Field-Test of a Date-Rape Drug Detection Device

Abstract

Drink Safe Technology Version 1.2 is an inexpensive color-change reagent test marketed internationally for use by consumers in settings such as a night club to detect potentially incapacitating concentrations of gamma-hydroxybutyric acid (GHB) and ketamine in beverages. The objective of this study was to compare product performance in the laboratory and performance in the hands of consumers in the field. Product performance in the laboratory adhered to the protocol defined by the manufacturer. Product performance in the hands of consumers in field settings allowed browsing participants to pipette an aliquot of their own drinks into randomly coded vials containing authentic drugs, or pure water, so as to yield the same concentrations of GHB or ketamine specified in the manufacturer-defined protocol, or blanks. Consumers were to proceed according to the directions printed on the product, and to record their results on a card with a code corresponding with the vial to which they had added an aliquot of their beverage. Diagnostic performance was calculated using two-way analysis. In the laboratory, Drink Safe Technology Version 1.2 reliably detected GHB and ketamine at concentrations specified by the manufacturer's protocol. The reactive color change denoting a positive test for GHB was rapid, but a positive test for ketamine required substantially more time to resolve. Nonetheless, test accuracy following the manufacturer's protocol in the laboratory was 100%. In the field, based on 101 paired-test results recorded by consumers, the test efficiency was 65.1%, sensitivity 50%, and specificity 91.6%. The product performed much better in the laboratory than it did in the hand of consumers in the field. There seems to be considerable potential for consumers to misinterpret a test result. The potential for consumers to record a false-negative test result for a spiked drink is cause for concern.

Full text

Journal of Analytical Toxicology, Vol. 31, July/August 2007
Technical Note
Field-Test of a Date-Rape Drug Detection Device
Dale W. Quest*
College of Nursing, University of Saskatchewan, Saskatoon
Joanne Horsley
Department of Campus Safety, University of Saskatchewan
Saskatchewan, Canada
Saskatoon, Saskatchewan, Canada
Abstract[
Drink Safe Technology Version 1.2 is an inexpensive color-change
reagent test marketed internationally for use by consumers in
settings such as a nightclub to detect potentially incapacitating
concentrations of y-hydroxybutyric acid (GHB) and ketarnine in
beverages. The objective of this study was to compare product
performance in the laboratory and performance in the hands of
consumers in the field. Product performance in the laboratory
adhered to the protocol defined by the manufacturer. Product
performance in the hands of consumers in field settings allowed
browsing participants to pipette an aliquot of their own drinks into
randomly coded vials containing authentic drugs, or pure water, so
as to yield the same concentrations of GHB or ketamine specified
in the manufacturer-defined protocol, or blanks. Consumers were
to proceed according to the directions printed on the product, and
to record their results on a card with a code corresponding with
the vial to which they had added an aliquot of their beverage.
Diagnostic performance was calculated using two-way analysis. In
the laboratory, Drink Safe Technology Version 1.2 reliably detected
GHB and ketamine at concentrations specified by the
manufacturer's protocol. The reactive color change denoting a
positive test for GHB was rapid, but a positive test for ketamine
required substantially more time to resolve. Nonetheless, test
accuracy following the manufacturer's protocol in the laboratory
was 100%. in the field, based on 101 paired-test results recorded
by consumers, the test efficiency was 65.1%, sensitivity 50%, and
specificity 91.6%. The product performed much better in the
laboratory than it did in the hands of consumers in the field. There
seems to be considerable potential for consumers to misinterpret a
test result. The potential for consumers to record a false-negative
test result for a spiked drink is cause for concern.
Introduction
The Society of Forensic Toxicologists' Drug-Facilitated
Sexual Assault Committee defines drug-facilitated sexual
assault as an offense whereby a person is subjected to non-
" Author to whom correspondence and reprint requests should be addressed:
Dr. Dale W. Quest, A-084 Medical Research Bldg., 103 Wiggins Rd., Saskatoon, SK,
S7N 5E4, Canada. E-mail: quest.d@usask.ca.
consensual sexual acts while he/she is incapacitated or un-
conscious due to the effects of alcohol and/or drugs and there-
fore prevented from resisting and/or unable to consent (1). It
is not a legal definition; there is no clear legal standard for in-
capacitation (2). The extent that drug-facilitated sexual assault
constitutes a public health problem remains poorly resolved.
