Illicit Drug Use and Smell and Taste Dysfunction: A National Health and Nutrition Examination Survey 2013–2014

Taste and smell dysfunction are suspected to be associated with substance use. However, representative epidemiological studies remain insufficient. This cross-sectional study explored the relationship between drug use (including cannabis or hashish, cocaine, heroin, and methamphetamine) and olfactory/gustatory dysfunction using data from the 2013–2014 National Health and Nutrition Examination Survey. In this study, participants who completed the smell examination with mean age of 59 were classified into four groups: cannabis users (n = 845), participants without cannabis use (n = 794), illicit drug users (n = 450), and participants without illicit drug use (n = 2000). Participants who completed the taste examination with mean age of 58 were also categorised into four groups: cannabis users (n = 810), participants without cannabis use (n = 714), illicit drug users (n = 428), and participants without illicit drug use (n = 1815). Logistic regression models investigated the association between cannabis or illicit drug use and smell or taste dysfunctions among study participants. Odds ratios and 95% confidence intervals were calculated. Finally, we did not find correlations between illicit drug use and dysfunction of taste or smell senses; our findings were consistent in many subgroup analyses. We recommend that further studies explore the mechanism and dose of illicit drug use that could have chemosensory impacts.


Introduction
Increasing evidence indicates that illicit drug use may lead to a substantial loss of life and disabilities [1]. Up until 2019, it was estimated that 5.4% of the global population used illegal drugs, and around 36.3 million people worldwide could be considered problematic drug users [2,3]. Additionally, according to the United Nations Office on Drugs and Crime report, cannabis is the most used drug, followed by opioids and amphetamines [2]. Moreover, drug abuse has noted tremendous impacts and health problems, including socioeconomic burdens, social and legal consequences, poverty and mental disorders, etc. [4,5]. Previous research indicated the need for excessive sensory stimulation as a predisposition to addictive behaviors. Exploring more sensory characteristics in people with addictive behaviors might provide the governments and policymakers with more information on this population [1,6,7].
Generally, olfactory dysfunction has been recognised as a heterogeneous condition, commonly caused by infection and trauma [8]. Olfactory loss is divided into two types: conductive and sensorineural dysfunction [9]. Impaired olfaction is disabling because it results in nutritional alterations in patients, increases the risk of injury, and lowers social relationships with reduced quality of life [10,11]. To date, taste and smell dysfunction are suspected to be associated with substance use. However, representative epidemiological studies remain insufficient. Smell and taste dysfunctions involve ageusia, hypogeusia, anosmia, hyposmia, etc. Ageusia refers to an absence of the sense of taste and hypogeusia refers to decreased taste sensitivity [12]. Furthermore, qualitative gustatory dysfunction is more frequent than quantitative dysfunction [13]. Moreover, anosmia and hyposmia, referring to a total and partial loss of the senses of smell, affect up to 25% of adults who are 50 years or older in America [14][15][16][17].
Currently, most studies explored the association of smell and taste dysfunction in substance users with alcohol use and smoking [18][19][20][21]. Some studies have shown a preserved level of olfaction without odour detection thresholds or discrimination deficits in alcoholic participants. Still, some even revealed impaired olfactory event-related potentials and olfactory dysfunction in patients with alcohol dependence [22,23]. However, only a few case series and animal models have reported and investigated the effects of chemosensory impairment in cannabis use [24,25]. Some studies also noticed that heroin addicts might be associated with a change in smell and taste function [26,27]. Nevertheless, there were equivocal results in reporting the connection between cocaine use and smell/taste dysfunction from animal models and humans [28,29].
Accordingly, relevant findings in previous studies have remained inconsistent. The research design investigating the association between chemosensory dysfunction and substance use has included cell research, animal models, and case-control studies. However, although the research focused on one illicit drug, the collected data were limited by small sample sizes and patient reluctance to discuss illicit drug use or resulting harms. Moreover, comprehensive epidemiological studies investigating the relationship between illicit drug use and smell and taste dysfunction have remained deficient. Therefore, this study explored the relationship between drug use (including cannabis or hashish, cocaine, heroin, and methamphetamine) and olfactory/gustatory dysfunction using information from a largescale cross-sectional database in the United States.

Database
Data from the National Health and Nutrition Examination Survey (NHANES) in the United States were used in this cross-sectional study. NHANES, a vital programme of the National Center for Health Statistics (NCHS), contains interviews and physical examination data of health and nutrition measurements. The database also includes information on about 5000 people from counties across the U.S. as a national and yearly representative sample. The research protocols for the NHANES received approval from the NCHS Research Ethics Review Board and written informed consent was obtained from each participant (https://www.cdc.gov/nchs/nhanes/irba98.htm, accessed on 1 April 2022). This study was exempted from a full review of the Institutional Review Board because the data used were de-identified secondary data from NHANES.

