Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: Addict Behav. 2020 Aug 11;112:106606. doi: 10.1016/j.addbeh.2020.106606

How do medical and non-medical use of z-drugs relate to psychological distress and the use of other depressant drugs?

VS Tardelli 1, TM Fidalgo 1, Martins SS 2
PMCID: PMC7572587  NIHMSID: NIHMS1623199  PMID: 32818728

Abstract

Background

Z-drugs are hypnotic drugs used for insomnia with considerable potential of abuse. We investigated the relationship of past-year medical and non-medical use of z-drugs with past-year: i) psychological distress; ii) medical use, non-medical use, and DSM-IV use disorder of Benzodiazepine (BZD) tranquilizers; iii) use and DSM-IV alcohol and marijuana use disorders, recreational drugs with a depressant effect on the central nervous system.

Methods

Data came from the 2015–2017 NSDUH n=128,740). Participants aged 18+ were asked if they had used any Z-Drug medically or non-medically in the past year. We investigated the associations between the three-level z-drugs variable with psychological distress and BZD, alcohol, and marijuana variables in multinomial logistic regression models.

Results

Past-year prevalences of z-drug use were 3.3% for medical and 0.5% for non-medical use. Medical and non-medical users of z-drugs had higher risk of psychological distress, compared to nonusers. Medical and non-medical users of z-drugs had higher risk of medical use, non-medical use, and BZD use disorders. Compared to non-users of z-drugs, medical users had higher risk of marijuana and alcohol use and alcohol use disorders, while non-medical users had higher risk of marijuana use and marijuana use disorders and alcohol.

Conclusion

Z-drug users have elevated risk of use and dependence of BZDs and recreational drugs, possibly due to concurrent prescribing and self-medication.

Keywords: Z-Drugs, hypnotics, psychological distress, distress, marijuana, alcohol, benzodiazepines, sedatives

Introduction

Z-Drugs (zolpidem, zopiclone, eszopiclone, and zaleplon) are non-benzodiazepine hypnotic drugs that act by enhancing the GABA-A receptor activity (Sanger, 2004). They are commonly prescribed to treat insomnia and were originally developed as a safer alternative to benzodiazepines (BZDs), especially for long-term use. Early clinical trials pointed the safety of their long-term prescriptions (Scharf, 1994), which boosted prescription rates, in particular among the elderly, throughout the past two decades (Kaufmann et al., 2016a). However, the safety of their prescription was not confirmed by later evidence, as z-drugs have been related to high risk of dementia and falls especially among the elder, much like BZDs (Brandt and Leong, 2017).

Also like BZDs, severe cases of substance use disorder from z-drugs are frequent in clinical practice, some of which require hospitalization to taper off the medication (Bajaj et al., 2019; Kapil et al., 2014). A convenience sample from the United Kingdom (n = 1,500) showed that 29.6% of the sample had already misused a BZD or a z-drug, of which 11.2% were at least weekly misusers and 3.4% of were daily misusers (Kapil et al., 2014). Populational data on use disorders of z-drugs are not available in the literature. Zopiclone was the second most frequent misused drug in that sample The abuse of z-drugs potentializes their dissociative symptoms and intensifies their long-term effects; moreover, their consumption in high dosages may have psychostimulant effects including euphoria and craving (Chattopadhyay et al., 2016) and a potentially severe withdrawal syndrome (Chiaro et al., 2018). Despite these risks, z-drugs are categorized as schedule IV drugs by the United States Department of Justice Enforcement Administration, meaning that they are considered of low potential of abuse.

A repeated cross-sectional analysis using survey data from the US National Ambulatory Medical Care Survey reported an increase in z-drug prescriptions between 1993 and 2010 that was not paralleled by a decrease in the prescription of BZDs (Kaufmann et al., 2016a). Other studies found this increase mostly attributable to growth in continuing prescriptions rather than new prescriptions (Kaufmann et al., 2016b, 2018). A study assessing Emergency Department visits related to adverse events of BZDs and non-BZDs hypnotics between 2004–2011 found that combinations among BZDs and z-drugs were significantly more related to severe outcomes (hospitalization, patient transfer or death) when compared to combinations of sedative-hypnotics other than BZDs and z-drugs (Kaufmann et al., 2017). Even after the Food and Drug Administration (FDA) demanded labeling changes for zolpidem in 2013 (Food and Drug Administration, 2013), new prescriptions of zolpidem did not decrease. However, there was a shift towards lower-dose formulations for initial zolpidem prescriptions (Kesselheim et al., 2017; Norman et al., 2017).

It is well-known that stress leads to drug use and vulnerability to drug use disorder (Sinha, 2008). The relationship between drug use and psychological distress is well established in the literature (Green et al., 2012; Gyawali et al., 2016; Newcomb et al., 1999). Causation is likely bidirectional, with self-medication (Khantzian, 1985) acting as a mediator in the causal effect of psychological distress on drug use while impaired functioning mediates the causal effect from drug use to psychological distress (Newcomb and Bentler, 1988). Classic examples are the relationship between cocaine and psychosis, with the former causing the latter (Tang et al., 2014), and the well-established causal effect of traumatic events on problematic drinking (Buckner et al., 2007; Cerda et al., 2011). Psychological distress has also been associated with medical (Schepis and McCabe, 2019) and non-medical (McHugh et al., 2017) use of BZDs.

The theory of self-medication states that individuals seek for psychoactive substances to relieve distress (Smith et al., 2017). Crum and colleagues have described an important association between the report of self-medication of mood disorders with alcohol and the onset of an alcohol dependence (Crum et al., 2013). Similar findings have been reported on the relationship between self-medication with marijuana and the development of marijuana use disorders (Moitra et al., 2015). Despite having diverse mechanisms of action and pharmacodynamics, z-drugs share a depressive effect on the CNS with several recreational depressive substances, with alcohol and marijuana as the most widely used (United Nations Office on Drugs and Crime, 2019). To our knowledge, the association of z-drugs with psychological distress and CNS depressant prescription and recreational drugs have not been investigated up to this point.

To fill this gap, we used survey data from the National Survey on Drug Use and Health (NSDUH) to i) describe the prevalence of medical and non-medical use of z-drugs in the general US population; ii) describe sociodemographic characteristics of medical and non-medical use of z-drugs; iii) investigate associations between medical and non-medical use of z-drugs with psychological distress and distinct patterns of BZD, alcohol, and marijuana use.

