Prevalence and Correlates of GHB Use among Adults in the United States

Abstract

Gamma-hydroxybutyrate (GHB) was a somewhat prevalent club drug two decades ago, but research has been lacking regarding the current epidemiology of use. In this study, prevalence and correlates of past-year GHB use were estimated based on noninstitutionalized US adults participating in the 2015–2020 National Survey of Drug Use and Health (N= 241,675). An estimated 0.05% (95% CI: 0.04–0.07) of adults in the US have used GHB in the past year. Compared to those age 18–25, those age 35–49 were at increased odds for use (aOR = 2.28, 95% CI: 1.02–5.10), and compared to heterosexual men, gay men in particular were at high odds for use (aOR = 27.82, 95% CI: 11.09–69.80). Those identifying as bisexual men (aOR = 3.39, 95% CI: 1.17–9.82) and lesbian/bisexual women (aOR = 3.37, 95% CI: 1.44–7.90) were also at high odds for use. Despite all drugs examined being significant correlates of use in bivariable models (ps < .001), only past-year use of methamphetamine (aOR = 15.68, 95% CI: 7.40–33.25), ketamine (aOR = 8.52, 95% CI: 2.67–27.20), and ecstasy (aOR = 4.41, 95% CI: 1.80–10.75) was associated with increased odds of use in the multivariable model. Results suggest that GHB use is currently rare, but gay men and people who use other party drugs are at particularly high risk for use.

Introduction

Gamma-hydroxybutyrate (GHB) is a drug with depressant effects that has been used recreationally for decades. Aside from national studies that estimate prevalence of use, poisonings, and deaths related to use in the general population, few recent national studies have examined risk factors for use. As such, epidemiological studies that estimate correlates of use using current data are needed in order to best inform prevention, treatment, and harm reduction efforts among people at highest risk for use.

GHB is a naturally occurring acid in mammals, which was first isolated over a half-century ago (Nicholson and Balster 2001). GHB became synthesized as a drug, typically into liquid form, and became widely available during the late 1980s and 1990s as a sleep aid and as a recreational drug. After thousands of hospitalizations and dozens of deaths related to use in the US and well-publicized instances of rape involving use (Nicholson and Balster 2001; Substance Abuse and Mental Health Services Administration [SAMHSA] 2004), in 2000, GHB was scheduled in the US, making the drug illegal to possess or sell. However, GHB is legally available in some countries. Specifically, sodium oxybate is a form of GHB commonly sold under brand names such as Xyrem, and it has been prescribed to effectively treat narcolepsy-cataplexy in the US, Canada, and Europe for about two decades (Xu et al. 2019), and it has been used to treat alcohol withdrawal syndrome in Italy and Austria for over 25 years (van den Brink et al. 2018). Most GHB use, however, appears to be nonmedical in nature.

GHB is used recreationally for its euphoric effects that do not typically result in a hangover (Nicholson and Balster 2001). This drug affects the body similar to the effects of alcohol (Gessa et al. 2000) and its effects are often compared to alcohol and ecstasy (Palamar and Halkitis 2006). The GHB high is of short duration (e.g., under an hour), but given the drug’s steep dose–response curve, it is easy for users to overdose and experience unconsciousness or a coma, which can last 1–5 hours (Nicholson and Balster 2001). Other adverse effects include dizziness, nausea, vomiting, confusion, agitation, and myoclonic seizures (Kam and Yoong 1998). The drug’s effects can be attributed to a release of dopamine, glutamate, and GABA, and agonism of GABA_B receptors in particular appears to be a main driver of effects (Felmlee, Morse, and Morris 2021). Recreational use appears to be most common among people who attend nightclubs and dance parties, and GHB has also been a leading “chemsex” drug, used to facilitate sex, mainly among men who have sex with men (MSM) (Griffin et al. 2019; Halkitis and Palamar 2006; Kelly, Parsons, and Wells 2006; Maxwell, Shahmanesh, and Gafos 2019).