Study Sample Selection
This study investigated illicit drugs and cannabis use associated with smell and taste dysfunction. Therefore, participants who underwent the Taste and Smell Examination and filled out the Drug Use Questionnaire from 2013 to 2014 were included. Those with incomplete smell and taste, as detected from the examination results, or those with incomplete responses to related drug use questions, were excluded from this study. Finally, participants who completed the smell examination were classified into four groups: cannabis users (n = 845), participants without cannabis use (n = 794), illicit drug users (n = 450), and participants without illicit drug use (n = 2000). Participants who completed the taste examination were also categorised into four groups: cannabis users (n = 810), participants without cannabis use (n = 714), illicit drug users (n = 428), and participants without illicit drug use (n = 1815). Cannabis and illegal drug users were identified based on their answers to the relevant questions in the Drug Use Questionnaire. Therefore, cannabis users were defined as participants who answered that they had ever used marijuana or hashish, and illicit drug users were identified as participants who responded that they had ever used cocaine, heroin, or methamphetamine. Flow diagram for study sample selection is displayed in Figure 1.

Study Sample Selection
This study investigated illicit drugs and cannabis use associated with smell and taste dysfunction. Therefore, participants who underwent the Taste and Smell Examination and filled out the Drug Use Questionnaire from 2013 to 2014 were included. Those with incomplete smell and taste, as detected from the examination results, or those with incomplete responses to related drug use questions, were excluded from this study. Finally, participants who completed the smell examination were classified into four groups: cannabis users (n = 845), participants without cannabis use (n = 794), illicit drug users (n = 450), and participants without illicit drug use (n = 2000). Participants who completed the taste examination were also categorised into four groups: cannabis users (n = 810), participants without cannabis use (n = 714), illicit drug users (n = 428), and participants without illicit drug use (n = 1815). Cannabis and illegal drug users were identified based on their answers to the relevant questions in the Drug Use Questionnaire. Therefore, cannabis users were defined as participants who answered that they had ever used marijuana or hashish, and illicit drug users were identified as participants who responded that they had ever used cocaine, heroin, or methamphetamine. Flow diagram for study sample selection is displayed in Figure 1.

Outcome Measurements
The definitions of smell and taste dysfunction were based on the smell and taste examination results. An odour identification test examined the ability to smell, and the tasting ability was measured using salt and quinine taste testing.

Smell Dysfunction Definition
An eight-item odour identification test was used to identify the participants' ability to smell. First, participants were presented with eight specific odours via scratching of test strips. Then, eight specific scents were presented in a fixed sequence of chocolate, strawberry, smoke, leather, soap, grape, onion and natural gas. Participants were finally requested to choose the correct odour of the four listed options for each scent. Those who failed to identify six or more odours were defined as participants with smell dysfunction.

Outcome Measurements
The definitions of smell and taste dysfunction were based on the smell and taste examination results. An odour identification test examined the ability to smell, and the tasting ability was measured using salt and quinine taste testing.

Smell Dysfunction Definition
An eight-item odour identification test was used to identify the participants' ability to smell. First, participants were presented with eight specific odours via scratching of test strips. Then, eight specific scents were presented in a fixed sequence of chocolate, strawberry, smoke, leather, soap, grape, onion and natural gas. Participants were finally requested to choose the correct odour of the four listed options for each scent. Those who failed to identify six or more odours were defined as participants with smell dysfunction.

Taste Dysfunction Definition
The 1-mM quinine whole-mouth taste test was used to determine participants' taste ability [30]. The participants were first asked to take the 10-mL quinine solution into their mouths without swallowing, after which they were asked to swish and spit out the solution. Subsequently, each participant identified the taste of the solution, and their mouths were rinsed with water afterward. Those who failed to identify the bitter taste of quinine in the 1-mM quinine whole-mouth taste test were defined as participants with taste dysfunctions.