The uncertainty is due in large part to the want of reliable
es-
timates
of the incidence of this crime and the prevalence of its
sequelae (3). There is evidence to suggest that the incidence of
drug-facilitated sexual assault has increased in one Canadian
locale (4). Estimates based on interviews, surveys, and statistics
suggest gross underreporting of sexual assaults in general and
drug-facilitated sexual assaults in particular (5-7). The nature
of the crime is such that amnesic properties of the drugs at-
tenuate assault victims' recollection of the assault, their
propensity to report and their reliability as witnesses. Case re-
ports suggest that the risk is real and that this crime can have
a profound negative impact on the lives of victims (8-14).
Public awareness and education initiatives have potential to re-
duce drug-facilitated assault if they inspire potential victims to
take rational, effective precautions to lower personal risk. A
caveat, described as the Melbourne effect, is the potential for
widespread media attention to embolden an increase in drug-
facilitated sexual assaults (4).
Fear of drug-facilitated sexual assault has generated a market
for inventions that claim to reduce potential victims' vulnera-
bility. These include tamper-resistant serving containers to
prevent drinks from being surreptitiously spiked (15,16) and di-
rect-to-consumer in vitro screening tests for beverages sus-
pected of being drugged. Drink Safe Technology Version 1.2 is
an inexpensive color-change reagent test marketed interna-
tionally and in Canada for use by consumers in settings such as
a nightclub or house party. It purports to detect potentially in-
capacitating concentrations of ~,-hydroxybutyric acid (GHB)
and/or ketamine, on suspicion that one of those drugs has
been surreptitiously added to the consumer's beverage. The
performance of Drink Safe Technology Version 1.2 under lab-
oratory conditions has been reported by the manufacturer
(17). An independent laboratory evaluation has also been pub-
lished (18). The objective of this study was to compare product
performance in the laboratory, to performance in the hands of
354 Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.

Journal of Analytical Toxicology, Vol. 31, July/August 2007
consumers in the field. The field testing component was
imbedded in a university campus date-rape information cam-
paign called
"It's
All About Awareness".
Experimental
Materials
Drink Safe Technologies
TM
Version 1.2 Date Rape Drug
Test Strips [UPC lots: 5534500001 (4 packages of 2 strips),
5534500002 (1 package of 10 strips), 5534500005 (1 package of
20 strips)] were provided as a gift by Drink Safe Technologies,
Inc. (Wellington, FL). Drink Safe Technologies Version 1.2
customized coasters (UPC lot: 5534500029) were purchased
from Drink Safe Canada, Inc. (Calgary, AB, Canada). The fol-
lowing drugs were used in this study: GHB sodium (Cat#
H3635, Sigma-Aldrich Chemical Company, Inc., St. Louis, MO,
Health Canada authorization #10258.05.05); ketamine hy-
drochloride (Cat# K2753, Sigma Aldrich Canada, Ltd., Oakville,
ON, Canada); flunitrazepam (Cat# F9261, Sigma-Aldrich
Chemical, Health Canada authorization # 10257.05.03); 1,4-bu-
tanediol (Cat#18960, Sigma-Aldrich Canada); and ~,-butyro-
lactone (Cat# 20740, Sigma-Aldrich Canada). The Milli-Q pu-
rification system for ultra-pure water was from Millipore
Canada, Ltd. (Mississauga, ON, Canada). The analytical bal-
ance (d -- 0.1 mg, model BL120S) was from Sartorius (Edge-
wood, NY). Digital camera (5.0 megapixel resolution, images
stored as JPEG 2592 x 1944) was a PowerShot A95 (Canon
Canada, Mississauga, ON, Canada).
