Surveillance by oral fluid of drugs subject to misuse among individuals under arrest

ABSTRACT

Background: Between 1988 and 2013 the U.S. government conducted surveillance of national drug misuse use trends by collecting voluntary urine specimens from individuals under arrest in major counties. It was discontinued for financial reasons. The program was the only national survey that used a bioassay to measure drug use. Other national drug surveys continue to be based entirely on self-reports of drug use.

Objective: Given the current opioid and incipient methamphetamine epidemics, this study aimed to demonstrate the feasibility of surveilling drugs subject to misuse among individuals under arrest using oral fluid collected anonymously by jail staff in one U.S. county. This method has never been previously employed with an offender population.

Methods: The subjects were adults arrested for any reason and booked in one Midwest county jail in the U.S. between July 2019 – January 2020 (N = 196; 145 males). Oral fluid specimens were provided for research purposes voluntarily and anonymously.

Results: 79% of individuals approached consented to participation. The most frequently detected drugs were cannabis (53%), methamphetamine (27%), cocaine (9%) and opioids (11%). Further, 74% tested positive for at least one drug; 36% tested positive for at least one illegal drug, 10% tested positive for at least one possibly illegal drug, and 54% tested positive for at least one legal drug (predominantly cannabis). (Tests for nicotine and ethanol were not included.)

Conclusion: The feasibility of collecting oral fluid from individuals under arrest in a jail setting to measure the prevalence of drugs subject to misuse was demonstrated.

KEYWORDS:

• oral fluid
• saliva
• offenders
• drug use
• drug testing

Introduction

Between 1988 and 2013 the U.S. government sponsored collection of data on drug misuse from adult male individuals under arrest in up to 35 selected counties nationwide during various periods. The initiative was termed Arrestee Drug Abuse Monitoring (ADAM). Data collection from individuals under arrest was voluntary and confidential, consisting of an interview, official record information, and a urine specimen that was tested for several drugs subject to misuse. ADAM was decreased to five counties involving only major cities in 2012–2013 and entirely discontinued in 2014 due to budget limitations (1).

ADAM was important as a major source of information about drug misuse trends in the U.S. because it assessed an offender population known to be highly susceptible to drug misuse (2); involved persons not often reached in surveys of the general, in-treatment or health care populations; and was the only national survey that used a bioassay (urinalysis) to measure drug use.

In particular, the current opioid misuse epidemic requires better knowledge about the extent and types of opioid misuse in vulnerable populations as well as sentinel data that could identify and predict local and national trends in the availability and use of illicit opioids. In that regard the President’s Commission on Combating Drug Addiction and the Opioid Crisis recommended improvements in “real time surveillance of the opioid crisis at the national, state, local and tribal levels.” Specifically, “the Commission recommends the federal government work with the states to develop and implement standardized rigorous drug testing procedures … ” including “reinstituting the Arrestee Drug Abuse Monitoring (ADAM) program … ” (3).

The purpose of this study was to demonstrate the feasibility of surveilling drugs subject to misuse among individuals under arrest using oral fluid collected anonymously by jail staff in one Midwestern (U.S.) county.

Methodology

Setting

The subject pool was individuals arrested for any reason and booked at a jail located in one Midwestern county in the United States (Calhoun County, MI), whose principal city is Battle Creek. As of 2018, the county’s population was 134,487; female 51%; white non-Hispanic 77%, black 10%, Hispanic 5%; median household income 47,204 USD; high school graduate 91% (4).

The goal was to recruit 200 eligible unduplicated persons for the study. This was not based on a power calculation, but comes from a negotiation about the time commitment the Sheriff’s Office was able to give to the conduct of the study.

Training

Before starting the recruitment and data collection, correctional deputies completed online and in-person trainings. Online training consisted of deputies completing human subjects protection Collaborative Institutional Training Initiative (CITI) trainings approved by the university, which included assessing risk, informed consent, privacy and confidentially, and research involving prisoners (5). The in-person trainings were conducted by the first and second authors and addressed five key steps to implementing the study (i.e., recruitment, consent procedures, collection of oral fluid samples, completing a Data Collection Sheet, and packaging oral fluid specimens for mailing). Deputies also reviewed a manual that summarized five steps involved in implementing the study.

Eligibility Criteria

The inclusion criteria were:

• booked at the Calhoun County (MI) jail. for any offense;

• 18 years of age or older;

• able to provide an oral fluid specimen within 6 hours of booking;

• able to understand the English-written consent form; and

• had not previously participated in the study.