Covariate Measurement
We reduced the influence of potential confounders by considering participants' age, gender, ethnicity, hypertension status, diabetes mellitus status, coronary heart disease Healthcare 2022, 10, 909 4 of 11 history, angina pectoris, heart attack susceptibility, stroke, persistent cold/flu over the last 12 months before the study, head injury/loss of consciousness, broken nose/serious injuries to the face or skull, two or more sinus infections, smoking status, heavy alcohol use, and overweight in the regression models. Ethnicity was categorised as Mexican American, other Hispanic, non-Hispanic white, non-Hispanic black, non-Hispanic Asian and other races. Participants were defined as being overweight or having a medical history of hypertension, diabetes mellitus, coronary heart disease, angina pectoris, heart attack and stroke, based on their self-report of doctors' diagnosis. Those taking insulin and diabetic pills were also defined as having diabetes mellitus. Furthermore, participants who smoke and heavily take alcohol, or those with persistent cold/flu over the last 12 months before the study, head injury/loss of consciousness, broken nose/serious injury to the face or skull and those with two or more sinus infections, were identified using self-reported questionnaires.

Statistical Analysis
This study used the SAS system (SAS System for Windows, V.9.4, SAS Institute Inc., Cary, NC, USA) to conduct all analyses. Chi-squared tests were first applied to investigate differences in gender, ethnicity, hypertension, diabetes mellitus, coronary heart disease, angina pectoris, heart attack, stroke, persistent cold/flu last 12 months before the study, head injury/loss of consciousness, broken nose/serious injury to face or skull, two or more sinus infections, smoking status, heavy alcohol use, and overweight, between participants with and without smell or taste dysfunction. Next, the independent t-test was performed to compare the difference in age between participants with and without smell or taste dysfunctions. Finally, logistic regression models investigated the association between cannabis or illicit drug use and smell or taste dysfunctions among study participants, after which odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. A two-sided p-value < 0.05 was used to define the statistical significance of this study.

Results
This study contained smell and taste dysfunction study groups. Table 1 displays study participants' demographic characteristics and comorbidities with and without smell dysfunctions. The relevant findings showed significant differences in age, gender, ethnicity, hypertension, diabetes mellitus, coronary heart disease, angina pectoris, heart attack, stroke, two or more sinus infections, smoking status, and overweight between participants with and without smell dysfunction. The results (Table 1) also showed significant differences between the participants with and without taste dysfunction based on age, ethnicity, and smoking status. Therefore, we considered all these factors in the regression model to eliminate potential bias. Table 2 presents the prevalence of cannabis and illicit drug use among participants with and without smell dysfunction to further investigate the cannabis and illicit drug use association with smell dysfunction. Participants with and without smell dysfunction comprised 40.8% and 52.7% of cannabis users, respectively. After adjustments, the adjusted OR for cannabis use was 0.67 (95% CI, 0.43-1.03) among participants with and without smell dysfunction. Furthermore, individuals with and without smell dysfunction included 15.6% and 18.8% of illicit drug users, respectively. However, the adjusted OR for illicit drug use was 0.85 (95% CI, 0.59-1.25) among study participants with and without smell dysfunction, respectively. Subsequently, we further estimated the adjusted ORs for cocaine (OR, 0.85; 95% CI, 0.58-1.26), heroin (OR, 0.69; 95% CI, 0.30-1.56), and methamphetamine (OR, 0.84; 95% CI, 0.47-1.50) use of the understudied participants, respectively. Overall, there was no significant difference in smell dysfunction between cannabis or illicit drug users and those who did not use cannabis or illicit drugs. Note: SD, standard deviation. a The sum of n did not equal to the total number because of the missing data.
The prevalence and ORs for cannabis and illicit drug use among study participants with and without taste dysfunction are displayed in Table 2. Study participants with and without taste dysfunction comprised 57.0% and 52.2% of cannabis users, respectively. However, the adjusted OR for cannabis use was 0.98 (95% CI, 0.71-1.35) among participants with and without taste dysfunction. Furthermore, although participants with and without taste dysfunction comprised 20.5% and 18.8% of illicit drug users, respectively, the adjusted OR for illicit drug use was 0.85 (95% CI, 0.63-1.15) among participants with and without taste dysfunction. We further evaluated the adjusted ORs for cocaine (OR, 0.92; 95% CI, 0.68-1.26), heroin (OR, 0.64; 95% CI, 0.32-1.27), and methamphetamine (OR, 0.79; 95% CI, 0.50-1.25) use of the participants. Summarily, no significant difference in taste dysfunction was observed between cannabis or illicit drug users and participants who did not use cannabis or illicit drugs.
Subsequently, to reduce the effect of gender, we performed stratified analyses to explore the relationship between cannabis or illicit drug use and smell or taste dysfunction in males and females (Table 3). After adjusting for various confounders, findings indicated no significant association between cannabis or illicit drug use and smell or taste dysfunctions in the male or female population. Then, due to the influence of age on smell and taste dysfunction, we further investigated the association between cannabis or illicit drug use and smell or taste dysfunctions among study participants according to the different age groups ( Table 4). The study results remained unchanged after adjusting for demographic characteristics and comorbidities. Additionally, there was no significant difference in smell or taste dysfunction between cannabis or illicit drug users and those who did not use cannabis or illicit drugs in participants aged 40-49, 50-59, and 60-69 years.   Note: CI = confidence interval; OR = odds ratio; a Logistic regression; b adjusted for gender, ethnicity, hypertension, diabetes mellitus, coronary heart disease, angina pectoris, heart attack, stroke, persistent cold/flu last 12 months, head injury/loss of consciousness, broke nose/serious injury to face/skull, two or more sinus infections, smoking status, heavy alcohol use, and overweight; c Study participants who ever used cocaine/heroin/methamphetamine. ** p ≤ 0.01. Furthermore, to consider the effect of the duration of drug use and the age of first drug use, we also investigated the association between cannabis/illicit drug use and smell/taste dysfunction among study participants after adjusting for age of first cannabis/illicit drug use, days used cannabis/illicit drug during the past 30 days, and other demographic characteristics and comorbidities (Supplemental Table S1). The findings remained unchanged after adjusting for relevant confounders.

