Background

This component of the monitoring project is targeting Emergency Department (ED) attendees between 17 and 75 years of age, who present to an ED for treatment late at night and in the early hours of the morning on weekends (i.e. Friday and Saturday from 9 p.m. to 4 a.m.). A systematic sampling strategy is used to select research participants from all patients presenting to the ED between these times. The sampling strategy has been developed from previous work, including Brubacher et al’s ED adverse drug reaction study (Hohl et al, 2007), Stockwell et al’s Australia alcohol-related ED visits study (McLeod et al, 2000; Stockwell et al, 2002) and Cherpitel et al’s number of US and international studies concerning alcohol use and injury (2005). This strategy has been designed to maximize the response rate, and is being tested in practice.

This project has ethical approval from the University of British Columbia (UBC) and the Vancouver Island Health Authority (VIHA) jointly with the University of Victoria (UVIC).

Pilot Study

The pilot study was conducted on 97 patients early in 2008 to test the survey instruments and protocol. As a result, improvements were made to the sampling strategy, the survey instruments, the study protocol and the consent form.

Locations and Partnerships

Two locations are used for this study – Vancouver General Hospital (VGH) in Vancouver and Royal Jubilee Hospital (RJH) in Victoria. VGH is the largest hospital in British Columbia and the major tertiary referral centre for the province. Services in all adult areas of medicine except obstetrics are provided at this site. RJH is one of two tertiary hospitals serving Vancouver Island (Victoria General Hospital is the other). These hospitals operate as one large facility across two sites, providing high-level trauma care and specialized services.

Securing these locations has been made possible by having direct collaboration with key ED medical staff at each site. Drs. Andrew MacPherson at RJH and Jeff Brubacher at VGH have been instrumental in moving this project forward. Other possible sites for expansion include St. Paul’s Hospital in Vancouver and Victoria General Hospital as both sites have shown interest.

Survey Instruments

The ED AOD Monitoring Survey was developed from similar surveys developed by the monitoring team in Australia (Stockwell et al, 2002) and Canada (Macdonald et al, 2005) which have also been incorporated within the ERCAAP dataset (Cherpitel et al, 2003). It was also designed to be comparable to the other survey instruments in the monitoring project in order to contribute to efforts to characterize substance use patterns and related harms among high risk populations in the participating sites.

The survey consists of questions pertaining to the reason for visit, the participant’s drug use history (lifetime use, past 12 months, one month,one week, and yesterday use), including specific alcohol use, cannabis use, and injection drug use. There are questions pertaining to drug-related harms and to alcohol or drug use in the six hours prior to onset of symptoms/injury, such as the AUDIT and ASSIST instruments listed below.

The Alcohol Use Disorders Identification Test (AUDIT) is a brief screening instrument developed by the World Health Organization for identifying hazardous and harmful patterns of alcohol consumption. The Emergency Department (ED) study uses AUDIT scores as a measure of risky alcohol use among ED attendees. Low AUDIT scores (1-7) are associated with low-risk consumption of alcohol. Moderate scores (8-15) indicate alcohol use in excess of low-risk guidelines and moderate-risk of harm, while high scores (16+) are indicative of harmful and hazardous drinking (i.e. high-risk). Extreme scores on the AUDIT (20+) are particularly indicative of dependence.

Risk levels for consumption of other substances were generated using The Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST), a brief screening questionnaire about use of psychoactive substances developed by the World Health Organization (WHO) and an international team of substance use researchers. Low-risk scores (1-3) suggest low risk of health and other problems from current pattern use, while moderate-risk scores (4-26) indicate an increased risk of health and other problems from current pattern of substance use. High-risk scores (27+) indicate that the individual is at risk of experiencing severe problems (health, social, financial, legal, relationship) and is likely to be dependent.

Finally, there are questions about the participant's demographic profile. Interviews take place at the patient's bedside for stretcher bound patients or in a private area for all others. Survey instruments were developed for Royal Jubilee Hospital and Vancouver General Hospital.