The eligibility of individuals for the study was determined in sequential order of booking. However, whether a given person under arrest was assessed for the study and approached to participate depended on the how busy the booking area was and whether a trained deputy was available to recruit a given individual and collect the data. Because of the conditions of jail operations, eligible individuals could usually be recruited for the study only at less busy times.

Subject Recruitment

For individuals eligible for the study, correctional deputies approached them using a recruitment script. The script stated the following: “I’m inviting you to participate in a voluntary research study being conducted by researchers from Western Michigan University. The purpose is to learn more about what substances people are using in Calhoun County. Two hundred people who have been arrested are being asked to voluntarily provide an anonymous saliva sample for drug testing. “Anonymous” means your name will not be on the sample. Also, the Sheriff’s Office will not get the test result. So there is no risk to you. A Deputy such as myself will obtain the saliva sample and then submit it to the researchers (show tube). In appreciation for helping out, you will receive two candy bars. If you are interested, I’ll read you the consent form.” Deputies kept a log of all individuals approached about the study to avoid duplicating participation, and the researchers never had access to this log. It was later discovered that data were collected from five individuals whom the deputies had marked “yes” on “participated in the study previously?” on their Data Collection Sheet. While this was likely an error in transcribing from the log to the sheet, we dropped such data to avoid possibly duplicate individuals in the study.

If individuals were interested in participating, the informed consent was administered. The correctional deputies provided possible participants the opportunity to read the consent form, have the form read to them, or have the deputy provide an overview of the consent form. Based on the person’s preference, the deputy administered the consent form. As they administered the consent form, deputies checked periodically whether the person understood the consent form. After administering it, the deputy offered to place a copy of the consent form with the possible participants’ personal belongings, which they would receive once released from jail.

After the consent form was administered, individuals were asked whether they would like to participate in the study. Individuals who reported “yes” were considered to have provided voluntary verbal consent and the deputy attested to that by signing the Data Collection Sheet. Deputies did not ask the reason for declining. However, if the individual offered a reason, then this information was documented. Individuals who reported “no” were considered to have declined to participate.

Verbal consent was used because the testing was anonymous and there was no need to record participants’ names. Federal regulations allow the Institutional Review Board (IRB) to waive the requirement for obtaining signed consent if it finds that “the only record linking the subject and the research would be the informed consent form and the principal risk would be potential harm resulting from a breach of confidentiality” (6).

Data Collection Procedures

Correctional deputies implemented the data collection procedures. Data collection consisted of (1) completing a Data Collection Sheet for each individual who was eligible to participate in the study and (2) the collection of the oral fluid specimen. The Data Collection Sheet included the following: certification of verbal consent including date and time, whether individual wanted a copy of consent form with their personal possessions, subject demographics, arrest charges (obtained from the booking log), and time of oral fluid collection; no individual subject identifying information was recorded. The study purposely did not ask the subjects for self-reports of substance use because deputies were collecting the data and it is likely that the recruited individuals would be reluctant to reveal such information.

Oral fluid was collected using the Quantisal device which is administered by the subject placing a stick with a small pad in their mouth until the pad is saturated (in 3 − 10 minutes). The stick and pad are then inserted into a plastic tube which is sealed; no subject identifying information was on the tube (7).

Once both the Data Collection Sheet was completed and the specimen was collected, the deputy placed a sticker with a unique research number on each item. The unique number was pre-generated and not linked with any subject identifier. The number linked the information on the sheet and the test results. The sheets and tubes were mailed at least weekly by overnight express to the laboratory; note oral fluid is not considered a biohazard.

The specimens were not refrigerated, although the Quantisal device does include a reagent buffer that acts as a preservative. The mean time between specimen collection and arrival at the laboratory was 4 days; they were tested on the day of arrival. Common drugs subject to misuse have been found to be stable at room temperature in buffered oral fluid specimens for 7 and 14 days, except for 6-monoacetylmorphine (6-MAM), a unique metabolite of heroin (8), which showed significant analyte decrease (- 28%) at 14 days (9).

Oral Fluid Analysis Procedures

The specimens were analyzed by a specialty laboratory – Forensic Fluids Laboratories Inc.

All specimens were analyzed for 19 psychoactive drugs or drug classes by enzyme-linked immunosorbent assay (ELISA) and 60 specific drugs by liquid chromatography/tandem mass spectrometry (LC/MS/MS), which included all the drugs analyzed by ELISA except for barbiturates and synthetic cathinones (‘bath salts”) for which LC/MS/MS tests were not available at the time. Thus, the study tested for 62 individual drugs or drug classes.