Methods
This project, which doubled as an educational event and an
independent consumer product field-test, underwent expe-
dited review by the University of Saskatchewan Biomedical
Research Ethics Board. Product performance was assessed in
the laboratory according to the manufacturer-defined protocol
(17). The laboratory readings were performed within an 80-cm
diameter workspace illuminated 555.4 • 35.6 Lux by two over-
head fixtures, 120 cm apart, each with two 32 W Sylvania E2y
Octron fluorescent tubes. This lighting configuration provides
a color rendering index (CRI) of 82 and a color temperature of
4100 Kelvin. Drug dilutions: 3.6339 g of GHB sodium salt
yields 3.0 g of ~,-hydroxybutyric free acid, which, dissolved in an
8-oz drink, will yield a 115.3 mmol/L concentration (the con-
centration used for testing the Drink Safe Technologies product
both in the laboratory and also the resulting concentration
after consumers added a standard 1000-1aL aliquot of their
beverage to a vial if it happened to contain a concentrated
GHB solution) and 1.1536 g of ketamine hydrochloride yields
I g of ketamine, and a 16.8 mmol/L concentration equates to
1.0 g of the free base in an 8-oz drink (the concentration used
for testing the Drink Safe Technologies product, both in the
laboratory and as the resulting concentration after consumers
added a standard 1000-~tL aliquot of their beverage to a vial that
happened to contain a concentrated ketamine solution). A
133 mmol/L 1,4-butanedioi concentration, a 127 mmol/L
y-butyrolactone concentration, and a 0.1 mmol/L flunitraze-
pam concentration were used in the laboratory to determine
whether those agents reacted with either of the Drink Safe
Technology Version 1.2 chromogenic reagents. Diagnostic
efficacy was calculated using conventional two-way (Bayesian)
analysis. Relative resolution times were compared by one-way
analysis of variance. Product evaluation in field settings al-
lowed browsing participants to pipette an aliquot of their own
drink into any of 101 coded vials containing authentic drug
standards or pure water so as to yield the same concentra-
tions of ?-hydroxybutyric acid or ketamine specified in the
manufacturer-defined protocol, or blanks. Consumers were
advised to proceed according to the directions printed on the
product and to record their results on a card with a code sticker
corresponding to the vial to which they had added an aliquot of
their beverage. Diagnostic efficacy was calculated using a two-
way analysis.
Results
In the laboratory, Drink Safe Technology Version 1.2 reliably
detected the concentrations of ~,-hydroxybutyrate and ketamine
specified by the protocol proposed by the manufacturer. The
product reliably excluded pure water samples. The volume
spotted on the test area influenced drying time, which affected
resolution time. Spotting 25-IJL samples, the reactive color
change from green to blue, denoting a positive test for ,/-hy-
droxybutyrate, resolved in 21 • 13 s; but a definitive color
change from pink to blue to denote a positive test for ketamine
required substantially more time to resolve (95% CI: 5.5, 6.5
rain). Resolution times are somewhat subjective. Spotting
larger volumes or a lack of familiarity for recognizing a positive
or negative reaction are likely to delay a confident test deter-
mination (Figure 1, see page 11A). There were no significant
differences in resolution times or accuracy between lots. About
12,000 of the Drink Safe Technologies customized coasters
were purchased for the
"It's
All About Awareness" campaign. A
stack of 250 coasters was removed for use in this project from
the partial supply remaining. We found that 0.48% of the stack
was defective, having no reagent present on any of the four
spots of the two test areas (Figure 2, see page 11A). Excluding
defective coasters, the test efficiency in the laboratory fol-
lowing the manufacturer's protocol was 100% (Table I). Al-
though 1,4-butanediol and ~,-butyrolactone are biotransformed
to GHB following ingestion, neither agent at the tested con-
centrations reacted with either of the Drink Safe Technology
Version 1.2 test reagent spots to elicit a blue color change, nor
did the test concentration of flunitrazepam result in a positive
color change.
