# Lesson 6: Stimulants, Caffeine, Tobacco and Vaping

*Companion-podcast transcript • Sarah & Kiffer*  
*Office Hours episode to listen to after working through the lesson*

---

**Sarah:** Welcome back to Office Hours. I'm Sarah.

**Kiffer:** And I'm Kiffer. With Lesson six, on stimulants, caffeine, tobacco and vaping, it makes sense to begin with the stimulants in the BC Coroners Service's toxicology figures, because they're easy to misread.

**Sarah:** Question one, and a short note first, because it's about deaths from the unregulated drug supply. Support is available through the nine eight eight Suicide Crisis Helpline, which you can call or text; the National Overdose Response Service, at one, eight eight eight, six eight eight, six six seven seven; and, for Indigenous people, the Hope for Wellness Help Line, at one, eight five five, two four two, three three one zero. Here's the question. Stimulants are now found in most toxic drug deaths in British Columbia. Does that make this a stimulant crisis?

**Kiffer:** Let's start with the newest numbers. In a release dated October 14, 2026, the BC Coroners Service reported preliminary toxicology for deaths so far in 2026 that underwent expedited testing. Fluorofentanyl, a fentanyl analogue from Lesson five, was detected in sixty-eight percent. Cocaine was detected in fifty-six percent, methamphetamine in fifty-three percent, and fentanyl itself in fifty percent.

**Sarah:** So someone could write a headline saying cocaine has overtaken fentanyl.

**Kiffer:** They could, and it would skip the top line. Fluorofentanyl is a fentanyl-type opioid, and on its own it was in sixty-eight percent of these deaths. So at least sixty-eight percent involved a fentanyl-type opioid, and possibly more, if some deaths had fentanyl without fluorofentanyl. Comparing cocaine with one opioid on the list makes the opioids look smaller than they are.

**Sarah:** Here's a question for everyone listening. Cocaine was detected in fifty-six percent of these deaths and methamphetamine in fifty-three percent. What's the smallest share of deaths that could have had both drugs detected? Pause the audio if you'd like more time. Take a few seconds.

*(Pause)*

**Kiffer:** Add the two and you get one hundred nine percent, which is more than everyone. Two groups can only fit inside one hundred percent if they overlap, and here the overlap has to be at least nine percent. So at least nine percent of these deaths had both cocaine and methamphetamine detected. The real overlap could be much larger, and these two numbers alone can't tell you how large.

**Sarah:** And the opposite trap is adding them and saying stimulants were in one hundred nine percent of deaths.

**Kiffer:** Right. Every row has the same denominator, deaths that underwent expedited testing, and one death can appear in several rows. From these figures, the share with cocaine or methamphetamine is somewhere between fifty-six and one hundred percent. For a single month the coroners have reported it directly: in October 2025, more than ninety percent of suspected unregulated drug deaths with expedited testing had a stimulant detected.

**Sarah:** So, back to the question. Is this a stimulant crisis?

**Kiffer:** Detection tells you a drug was present at death. It can't tell you which drug caused the death, how much was taken, or whether the person meant to take it, since the supply is unpredictable. Nationally, the Public Health Agency of Canada reported in September 2026 that sixty-two percent of apparent opioid toxicity deaths from January to March 2026 also involved a stimulant. So I'd describe it as a crisis of a toxic, unpredictable supply in which opioids and stimulants are often found together. Calling it only an opioid crisis hides the stimulants, and calling it a stimulant crisis would hide the opioids that are still in most of these deaths.

**Sarah:** Why does the stimulant side matter for responses, though, if the opioid is often what stops someone's breathing?

**Kiffer:** Two reasons from the pharmacology. Stimulants carry their own risks to the heart, the blood vessels and the mind, so a person using both faces several kinds of strain at once. And the route has shifted. In preliminary data released in December 2025, the coroners reported smoking as the most common mode of consumption among 2025 deaths, at sixty-five percent, with injection at ten percent. Services built around injection and opioids alone will miss much of that pattern, which HSCI 402 takes up.

**Sarah:** Question two is one I want to try myself. In McKetin and colleagues' 2013 study of two hundred seventy-eight people with methamphetamine dependence, months with sixteen or more days of use had an odds ratio of 11.2 for psychotic symptoms, compared with months of no use. What does that mean for a person's chances?

