Language Matters in Combatting the Opioid Epidemic: Safe Consumption Sites Versus Overdose Prevention Sites

Opioid overdoses are the leading cause of injury-related death in the United States and contribute to reversals in life expectancy gains.¹ In the face of escalating mortality, awareness is growing that wide-scale adoption of multiple evidence-based approaches—including harm reduction strategies—will be critical to reversing the epidemic. Harm reduction approaches such as syringe services and naloxone distribution programs aim to minimize negative health, social, and economic consequences of drug use for people who use drugs. They have been shown to reduce overdose deaths, HIV transmission, and hepatitis C–related risk behaviors among individuals using drugs.

Safe consumption sites where individuals can legally use preobtained drugs under medical supervision is another evidence-based harm reduction approach being considered in various locations, including San Francisco, California; Seattle, Washington; Baltimore, Maryland; and Philadelphia, Pennsylvania. Such sites are operating in Canada and Western Europe and have been shown to decrease the harms of opioid use—including reducing overdose deaths, decreasing HIV and hepatitis C infection, and lowering rates of public syringe disposal—and they have not led to increases in crime or drug use in surrounding neighborhoods.^2,3 They offer opportunities to connect individuals to primary care, detoxification, drug and HIV treatment, and housing. Although “safe consumption sites” is one of the most common terms to describe these facilities, they are also known—particularly outside the United States—as supervised injecting facilities, overdose prevention sites, or drug consumption rooms.

Safe consumption sites have not been implemented in the United States, in part because of low public support; a recent national survey found that only 29% of adults supported legalizing safe consumption sites,⁴ suggesting a potential role for strategic communications efforts to drive up public support. Small changes in language—known as framing effects—can sometimes shift attitudes fairly dramatically, including on controversial public health topics. To explore the role of strategic communications, we fielded two Web-based opinion surveys to examine whether changing the name “safe consumption sites” to “overdose prevention sites” could increase public support.

2017 WEB-BASED SURVEYS

We fielded two separate Web-based surveys—in July 2017 and November 2017—using the opinion research firm GfK. GfK uses a probability-based panel of 50 000 US adults, including those in cell phone–only and non-Internet households, which the firm provides with a Web-enabled device. It recruits panelists using equal-probability sampling, with a published sample frame of residences covering 97% of households.

In both surveys, respondents read three sentences:

1. Opioids are a type of drug that includes heroin, synthetic opioids like fentanyl, and prescription opioids like Percocet and Oxycontin.

2. In 2015, more than 33,000 people from across the country died from an opioid overdose.

3. Rates of opioid overdose in the United States have quadrupled since 1999 and there are currently over 90 opioid overdose deaths each day.

Next, respondents to the first survey read this definition:

Safe consumption sites are places where people who use drugs can bring in previously purchased opioids and other drugs and legally use them under medical supervision. These sites have been suggested in a number of different locations as a way to address the opioid epidemic.

Respondents to the second survey read the identical definition except “overdose prevention site” was used instead of “safe consumption site.” All respondents were asked, “Do you support or oppose the legalization of safe consumption sites [overdose prevention sites] in your community?”

A total of 1004 respondents completed survey 1 and 1010 respondents completed survey 2, with completion rates of 70% and 59%, respectively. All respondents were aged 18 years or older. On the basis of sociodemographic characteristics, GfK created survey weights to account for known selection deviations and survey nonresponse so that estimates were representative of the US population. The protocol was determined exempt by the Johns Hopkins Bloomberg School of Public Health institutional review board. We used the Pearson χ² test to examine differences in proportions of respondents supporting safe consumption sites versus overdose prevention sites. On the basis of state of residence, we divided respondents for both surveys into tertiles by states with high, medium, and low drug overdose rates using 2015 Centers for Disease Control and Prevention data.¹

SUPPORT FOR OVERDOSE PREVENTION SITES

Forty-five percent of respondents supported legalizing overdose prevention sites compared with 29% supporting safe consumption sites, a statistically significant 16-percentage-point difference. Support was significantly higher for overdose prevention sites among most groups, including women, men, nearly all age groups, non-Hispanic Whites and other non-Hispanic groups, and nearly all educational and household income levels, employment categories, and regions (Table A, available as a supplement to the online version of this article at http://www.ajph.org). Support levels for safe consumption sites versus overdose prevention sites were not significantly different among those with less than a high school education (31.1% vs 31.6%) or with incomes below $25 000 (32.9% vs 33.2%). Support was significantly higher (P < .001) for overdose prevention sites than for safe consumption sites independent of whether respondents lived in states with low, medium, or high drug overdose rates (Table 1). Compared with support in regions hardest hit by overdose deaths, support for overdose prevention sites was higher in regions with the lowest overdose death rates (39% vs 48%; P = .03) and in midlevel regions (39% vs 47%; P = .06).

TABLE 1—

Support for “Overdose Prevention Sites” vs “Safe Consumption Sites” Among Respondents in States With High, Medium, and Low Drug Overdose Rates: United States, 2017

States by Tertile of Overdose Deaths	Overdose Prevention Site, %	Safe Consumption Site, %	Difference, Percentage Pointa
Highest tertile	39	26	13***
Medium tertile	47	29	18***
Lowest tertile	48	30	18***

Note. Drug overdose rates include all illicit drugs. The Centers for Disease Control and Prevention does not publish opioid-specific overdose rates for all states because about half of states do a poor job reporting on the specific substance involved in drug overdose deaths.

Source. Rudd et al.¹

^aThe null hypothesis that the differences were jointly equal across tertiles was not rejected (P = .583).

^***P < .001.

MINOR WORDING CHANGE

Use of the term “overdose prevention sites” garnered significantly higher support among a national sample of US adults compared with the term “safe consumption sites.” The magnitude of the difference was large—16 percentage points—suggesting that a fairly minor wording change could matter in shifting public support for this harm reduction intervention.

These findings raise the larger question of why so few Americans support this evidence-based policy to combat drug overdose deaths. One likely explanation is pervasive, persistent stigma toward people who use drugs. Rates of social stigma toward this group are significantly higher even than stigma reported toward individuals with mental illness.⁵ Stigma shapes beliefs about appropriate responses to the opioid epidemic. For example, one study found that high stigma ratings are associated with lower support for public health–oriented policies and higher support for punitive policies aimed at individuals using drugs (e.g., prosecuting women if there is evidence of narcotic use during pregnancy).⁶ Results suggest the importance of better understanding how exposure to the epidemic (e.g., living in a hard-hit state, direct experience with drug use) and other individual characteristics (e.g., education level) affect support for harm reduction.

A small but growing literature suggests that strategic communication approaches have the potential to reduce stigma and increase policy support to combat the epidemic.⁷ The public might be more amenable to the concept of an overdose prevention site because it is, in communication parlance, a consequence frame. The term overdose prevention site, by stressing the notion of how many people are dying, has the potential to expand the subset of the population willing to view the policy as acceptable. By contrast, the term safe consumption site emphasizes making an illegal activity safer for a highly stigmatized population. It is important to note that advocates might not have full control over the language used in public discourse to describe harm reduction.

LANGUAGE MATTERS

The notion that language matters is increasingly being recognized with the encouragement of “person-first language”; for example, in the Associated Press’s 2017 Stylebook directing reporters to avoid words like “addict” and “junkie” and in a 2017 White House memorandum to federal agencies on changing the language of addiction. Moving forward, it is critical to understand how communication campaigns can decrease stigma toward people using drugs and whether this can, in turn, increase public support for historically controversial, evidence-based approaches to combating the overdose epidemic, including harm reduction measures.