Methods

Sample

NSDUH is a yearly representative survey of the non-institutionalized civilian US population designed to gather nationwide information on the prevalence of substance use and related variables. The NSDUH includes individuals aged 12 or older in the 50 states and Washington DC, oversampling younger age groups (Center for Behavioral Health Statistics and Quality, 2016b, 2017, 2018). Data were collected using face-to-face household interviews with computer-assisted interviewing and audio computer-assisted survey instruments (ACASI) to ensure adequate privacy while reporting sensitive information. Weighted interview response rates among adults for the years 2015–2017 ranged between 66.3% and 68.4% (Center for Behavioral Health Statistics and Quality, 2016b, 2017, 2018). Sampling weights accounted for selection probability, non-response, and population distribution. In the present study, we performed a cross-sectional analysis of aggregated data from 2015, 2016, and 2017 public use files, which provided us a sample of 128,740 individuals aged 18 or older.

Measures

Participants were asked if they had used zolpidem (medically and non-medically), eszopiclone (medically and non-medically), and zaleplon (only medically – non-medical use of it was not asked) during the past 12 months. Non-medical use was defined as any z-Drug use in ay other way than directed by the doctor during this time. Medical use referred to those who reported past-year z-drug use and did not report using in any way other than directed. We combined past-year medical and non-medical users into a single z-drug 3-level past-year use variable, coded as follows: 0-no use of any z-Drugs; 1-medical use only; 2-non-medical use. Those who reported both medical and non-medical use in the past year were included in the non-medical category.

Our independent variables were past-year psychological distress and drug-related variables. Psychological distress is defined by a score of 13 or higher in the Kessler Psychological Distress Scale (Kessler et al., 2010) at any time in the past year. The drug-related variables were selected as follows: past-year use of marijuana, alcohol, and tranquilizers; past-year non-medical use of tranquilizers; past-year DSM-IV (American Psychiatric Association, 1994) abuse/dependence (use disorders) of marijuana, alcohol, and tranquilizers. All the variables described are binary and coded as yes/no. Among the demographic variables available in the datasets, we used the following control variables: age (18–25, 26–34, 35–49, 50–64, 65+), gender (male or female), education level (less than high school, high school graduate, some college, college graduate or more), race/ethnicity (Non-Hispanic [NH]white, NH Black, Hispanic, and other), population density (1,000,000 persons or more, less than 1,000,000 persons, and rural) and survey year (2015, 2016, and 2017), chosen by plausibility and submitted to stepwise selection.

Analysis

We described prevalences of the three z-drug categories of past-year medical and non-medical use of z-drugs overall and across the psychological distress variable, the drug-related variables, and the demographic categories described above.

We then ran multinomial logistic regression models with the three-level z-drug as the dependent variable to assess associations with the psychological distress and drug-related variables of interest as described above to obtain adjusted Relative Risk Ratio (aRRR) estimates (StataCorp, 2015).

We chose to run three different multinomial logistic regression models in order to avoid excessive collinearity between variables in one same model. The three-category z-drug variable was the dependent variable in all of the three models. Our reference categories for the z-drug variable were 0-No use of any z-drugs, to compare medical users and non-medical users to non-users, and 1-Past-year medical use of z-drugs, to compare non-medical users to medical users. The first model included the psychological distress variable and any past-year use of marijuana, tranquilizers, and alcohol, besides the demographic covariates. The second model included the psychological distress variable and any past-year non-medical use of tranquilizers, besides the demographic covariates. The third model included the psychological distress variable and past-year DSM-IV use disorders variables for marijuana, tranquilizers, and alcohol. We also ran sensitivity analyses separating our sample by gender to investigate possible disparities between women and men and whether our total estimates were predominantly due to a gender than to the other. Differences between each gender and the total results are discussed.

To take into account the complex sampling design, all analyses were based on Taylor series approximations (StataCorp, 2015) to estimate variances. Analyses were weighted using the provided person-level analysis weights that account for the selection probability at specific stages or adjustment factors. As three NSDUH survey years were used (2015, 2016, and 2017), we divided the person-level analysis weights by the number of years (3) used in the analyses.

Results

Past year prevalences of Z-Drug use were 3.3% for medical use and 0.5% for non-medical use. Past-year medical and non-medical use of z-drugs were higher among respondent who reported psychological distress in the past year. Non-medical use of z-drugs was notably higher among respondents DSM-IV marijuana, tranquilizers, and alcohol use disorders, as well as among those who reported non-medical use of tranquilizers in the past year. Medical use of z-drugs was also more prevalent among medical users, non-medical users, and those with tranquilizer use disorders. Prevalence across demographic groups are described in Table 1.

Table 1.

Past-year medical and non-medical use of z-drugs by characteristics of participants aged 18 or older (n= 128,740), US National Survey on Drug Use and Health (NSDUH) 2015–2017.

Past year Z-Drugs use statusa
No Use
 Medical Use
 Non-Medical Use
 Total
n = 123,832 (96.2%) n = 4,301 (3.3%) n = 607 (0.5%) n = 128,740 (100%)
Past-year Psychological Distress 10.3 20 32.6 10.8
Past-year Marijuana use 14.1 17.4 44.2 14.4
Past-year Marijuana Use Disorder 1.4 1.3 10.5 1.4
Past-year BZD use 13.8 48.7 60.3 15.5
Past-year non-medical BZD use 2.0 5.8 39.4 2.3
Past-year BZD Use Disorder 0.2 0.8 8.6 0.3
Past-year Alcohol use 69.6 75.4 85.1 69.9
Past-year Alcohol Use Disorder 5.9 7.7 23.4 6
Demographics
Gender
 Male 48.6 40.2 42.6 48.2
 Female 51.4 59.8 57.4 51.8
Age (years old)
 18–25 14.5 4.8 18.9 14.1
 26–34 16.1 9.9 24.1 15.9
 35–49 24.8 23.9 23.7 24.8
 50–64 25.0 35.9 23.9 25.5
 65 or older 19.5 25.5 9.5 19.7
Race/Ethnicity
 NHb White 63.6 78.3 80.6 64.3
 NH Black 12.0 8 5.2 11.8
 Hispanic 16.1 8.9 10.6 15.8
 Other 8.2 4.9 3.6 8.1
Total Family Income
 Less than $20.000 17.1 15.2 14.7 17
 $20.000–$49.999 30 25.3 24.9 29.8
 $50.000–$74.999 16.1 16.8 17.8 16.2
 $75.000 or more 36.8 42.8 42.6 37
Education
 Less than high school 13.3 8.3 8 13.1
 High School Graduate 25.2 22.5 18 25
 Some College/ Associate Degree 30.8 31.4 34.3 30.8
 College Graduate 30.7 37.7 39.7 31.1
Population Density
 Segment in a CBSAc with a million or more persons 54.1 51.7 57.1 54
 Segment in a CBSA with fewer than a million persons 40.1 42.5 40 40.2
 Segment not in a CBSA 5.8 5.8 2.9 5.8
Survey Year
 2015 32.9 35.8 34.4 33.1
 2016 33.2 34.7 35.8 33.3
 2017 33.8 29.5 29.8 33.7
Open in a new tab
a:

Past-year medical and non-medical use, coded as follows: 0 - no use of any z-drugs; 1 - medical use only; 2 - non-medical use. Those who reported medical and non-medical use were included on the non-medical category. Race/ethnicity category “Other” included NH native Americans/Alaska natives, NH native Hawaiians/other pacific islanders, NH Asians and more than one race

b:

non-hispanic

c:

core-based statistical area.

In the multinomial regression model with past-year use variables, medical (aRRR=1.75, CI=[1.56,1.97]) and non-medical use (aRRR=1.99, CI=[1.63,2.43]) of z-drugs were associated with past-year psychological distress. Compared to non-z-drug users, medical users of z-drugs reported higher risk of past-year use marijuana (aRRR=1.29, CI=[1.16,1.44]), tranquilizers (aRRR=4.70, CI=[4.24,5.20]), and alcohol (aRRR=1.27, CI=[1.13,1.42]) use, whereas non-medical users of z-drugs reported higher risk of past-year marijuana (aRRR=2.78, CI=[2.16,3.56]) and tranquilizers (aRRR=6.56, CI=[5.13,8.38]) use, but not alcohol use. When comparing non-medical users of z-drugs to medical users, the association with psychological distress was not significantly different. Compared to medical users, non-medical users had higher risk of past-year marijuana (aRRR=2.15, CI=[1.65,2.81]) and tranquilizers (aRRR=1.39, CI=[1.07,1.83]) use, but not alcohol use.

In the model with the past-year non-medical use of tranquilizers variable, medical users (aRRR=2.59, CI=[2.31,2.90]) and non-medical users (aRRR=2.29, CI=[1.84,2.84]) of z-drugs had higher risk of past-year psychological distress compared to those with no past-year use of z-drugs. Past-year non-medical use of tranquilizers was higher among past-year medical users (aRRR=3.20, CI=[2.60,3.93]) and non-medical users (aRRR=24.29, CI=[18.32,32.20]) of z-drugs. When using medical users as the reference category, non-medical users did not significantly differ regarding past-year psychological distress. Non-medical users of z-drugs had a higher risk of non-medical use of tranquilizers (aRRR=7.60, CI=[5.31,10.87]) when compared to medical users of z-drugs.

In the model with past-year use disorders variables, medical users (aRRR=2.65, CI=[2.37,2.95]) and non-medical users (aRRR=2.59, CI=[2.12,3.15]) of z-drugs had higher risk of past-year psychological distress when compared to those with no past-year use of z-drugs. Medical users had higher risk of past-year tranquilizers (aRRR=3.25, CI=[1.98,5.35]) and alcohol (aRRR=1.28, CI=[1.18,1.61]) use disorders, but not marijuana. Non-medical users of z-drugs were at higher risk of past-year marijuana (aRRR=3.28, CI=[2.14,5.03]), tranquilizers (aRRR=16.09, CI=[10.27,25.22]), and alcohol (aRRR=2.66, CI=[1.99,3.55]) use disorders. When comparing past-year non-medical to medical users of z-drugs, the risk of reporting past-year psychological distress did not differ significantly across the two groups. Non-medical users were at higher risk of past-year marijuana (aRRR=2.82, CI=[1.64,4.82]), tranquilizers (aRRR=4.95, CI=[2.79,8.77]), and alcohol (aRRR=1.92, CI=[1.40,2.66]) use disorders compared to medical users of z-drugs. Information in detail is provided on Table 2.

Table 2.

Past-year medical and non-medical z-drugs users compared to non-users and medical users compared to non-medical users among participants aged 18 or older (n = 128,740) across psychological distress, BZDs, alcohol, and cannabis variables, US National Survey on Drug Use and Health (NSDUH) 2015–2017.