**Kiffer:** To answer that, you need a starting chance for months without use, so let's make one up. Suppose, hypothetically, that in months without use, ten percent of participants had psychotic symptoms. Using the odds ratio of 11.2, what's the chance of symptoms in a heavy-use month?

**Sarah:** Okay. An odds ratio of 11.2 means about eleven times as likely. So ten percent times 11.2 is one hundred twelve percent.

**Kiffer:** What does your sanity check say about that?

**Sarah:** That it's impossible. Nobody has a one hundred twelve percent chance of anything. So my mistake was multiplying the probability. An odds ratio multiplies the odds.

**Kiffer:** Right. Odds are the chance something happens divided by the chance it doesn't, and you turn odds back into a chance by dividing the odds by one plus the odds. So here's the question for everyone. Convert the ten percent into odds, multiply by 11.2, and convert back. What's the chance of symptoms in a heavy-use month? Take a few seconds.

*(Pause)*

**Sarah:** Ten percent divided by ninety percent is about 0.111. Times 11.2 gives about 1.24. Then 1.24 divided by one plus 1.24, which is 2.24, comes to about 0.55. So about fifty-five percent.

**Kiffer:** That's right for our made-up starting point. Fifty-five percent is about five and a half times ten percent, so "eleven times as likely" would have doubled the effect. When an outcome is common, the odds ratio runs well ahead of the ratio of probabilities. If the starting chance were one percent, the same odds ratio would give about ten percent, which is close to eleven times, because odds and probabilities are nearly equal for rare outcomes.

**Sarah:** Now the bigger question. Does this study show that methamphetamine causes psychosis?

**Kiffer:** It's one of the stronger observational designs you'll meet. Each person is compared with themselves, so stable things such as genes or childhood experiences can't explain the difference between months. And the dose-response pattern points the same way: odds about four times as high with one to fifteen days of use, and 11.2 times as high with sixteen or more.

**Sarah:** But a within-person comparison only removes the things that stay the same. Sleep changes from month to month, and so do stress and the use of other drugs.

**Kiffer:** Yes, and lost sleep during heavy-use months is a real alternative explanation. The order of events within a month is also hard to pin down. If early symptoms, or the distress around them, led someone to use more that month, the association would look much the same. Then there's who was studied: people with methamphetamine dependence in Sydney and Brisbane, with no lifetime diagnosis of schizophrenia or mania. The study says little about occasional use.

**Sarah:** So the fair summary is that heavy use is followed by a steep, dose-related rise in psychotic symptoms in the same people, the evidence points to the drug contributing, and things that travel with heavy use, like lost sleep, may contribute too.

**Kiffer:** That's a careful summary, and it's the same reasoning you'll want for cannabis and psychosis in Lesson seven. It also helps to remember that these symptoms usually ease when a person reduces or stops use, although they persist or return in a minority.

**Sarah:** Question three. Nicotine is the reason people keep smoking, so isn't nicotine what makes cigarettes deadly?

**Kiffer:** That's a common belief about smoking, and it gets the roles mixed up. In a 2010 review, Benowitz put it in two sentences: smoking-caused disease comes from exposure to toxins in tobacco smoke, nicotine plays a minor role, if any, in causing those diseases, and addiction to nicotine is the proximate cause. Nicotine keeps the exposure going for decades, and combustion is what makes that exposure deadly.

**Sarah:** And the scale is hard to take in. Health Canada, citing the Canadian Centre on Substance Use and Addiction, attributed forty-six thousand three hundred sixty-six deaths in Canada to tobacco in 2020, nearly two in three deaths attributed to substance use.

**Kiffer:** Which is why the patch and the cigarette make such a useful pair. Both deliver the same molecule. A patch releases nicotine slowly through the skin and has low potential for problematic use. A cigarette gets nicotine to the brain in ten to twenty seconds, many times a day. It's the same principle that separates lisdexamfetamine taken by mouth from smoked methamphetamine: speed and form shape the risk as much as the molecule does.

**Sarah:** Here's something I've never quite pictured. Nicotine builds up over six to eight hours of regular smoking, yet its half-life is only about two hours. How can both be true?