Prevalence estimates of GHB use in various countries continue to suggest that GHB use is rare in the general population. In the US, its use appeared to be most prevalent in the early 2000s during the peak of popularity of other club drugs such as ecstasy, ketamine, and methamphetamine (Johnston et al. 2022). Past-year use among high-school seniors in the US was estimated to be 1.4–2.0% in 2000–2004, and the estimated use in this population decreased and has been <1% since 2015 (Johnston et al. 2022). Internationally, use has also been shown to be rare. Estimated past-year use among those ages ≥14 in Australia has remained ~0.1% from 2004 to 2019 (Australian Institute of Health and Welfare 2020). In 2018, an estimated 0.3% of the people of ages 16–64 in Norway had used (0.7% of the people of ages 16–34), and 0.1% of the people of ages 15–64 in Poland had used (0.2% among the people of ages 15–34) (European Monitoring Centre for Drugs and Drug Addiction [EMCDDA] 2020). National surveys in England and Wales have not queried GHB use in a decade, but in 2011/12, an estimated 0.13% of the people of ages 16–59 had used in the past year (Home Office 2012). While prevalence of use appears to remain low in the general population, small studies have found that gay men in particular are at high risk for use (Griffin et al. 2019; Kelly, Parsons, and Wells 2006). Further, the use appears to be increasing among certain subpopulations. Specifically, a recent study of nightclub and festival attendees in New York City estimated that between 2016 and 2019, past-year use increased from 1.0% to 4.2% (Palamar and Keyes 2020).

With respect to poisonings and deaths involving GHB use, mentions of GHB use during emergency department visits appeared to peak in the US in 2000, with 4,969 related visits that year, with mentions decreasing to 2,406 in 2011 – the last year national drug-related hospitalization data were collected systematically in the US (SAMHSA 2004, 2013). Reports of poisonings related to GHB, its precursors, and analogs to Poison Control centers in the US were highest during the first year of this drug’s inclusion in their annual report in 2002 with 1,386 poisonings reported (Watson et al. 2003). The number of reported poisonings dropped to a nadir of 448 in 2008, and reports have slowly increased in recent years to 776 in 2020 (Bronstein et al. 2009; Gummin et al. 2021). In 2019, GHB was the fifth most common drug involved in poisonings in hospitals monitored by the European Drug Emergencies Network (Euro-DEN) Plus, present in 10.6% of the acute presentations and in 27% of critical care admissions (EMCDDA 2021). Data on GHB-related deaths are not available from the US, but data from England and Wales suggest there was a recent increase in related deaths to 28 in 2020 – an exponential increase compared to the number of related deaths in the early 2000s (Office for National Statistics 2021). Another recent study also determined that between 2012 and 2017, GHB-involved ambulance pick-ups in Victoria, Australia, increased from 8.8 to 21.7 per 100,000 population (Arunogiri et al. 2020).

Data on drug treatment admissions add to information on use, poisonings, and deaths. Drug treatment data in the US do not include data regarding GHB use, but in England, new treatment presentations involving GHB increased from 19 in 2005/06 to 389 in 2015/16 and has remained somewhat stable from 2016 to 2019 with 280–314 new presentations per year (Advisory Council on the Misuse of Drugs 2020).

Altogether, national statistics suggest that GHB use is rare, but there have been recent increases in use among some subpopulations (e.g., nightclub attendees) and related poisonings and deaths have recently increased in some areas. While prevalence estimates of use are informative, given that harms related to use of this drug appear to be increasing, analysis of new national data to determine correlates of use is needed. This analysis seeks to estimate the prevalence and demographic and drug use-related correlates of GHB use among the general adult population in the US in order to inform prevention, intervention, and harm reduction efforts for subpopulations that may be at current risk for use.