Past-Year Z-Drugs use (n =128,740)
Medical Usersa (n=7,562) versus non-usersa (n = 123,832) Non-Medical Users (n=2,175) versus non-users (n = 123,832) Non-Medical Users (n=2,175) versus Medical Users (n=7,562)
aRRRb 95% CIc aRRR 95% CIc aRRR 95% CIc
Model with past-year use of depressant drugs
Past-Year psychological Distress
1.75** (1.56, 1.97) 1.99** (1.63, 2.43) 1.14 (0.92, 1.41)
Past-year use of Cannabis
1.29** (1.16, 1.44) 2.78** (2.16, 3.56) 2.15** (1.65, 2.81)
Past-year use of tranquilizers
4.70** (4.24, 5.20) 6.56** (5.13, 8.38) 1.39* (1.07, 1.83)
Past-year use of alcohol
1.27** (1.13, 1.42) 1.37 (0.94, 1.98) 1.07 (0.74, 1.57)
Gender
Male 0.87** (0.79, 0.96) 0.92 (0.75, 1.13) 1.06 (0.84, 1.33)
Education (ref=Less than high school)
High School Graduate 1.28** (1.07, 1.55) 1.03 (0.68, 1.57) 0.80 (0.51, 1.25)
Some College 1.40** (1.18, 1.66) 1.27 (0.79, 2.06) 0.91 (0.56, 1.49)
College Graduate 1.68** (1.42, 1.97) 1.77* (1.07, 2.92) 1.05 (0.63, 1.77)
Race (ref=NH White)
NHc Black 0.80** (0.68, 0.93) 0.47** (0.28, 0.78) 0.59 (0.34, 1.02)
Hispanic 0.72** (0.61, 0.85) 0.70 (0.48, 1.01) 0.96 (0.67, 1.38)
Other 0.70** (0.56, 0.89) 0.47** (0.29, 0.76) 0.66 (0.37, 1.18)
Age (ref=18–25 yod)
26–34 yo 1.80** (1.57. 2.06) 1.23 (0.99, 1.52) 0.68** (0.54, 0.87)
35–49 yo 2.92** (2.56, 3.32) 0.92 (0.72, 1.18) 0.32** (0.24, 0.42)
50–64 yo 4.42** (3.80, 5.14) 1.02 (0.71, 1.48) 0.23** (0.16, 0.34)
65+ yo 4.37** (3.84, 4.96) 0.64 (0.37, 1.10) 0.15** (0.09, 0.25)
Population Density (ref=Large Metropolitan Area)
Small Metropolitan Area 1.02 (0.93, 1.12) 0.91 (0.69, 1.20) 0.89 (0.68, 1.18)
Rural Area 0.95 (0.80, 1.14) 0.52* (0.31, 0.88) 0.55* (0.31, 0.98)
Survey Year (ref=2015)
2016 0.95 (0.86, 1.05) 1.03 (0.77, 1.39) 1.09 (0.80, 1.47)
2017 0.78** (0.70, 0.88) 0.82 (0.60, 1.13) 1.05 (0.74, 1.48)
Model with past-year non-medical use of tranquilizers
Past-Year psychological Distress
2.59** (2.31, 2.90) 2.29** (1.84, 2.84) 0.88 (0.69, 1.13)
Past-year non-medical use of tranquilizers
3.20** (2.60, 3.93) 24.29** (18.32, 32.20) 7.60** (5.31, 10.87)
Gender
Male 0.76** (0.68, 0.84) 0.81* (0.66, 1.00) 1.07 (0.85, 1.34)
Education (ref=Less than high school)
High School Graduate 1.34** (1.12, 1.61) 1.08 (0.71, 1.65) 0.81 (0.52, 1.26)
Some College 1.54** (1.29, 1.84) 1.39 (0.86, 2.22) 0.90 (0.55, 1.48)
College Graduate 1.78** (1.29, 1.84) 1.94* (1.20, 3.15) 1.09 (0.66, 1.80)
Race (ref=NH White)
NHc Black 0.66** (0.56, 0.78) 0.45** (0.27, 0.74) 0.68 (0.39, 1.18)
Hispanic 0.62** (0.52, 0.73) 0.64* (0.44, 0.93) 1.04 (0.71, 1.51)
Other 0.54** (0.43, 0.68) 0.39** (0.24, 0.63) 0.71 (0.40, 1.27)
Age(ref=18–25 yod)
26–34 yo 1.97** (1.72, 2.25) 1.40** (1.13, 1.73) 0.71** (0.56, 0.90)
35–49 yo 3.35** (2.93, 3.83) 1.17 (0.93, 1.48) 0.35** (0.27, 0.46)
50–64 yo 5.16** (4.44, 6.00) 1.36 (0.92, 1.99) 0.26** (0.18, 0.37)
65+ yo 4.97** (4.40, 5.61) 0.82 (0.49, 1.38) 0.17** (0.10, 0.28)
Population Density (ref=Large Metropolitan Area)
Small Metropolitan Area 1.02 (0.93, 1.11) 0.90 (0.68, 1.18) 0.88 (0.67, 1.16)
Rural Area 0.92 (0.78, 1.08) 0.51* (0.30, 0.87) 0.56 (0.31, 1.01)
Survey Year (ref=2015)
2016 0.95 (0.86, 1.03) 1.01 (0.75, 1.35) 1.07 (0.79, 1.44)
2017 0.78** (0.70, 0.87) 0.82 (0.59, 1.13) 1.05 (0.74, 1.48)
Model with past-year use disorders of depressant drugs
Past-Year psychological Distress
2.65** (2.37, 2.95) 2.59** (2.12, 3.15) 0.97 (0.79, 1.21)
Past-year Cannabis use disorders
1.17 (0.87, 1.56) 3.28** (2.14, 5.03) 2.82** (1.64, 4.82)
Past-year tranquilizers use disorders
3.25** (1.98, 5.35) 16.09** (10.27, 25.22) 4.95** (2.79, 8.77)
Past-year alcohol use disorders
1.38** (1.18, 1.61) 2.66** (1.99, 3.55) 1.92** (1.40, 2.66)
Gender
Male 0.75** (0.68, 0.83) 0.75** (0.61, 0.92) 1.00 (0.81, 1.24)
Education (ref=Less than high school)
High School Graduate 1.34** (1.12, 1.61) 1.09 (0.71, 1.67) 0.82 (0.52, 1.29)
Some College 1.55** (1.30, 1.84) 1.44 (0.90, 2.31) 0.93 (0.57, 1.53)
College Graduate 1.78** (1.50, 2.10) 1.92* (1.17, 3.14) 1.08 (0.65, 1.80)
Race (ref=NH White)
NHc Black 0.64** (0.55, 0.76) 0.35** (0.22, 0.58) 0.55* (0.32, 0.94)
Hispanic 0.61** (0.52, 0.72) 0.57** (0.39, 0.82) 0.93 (0.64, 1.34)
Other 0.54** (0.43, 0.67) 0.32** (0.19, 0.53) 0.59 (0.32, 1.08)
Age(ref=18–25 yod)
26–34 yo 1.93** (1.68, 2.22) 1.40** (1.13, 1.75) 0.73** (0.58, 0.92)
35–49 yo 3.24** (2.83, 3.69) 1.05 (0.82, 1.34) 0.32** (0.25, 0.43)
50–64 yo 4.96** (4.27, 5.77) 1.15 (0.77, 1.69) 0.23** (0.16, 0.33)
65+ yo 4.77** (4.19, 5.44) 0.67 (0.39, 1.14) 0.14** (0.08, 0.24)
Population Density (ref=Large Metropolitan Area)
Small Metropolitan Area 1.02 (0.93, 1.11) 0.90 (0.68, 1.19) 0.89 (0.67, 1.17)
Rural Area 0.91 (0.78, 1.07) 0.48** (0.28, 0.83) 0.53* (0.29, 0.97)
Survey Year (ref=2015)
2016 0.95 (0.87, 1.03) 1.02 (0.76, 1.37) 1.08 (0.80, 1.46)
2017 0.78** (0.70, 0.87) 0.81 (0.58, 1.12) 1.04 (0.73, 1.47)
Open in a new tab
a:

Past-year medical and non-medical use, coded as follows: 0 - no use of any z-drugs; 1 - medical use only; 2 - non-medical use. Those who reported medical and non-medical use were included on the non-medical category. Model controlled for covariates gender, age, race/ethnicity, education, population density and survey year

b:

adjusted Relative Risk Ratio

c:

95% confidence interval.