BAC Testing

In addition to the self-reported use of alcohol, blood alcohol concentration (BAC) is estimated indirectly by measuring the amount of alcohol in one’s breath. The breath test used is either the Alco-Sensor IV - Blue Dot or the Alco-Sensor IV FST breathalyzer. These instruments are calibrated before each shift, using a water-based alcohol solution of 0.1%, to ensure accuracy of readings. For the breathalyzer test, a subject blows into a sterile disposable mouthpiece for 5 to 10 seconds and the machine automatically estimates the BAC.

These units have been chosen for the project for their ease of use, portability, and unobtrusiveness. Similar devices are used by law enforcement for roadside breath-testing, as well as in various other markets such as workplace testing, emergency departments, occupational health centres, and drug and alcohol treatment centres. Previous emergency department studies (e.g. Macleod et al, 2000) have confirmed that BAC tests correlate well with self-reported alcohol consumption, especially when the delay between the last drink and a breath test is taken into account. Delays of longer than two hours, however, will generate increasing numbers of false negative results when BAC data alone is relied on. Published formulae are available to calculate the BAC predicted from a given amount of alcohol consumption, making allowance for the delay between the last drink and a breath test (Blaze-Temple et al, 1988). On balance, most studies find self-reported prior alcohol consumption is the most predictive measure, and that BAC readings are a means of confirming the reliability of the self-reported general terms (Stockwell et al, 2007).

Saliva Testing

A saliva/sweat drug test is administered, using a Securetec Detektions-Systeme AG Drugwipe5 five-drug sensing test strip. This is a self-contained testing strip which displays the presence of metabolites of drugs via the development of coloured lines in the strip's detection zone: amphetamine-like substances (including amphetamine, methamphetamine, ecstasy), cocaine, opiates (heroin and morphine), cannabis and benzodiazepines. A separate sterile Drugwipe is used for each consenting subject. If a saliva sample is difficult to take, the Drugwipe test can use a sweat sample.

Traditionally, urine testing has been the standard method of detecting the presence of commonly used illicit substances (cocaine, cannabis, opiates/opioids, benzodiazepines, barbiturates, amphetamines and methamphetamine). However, based on previous research on methods of drug testing, oral fluid testing has been chosen for the ED monitoring project over the traditional method of urinalysis for a number of reasons. In particular, oral fluid testing is less invasive, is capable of producing results quickly, and is sensitive to recent substance use (Verstraete, 2004; Cone, 2006). When compared to urinalysis, oral fluid testing has proven to be as accurate, with opiates having sensitivity between 91-99%, and 98% for cocaine, and 86% for methamphetamine (Barrett et al., 2001; Bennet, 2003; Wish & Yacoubian, 2002). While drugs can be detected in urine and hair for weeks (in salivary glands, the detection window is about 12 to 24 hours), the present study is most concerned with recent rather than historical use (Verstraete, 2004). Oral fluid testing detects drug use within hours, and as such is a better indicator than urinalysis of recent drug use, and thus more applicable to the ED monitoring project.

Furthermore, it was decided to use disposable saliva strips, a point-of-collection device (POC), rather than a laboratory device which requires samples to be sent for analysis in laboratories (which normally takes 24 to 72 hours). POC devices fit the objectives of this component as they are inexpensive, and give instant results, with good sensitivity and specificity (Yacoubian & Wish, 2006). As well, there are few laboratories in Canada that perform oral fluid analysis.

Collection of Demographic Data

In addition to the survey, breathalyzer and saliva test results, interviewers record basic demographic (age, gender, occupation, postal code, city of address) and clinical (chief complaint, disposition, final diagnosis, results, if available, of serum alcohol level or urine drug testing performed for clinical purposes) information from the Emergency Department Information System (EDIS). The format and response options of these questions have been made to correspond as closely as possible to what is planned for the Canadian Alcohol and Other Drug Use Monitoring Survey for future comparative analyses. For more information, see demographic tables for Vancouver, Victoria and both sites.

<< Overview