Of 101 paired field test results recorded by consumers, there
were three instances in which paired tests of the same drink did
not concur; this was perhaps due to uneven carryover of
reagent between spots in one but not both test areas, but more
likely due to applying different volumes of the drink sample to
the two test areas, such that the areas did not resolve at the
same rate. Noting that all five pairs of false-positive responses
were associated with Bombay Sapphire London Dry Gin, the
355

potential for that spirit to react with the GHB reagent spot was
confirmed in the laboratory to elicit a distinct but transient
bluish color change. The proportion of vials that actually con-
tained ketamine or GHB (i.e., prevalence) was 40.2%. On the
basis of theoretical results in relation to results recorded by
consumers, the test efficiency was 65.1%, sensitivity 50%,
specificity 91.6%, and Likelihood Ratios: (+) 5.950, (-) 0.546
(Table II).
Discussion
The product performed substantially better in the laboratory
than it did in the hands of consumers in the field. There is con-
siderable potential for consumers to misinterpret a test re-
Table I. Performance Characteristics of Drink Safe
Technologies Version 1.2 Test Strips and Coasters in the
Laboratory following the Protocol Specified by the
Manufacturer*
+
+ 146
0
Sensitivity = TP/(TP + FN)
Specificity = TN/(TN = FP)
Likelihood Ratio (+) = Sens/(1 - Spec)
Likelihood Ratio (-) = (1 - Sens)/Spec
Efficiency = (TP + TN)/(TP + TN + FP + FN)
0
146
95%CI
0.997 0.968,1.000
0.997 0.968,1.000
100%
* Pooled results from ~our different UPC lots comprising 146 units, 2 tests per unit,
1 card or coaster from each lot provided a wet and a dry control.
Table II. Performance Characteristics of Drink Safe
Technologies Version 1.2 Custom Coasters in the Hands
of Consumers in the Field*
+
20 5
20 56
Specificity = TN/(TN = FP) 0.918
Likelihood Ratio (+) = Sens (1 - Spec) 6.100
Likelihood Ratio (-) = (1 - Sens)/Spec 0.545
Efficiency = (TP + TN)/(TP + TN + FP + FN) 0.651
95% Cl
0.822, 0.964
0.553, 0.748
* Pooled results from four different UPC lots comprising 146 units, 2 tests per unit,
I card or coaster from each lot provided a wet and a dry control.
356
Journal of Analytical Toxicology, Vol. 31, July/August 2007
sult. This field test supported what Meyers and Almirall had
predicted: that the time it takes for the color reaction to occur
could hinder the proper interpretation of the test (18). They an-
ticipated that if the reaction is expected within a few minutes,
a person might prematurely interpret the test to be negative
when, in reality, a positive reaction had not yet resolved. Com-
pared to laboratory conditions, the lighting available to read a
test strip in a typical nightclub would be less optimal. Lack of
experience or visual reference to aid interpretation of a positive
color change may further compromise the performance of
Drink Safe Technology Version 1.2 in the hands of consumers.
There were few false positives. The consequence of a false pos-
itive is to compel disposal of an unadulterated beverage. A
false positive could also trigger false accusations. The greater
probability for a largely university student population to falsely
conclude a negative test on a spiked drink is a safety concern
for those who would rely on this product to reduce their risk.
Implications for Clinical Practice
At best, the Drink Safe Technology Version 1.2 product is ca-
pable of detecting only 2 of the over 50 parent drugs listed by
the Society of Forensic Toxicologists (SOFT) Drug-Facilitated
Sexual Assault Committee website (19). Perhaps no technology
will protect those who venture into nightclubs and private
parties better than being aware of the risk, being selective
about the places they party and the people with whom they
party. There is safety in the company of conscientious friends
who know what to watch for and can be trusted to watch out
for each other. A supplementary role for the Drink Safe Tech-
nology Version 1.2 test kit to enhance personal safety is ques-
tionable.
Acknowledgments
The authors gratefully acknowledge the enthusiastic support
of the University of Saskatchewan Students' Union (USSU),
the USSU Women's Centre, and the management and staff of
the USSU owned and operated campus nightclub, Louis', for or-
ganizing and promoting the
"It's
All About Awareness" cam-
paign events at which the field-testing was conducted; Dr.
