Last summer, a study published in the Proceedings of the National Academy of Sciences sparked a new round of worries about the dangers of smoking pot—especially for those who start smoking at younger ages. The study found that consistent marijuana use gradually eroded cognitive functioning and IQ, and with the legalization of recreational marijuana in Colorado and Washington, it’s made an appearance in a number of articles arguing that legalized pot poses a serious health hazard. Today, though, a new study published in the very same journal—and using the very same data set—suggests that the case against marijuana is a little less cut-and-dry.
Ole Røgeberg, a researcher at the Frisch Centre for Economic Research in Norway, analyzed the same survey results and found that the declines in cognitive abilities could be entirely attributed to socioeconomic factors. As a result, “the true effect” of marijuana use, he argues, “could be zero.”
Røgeberg is careful to note that his reinterpretation of the data doesn’t entirely discredit the original study, but he does write that its “methodology is flawed and the causal inference drawn from the results premature.”
Both the new and old studies draw upon a data set of 1,037 individuals from Dunedin, New Zealand, who were followed from their birth (either in 1972 or 1973) until they turned 38. At the ages of 18, 21, 26, 32 and 38, each of them were interviewed and scored for marijuana use. The original study found that IQ decline increased proportionately with cannabis dependence—especially for those who started smoking earlier on—and the authors concluded that using the drug was the cause of the decline.
Røgeberg, though, dug a little deeper into the data. He found those who started using marijuana during adolescence were disproportionately likely to have poor self-control and conduct problems in school—both factors that are themselves correlated with low socioeconomic status. In particular, members of the study with these traits were more likely to come from a Maori background, a group indigenous to New Zealand that has much higher unemployment, poverty and incarceration rates than the country’s population as a whole.
Numerous other studies have shown that low socioeconomic status adolescents are more likely to experience steeper IQ declines during adulthood. (Researchers hypothesize this is a result of being exposed to less intellectually stimulating environments.) As a result, Røgeberg wondered, could socioeconomic factors explain the IQ declines originally attributed to marijuana?
In his simulation, he tested whether socioeconomic environmental factors (dropping out of school, being exposed to less stimulating environments, and so on) could conceivably drive the same IQ declines reported in the group without turning to marijuana as an explanation. His statistical analysis found that these other factors could indeed completely account for the cognitive declines observed.
For support, he also points to a 2002 Canadian study that also asked whether long-term marijuana use impacted IQ, but with data entirely from middle-class survey participants. That paper found that IQ only decreased for current cannabis users, and when even heavy users stopped smoking, their IQ rebounded. Since that study largely excluded socioeconomic factors and did not find a permanent trend, he feels that it supports his argument that such factors play a major role.