Discussion
To date, many medications, such as antimicrobials, antihypertensives, antidepressants, antipsychotics, antineoplastics, agents in cigarette smoke, and ethanol use, have been associated with chemosensory disturbances [31,32]. The potential olfactory dysfunction mechanisms based on these substances may be due to olfactory epithelium damage or central nervous system damage. Based on clinical experiences, this study initially hypothesised that illicit drug use might be associated with smell and taste function changes. However, we observed no association between participants with smell or taste dysfunction and cannabis or illicit drug use history. Notably, the association did not remain statistically significant after adjustments for confounders, including age, race, gender, or other physical comorbidities in males, females, and participants with different age groups.
Previous studies have investigated the relationship between substance use and smell/taste function. However, most current research only examined the impact of smoking and alcohol on taste or smell functions. Therefore, potential impacts on olfactory function have remained controversial. For instance, a study enrolled 48 drug addicts between the ages of 16 and 48.
Although 52.1% of the participants showed disturbances in olfactory performance, 16.7% were diagnosed with ageusia. Additionally, olfactory problems were detected in those who took drugs intravenously and smoked or inhaled drugs [33]. Another study enrolled 21 smokers and 59 non-smokers. Although their results showed smaller olfactory bulb volumes in smokers than in non-smokers, no difference in olfactory function between the two groups was observed [18]. Moreover, though Schriever et al. suggested that smoking negatively affects the olfactory system before it becomes obvious in decreased olfactory function, Vennemann et al. noted that heavy smokers (20 or more cigarettes per day) had an increased risk for impairments in smell and taste senses [19]. Therefore, since the temporal lobe mediates olfactory processing, and a correlation between general olfactory function and olfactory bulb volumes has been proposed, factors that affect the optimal function of this system could result in issues [9].
Our study focused on the relationship between cannabis, illicit drug use, and smell or taste function changes. Some prior studies have also investigated the relevant issue of cannabis. As reported, cannabinoids are involved in the neuromodulatory regulation of the sensory systems [34,35]. The active ingredient of cannabis, delta-9-tetrahydrocannabinol (THC), has also been reported to palliate symptoms in cancer patients [36][37][38]. Furthermore, it stimulates the orosensory reward pathway and enhances food enjoyment. Although cannabinoid type-1 receptors are located in different brain olfactory areas, including the olfactory epithelium and bulb [39,40], cannabinoid receptor agonists increase the hedonic reactions to sweet taste and reduce the aversive responses to quinine [24,35,36,41]. However, participants in that research were cancer patients who were explicitly administered THC. Case reports from students with an experience of marijuana intoxication also reported more vivid taste sensations and a richer sense of smell when intoxicated [42]. Similarly, a recent study showed that cannabidiol (CBD) protects against the psychoactive effects of THC [43]. Moreover, Woelfl et al. conducted a randomised controlled trial, which showed that CBD intake did not affect healthy volunteers, and a single dose of CBD before taking THC administration was insufficient to mitigate the impact of THC [44]. Nevertheless, cannabis contains over 400 chemical entities, with the ratios of each compound in recreational cannabis making it challenging to investigate the effect of drug users in real-world situations [45,46]. Our study did not notice a change in the smell or taste of individuals who ever used cannabis.
Additionally, our study did not find correlations between illicit drug use and dysfunction of taste or smell. Some previous research also investigated relevant issues. For instance, cocaine is usually administered illegally from the nasal route, and abusers often complain of decreased olfaction. Clinical pathology that contributes to reduced olfaction includes immune-mediated diseases and even nasal defects that require surgical reconstruction [47,48]. However, no olfactory or gustatory function tests were conducted on those users. Roebber et al. suggested that cocaine exposure did not change the taste sensitivity of animal models of mice [28]. Another study by Gordon et al. reported from three olfaction tests that most cocaine abusers did not develop permanent olfactory dysfunction [49]. Beidler and Smallman also proposed that nerves innervating the taste buds could regenerate after abrasion, making the taste buds retain function [50]. Therefore, there might be a restoration of chemosensory perception after quitting illicit drug use. Nevertheless, exploring the underlying mechanisms of individuals after using illicit drugs is needed.
Furthermore, Perl et al. noticed that heroin addicts estimated sweet tastes and savoury smells as more pleasant, and bitter, sour or putrid tastes and odours were considered less unpleasant by them than detoxified former addicts and healthy controls [26]. A randomised controlled trial with a small sample size measured sweet and salt taste perceptions of heroin users, recently detoxified subjects, and healthy volunteers. The results showed that heroin users and recently detoxified subjects had significantly greater measures of taste perception, and even this effect could be reversed by opiate antagonists [27]. These results suggest that heroin addicts might have altered taste and odour hedonic brain mechanisms. Findings in previous literature were similar to the trends found in our study. We observed that individuals with and without smell dysfunction included 2.4% and 3.2% of heroin users, respectively. Even though the relationship did not reach statistical significance, heroin users were more sensitive to odours. With amphetamine, a scarce study investigated the association between smell or taste dysfunction and amphetamine use.
This study had some unique strengths. First, it included a large-scale sample size from a population with a history of illicit drug use and relatively accurate examination results on smell and taste function. Second, we used both self-reported questionnaires and examinations (including an 8-item odour identification test and 1-mM quinine wholemouth taste identification test) to identify the smell and taste dysfunctions, eliminating recall bias from respondents. Finally, this study investigated the relationships between smell/taste dysfunction and illicit drugs, including cannabis or hashish, cocaine, heroin, and methamphetamine. Moreover, based on our extensive literature search, this research is the first to comprehensively study the relationship between different illicit drugs and smell/taste dysfunction. Notwithstanding, several limitations were encountered in this study. First, we could not explore the causal relationships because of the cross-sectional nature of this study. Second, this study did not consider certain medications that may affect smell and taste because the database lacked detailed information.