Mikelis Bickis, Associate Professor, Department of Mathematics
& Statistics, provided expert advice that improved the reporting
and presentation of the results; and Karen Leedahl, P.Eng,
Electrical Engineer, described our lighting design specifica-
tions for us. Her support was valuable and much appreciated.
References
1. Society of Forensic Toxicologists. Drug-Facilitated Sexual Assault
Committee. http://www.soft-tox.org/docs/2005 %20Survey
%20of%20Laboratories%20doing%20DFSA%20analysis.pdf,
February 2007.

Journal of Analytical Toxicology, Vol. 31, July/August 2007
2. IAPC National Law Enforcement Policy Center. Investigating
Sexual Assaults: Concepts and Issues Paper. http://www.
theiacp.org/documents/pdfs/RC D/InvestigatingSexual
AssaultsPaper.pdf, rev. July 2005.
3. N. Fitzgerald and K.J. Riley. Drug-facilitated rape: looking for
the missing pieces.
Nat. Inst. Justice J.
243:8-15 (2000)
(http://www.ncjrs.gov/pdffiles1/jr000243c.pdf).
4. M.J. McGregor, J. Erickson, L.A. Ronald, P.A. Jannssen, A. Van
Vliet, and M. Schulzer. Rising incidence of hospital-reported
drug-facilitated sexual assault in a large urban community in
Canada: a retrospective population-based study.
Can. J. Publ.
Health
95(6): 441-445 (2004).
5. Statistics Canada. Statistical Trends 2006: Measuring Violence
Against Women. http://www.statcan.ca/english/research/
85-570-XlE/85-570XlE2006001 .pdf, December 2006.
6. Northern Territory Government, Department of Health and Com-
munity Services, Women's Health Strategy Unit. Drug-Facilitated
Sexual Assault in Darwin Urban, April 2004.
7. L. Ledray. Date rape drug alert.
J. Emerg. Nurs.
22(1): 80 (1996).
8. J. Gauntlett-Gilbert, J. Keegan, and J. Petrak. Drug-facilitated
sexual assault: cognitive approaches to treating the trauma.
Behav.
Cognit. Psychother.
32:215-223 (2004).
9. L. Slobodian. Spiked drink changed her life forever.
Calgary
Herald.
September 18, 2005, p B1.
10. J. van Rassel. Drug rapes documented.
Calgary Herald.
September
7, 2003, p C3.
11. M. Roberts. Few women using new date rape drug test.
The
Province.
January 18, 2004, p A20.
12. M. Roberts. Raped by a 'complete coward'.
The Province.
November 21,2004, p B3.
13. H. Hoy. Victims of date rape silenced by shame, attacks by those
they trust leave scars.
Toronto Star.
October 29, 2005, p L01.
14. P. Sturman.
Drug Assisted Sexual Assault: A Study for the Home
Office under The Police Research Scheme.
Home Office, London
Metropolitan Police, 2004.
15. Glass aims to prevent spiking of drinks with date-rape drug.
Na-
tional Post,
October 26, 2005, p A5.
16. Two fighting the date-rape scourge. 'Billglass' filled at base, sealed
by the bartender.
The Province.
December 7, 2005, p C33.
17. Drink Safe Technology, Inc. The Official Drink Safe Technology
Website, http://www.drinksafetech.com.au/pages/usefulinfo/foren-
sics.html, December 2006.
18. J.E. Meyers, and J.R. Almirall. A study of the effectiveness of com-
mercially available drink coasters for the detection of "date rape"
drugs in beverages.
J. Anal. Toxicol.
28(8): 685-688 (2004).
19. Society of Forensic Toxicologists (SOFT) Drug-Facilitated Sexual
Assult (DFSA) Committee. Recommended Maximum Detection
Limits for Common DFSA Drugs and Metabolites in Urine Sam-
ples. http://www.soft-tox.org/docs/SOFT%20D FSA%20List.pdf,
February 2007.
Manuscript received January 24, 2007;
revision received February 8, 2007.
357