The vendors for the ELISA text kits for each drug screened are given in Table 1. The LC/MS/MS analysis was conducted with Waters TQD Mass Spectrometer and Waters Acquity Liquid Chromatography System.

Table 1. Vendors for ELISA screening kits

ELISA Screening Panel	Vendor	Catalog Number
Amphetamine	Immunalysis	209–0480
Methamphetamine	Immunalysis	211–0480
THC	Immunalysis	224–0480
Cocaine	Immunalysis	212–0480
Opiates	Immunalysis	207–0480
Benzodiazepines	Immunalysis	214–0480
Barbiturates	Neogen	130615
Methadone	OraSure	114211 C
PCP	Immunalysis	208–0480
Oxycodone	Immunalysis	221–0480
Fentanyl	Immunalysis	218–0480
Buprenorphine	Immunalysis	236–0480
Ketamine	Immunalysis	240–0480
Tramadol	Immunalysis	225–0480
Carisoprodol	Immunalysis	231–0480
Methylphenidate	Immunalysis	219–0480
Mitragynine	Randox	MT3489
Synthetic Cannabinoids	Randox	PAC10046
Synthetic Cathinones	Immunalysis	BSA-1000

All drugs were analyzed by LC/MS/MS whether or not they were positive by ELISA. The testing included these individual drugs or drug classes susceptible to misuse: methamphetamines; cocaine; cannabis; synthetic cannabinoids; opioids; benzodiazepines; synthetic cathinones (“bath salts”); selective serotonin reuptake inhibitors (SSRIs); tricyclic anti-depressants (TCAs); and various individual substances of interest such as 6-monoacetylmorphine (6-MAM; fentanyl; ketamine (“Special K”); mitragynine (“kratom”); 3,4-methylenedioxy-methamphetamine (MDMA – “Ecstasy,” “Molly”) and tramadol. A complete list of the drugs and drug classes for which the study tested is in Table 4 (including the footnotes), except that all the individual synthetic cannabinoids are not listed. Common psychoactive drugs are generally detectable in oral fluid for 24 hours and sometimes longer after last use (10).

Table 2. Sample characteristics

	%	(n)
Gender^a		(191)
Female	24.1	(46)
Male	75.9	(145)
Age		(189)
18–29	42.9	(81)
30–39	29.6	(56)
40–49	17.5	(33)
50–59	6.3	(13)
60 and over	3.7	(7)
Race/Ethnicity		(191)
White, non-Hispanic	55.0	(105)
Black, non-Hispanic	29.8	(57)
Hispanic	94	(10)
Biracial/Native American/Arabic	5.8	(14)

^aIncludes one individual who identified as trans female and one individual who identified as trans male.

Table 3. Charges at time of booking (n= 196)

	%	(n)
Charges^a
Failure to appear	30.1	(59)
Violent	22.4	(44)
Driving while impaired	11.2	(15)
Driving impaired by alcohol	3.6	(7)
Driving impaired by drugs	0.5	(1)
Driving impaired by alcohol or drugs (indefinite)	7.1	(16)
Friend of the court (e.g., child welfare)	9.7	(17)
Probation/parole/bond violation	8.7	(18)
Controlled substance(s)	8.2	(19)
Disorderly	7.7	(20)
Property	7.7	(20)
Traffic violation (excluding impairment)	4.6	(9)
Weapons violation	1.5	(3)

^aUp to two charges were coded for the table. Only eight individuals had more than two charge categories; the most serious were included.

Table 4. Drug positives at time of booking^a (n = 196)

	% (n)	95% CI
Cannabis^b	53.1 (104)	[45.8, 60.3]
Methamphetamine	26.5 (52)	[20.5, 33.3]
Cocaine^c	9.2 (18)	[5.5, 14.1]
Opioids	10.7 (21)	[6.8 15.9]
Fentanyl^d	3.6 (7)	[1.4, 7.2]
6-MAM (heroin)	2.0 (4)	[0.6, 5.1]
Buprenorphine	2.0 (4)	[0.6, 5.1]
Methadone^e	1.5 (3)	[0.3, 4.4]
Morphine (excluding 6-MAM)	1.0 (2)	[0.1, 3.6]
Hydrocodone	0.5 (1)	[0.01, 2.8]
Tramadol	0.5 (1)	[0.01, 2.8]
Other opiate	0.5 (1)	[0.01, 2.8]
Synthetic cathinones (“bath salts”)	1.5 (3)	[0.3, 4.4]
Selective serotonin reuptake inhibitors (SSRI)^f	1.0 (2)	[0.1, 3.6]
Benzodiazepines^g	1.0 (2)	[0.1, 3.6]
Synthetic cannabinoids	0.5 (1)	[0.01, 2.8]
Tricyclic anti-depressants (TCA)	0.5 (1)	[0.01, 2.8]
Mitragynine (“kratom”)	0.5 (1)	[0.01, 2.8]