Conclusions
In conclusion, our research provided insight into the relationship between illicit drug use and olfactory/gustatory functions, and this association did not remain statistically significant after adjustments for confounders, including age, race, gender, or other physical comorbidities in males, females and participants with different age groups. With the diversity and complexity of chemical compounds in each illicit drug, studies with multidisciplinary teamwork may be warranted to objectively assess the nature of chemosensory effects caused by each illicit drugs. Therefore, we recommend that future studies explore the mechanism and dose of illicit drug use that could have chemosensory impacts in humans.

Supplementary Materials:
The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/healthcare10050909/s1, Table S1. Association between cannabis/illicit drug use and smell/taste dysfunction among study participants after adjusting for age of first cannabis/ illicit drug use, days used cannabis/ illicit drug during the past 30 days and other demographic characteristics and comorbidities. Funding: This study was supported by grants from Tri-Service General Hospital, Taipei, Taiwan (TSGH-E-111272) and Yang Ji Mental Hospital, Keelung, Taiwan. The funding source had no role in the study design; collection, analysis and interpretation of the data; preparation, review or approval of the manuscript and decision to submit the manuscript for publication.
Institutional Review Board Statement: Data from the National Health and Nutrition Examination Survey (NHANES) in the United States were used in this cross-sectional study. This study was exempted from a full review of the Institutional Review Board because the data used were deidentified secondary data from NHANES.

Informed Consent Statement:
The research protocols for the NHANES received approval from the NCHS Research Ethics Review Board and written informed consent was obtained from each participant (https://www.cdc.gov/nchs/nhanes/irba98.htm, accessed 1 April 2022).

Data Availability Statement:
Publicly available datasets were analyzed in this study. This data can be found here: https://wwwn.cdc.gov/nchs/nhanes/Default.aspx, accessed 1 April 2022.