*

p < 0.05

**

p < 0.01

Demographics

In all three models, men were at lower risk of medical use of z-drugs when compared to those with no past-year use. Also, medical use was generally associated with higher education, white race/ethnicity, and older ages when compared to no past-year use. Non-medical use was also related to white ethnicity in all models as compared to no use. Those in rural areas were at lower risk of non-medical use compared to those with no use. Finally, non-medical use was importantly associated with lower age, with lower risk from lower age groups to higher age groups when compared to medical use. Those at rural areas had lower risk of non-medical use also when compared to medical users. See Table 2.

Sensitivity analysis – the role of gender

We ran separate analyses restricted to men and women to assess whether these groups would reveal disparities between each other and with the total sample regarding the outcomes of interest.

Men

With regards to the model with past-year use variables among men, past-year medical use of z-drugs was not associated with past-year use of marijuana (aRRR=1.19, CI=[0.97,1.45]) and alcohol (aRRR=1.17, CI=[0.93,1.46]) when compared to no use of z-drugs, unlike findings in the total sample, where such associations were statistically significant.

All other outcome variables in all the models did not differ from the total sample in statistical significance.

Women

In the model with past-year use variables among women, past-year non-medical use of z-drugs was not associated with past-year use of BZDs, as compared to those who used z-drugs medically (aRRR = 1.18, CI=[0.82, 1.69]), different from findings in the total sample.

In the model with dependence variables among women, past-year medical use of z-drugs was not associated with alcohol use disorder (aRRR=1.12, CI=[0.91,1.37]) when compared to no past-year use of z-drugs, differing from the total sample where this association was statistically significant. Also, past-year non-medical use of z-drugs was not associated with alcohol use disorder when compared to medical use of z-drugs (aRRR=1.97, CI=[1.29, 3.01]), as opposite to the model with men and women.

All other outcome variables in all the models did not differ from the total sample in statistical significance.

Discussion

In this cross-sectional analysis of NSDUH data from 2015–2017, we found that 3.8% of the American population 18+ used a z-drug in the past year, which accounts for about 12 million Americans. Within this population, our main findings were that: i) both medical and non-medical use of z-drugs were associated with psychological distress in the past year; nonetheless, non-medical users did not differ from medical users with regards to past-year psychological distress. ii) Past-year medical and non-medical users of z-drugs were at higher risk than non-users of past-year use of tranquilizers, non-medical use of tranquilizers, and tranquilizers use disorders. Also, non-medical users of z-drugs had a significant and markedly higher risk of past-year use, non-medical use and use disorders of tranquilizers, a NSDUH variable that includes most BZDs, compared to medical users of z-drugs. iii) Medical and non-medical users had a significantly higher risk of nearly all the drug-related variables tested, including past-year use and use disorders of marijuana, tranquilizers, and alcohol, and non-medical use of tranquilizers. Non-medical users were at higher risk than medical users of any past-year use disorders of all of the depressant drugs assessed in the models. Besides, non-medical z-drug users were also at higher risk of non-medical use of tranquilizers and past-year use of tranquilizers and marijuana, when compared to medical z-drug users.

We found that both past-year medical and non-medical use of z-drugs were associated with past-year psychological distress compared to those with no past-year use of z-drugs, but no significant difference was found comparing medical to non-medical users. The Psychological distress variable in the NSDUH is derived from the K6 score (Kessler et al., 2010), a screening scale for severe mental illness. Z-drugs are only approved by the FDA for the treatment of insomnia, a condition that is comorbid in as much as 40.5% of patients with major depressive disorder and 29.9% of patients generalized anxiety disorder (Stewart et al., 2006), two highly prevalent mental disorders. The K6 scale does not directly address sleep complaints but has proven to be a useful screening tool for severe mental disorders that are knowingly related to insomnia as those described above. As such, the elevated prevalence of medical use of z-drugs in the US indicates that insomnia is a common complaint among patients who seek for medical healthcare. The association between the medical use of z-drugs and psychological distress (assessed by the K6 scale) indicates that many individuals who are being prescribed z-drugs struggle with a major mental disorder.

Interestingly, our findings suggest that distress is evenly present among z-drug past-year users regardless of medical use status. Sociodemographic differences between the medical and non-medical users may partially explain this finding: we have found that non-medical users are younger and more likely to live at a large urban center, which is in line with findings from another study (Schepis and McCabe, 2019). Such factors suggest that the relationship between z-drug use and psychological distress may be complex and multifactorial, as we cannot merely picture non-medical use as a more severe presentation of the same factors that lead to medical use as we are dealing with two different populations. Also noteworthy is the fact that white people make more medical and non-medical use of z-drugs compared to all other ethnicities. This may reflect racial/ethnic disparities in access to healthcare, an issue that has been extensively discussed on the literature (Richardson and Norris, 2010) that has been identified by previous studies in BZDs and prescription opioids (Tardelli et al., 2019). Although the regression models show that men are significantly less likely than women to report medical use of z-drugs, no gender differences were found regarding non-medical use. There were only slight differences between the overall and the gender-specific models. Those were most probably attributable to lack of power, with minimal clinical relevance.

Past-year medical and non-medical users of z-drugs were at dramatically higher risk of past-year BZD use in all patterns, compared to those with no past-year z-drug use. Moreover, non-medical use of z-drugs was also associated with higher BZD use in all patterns when compared to those with medical use. This indicates that the association becomes more consistent for more severe and harmful patterns of use of both z-drugs and BZDs. According to a recent study, 12.6% of the American population used a BZD in the past year (Tardelli et al., 2019). Long-term prescriptions of both BZDs and z-drugs are rising simultaneously (Kaufmann et al., 2018), especially among the elderly (Olfson et al., 2015). Because z-drugs are seen as safer hypnotics than BZDs, attempting to replace BZDs with z-drugs is still a common practice by clinicians. Since both classes are hard to taper off (Belanger et al., 2009), this likely explains simultaneous raises in their long-term and non-medical use (Kaufmann et al., 2018). Kaufmann and colleagues reported a rise in the prescriptions of both z-drugs and BZDs from 1993 to 2010 (Kaufmann et al., 2016a). Our findings suggest that a large number of individuals are co-prescribed BZDs and z-drugs. It is known that a substantial proportion of non-medical users of BZDs started with medical use (McCabe et al., 2014) and that regular prescribers are the most common source of BZD diversion (Schmitz, 2016). Concurrent use of BZDs and z-drugs is highly related to undesirable outcomes such as hospitalization and death (Kaufmann et al., 2017) and was also found to be related to the risk of suicide (Sung et al., 2019). Hence, the development of pharmacologic and behavioral strategies to deal with insomnia and z-drugs/BZDs dependence should be prioritized.