^aDrugs and drug metabolites for which specific tests were conducted and had no positives are as follows: phencyclidine (PCP), ephedrine (Bronkaid, Sudafed), MDMA (Ecstasy, Molly), MDA (Adam), MDEA (Eve), alprazolam (Xanax), clonazepam (Klonopin), 7-aminoclonazepam, lorazepam (Ativan), propoxyphene (Darvon, Darvocet), norbuprenorphine, oxycodone, propoxyphene, carisoprodol (Soma), ketamine, methylphenidate (e.g, Ritalin), and barbiturates.

^bdelta09-tetrahydrocannabinoid (THC)

^cIncludes positives for the parent drug, benzoylecgonine (BZE), ecgonine methyl ester (EME), and cocaethylene.

^dIncludes norfentanyl

^eIncludes EDDP (methadone metabolite)

^fIncludes gluoxetine (e.g., Prozac, Sarafem) and sertraline (i.e., Zoloft)

^gIncludes positive benzodiazepine ELISA screen, diazepam, nordiazapam, temazepam

The study was approved by the Institutional Review Board (IRB) of Western Michigan University. It was funded by internal funding from the first author’s organization. Funding covered the cost of testing and reimbursement for deputies’ time.

Results

Subject recruitment

Data were collected for approximately 6 months between July 2019 – January 2020; 257 individuals were approached and 203 provided data, resulting in a 79% participation rate. Most of those individuals who declined did not give a reason and were deliberately not asked for a reason. Of 203 individuals providing data, seven were identified as ineligible for the study upon inspection of their information (five duplicate individuals, two aged 17), leaving an analytic sample of 196.

Subject characteristics

Table 2: Males, black non-Hispanics, Hispanics, and younger adults were overrepresented in the offender sample as compared with the general population in Calhoun County but were represented by sex and race/ethnicity similarly to jail populations nationally (11).

Table 3: Failure to appear (FTA) was the most frequent arrest charge. We did not attempt to classify reasons for FTA, because our main interest was in documenting alleged offenses committed at the time of the arrest. Violent offenses were the second most frequent charge and driving while impaired (combining all categories) the third most frequent. Possession of controlled substances was relatively infrequent.

Oral fluid test results

Table 4: This table shows the occurrences of specific drugs and drug classes detected by the testing. The most frequently detected drug was a newly legal drug available recreationally in Michigan, cannabis (53.1%). Next in prevalence were methamphetamine (26.5%), cocaine (9.2%), and opioids (10.7%). Among opioids, fentanyl was at 3.6% and 6-MAM, an indicator of heroin use, was at 2.0%. Relatively new illegal drugs subject to misuse were detected rarely (e.g., synthetic cathinones at 1.5% and synthetic cannabinoids at 0.5%) or not at all (e.g., Ecstasy/Molly).

Mitragynine, popularly known as kratom, was detected in only one specimen (0.5%).

We examined the relations among opiate-, 6-MAM- and fentanyl- positive specimens. “Opiates” are defined as opioids without synthetic composition, i.e., solely opium-derived. Note that 6-MAM will cause an opiate-positive ELISA screening test, but fentanyl will not. Among the seven opiate-positives, four were also positive for 6-MAM, indicating that these four opiate-positives were attributable to heroin use, since 6-MAM is a unique metabolite of heroin (8). Also, among seven subjects testing positive for opiates, only one also tested positive for fentanyl, indicating that fentanyl and opiates were usually not being used as a mixture or otherwise close in time. Among the seven subjects testing positive for fentanyl, only one was also positive for 6-MAM, indicating that fentanyl and heroin were usually not being used as a mixture or otherwise close in time.

Table 5: This table summarizes the drug use data in Table 4 focusing on illegality. “Possibly illegal drug use” refers to drugs that are available by prescription, some widely available, but also often misused by individuals who do not have a prescription, but may procure them from others who do have a prescription, including family members, or from the black market.

