Good morning. Thank you very much to the committee for having me here today.
I'm Stacey Hannem. I'm an associate professor of criminology at Wilfrid Laurier University at the Brantford campus.
I've been thinking about, writing about, and talking about ion scanners since 2005, when I was working on my doctoral research. My doctoral research was intended to focus on the difficulties in issues faced by families who have an incarcerated loved one. I expected very much to find issues of stigma, financial issues, and emotional distress. What I did not expect to find was that our own correctional institutions were so drastically compounding these issues for families.
The IONSCAN technology came up in my research first in 2005. In 2015, I worked with the Canadian Families and Corrections Network to do an online survey and follow-up interviews with families of incarcerated people in Canada about their mental health issues. Once again, the ion scanner, 10 years later, showed up in my data as a still significant issue.
Because of this issue of false positives and the questions being raised by families, I started to look in more detail into the technology itself. I am not a scientist—my Ph.D. is in sociology—but I spent quite a bit of time and effort in studying the scientific literature and consulting with colleagues in physics and chemistry to better understand it and to give myself a layperson's understanding of the technology. That is what I would like to offer the committee today: a better understanding of how the technology works and why it is subject to false positives.
The science underlying the IONSCAN technology first emerged in the 1970s and has not been significantly refined since that time. It was originally devised as a means of detecting trace particles of explosives. The newer machines have since been calibrated to look for drugs. Every member of this committee will be familiar with the IONSCAN machine. It is the very same machine that is used in airports across this country and internationally to scan for explosives, but in Canadian prisons they're used to look for drugs.
The principle of ion mobility spectrometry is quite simple. The sample is swabbed, as Anne has said, from a zipper, a ring, or your glasses, and the trace particles collected on the swab are then put into the machine and ionized: they are subjected to a radioactive source that will create either a positively or a negatively charged ion. The ionization process breaks down that substance into all of its constituent particles. The ions that belong inside the molecule are of varying sizes and weights, if you remember your high school chemistry, and all of these ionized particles will travel at different speeds.
The ionized particles are put into the machine. They travel through what's called a “drift tube”. This is literally a short tube that has an electromagnetic plate at the far end. The IMS machine measures the amount of time, down to the nanosecond, that these various particles take to reach the far end of the drift tube. The drift tube has an inert gas inside that is designed to create friction and to collide with the particles. All of the various sizes of the particles will reach the drift tube at varying times, depending on how big or heavy they are.
The device measures all of this and generates a drift spectrum that looks like a little graph, which shows the drift times of all the composite ions. It will compare that generated graph to the known drift spectrums of the substances it is searching for—in this case, illicit drugs. The IMS machine will alarm if the drift times of the ion mobility spectrum of the sample match one of the substances in its database with a sufficient threshold.
That's how it works. The issue of false positives, of course, is the reason why this was brought to Parliament. False positives were highlighted in 2008 in the United States, when the U.S. Federal Bureau of Prisons suspended the use of ion scanners in all of its prisons. There was a lawsuit alleging that a prisoner's family members had been unfairly denied visits due to false positive readings. When they resumed the use of the IONSCAN in 2009, the Federal Bureau of Prisons in the United States no longer authorized the machine for screening visitors, only for screening mail, work areas, and prisoner belongings.
CSC's own 2011 report by Johnson and Dastouri found that the devices are, quote, “oversensitive and are limited in their ability to detect certain forms of drugs”. The use of IONSCAN to identify people suspected of trafficking drugs is complicated by the fact that the technology itself has several known—although often unacknowledged—shortcomings.
Its first shortcoming is that, unlike human fingerprints, ion mobility spectrums that are generated and read by the IONSCAN machines are not unique identifiers; that is, the ion mobility spectrums of closely related or similar chemical compounds may be so nearly identical that the technology is unable to differentiate them. According to the original patent by Smiths Detection for the IONSCAN, Ranitidine, which is a common antacid often known as Zantac, tests positive for cocaine. The only way to identify this as a false positive is to run a second scan in negative ion mode, and in that case, Ranitidine will ring and cocaine will not.
Smiths Detection suggests in its patent that users should operate two machines, one in positive ion mode and one in negative ion mode, in order to control for this. To my knowledge, (a) CSC doesn't actually operate that way, and (b) it's a very simplistic kind of fix, because it assumes that Ranitidine is the only other substance that would ring positive for cocaine and it is not.
A number of legal pharmaceutical medications and other substances are known to alarm the IONSCAN machines because they have a similar chemical composition or share some chemical compound with illicit drugs. Correctional staff admit to keeping a short list of medications on other substances known to set off alarms: asthma inhalers, nitroglycerin for heart problems, and Adderall for ADHD will all alarm as methamphetamines; antifungal creams used to treat athlete's foot or thrush will alarm as an opiate, as will poppy seeds from bagels and other baked goods; and, chlorine baby wipes and some perfumes and lotions are prone to alarm as various types of drugs. Research by Dussy et al in 2008 found that several types of detergents will alarm as heroin.
This is the number one problem: these are not unique identifying markers. Therefore, this is quite prone to false positives.
Second, as Anne has already mentioned, there's also a high potential for cross-contamination and for individuals to inadvertently come into contact with trace amounts of drugs.
Sometimes an alarm is not chemically false. What the machine is picking up is actually a very small particle of illicit drugs, but that doesn't mean that the individual came into contact with that drug substance on purpose, nor that it's present in significant enough quantities to warrant concern about visiting a family member. Many banknotes contain micrograms of illicit drugs—cocaine being the most common—which is far more than the threshold required to alarm the IONSCAN. Some studies have found that up to 90% of banknotes will test positive for cocaine. Handling cash, as Anne has said, can cause people to alarm.
A third issue, of course, is the assumption that the people who alarm positive are attempting to traffic, and there's no evidence to suggest that.
Also, the IONSCAN is subject to the possibility of operator error. It needs to be very carefully cleaned in the event of a positive alarm. I witnessed this problem at an airport just recently. One positive alarm and then the failure to subsequently clean the machine effectively will result in subsequent positive alarms. If an alarm is activated in CSC, the operator is required to change their gloves, clean the countertop surfaces, and run clean swabs until the machine no longer alarms, but when the visiting room is busy, these things don't always happen.
In 2006, when CSC conducted an internal audit of its drug interdiction activities, they noted that in five of 11 audited institutions, the IMS device policy and procedures were not being followed. Operator error is a problem.
A last but kind of unlikely issue is the fact that the operator can alarm the machine at will. The device has to be calibrated on a regular basis with exposure to the subject substances, which means they have pieces of the drugs there to calibrate the machines. In her 2016 memoir, Diane Schoemperlen, a noted Canadian author, recounted that in her final visit to her now ex-partner at a medium security institution the correctional officer responsible for running the ion scanner set the machine off on purpose, as a joke. This raises questions about how it could possibly be used to target people who are already under suspicion.
CSC does not have any documentation about the reliability of the machine. They rely on the idea that the manufacturer says that the false positive rate is below 0.1%, but they continue to note that there are instances of false positives and they're quite aware of it.
I have more information on the reliability question, but I'm happy to answer on that during the question period.