Methods

Procedure

Aggregated data from respondents of age ≥18 were examined from the 2015–2020 National Survey on Drug Use and Health (NSDUH), an annual cross-sectional survey of a nationally representative sample of noninstitutionalized individuals in the US (Center for Behavioral Health Statistics and Quality [CBHSQ] 2020). The survey is conducted in all 50 states and the District of Columbia, and a multistage sampling design is utilized to survey participants. Participants are surveyed in their homes, which can include noninstitutional group quarters (e.g., group homes, college dormitories, and military installations). Individuals with no permanent residence are eligible and surveyed if they utilize homeless shelters or if they are long-term residents of rooms in hotels. Surveys are administered via computer-assisted personal interviewing where the respondents listen to and read the questions. The aggregate sample size was 241,675 and the weighted interview response rates from these years ranged from 60.4% to 68.4%. This analysis was limited to 2015–2020 as 2015 was the first year of redesign of the NSDUH survey (SAMHSA 2015).

Measures

Participants were asked whether they have ever used GHB, “also called ‘G,’ ‘Georgia Home Boy,’ ‘Grievous Bodily Harm,’ ‘Liquid G,’ or gamma hydroxybutyrate.” Those reporting use were asked about recency of use (i.e., lifetime but not past-year use, past-year but not past-month use, and past-month use). This analysis focused on use in the past year (yes vs. no). Similarly, participants were asked about past-year use of various other drugs including cannabis, cocaine, ecstasy (MDMA, Molly), LSD, methamphetamine, heroin, and ketamine. They were also asked about misuse of prescription opioids, tranquilizers, sedatives, and stimulants. With respect to demographic characteristics, participants were asked about their age, sex, race/ethnicity, education, and sexual orientation. A variable was further coded combining sex and sexual orientation for analysis.

Statistical analyses

First, the prevalence of past-year GHB use was estimated. To determine whether there have been shifts in use across years, logistic regression was used examining GHB use as a linear function of time (year) as a continuous predictor. Next, Rao-Scott chi-square was used to compare the prevalence of GHB use according to each level of each independent variable in a bivariable manner. Then, all independent variables were fit into a multivariable logistic regression model, which generated adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for each variable. These variables included survey age, sex, race/ethnicity, education, sexual orientation, past-year use of cannabis, cocaine, ecstasy, LSD, methamphetamine, heroin, and ketamine, and past-year misuse of prescription opioids, tranquilizers, sedatives, and stimulants. Year of survey administration was included in this model. Survey commands were used (using Stata SE 17) and weights were utilized to account for the complex survey design, non-response, selection probability, and population distribution CBHSQ 2020). A test was also conducted to confirm that multicollinearity was not present (variance inflation factor range: 1.02–2.62). This analysis was exempt from review by the New York University Langone Medical Center institutional review board.

Results

An estimated 0.05% (95% CI: 0.04–0.07) of adults in the US have used GHB in the past year (based on 141 participants reporting use). Estimated use has remained stable across the years (ranging from 0.03% to 0.07%) with no linear, quadratic, or cubic trend detected. Table 1 presents bivariable and multivariable correlates of use. Compared to those age 18–25, those age 35–49 were at increased odds for use (aOR = 2.28, 95% CI: 1.02–5.10), and compared to heterosexual men, gay men in particular were at high odds for use (aOR = 27.82, 95% CI: 11.09–69.80). Those identifying as bisexual men (aOR = 3.39, 95% CI: 1.17–9.82) and lesbian/bisexual women (aOR = 3.37, 95% CI: 1.44–7.90) were also at high odds for use. Despite all drugs examined being significant correlates of use in bivariable models (ps < .001), the three drugs in the model to remain significant risk factors were methamphetamine (aOR = 15.68, 95% CI: 7.40–33.25), ketamine (aOR = 8.52, 95% CI: 2.67–27.20), and ecstasy (aOR = 4.41, 95% CI: 1.80–10.75).

Table 1.

Correlates of reporting past-year GHB use.