Past-year medical users of z-drugs were at higher risk of marijuana and alcohol use and alcohol use disorders compared to those with no past-year use of z-drugs. Past-year non-medical users of z-drugs were at higher risk of marijuana use and marijuana and alcohol use disorders, when compared to those with no past-year use of z-drugs, and of marijuana use and marijuana and alcohol use disorders, when compared to past-year medical users. The prevalence of alcohol use is high among users of sedative-hypnotic medications and vice-versa (Ilomaki et al., 2013). Alcohol has also been reported as a highly frequent aid to achieve sleep, outnumbering hypnotic medications in a Japanese survey (Kaneita et al., 2007). Zolpidem has been regarded as a possible approach to sleep disturbances during marijuana withdrawal (Vandrey et al., 2011), and interestingly medical marijuana is perceived as a potentially adequate substitute for benzodiazepine hypnotics (Lucas and Walsh, 2017). BZDs and z-drugs are used medically for the treatment of insomnia. BZDs, but not z-drugs (Fava et al., 2009), are also used for the treatment of anxiety (Davidson, 2004; Lydiard et al., 2010). Alcohol (Turner et al., 2018) and marijuana (Sarvet et al., 2018) are commonly used as self-medication for anxiety and mood disorders. The association between medical and non-medical use of z-drugs with psychological distress (defined as the presence of unspecific impairing mental health symptoms) found by this study combined with consistent literature associating alcohol and marijuana with psychological distress (Hill and Angel, 2005; Mathews et al., 2011; Moitra et al., 2015) support that distress might be the common cause underlying the association between medical and non-medical use of z-drugs and marijuana and alcohol outcomes. Noteworthy is the finding that non-medical users of z-drugs were at much higher risk of marijuana and alcohol dependence compared to medical users of z-drugs, despite not being at significantly higher risk of distress. Other shared risk factors apart from distress and mental disorders should therefore play a significant role in the onset of harmful use of recreational and prescription drugs, such as genetic susceptibility, ethnicity, and socioeconomic status. Prescribers must be as cautious when initiating z-drugs as when starting BZDs, assessing thoroughly present and past mental health conditions, and use/dependence of recreational drugs.

The study has several strengths. The complex survey design allows the extrapolation of the findings for the entire United States population and assures their external validity. Also, we ran different multinomial logistic regression models to avoid multicollinearity and ensure models’ fit. This is the first paper to our knowledge that assesses z-drugs separate from BZDs and investigates their relationship with each other, with distress, and recreational drugs using a nationwide database in the United States.

Limitations are noted. The NSDUH tranquilizer variables comprise most of the BZDs available in the market. However, BZDs with shorter half-lives Temazepam and Triazolam are categorized into the “sedatives” category together with z-drugs and Phenobarbital and were not included in this analysis. Also, the codebook does not provide a variable for non-medical use of zaleplon as it did for zolpidem and eszopiclone (Center for Behavioral Health Statistics and Quality, 2016a) and also does not contain variables for any z-drug use disorders. As such, non-medical use and use disorders of z-drugs were encompassed by one single variable, different BZDs. Moreover, despite discussing possible causal relations between the variables investigated using criteria such as strength, temporality, and plausibility (Lucas and McMichael, 2005), this is a cross-sectional analysis that only report measures of association, and causality cannot be determined. Cross-sectional analyses are unable to establish causality due to the temporality of the measures, that were collected in the same time-frame. We are only able in this paper to report associations that can raise hypotheses for future studies with adequate longitudinal design to investigate causal relationships.

Conclusion

Future policies should guide clinicians to be especially cautious when prescribing z-drugs, particularly among those with concurrent use of BZDs, alcohol, and marijuana due to potential harm of concurrent use with z-drugs. Z-drugs have considerable abuse liability, and that should be considered when initiating prescriptions. Z-drugs should not replace BZDs for the treatment of insomnia, as both classes are challenging to taper off, and this decision may lead to harmful concurrent use. Clinicians should actively ask about the non-medical use of z-drugs and assess harmful use of BZDs, alcohol, and marijuana in all z-drug users. Moreover, other types of treatment for insomnia, such as behavioral therapy, should be encouraged.

  • 3.8% of the Americans (12 million) used a z-drug in the past year.

  • Z-drug medical and non-medical use are associated with psychological distress.

  • Z-drug use is highly correlated with BZD use, misuse, and use disorders.

  • Z-drug use is related to use and dependence of marijuana and alcohol.

  • Distress may confound the relation between z-drugs and BZD, alcohol, and marijuana.

Acknowledgments

The authors thank the National Survey on Drug Use and Health (NSDUH) respondents and Substance Abuse Mental Health Service Administration (SAMHSA) for the publicly available files. The results herein do not reflect the position of SAMHSA. This study was partially funded by NIDA R01 DA037866 (Martins, PI).

Role of Funding Source

This study was partially funded by R01DA037866 (Martins, PI).

