This June, Dr. Carl Hart, a neuroscientist at Columbia University, published "High Price," a memoir about how he came to devote himself to science after growing up around drugs and gangs. With his publisher's permission, we republish an excerpt below.
Dozens [of writers] lined up to meet with the police officer and attorney. They wanted to hear more about how methamphetamine caused gay men to engage in sexual practices that increased HIV rates; how it kept people up for consecutive days without sleeping; how the drug made people behave irrationally; and how it ruined people’s teeth and made them unattractive.
While some of the writers were undoubtedly there simply seeking a sexy story to sell, I think most genuinely wanted to learn about the drug and, if needed, to warn the public about its dangers. They weren’t thinking about separating anecdote from evidence. They had just heard from a U.S. attorney and a cop that this drug was nasty stuff. The government invited both of these individuals as experts on the topic. As a result, there didn’t seem to be a need to separate fact from fiction. Of course the information was factual. Otherwise, it would not have been presented in a forum sanctioned by the government, would it?
I pondered this and other questions as I rode the subway back to my lab. Why was my data so inconsistent with the stories told by the other panelists? Was I out of touch with the way people use drugs in the real world? Maybe the doses that I tested were too low, I thought. I had intentionally started off with low doses to ensure the safety of my research participants.
At that point, the largest dose I had given was 20 mg, which is considerably lower than doses reportedly used by methamphetamine addicts. Perhaps the individuals described by the prosecutor and police officer used much larger doses than those tested in my studies. This might explain our disparate conclusions. I also thought about how methamphetamine is typically used outside the lab—snorted, injected, or smoked. This ensures that the drug hits the brain more quickly and produces more intense effects.
In my studies, it was swallowed. When taken this way it produces the least intense effects. Given these caveats, I questioned whether the data collected in my studies was relevant to the situation in the real world. I figured the hysteria about methamphetamine had to reflect something about reality and that my studies, up until that point, had not captured it.
Over the next seven years, I went about trying to resolve this issue. I searched the human literature to see if anyone had studied larger methamphetamine doses when the drug was snorted, smoked, or injected. There was virtually nothing. I thought about José Martí’s famous quote in his 1882 essay “On Oscar Wilde”: “A knowledge of different literatures is the best way to free one’s self from the tyranny of any of them.”
So I read the literature on animal studies looking for information that might be relevant to human addiction. These studies showed that the drug caused extensive damage to certain brain cells and produced severe learning and memory problems. Aha, I got it! Finally, here was some data that was in line with popular anecdotes about methamphetamine.
But as I looked more closely it became clear that the animal results had serious limitations and might not be applicable to people. For one, the amounts of methamphetamine given to animals are far more than amounts taken by methamphetamine addicts. If one gave similarly high doses of caffeine or nicotine to animals, the same serious toxic effects would be seen.
But when animals were given methamphetamine doses comparable to those used by people, the destructive effects were not observed. During my graduate education, the notion that methamphetamine damaged brain cells was an unquestioned fundamental truth in drug research. Now this basic belief needed to be qualified, making it difficult to extrapolate to people.
Next, I studied the literature on the long-term effects of methamphetamine in addicts. These were people who had used the drugs for many years. In these studies, abstinent methamphetamine addicts and a control group (usually non–drug users) completed a comprehensive set of cognitive tests over the course of several hours, and the results were compared to determine whether the cognitive functioning of the methamphetamine addicts was normal.
Of course, normality is a relative concept that is determined by not only comparing performance of the methamphetamine group with the performance of a control group but also comparing the methamphetamine group’s scores with those from a normative dataset, taking into consideration the individual’s age and level of education.
These requirements are important because they allow us to take into account the relative contribution of age and education in terms of the individual’s score and adjust the score accordingly. Simply stated, it would be inappropriate to compare the vocabulary scores of a sixteen-year-old high school dropout with those of a twenty-two-year-old college graduate. The older college graduate would be expected to outperform the younger dropout.
Study after study found that methamphetamine addicts had severe cognitive impairment. In one study by Sara Simon and colleagues, the apparent impairments were so bad that it led them to warn: “The national campaign against drugs should incorporate information about the cognitive deficits associated with methamphetamine. . . . Law enforcement officers and treatment providers should be aware that impairments in memory and in the ability to manipulate information and change points of view (set) underlie comprehension . . . methamphetamine abusers will not only have difficulty with inferences . . . but that they also may have comprehension deficits . . . the cognitive impairment associated with [methamphetamine abuse] should be publicized. . . .”
As I read this and similar papers more critically, I noticed something intriguing. While it was true that the controls had outperformed methamphetamine addicts on a few tests, the performance of the two groups wasn’t different on the majority of tests. More important, when I compared the cognitive scores of the methamphetamine addicts in the Simon study against scores in a larger normative dataset, none of the methamphetamine users’ scores were outside the normal range. This meant that the cognitive functioning of the methamphetamine users was normal.
This should have tempered the researchers’ conclusions and prevented them from stating such dire warnings. What’s more, the methamphetamine literature was filled with similar unwarranted conclusions; as a result, the apparent methamphetamine addiction–cognitive impairment link has been widely publicized—numerous articles have appeared in scientific journals and the popular press.
The reporting of brain imaging findings has been especially misleading. On July 20, 2004, for example, the New York Times printed an article titled, THIS IS YOUR BRAIN ON METH: A “FOREST FIRE” OF DAMAGE. It stated, “People who do not want to wait for old age to shrink their brains and bring on memory loss now have a quicker alternative—abuse methamphetamine . . . and watch the brain cells vanish into the night.”
This conclusion was based on a study that used magnetic resonance imaging (MRI) to compare brain sizes of methamphetamine addicts with non-drug- using healthy people. The researchers also looked at the correlation between memory performance and several brain structural sizes. They found that methamphetamine users’ right cingulate gyrus and hippocampus were smaller than those of controls by 11 and 8 percent, respectively. Memory performance on only one of four tests was correlated with hippocampal size (that is, individuals with larger hippocampal volume performed better).
As a result, the researchers concluded, “chronic methamphetamine abuse causes a selective pattern of cerebral deterioration that contributes to impaired memory performance.” This interpretation, as well as the one printed in the Times article, is inappropriate for several reasons.
First, brain images were collected at only one time point for both groups of participants. This makes it virtually impossible to determine whether methamphetamine use caused “cerebral deterioration,” because there might have been differences between the groups even before methamphetamine was ever used.
Second, the non–drug users had significantly higher levels of education than methamphetamine users (15.2 versus 12.8 years, respectively); it is well established that higher levels of education lead to better memory performance. Third, there were no data comparing methamphetamine users with controls on any memory task. This, in itself, precludes the researchers from making statements regarding impaired memory performance caused by methamphetamine.
Nonetheless, the only statistically significant cognitive finding was a correlation of hippocampal volume and performance on one of the four tasks. This finding is the basis for the claim that methamphetamine users had memory impairments, because the hippocampus is known to play a role in some long-term memory; but other brain areas are also involved in processing long-term memory. The size of these other areas was not different between the groups.
Finally, the importance on everyday functioning of the brain differences is unclear because an 11 percent difference between individuals, for example, is most likely within the normal range of brain structure sizes.