A couple of years ago, Maine's Republican governor, Paul LePage, vetoed a bill allowing pharmacists to dispense naloxone, an opiod antagonist that can save people's lives by reversing overdoses, without a prescription. "Naloxone does not truly save lives; it merely extends them until the next overdose," LePage wrote in his veto letter. "Creating a situation where an addict has a heroin needle in one hand and a shot of naloxone in the other produces a sense of normalcy and security around heroin use that serves only to perpetuate the cycle of addiction."
Maine legislators, who overrode LePage's veto, apparently disagreed with his objections. But a paper recently published by the National Bureau of Economic Research lends empirical support to LePage's argument about naloxone's impact on risky behavior, if not moral support to the implication that drug use should be as dangerous as possible for the sake of deterrence.
University of Virginia economist Jennifer Doleac and University of Wisconsin at Madison economist Anita Mukherjee wondered whether wider availability of naloxone, in addition to reversing overdoses that would otherwise be fatal, might encourage opioid use by making it less dangerous. Depending on how big that effect is, Doleac and Mukherjee write, "expanding Naloxone access might not in fact reduce mortality." Although "the risk of death per opioid use falls," they say, "an increase in the number or potency of uses means the expected effect on mortality is ambiguous."
Economists call this kind of effect "moral hazard": When people are protected from the consequences of their risky behavior, they may be more inclined to take risks. A classic example is seat belts, which protect motorists from potentially fatal injuries but may thereby encourage riskier driving. While "the moral hazard from seatbelts" seems to be "small relative to the safety-improving effect of seatbelts," Doleac and Mukherjee observe, research suggests that "automobile insurance, which also incentivizes riskier driving through moral hazard, causes a large increase in traffic fatalities." They also note evidence that new HIV treatments encourage riskier sexual practices.
To investigate the net impact of naloxone, Doleac and Mukherjee looked at what happened after states adopted laws aimed at encouraging its use by making it available without a doctor's prescription or by providing legal immunity to people who prescribe or administer it. Every state had enacted some such law by 2017. Over all, Doleac and Mukherjee found, the adoption of naloxone laws was associated with an increase in opioid-related theft, an increase in opioid-related emergency room visits, and no reduction in opioid-related deaths. "While Naloxone has great potential as a harm-reduction strategy," they conclude, "our analysis is consistent with the hypothesis that broadening access to Naloxone encourages riskier behaviors with respect to opioid abuse."
The increase in opioid-related theft, which amounts to something like five more thefts per 1 million residents, was neither large nor statistically significant by the conventional standard. Doleac and Mukherjee say the results "suggest that any social costs of Naloxone laws—in terms of additional property crime—are small." By contrast, the increase in opioid-related E.R. visits, 266 per 100,000 residents in each quarter, "is large and consistent with the hypothesis that Naloxone access increases the abuse of opioid drugs."
More E.R. visits do not necessarily translate into more deaths, since increased use of naloxone means any given overdose, whether or not it results in a trip to the hospital, is less likely to be fatal. "On average across all urban areas, we find that these laws have no signicant impact on the opioid-related death rate," Doleac and Mukherjee report. "Thus, while the risk per use has gone down due to Naloxone access, the number of uses increases enough that we find no net effect on opioid-related mortality."
That overall finding masks regional differences. In the Midwest, Doleac and Mukherjee say, naloxone laws were associated with a 14 percent increase in opioid-related deaths. Mortality also rose in the South, while it fell in the West and Northeast, but none of those changes was statistically significant. In rural areas outside of the Midwest, naloxone laws were associated with declines in opioid-related deaths, although the results were "generally statistically insignificant." The analysis also finds reduced mortality in the 25 largest cities, an encouraging but still statistically insignificant finding.
Dolerac and Mukherjee suggest one reason for the variability might be the availability of drug treatment. "It appears that Naloxone access increases opioid-related mortality in places with limited treatment and decreases it in places with more treatment," they write. "We do not have enough statistical power to be sure that these effects are statistically different from one another, but this pattern is consistent with the hypothesis that broadening Naloxone access has less detrimental effects in places with more resources available to help those suffering from addiction."
Dolerac and Mukherjee's main finding, that greater access to naloxone was not associated with a decrease in opioid-related deaths, contrasts with the conclusions of an NBER paper published last year. In that study, University of Colorado at Denver economist Daniel Rees and his co-authors found that the adoption of naloxone laws was associated with a decrease in opioid-related deaths of 9 to 11 percent. Rees et al. used annual state-level data through 2014, while Doleac and Mukherjee used monthly city- and county-level data through 2015.
Assuming that Doleac and Mukherjee's results hold up, do they prove Paul LePage right? "Our findings do not necessarily imply that we should stop making Naloxone available to individuals suffering from opioid addiction, or those who are at risk of overdose," they write. "They do imply that the public health community should acknowledge and prepare for the behavioral effects we find here."
More to the point, the collectivist calculus of public health tends to obscure the moral issue raised by legal obstacles that make naloxone harder to obtain. The morally relevant level of analysis is not "society as a whole" but the individual who wants naloxone and the state that stands in his way. Naloxone indisputably saves people's lives, and it would be unconscionable to block access to it based on speculation about how the availability of that lifesaving option might affect other people's behavior. That is like banning seat belts or HIV treatment because the extra assurance they provide might encourage some people to behave more recklessly.
This is the logic of prohibition, which endangers the lives of drug users to deter people who otherwise might join them. One way it does that is by making drug potency unpredictable, which makes overdoses more likely, thereby increasing the need for naloxone. LePage is not wrong to think that making naloxone hard to get is consistent with this plan. He is wrong to think the plan is morally defensible.