Because, what's "highly addictive" as opposed to "addictive" or "slightly addictive"? At the very least you'd have to compare heroin to other drugs using statis

Because, what's "highly addictive" as opposed to "addictive" or "slightly addictive"? At the very least you'd have to compare heroin to other drugs using statistics to see how many people use the drug, how many times they use it, and what dosage, how many become addicted, and how many are able to break that addiction. And even then, that doesn't prove the drug itself is more addictive than other, because there are a lot of other possible explanations for why people might become addicted to one drug over another than the drug itself. There's a likely a large social aspect which has not been adequately studied: the Rat Park experiments[1] demonstrate give strong evidence that environmental factors are a key part of addiction. It's possible that people who are currently addicted to heroin would be addicted to any drug with a similar social stigma and legal status, because they would seek out behaviors they view as self-destructive, regardless of whether those behaviors actually are self-destructive or not. Do we include analysis of social aspects in what we consider to be highly addictive? And finally, there's a genetic component to addiction, especially with alcohol and benzodiazepines. Do we consider this, or are we just folding that into a flat study of the general population? In short, saying something is "highly addictive" isn't a scientific claim because we don't have any definition of what "highly addictive" would mean. We can't even start to test the assertion because we haven't decided what the assertion means. [1] https://en.wikipedia.org/wiki/Rat_Park https://en.wikipedia.org/wiki/Rat_Park