Footnotes

Conflicts of interest

No conflict declared.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. American Psychiatric Association, 1994. Diagnostic and statistical manual of mental disorders (4th edition), 4th ed. Author. [Google Scholar]
  2. Bajaj V, Kalra I, Bajaj A, Sharma D, Kumar R, 2019. A case of zolpidem dependence with extremely high daily doses. Asia Pac Psychiatry 11(4), e12356. [DOI] [PubMed] [Google Scholar]
  3. Belanger L, Belleville G, Morin C, 2009. Management of Hypnotic Discontinuation in Chronic Insomnia. Sleep Med Clin 4(4), 583–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brandt J, Leong C, 2017. Benzodiazepines and Z-Drugs: An Updated Review of Major Adverse Outcomes Reported on in Epidemiologic Research. Drugs R D 17(4), 493–507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buckner JD, Keough ME, Schmidt NB, 2007. Problematic alcohol and cannabis use among young adults: the roles of depression and discomfort and distress tolerance. Addict Behav 32(9), 1957–1963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Center for Behavioral Health Statistics and Quality, 2016a. 2015 National Survey on Drug Use and Health Public Use Codebook. Substance Abuse and Mental Health Services Administration, Rockville, MD. [Google Scholar]
  7. Center for Behavioral Health Statistics and Quality, 2016b. 2015 National Survey on Drug Use and Health: Methodological summary and definitions; Rockville, MD. [Google Scholar]
  8. Center for Behavioral Health Statistics and Quality, 2017. 2016 National Survey on Drug Use and Health: Methodological summary and definitions. Substance Abuse and Mental Health Services Administration, Rockville, MD. [Google Scholar]
  9. Center for Behavioral Health Statistics and Quality, 2018. 2017 National Survey on Drug Use and Health: Methodological summary and definitions. Substance Abuse and Mental Health Services Administration, Rockville, MD. [Google Scholar]
  10. Cerda M, Tracy M, Galea S, 2011. A prospective population based study of changes in alcohol use and binge drinking after a mass traumatic event. Drug Alcohol Depend 115(1–2), 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chattopadhyay AC, Shukla L, Kandasamy A, Benegal V, 2016. High-dose zolpidem dependence - Psychostimulant effects? A case report and literature review. Ind Psychiatry J 25(2), 222–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chiaro G, Castelnovo A, Bianco G, Maffei P, Manconi M, 2018. Severe Chronic Abuse of Zolpidem in Refractory Insomnia. J Clin Sleep Med 14(7), 1257–1259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Crum RM, Mojtabai R, Lazareck S, Bolton JM, Robinson J, Sareen J, Green KM, Stuart EA, La Flair L, Alvanzo AA, Storr CL, 2013. A prospective assessment of reports of drinking to self-medicate mood symptoms with the incidence and persistence of alcohol dependence. JAMA Psychiatry 70(7), 718–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Davidson JR, 2004. Use of benzodiazepines in social anxiety disorder, generalized anxiety disorder, and posttraumatic stress disorder. J Clin Psychiatry 65 Suppl 5, 29–33. [PubMed] [Google Scholar]
  15. Fava M, Asnis GM, Shrivastava R, Lydiard B, Bastani B, Sheehan D, Roth T, 2009. Zolpidem extended-release improves sleep and next-day symptoms in comorbid insomnia and generalized anxiety disorder. J Clin Psychopharmacol 29(3), 222–230. [DOI] [PubMed] [Google Scholar]
  16. Food and Drug Administration, 2013. Risk of next-morning impairment after use of insomnia drugs; FDA requires lower recommended doses for certain drugs containing zolpidem (Ambien, Ambien CR, Edluar, and Zolpimist). Silver Spring, MD: Drug Safety Communications. [Google Scholar]
  17. Green KM, Zebrak KA, Robertson JA, Fothergill KE, Ensminger ME, 2012. Interrelationship of substance use and psychological distress over the life course among a cohort of urban African Americans. Drug Alcohol Depend 123(1–3), 239–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gyawali B, Choulagai BP, Paneru DP, Ahmad M, Leppin A, Kallestrup P, 2016. Prevalence and correlates of psychological distress symptoms among patients with substance use disorders in drug rehabilitation centers in urban Nepal: a cross-sectional study. BMC Psychiatry 16(1), 314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hill TD, Angel RJ, 2005. Neighborhood disorder, psychological distress, and heavy drinking. Soc Sci Med 61(5), 965–975. [DOI] [PubMed] [Google Scholar]
  20. Ilomaki J, Paljarvi T, Korhonen MJ, Enlund H, Alderman CP, Kauhanen J, Bell JS, 2013. Prevalence of concomitant use of alcohol and sedative-hypnotic drugs in middle and older aged persons: a systematic review. Ann Pharmacother 47(2), 257–268. [DOI] [PubMed] [Google Scholar]
  21. Kaneita Y, Uchiyama M, Takemura S, Yokoyama E, Miyake T, Harano S, Asai T, Tsutsui T, Kaneko A, Nakamura H, Ohida T, 2007. Use of alcohol and hypnotic medication as aids to sleep among the Japanese general population. Sleep Med 8(7–8), 723–732. [DOI] [PubMed] [Google Scholar]
  22. Kapil V, Green JL, Le Lait C, Wood DM, Dargan PI, 2014. Misuse of benzodiazepines and Z-drugs in the UK. Br J Psychiatry 205(5), 407–408. [DOI] [PubMed] [Google Scholar]
  23. Kaufmann CN, Spira AP, Alexander GC, Rutkow L, Mojtabai R, 2016a. Trends in prescribing of sedative-hypnotic medications in the USA: 1993–2010. Pharmacoepidemiol Drug Saf 25(6), 637–645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kaufmann CN, Spira AP, Alexander GC, Rutkow L, Mojtabai R, 2017. Emergency department visits involving benzodiazepines and non-benzodiazepine receptor agonists. Am J Emerg Med 35(10), 1414–1419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kaufmann CN, Spira AP, Depp CA, Mojtabai R, 2016b. Continuing Versus New Prescriptions for Sedative-Hypnotic Medications: United States, 2005–2012. Am J Public Health 106(11), 2019–2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kaufmann CN, Spira AP, Depp CA, Mojtabai R, 2018. Long-Term Use of Benzodiazepines and Nonbenzodiazepine Hypnotics, 1999–2014. Psychiatr Serv 69(2), 235–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kesselheim AS, Donneyong M, Dal Pan GJ, Zhou EH, Avorn J, Schneeweiss S, Seeger JD, 2017. Changes in prescribing and healthcare resource utilization after FDA Drug Safety Communications involving zolpidem-containing medications. Pharmacoepidemiol Drug Saf 26(6), 712–721. [DOI] [PubMed] [Google Scholar]