**Kiffer:** Let's work it with made-up numbers, using the same half-life arithmetic as the caffeine calculation. Imagine someone who smokes one cigarette every hour, and say each cigarette adds one unit of nicotine to the blood. With a two-hour half-life, how much of a unit is left an hour later?

**Sarah:** One hour is half a half-life, so it's one half raised to the power of one half. That's about 0.71. So just after the second cigarette, there's 0.71 left from the first plus one new unit, which makes 1.71. After the third, 1.71 times 0.71, plus one, is about 2.21.

**Kiffer:** Right. Keep going and you get about 2.56, then 2.81, 2.99, 3.11 and 3.2 after the eighth cigarette. What do you notice about those steps?

**Sarah:** They keep climbing, but by less each hour. The first step added 0.71, and the last one added less than a tenth.

**Kiffer:** It levels off at about 3.4 units, where the amount cleared in an hour equals the one unit each cigarette adds. After eight cigarettes, the level is already at about ninety-four percent of that plateau. That's the build-up over six to eight hours.

**Sarah:** And in this picture, a person whose liver clears nicotine more slowly keeps more after each hour, so fewer cigarettes hold the same level. That fits the finding that people with slower forms of the liver enzyme CYP2A6 tend to smoke fewer cigarettes a day.

**Kiffer:** It does. Now the overnight part, and this one's for everyone. Say the last cigarette is at eleven at night, with the level at the plateau. With a two-hour half-life, what fraction of that level is left at seven the next morning? Take a few seconds.

*(Pause)*

**Sarah:** Eight hours is four half-lives, so the level halves four times: a half, a quarter, an eighth, a sixteenth. So there's still some nicotine in the body through the night, but by morning only about six percent of the evening level remains, about 0.21 units, which is a fifth of what a single cigarette adds.

**Kiffer:** And overnight is when the desensitized receptors recover, so the person wakes into craving and withdrawal that the first cigarette relieves. That's the daily cycle in one calculation. The short half-life lets the level fall a long way overnight, and frequent dosing keeps it high all day. Real smoking is messier, because people adjust each puff, liver enzymes vary from person to person, and nicotine stored in body tissues is released slowly, which keeps real overnight levels somewhat higher than this simple model, but the shape holds.

**Sarah:** Question four. In the Cochrane living review published in November 2025, nicotine e-cigarettes had a risk ratio of 1.55 for quitting compared with nicotine replacement therapy, with high-certainty evidence. What does that look like in people?

**Kiffer:** The review gives us a starting point. Its summary assumes that about six in every hundred people using nicotine replacement therapy, with limited support, have stopped smoking at six months or later. So here's the question for everyone. Using the risk ratio of 1.55, how many in a hundred would you expect to stop with nicotine e-cigarettes? Take a few seconds.

*(Pause)*

**Sarah:** Six times 1.55 is 9.3, so about nine in a hundred. That's about three more people per hundred, which matches the review's own estimate of three additional quitters per hundred. It also means that about ninety-one of every hundred people given e-cigarettes would still not have quit at six months. And 3.3 extra quitters per hundred works out to about one extra for every thirty people, so roughly thirty people would need to use e-cigarettes, compared with nicotine replacement therapy, for one extra person to stop smoking.

**Kiffer:** That's all correct, and the review's plain-language summary gives a range of eight to eleven in a hundred for nicotine e-cigarettes, compared with six in a hundred for nicotine replacement therapy.

**Sarah:** So it's a real effect, and a modest one for each person.

**Kiffer:** Modest for each person, and possibly large across a population, because about twelve percent of people in Canada aged twelve and older smoked in 2022. The speed of the change is worth noticing too. In 2018, the National Academies found only limited evidence that e-cigarettes help people stop smoking, and too little to compare them with approved treatments. By 2025, the comparison with nicotine replacement therapy was rated high certainty.

**Sarah:** Which brings us to where we disagree. I'm more cautious than Kiffer about tobacco harm reduction, and my reason is young people. Using Hammond and colleagues' revised estimates, past-thirty-day vaping among Canadian sixteen to nineteen year olds rose from 8.4 percent in 2017 to 12.1 percent in 2018 and 17.8 percent in 2019. And in the Canadian Student Alcohol and Drugs Survey for 2023 to 2024, fifteen percent of students in grades seven to twelve in nine provinces had vaped in the past thirty days, compared with four percent who had smoked.