Table 5. Drug use by legality (n = 196)

	%	(n)	95% CI
Any illegal drug use^a
No illegal drug use	64.3	(126)	[57.1, 71.0]
Yes, illegal drug use	35.7	(70)	[29.0, 42.9]
Any possibly illegal drug use (i.e. possibly unprescribed)^b
No possibly illegal drug use	90.3	(177)	[85.3, 94.1]
Yes, possibly illegal drug use	9.7	(17)	[5.9. 14.7]
Any legal drug use^c
No legal drug use	45.9	(90)	[38.8, 53.2]
Yes, legal drug use	54.1	(106)	[46.8, 61.2]
Overall drug use^d
No drug use	26.5	(52)	[20.5, 33.3]
Used at least one drug	73.5	(144)	[66.7, 79.5]

^aRefers to any of the illegal drugs listed in Table 4 (i.e., methamphetamine, cocaine, fentanyl, 6-MAM (heroin), synthetic cathinones, and synthetic cannabinoids

^bSome drugs were considered “possibly illegal” because they may have been taken illegally or without a valid prescription for the individual. The following drugs in table 4 are defined as possibly illegally used: Morphine (excluding 6-MAM), buprenorphine, methadone, hydrocodone, tramadol, selective serotonin reuptake inhibitors (SSRI), benzodiazepines, and tricyclic anti-depressants (TCA).

^cRefers to any of the following legal drugs: cannabis (THC) or mitragynine.

^dRefers to any drug use (i.e., illegal, possibly illegal, or legal).

Fentanyl, although technically available by prescription, is very rarely prescribed to outpatients because for outpatients it is only authorized for breakthrough cancer pain (12).

Thus, classifying fentanyl as a presumptively illegal drug for this article seemed to be the most accurate characterization.

As shown in Table 5, 35.7% of the individuals under arrest tested positive for at least one illegal drug, 9.7% tested positive for at least one possibly (for them) illegal drug and 54.1% tested positive for at least one legal drug (predominantly cannabis). When combining illegal, possibly illegal, and legal drugs, 73.5% of the individuals tested positive for at least one of the drugs measured in this study.

Discussion

This method of drug surveillance has practical and scientific benefits. As compared with standard urine testing, oral fluid is a simpler, less intrusive and more dignified method of specimen collection, as well as less susceptible to tampering (13–16). These considerations could prove useful in criminal justice and other restrictive settings or in broad-based epidemiological research that is usually limited to self-reporting of drug misuse (17). Several studies have demonstrated substantial underreporting of drug misuse in community samples, particularly of highly stigmatized drugs such as cocaine and heroin (18–21).

Although there was a potential that the participation rate might be low because the data were being collected by correctional deputies, that possibility was not substantiated in the current study; the participation rate for individuals approached (79%) was about the same as for ADAM (82%) where research assistants collected the data (1). Although the jail was reimbursed for deputies’ time, this approach to data collection was likely more cost-effective than stationing research assistants at the jail, especially because the goal was to collect specimens in “real time,” i.e., within 6 hours of booking, instead of within a 48-h window as in ADAM. The latter time lag could adversely affect drug detection rates, whether for oral fluid or urine (10).

Cannabis has historically been the most common illegal drug misused in the U.S (17) and the recent state trend for its legalization may increase the prevalence of its use. The possession of small amounts of cannabis became legal in Michigan in December 2018, before the period of the study. Thus, we expected a high prevalence of cannabis use in the sample, which was confirmed (53%).

Following cannabis, methamphetamine was the second most prevalent drug detected at 27%. This finding is similar to results reported by the National Forensic Laboratory Information System (NFLIS) drug monitoring system for law enforcement operations, which reported 33% of drug-related investigations involved methamphetamine, the first year it exceeded cannabis (22). Methamphetamine has been trending upward in law enforcement investigations since 2011 as well in national urine test data for substance use disorder, pain management and primary care settings since 2013 (23).

Although the numbers are small, the current study indicates that fentanyl is often being misused independently of opiates such as heroin.

A variety of drugs of recent concern were found rarely or not at all in the testing, i.e., synthetic cathinones (“bath salts”), synthetic cannabinoids (synthetic cannabis), MDMA (“Ecstasy/Molly”), ketamine (“Special K”), tramadol and mitragynine (“kratom”). These findings indicates either that their prevalence in this population is too low to be detected with this size sample, that their misuse is a regional phenomenon not currently present in southwest Michigan, that their misuse is not currently a widespread problem, or that our testing did not detect the specific agents that are being misused in this area. Specifically, with respect to synthetic cathinones, these were only tested by ELISA and the screen may miss agents in this class due to their continually changing profile (24). Several of the above substances also appear only at very low rates in the latest national forensic laboratory test data, i.e., synthetic cathinones and cannabinoids; MDMA; and tramadol (22).

