Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
So much has happened in America in recent months that one big positive story has probably got less attention than it deserves. A few weeks ago, the US Centers for Disease Control and Prevention published data showing that fentanyl deaths have been declining for 11 straight months, and now stand 20 per cent below their 2023 peak.
This is a stunning turnaround in the opioid epidemic that has been ravaging the country for more than a decade, killing hundreds of thousands, and has reduced US male life expectancy by a whole year.
But what’s behind the reversal? And what does the answer mean for how confident we can be that the decline will continue?
Some officials say new drug treatment initiatives and products like naloxone deserve the credit. If true, this would be hugely welcome news, and would mean officials now have a proven strategy that can be rolled out more widely until the epidemic is fully in remission.
But there are two problems. The first is that there is very limited evidence that these interventions really do move the needle. And the second is that these explanations don’t fit the data.
One of the striking things about the reversal in fentanyl deaths is that it is spreading smoothly from the east coast of the US to the west. Charles Fain Lehman, a crime researcher with the Manhattan Institute, notes that this is an almost perfect echo of the drug’s initial spread a decade ago.
By contrast, interventions from the government and health officials have generally either taken the form of blanket national policies or a hodgepodge of efforts at the state or local level with no geographical pattern.
The evidence suggests that whatever is driving this trend is probably coming not from the top down, but from the bottom up: from suppliers, substances and users.
There are several plausible theories as to what’s happening. One is simple: there is growing evidence that the supply of fentanyl is down, and when there’s less fentanyl around, fewer people die. Data from Ohio shows deaths rose and are now falling in almost perfect lockstep with supply. My analysis of national data shows the same pattern, and I also find that just like the deaths, the dip in supply started in the east.
Why would supply be down? A branch of Mexico’s Sinaloa cartel thought to have been one of the largest fentanyl suppliers has reportedly banned its production in a bid to ease pressure from law enforcement. There is some scepticism as to how real or lasting this stance will be, given the economic incentives, but for now the dip in circulation looks real.
Another supply theory is particularly intriguing. In recent years, gangs have begun mixing fentanyl with the horse tranquilliser xylazine, whose properties alter the fentanyl high. This is believed to postpone withdrawal symptoms in some cases, meaning users take fewer hits, reducing the risk of a deadly overdose. Xylazine is also thought to be one factor behind the shift among users from injecting to smoking fentanyl, which tends to reduce its lethality.
This is also another perfect fit for the east-west pattern. According to my analysis of state-level data on drug supply and deaths, the states that are furthest into their declines in fentanyl mortality are those where xylazine has been circulating for the longest.
To be clear, these theories remain smoking guns rather than definitive conclusions. A third possibility is the idea put forward by Nabarun Dasgupta and colleagues at the US academic collective Opioid Data Lab, that much like an infectious disease epidemic, fentanyl has now worked its way through its susceptible population. Some died, others figured out ways to use without overdosing, so the remaining fentanyl-naive drug-using population has shrunk. The wave-style dynamics at work here mean it would also theoretically fit the east-west pattern.
In short, we don’t know exactly why deaths are coming down, but on the balance of evidence, policy is probably playing a smaller role than might be assumed.
Tens of thousands of lives saved is a hugely positive development, but the possibility that it’s down to a chance narcotic innovation that on this occasion may have saved lives instead of ending them demonstrates how unpredictable the drug landscape can be.