	Full Sample % (95% CI) N = 241,675	No GHB Use % (95% CI) N = 241,534	GHB Use % (95% CI) N = 141	aOR (95% CI)
Age
18–25	13.8 (13.6–14.0)	13.8 (13.6–14.0)	20.2 (13.1–29.8)c	1.00
26–34	16.0 (15.8–16.2)	16.0 (15.8–16.2)	24.5 (15.9–35.9)	1.06 (0.53–2.14)
35–49	24.6 (24.4–24.9)	24.6 (24.4–24.9)	47.1 (33.6–61.1)	2.28 (1.02–5.10)a
≥50	45.6 (45.1–46.0)	45.6 (45.1–46.0)	8.2 (2.6–23.2)	0.40 (0.08–1.95)
Sex/Sexual Orientation
Heterosexual Male	46.4 (46.1–46.8)	46.4 (46.1–46.8)	28.3 (17.0–43.1)c	1.00
Heterosexual Female	48.3 (47.9–48.6)	48.3 (47.9–48.6)	25.3 (15.2–39.1)	1.56 (0.71–3.43)
Gay Male	1.1 (1.1–1.2)	1.1 (1.1–1.2)	32.7 (20.9–47.1)	27.82 (11.09–69.80)c
Lesbian/Bisexual Female	3.3 (3.1–3.4)	3.2 (3.1–3.4)	10.5 (5.4–19.6)	3.37 (1.44–7.90)b
Bisexual Male	0.9 (0.9–1.0)	0.9 (0.9–1.0)	3.2 (1.3–7.8)	3.39 (1.17–9.82)a
Race/ethnicity
Non-Hispanic White	63.7 (63.1–64.2)	63.7 (63.1–64.2)	61.9 (49.9–72.6)b	1.00
Non-Hispanic Black	11.9 (11.5–12.3)	11.9 (11.5–12.3)	5.2 (2.6–10.4)	0.40 (0.16–1.00)
Hispanic	16.2 (15.7–16.6)	16.1 (15.7–16.6)	27.1 (17.4–39.7)	1.61 (0.84–3.10)
Other/Mixed	8.3 (8.1–8.5)	8.3 (8.1–8.5)	5.7 (2.9–11.1)	0.68 (0.33–1.41)
Education
College Degree or Higher	31.4 (31.0–31.9)	31.4 (31.0–31.9)	33.5 (21.9–47.6)	1.00
Less than High School	12.5 (12.2–12.7)	12.5 (12.2–12.8)	6.1 (3.3–11.0)	0.39 (0.13–1.23)
High School	25.3 (25.0–25.6)	25.3 (25.0–25.6)	32.3 (21.8–44.8)	1.45 (0.64–3.31)
Some College	30.8 (30.5–31.1)	30.8 (30.5–31.1)	28.1 (19.5–38.7)	0.78 (0.35–1.73)
Past-Year Other Drug Use
Cannabis	15.9 (15.7–16.1)	15.9 (15.7–16.1)	64.4 (51.7–75.4)c	1.71 (0.69–4.23)
Cocaine	2.1 (2.0–2.2)	2.1 (2.0–2.1)	42.2 (30.2–55.3)c	1.46 (0.41–5.16)
Ecstasy	1.0 (0.9–1.0)	1.0 (0.9–1.0)	37.8 (25.8–51.4)c	4.41 (1.80–10.75)b
LSD	0.8 (0.7–0.8)	0.8 (0.7–0.8)	25.2 (16.4–36.8)c	1.13 (0.32–3.94)
Methamphetamine	0.7 (0.7–0.8)	0.7 (0.7–0.8)	43.1 (32.7–54.3)c	15.68 (7.40–33.25)c
Heroin	0.3 (0.3–0.4)	0.3 (0.3–0.4)	14.9 (7.7–27.0)c	1.43 (0.47–4.32)
Ketamine	0.2 (0.1–0.2)	0.2 (0.1–0.2)	25.8 (16.4–38.2)c	8.52 (2.67–27.20)b
Prescription Opioids	4.0 (3.9–4.1)	4.0 (3.9–4.1)	31.2 (21.3–43.1)c	1.23 (0.50–3.01)
Tranquilizers/Sedatives	2.5 (2.4–2.6)	2.5 (2.4–2.6)	34.4 (23.9–46.6)c	1.81 (0.82–3.98)
Prescription Stimulants	2.0 (1.9–2.1)	2.0 (1.9–2.0)	29.7 (19.4–42.5)c	(1.33 (0.68–2.60)

aOR = adjusted odds ratio; CI: confidence interval.