  28. Kessler RC, Green JG, Gruber MJ, Sampson NA, Bromet E, Cuitan M, Furukawa TA, Gureje O, Hinkov H, Hu CY, Lara C, Lee S, Mneimneh Z, Myer L, Oakley-Browne M, Posada-Villa J, Sagar R, Viana MC, Zaslavsky AM, 2010. Screening for serious mental illness in the general population with the K6 screening scale: results from the WHO World Mental Health (WMH) survey initiative. Int J Methods Psychiatr Res 19 Suppl 1, 4–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Khantzian EJ, 1985. The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. Am J Psychiatry 142(11), 1259–1264. [DOI] [PubMed] [Google Scholar]
  30. Lucas P, Walsh Z, 2017. Medical cannabis access, use, and substitution for prescription opioids and other substances: A survey of authorized medical cannabis patients. Int J Drug Policy 42, 30–35. [DOI] [PubMed] [Google Scholar]
  31. Lucas RM, McMichael AJ, 2005. Association or causation: evaluating links between “environment and disease”. Bull World Health Organ 83(10), 792–795. [PMC free article] [PubMed] [Google Scholar]
  32. Lydiard RB, Rickels K, Herman B, Feltner DE, 2010. Comparative efficacy of pregabalin and benzodiazepines in treating the psychic and somatic symptoms of generalized anxiety disorder. Int J Neuropsychopharmacol 13(2), 229–241. [DOI] [PubMed] [Google Scholar]
  33. Mathews RR, Hall WD, Gartner CE, 2011. Depression and psychological distress in tobacco smokers and people with cannabis dependence in the National Survey of Mental Health and Wellbeing. Med J Aust 195(3), S12–15. [DOI] [PubMed] [Google Scholar]
  34. McCabe SE, West BT, Teter CJ, Boyd CJ, 2014. Trends in medical use, diversion, and nonmedical use of prescription medications among college students from 2003 to 2013: Connecting the dots. Addict Behav 39(7), 1176–1182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McHugh RK, Votaw VR, Bogunovic O, Karakula SL, Griffin ML, Weiss RD, 2017. Anxiety sensitivity and nonmedical benzodiazepine use among adults with opioid use disorder. Addict Behav 65, 283–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Moitra E, Christopher PP, Anderson BJ, Stein MD, 2015. Coping-motivated marijuana use correlates with DSM-5 cannabis use disorder and psychological distress among emerging adults. Psychol Addict Behav 29(3), 627–632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Newcomb MD, Bentler PM, 1988. Consequences of adolescent drug use: Impact on the lives of young adults. Sage Publications, Inc. [Google Scholar]
  38. Newcomb MD, Vargas‐Carmona J, Galaif ER, 1999. Drug problems and psychological distress among a community sample of adults: Predictors, consequences, or confound? Journal of Community Psychology 27(4), 405–429. [Google Scholar]
  39. Norman JL, Fixen DR, Saseen JJ, Saba LM, Linnebur SA, 2017. Zolpidem prescribing practices before and after Food and Drug Administration required product labeling changes. SAGE Open Med 5, 2050312117707687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Olfson M, King M, Schoenbaum M, 2015. Benzodiazepine use in the United States. JAMA Psychiatry 72(2), 136–142. [DOI] [PubMed] [Google Scholar]
  41. Richardson LD, Norris M, 2010. Access to health and health care: how race and ethnicity matter. Mt Sinai J Med 77(2), 166–177. [DOI] [PubMed] [Google Scholar]
  42. Sanger DJ, 2004. The pharmacology and mechanisms of action of new generation, non-benzodiazepine hypnotic agents. CNS Drugs 18 Suppl 1, 9–15; discussion 41, 43–15. [DOI] [PubMed] [Google Scholar]
  43. Sarvet AL, Wall MM, Keyes KM, Olfson M, Cerda M, Hasin DS, 2018. Self-medication of mood and anxiety disorders with marijuana: Higher in states with medical marijuana laws. Drug Alcohol Depend 186, 10–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Scharf G, 1994. From electrostatics to optics : a concise electrodynamics course. Springer-Verlag, Berlin; New York. [Google Scholar]
  45. Schepis TS, McCabe SE, 2019. Prescription Tranquilizer/Sedative Sources for Misuse in Older Adults. Subst Use Misuse 54(11), 1908–1912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schmitz A, 2016. Benzodiazepine use, misuse, and abuse: A review. Ment Health Clin 6(3), 120–126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sinha R, 2008. Chronic stress, drug use, and vulnerability to addiction. Ann N Y Acad Sci 1141, 105–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Smith LL, Yan F, Charles M, Mohiuddin K, Tyus D, Adekeye O, Holden KB, 2017. Exploring the Link Between Substance Use and Mental Health Status: What Can We Learn from the Self-medication Theory? J Health Care Poor Underserved 28(2S), 113–131. [DOI] [PubMed] [Google Scholar]
  49. StataCorp, 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP. [Google Scholar]
  50. Stewart R, Besset A, Bebbington P, Brugha T, Lindesay J, Jenkins R, Singleton N, Meltzer H, 2006. Insomnia comorbidity and impact and hypnotic use by age group in a national survey population aged 16 to 74 years. Sleep 29(11), 1391–1397. [DOI] [PubMed] [Google Scholar]
  51. Sung HG, Li J, Nam JH, Won DY, Choi B, Shin JY, 2019. Concurrent use of benzodiazepines, antidepressants, and opioid analgesics with zolpidem and risk for suicide: a case-control and case-crossover study. Soc Psychiatry Psychiatr Epidemiol 54(12), 1535–1544. [DOI] [PubMed] [Google Scholar]
  52. Tang Y, Martin NL, Cotes RO, 2014. Cocaine-induced psychotic disorders: presentation, mechanism, and management. J Dual Diagn 10(2), 98–105. [DOI] [PubMed] [Google Scholar]
  53. Tardelli VS, Fidalgo TM, Santaella J, Martins SS, 2019. Medical use, non-medical use and use disorders of benzodiazepines and prescription opioids in adults: Differences by insurance status. Drug Alcohol Depend 204, 107573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Turner S, Mota N, Bolton J, Sareen J, 2018. Self-medication with alcohol or drugs for mood and anxiety disorders: A narrative review of the epidemiological literature. Depress Anxiety 35(9), 851–860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. United Nations Office on Drugs and Crime, 2019. World Drug Report 2019.
  56. Vandrey R, Smith MT, McCann UD, Budney AJ, Curran EM, 2011. Sleep disturbance and the effects of extended-release zolpidem during cannabis withdrawal. Drug Alcohol Depend 117(1), 38–44. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES