in press coverage of the study, which presented the MRI scans as evidence that smoking pot causes brain damage. News outlets claimed the study found that "marijuana re-shapes brains of users" (NBC News), that "even casually smoking marijuana can change your brain" (The Washington Post), that "casual pot use impacts brains of young adults" (The Oregonian), that "recreational pot use" is "harmful to young people's brains" (Time), that "casual marijuana use" is "bad for young adults" (The Times of India), and that "even 'casual' marijuana use can knacker bits of your brain" (Gizmodo UK). A Medical News Today headline quoted the researchers as saying "casual marijuana use changes the brain," although that statement does not appear in the article under the headline, in the study itself, or in press releases about the study issued by Northwestern University, Massachusetts General, and the Society for Neuroscience, which publishes The Journal of Neuroscience. Similarly, an MSN NZ headline had the study claiming that "cannabis use 'alters brain regions,'" another phrase that is absent from the study and the press releases.
Although they seem to have been misquoted in some cases, the researchers themselves are partly responsible for the misrepresentation of their findings. As John Gever points out in a coolheaded MedPage Today analysis, the study says "the left nucleus accumbens was consistently affected by cannabis use," even though the authors acknowledge elsewhere that their data cannot prove causation. One of the study's authors, Northwestern University psychiatrist Hans Breiter, goes even further in the Society for Neuroscience press release, saying, "This study raises a strong challenge to the idea that casual marijuana use isn't associated with bad consequences." Gever responds: "Um, no, it doesn't—not without before-and-after MRI scans showing brain structure changes in users that differ from nonusers and documentation of functional impairments associated with those changes."
Yet Breiter goes further still in the Northwestern press release. "People think a little recreational use shouldn't cause a problem, if someone is doing OK with work or school," he says. "Our data directly [say] this is not the case." No, they don't, as the study itself concedes. In the Massachusetts General press release, Breiter claims the study, which was supported by the National Institute on Drug Abuse, has policy implications: "Our findings—which need to be followed up with longer-terms studies—raise serious concerns about efforts to legalize recreational marijuana use, particularly for young adults."
What is the nature of the "bad consequences" and "problem[s]" that Breiter fears? "These two brain regions have been broadly implicated in processes underlying addiction," he says, "so it's a real problem that people claiming their marijuana use does not negatively impact their lives show significant changes in these structures." Harvard psychologist Jodie Gilman, the study's lead author, offers similar comments in the Northwestern press release. "It may be that we're seeing a type of drug learning in the brain," she says. "We think when people are in the process of becoming addicted, their brains form these new connections." It is not clear what the practical significance of "these new connections" might be, since Gilman et al. emphasize that their subjects "were not dependent" on marijuana.
Presumably Gilman and Breiter are not suggesting that anyone who smokes pot once a week is destined for addiction, which seems inconsistent with research on patterns of cannabis consumption. Data from the National Comorbidity Survey indicate that 9 percent of cannabis consumers experience "dependence" at some point in their lives. According to the National Survey on Drug Use and Health, about 18 million Americans have used marijuana in the last month, while about 3 million qualify as "dependent" at some point during a given year.
While the researchers were not always careful in explaining the significance of their findings, the misunderstandings reflected in the press coverage got a strong boost from the publicity departments at Northwestern and Massachusetts General, which picked headlines for their press releases that encouraged reporters to conflate correlation with causation and differences with damage:
Those summaries are actually more misleading than a lot of the headlines in the general press, which is saying something.
Addendum: Pete Guither notes a scathing assessment of Gilman et al.'s study by U.C.-Berkeley computational biologist Lior Pachter, who calls it "quite possibly the worst paper I've read all year."