^ap < .05,

^bp < .01,

^cp < .001

Discussion

GHB use appears to be very rare in the US, with use far less prevalent than in the early 2000s (Johnston et al. 2022). However, this study determined some subpopulations that are still at high risk for use. Men and women of sexual minority status in general were found to be at higher risk than heterosexual men, but gay men in particular are at high risk for use. This result corroborates both older and newer studies of clubgoers that found that gay men were at highest risk for use compared to men and women of other sexual orientations (Griffin et al. 2019; Kelly, Parsons, and Wells 2006). Chemsex has been found to be a driving factor of use among MSM (Halkitis and Palamar 2006; Maxwell, Shahmanesh, and Gafos 2019), so it is likely that a large portion of current use among gay men is in sexual contexts.

Past-year use of methamphetamine, ketamine, and ecstasy was also found to be associated with increased risk for GHB use. These three drugs were also the three most common party drugs combined with GHB among gay and bisexual clubgoers two decades ago (Halkitis and Palamar 2006), and a recent study of ambulance pick-ups in Victoria, Australia, found that co-use of methamphetamine, another common chemsex drug, was common (Arunogiri et al. 2020). Despite the risk for use related to sexual orientation and drug use remaining somewhat similar over the years, a unique finding of this study is that use is estimated to be most prevalent among older adults of ages 35–49. This is somewhat unique because many other studies have found that the average age of people who use or initiate GHB, who require treatment for GHB use, or who experience GHB-related mortality tend to be between their 20s or early 30s, although the average age of use may be slightly older among gay and/or bisexual men (Corkery et al. 2015; Dijkstra et al. 2021). Further, despite this younger average age described in many other studies, there isin fact still a small proportion of overdoses of users in their 40s and 50s (Abid et al. 2022; Chin et al. 1998). This finding is also unique because use of most other illegal drugs tends to be most prevalent among younger adults (CBHSQ 2020). More research is needed to determine why GHB use is more prevalent in the US among older adults.

Despite the low estimated prevalence of GHB use in the general population, it is important to monitor use of this drug as prevalence of use appears to be increasing in certain populations such as nightclub and festival attendees (Palamar and Keyes 2020). Further, populations at risk – particularly gay men – could benefit from education about this drug. While many people in nightclub-attending and gay populations appeared to be educated about this drug years ago, generational forgetting can occur, and it is possible for new cohorts to be introduced to this potentially dangerous drug in which they are unfamiliar with its effects and potential deleterious outcomes.

This study has limitations. Institutionalized individuals and the homeless who do not use shelters were underrepresented. Due to no lesbian women reporting GHB use, lesbian and bisexual women needed to be collapsed into a single category. Data were limited to adults because sexual orientation was not queried among adolescents in the full sample. The outcome variable (GHB use) was very rare, with only 141 participants reporting use. Modeling such a rare outcome using regular logistic regression (with survey commands for national weighted data) led to inflated 95% CIs. Firth and rare event logistic regressions were used to test the reliability of estimates (using unweighted data) and the model results were similar with slightly lower aORs and often slightly tighter 95% CIs. It should be noted, however, that tranquilizer/sedative use was a significant correlate in these models but not in the model presented based on weighted data. Independent variables included in the analysis were limited to factors assessed in the NSDUH survey and greater detail on GHB use (e.g., frequency of use) could not be examined. As such, this study lacks data on severity of use, so results can only inform who is at risk for any recent use. NSDUH did not ask about GHB prodrugs gamma-butyrolactone (GBL) or 1,4-butanediol (1,4-BD), so it is unknown whether any participants reported use as GHB. Further, although GBL metabolizes into GHB once ingested, it may be noteworthy that some participants may have used GBL thinking it was GHB (Busardò et al. 2018). Finally, it is assumed that reported use was nonmedical use, and misuse of prescription formulations such as Xyrem was likely underreported.

This was among the first papers to estimate trends and correlates of use of GHB among a national sample. GHB use is rare, but some subpopulations are at high risk for use. Gay men in particular remain at high risk, so prevention, intervention, and harm reduction efforts should be aimed toward this population in particular.