The current study is the first to drug test a general sample of individuals under arrest since the discontinuation of the ADAM program in 2013. Also, while ADAM only tested for 10 individual drugs or drug classes, the current study tested for 62 drugs or drug classes, providing a more comprehensive profile that includes newly emergent drugs of concern (25).

For this exploratory pilot study, the researchers decided to apply all the testing available at the laboratory for drugs susceptible to misuse. We cast the net widely and did not attempt to target certain drugs based on what might be expected to be found locally or even nationally. However, this might not be the most efficient procedure if such jail-based drug surveillance became more “routine.”

ADAM was discontinued due to financial reasons. Although conducting a formal differential cost analysis between the ADAM and current data collection methods was beyond the scope of this pilot study, the current method is likely to have a cost advantage. In ADAM dedicated research assistants were stationed at jails to recruit and collect data from prospective subjects, which must have led to considerable “down time” awaiting eligible subjects. Moreover, jail staff, whose time should be reimbursed, must also be involved in the collection of urine specimens, e.g., in escorting subjects to rest rooms as needed and otherwise monitoring subjects’ interaction with the research assistants, although it is unclear whether the jails received any reimbursement for staff expenses in ADAM. In the current study, the only on-site expense was reimbursement of correctional deputies’ time in recruiting and collecting oral fluid from subjects; “down time” was not an issue. Incidentally, currently the costs of analyzing urine and oral fluid for drugs can be considered equivalent (26).

The current study pilots a procedure that adds an informative source for drug surveillance that is currently not being utilized nationally. The NFLIS reports on toxicology of presumably illegal drug samples seized in the field (22), but other sources of data are needed, such as self-report surveys and biomarkers of use in the general population and special populations. Studies of biomarkers of drug use in the special population of individuals under arrest provide a unique source of drug surveillance data that however must be interpreted in the context of all other community sources of drug use data.

Limitations

The current feasibility study is subject to several limitations. First, it relied on a relatively small convenience sample in a limited geographical area. Second, the sample may be censored because individuals with detectable drugs may have been less likely to agree to participate. Therefore, the results are not generalizable nationally and may not even be entirely representative of drug misuse among individuals under arrest in the specified county.

Third, variability exists in oral fluid’s window of detection across different drugs, which could lead to bias in the determination and comparisons of prevalence rates across drugs included in this study (10,27).

Fourth, although laboratory results indicate that oral fluid testing can detect a wide variety of specific drugs (28), it is also possible that some drugs were not sufficiently detected, e.g., synthetic cathinones and synthetic cannabinoids, which frequently change in their chemical formulations (24,25).

Fifth, Oral fluid does not yield information on patterns of drug use. For example, occasional misuse of a drug that was detected on the day after versus chronic misuse are not distinguishable by the study.

Finally, tests for certain common legal substances were not included, i.e, nicotine and ethanol.

Conclusions

The advantages of drug surveillance of individuals under arrest to states and counties would be to provide a unique source of objective (bioassay) information on drug misuse trends within a local sentinel population vulnerable to drug misuse. Drug misuse among individuals under arrest may be a predictor of potentially wider spread of specific drugs into a community. But additional research first needs to examine in more detail whether oral fluid is a valid method to assess drug misuse in this or other populations of epidemiological interest. If valid then this approach could track local trends in the prevalence and types of drug misuse. Introduction of new drug compounds into the community and ingestion of dangerous combinations of, say, opioids or amphetamines with other drugs could be identified and tracked. The changing needs for certain types of drug addiction treatment could be identified. Compiling local drug surveillance data from individuals under arrest could contribute to a national profile of community drug misuse trends in nearly real time.

Disclosure of interest

The authors report no conflicts of interest.

Acknowledgements

The cooperation of the Calhoun County Sheriff’s Office in the conduct of this study is deeply appreciated, particularly Randy Hazel, Chief Deputy; Holly Thomas, Captain; Kevin Hirakis, Lieutenant; and all of the correctional sergeants and deputies. We also appreciate the support of the current Sheriff, Steven Hinkley, for publication of this study.

Additional information

Funding

No sponsored funding.