**Kiffer:** Those figures matter, and it's worth knowing why the 2018 figure is 12.1. The first published estimate was 14.6 percent, and the authors later reweighted their online surveys so that their smoking trends matched national surveillance. Survey estimates can move when the weighting changes, so check which version you're quoting. Either way, youth vaping in Canada rose.

**Sarah:** For a teenager who has never smoked, the comparison is with no nicotine at all. Then vaping adds dependence, the National Academies found substantial evidence that vaping increases the risk of ever smoking cigarettes among young people, and the long-term risks of vaping are still unknown. In June 2024, Health Canada described youth vaping as having levelled off since 2021 while remaining high, with flavours a major contributor to the earlier rise.

**Kiffer:** Where I'd start is the adult who already smokes. For that person, the comparison is with continuing to smoke, and that's where the evidence is strongest. The National Academies rated it conclusive that switching completely reduces exposure to numerous toxicants and carcinogens, and the cessation trials are now high certainty. With tobacco responsible for more than forty-six thousand deaths in Canada in 2020, even modest gains in quitting matter.

**Sarah:** And the youth evidence?

**Kiffer:** I take it seriously, and I'd also name what kind of evidence it is. Young people can't be randomly assigned to vape, so the link with later smoking comes from observational studies, and the same things that lead a teenager to try vaping may also lead them to smoke. Youth smoking also hasn't surged. In Hammond and colleagues' surveys, smoking among Canadian sixteen to nineteen year olds has declined very slowly since 2017.

**Sarah:** A slow decline during years of high youth vaping can't tell us what would have happened without vaping, though. Smoking might have fallen faster.

**Kiffer:** That's fair, and it cuts both ways. The trend can't prove harm, and it can't rule harm out. So here's where I land. I'd keep strong youth protections on flavours, marketing and sales, and I'd want adults who smoke to hear the message Health Canada already gives on its website: switching completely to vaping is less harmful than continuing to smoke, and it isn't harmless.

**Sarah:** I'd lean further toward the medical route. In January 2025, the Council of Chief Medical Officers of Health said that approved smoking cessation methods do not at present include nicotine vaping products, encouraged manufacturers to seek approval, and recommended banning flavoured vaping products. I find that persuasive. I'd like e-cigarettes to go through approval as quit aids before anyone presents them that way, with flavour restrictions kept tight.

**Kiffer:** On nicotine pouches, I think we're closer. Health Canada authorized its first pouch in 2023, for adults trying to stop smoking, and as nicotine replacement therapy it was at first sold with no age restriction. In 2024, the federal health minister said the pouches were reaching young people, and from August 28, 2024, they could be sold only by a pharmacist or under a pharmacist's supervision, from behind the counter, in mint or menthol flavours. A 2025 review of seven small trials found that pouches may cut the number of cigarettes people smoke, but none of the trials showed a clear increase in quitting. A 2026 review judged pouches considerably less toxic than cigarettes, and it also reported that pouch use by people who smoke daily more often ends in dual use than in stopping. With the adult benefit unproven and reports that young people were using them, I think the pharmacy-only rule is easy to defend while better trials are done.

**Sarah:** I'm still more cautious than Kiffer about vaping, but we agree on the practical rule. Before you weigh any claim in this debate, ask whose risk it's about, compared with what, and how certain the evidence is. Then say the answers out loud when you make the argument, because "less harmful than smoking" and "harmful compared with never using nicotine" can both be true of the same product.

**Kiffer:** Let's pull it together with three things to take away.

**Sarah:** First, a toxicology percentage tells you a drug was present at death. Rows that share a denominator can overlap, so they can't be added, and none of them shows cause or intent.

**Kiffer:** Second, an odds ratio multiplies odds. When an outcome is common, convert to a probability before you say how much more likely something is, and check that your answer stays below one hundred percent.

**Sarah:** And third, name the population and the comparison. Switching completely is less harmful compared with continuing to smoke, and starting adds risk compared with never using nicotine.

**Kiffer:** If you'd like more practice, rework the methamphetamine odds ratio problem from this episode with a starting chance of twenty percent, once with the odds ratio of 11.2 and once with 4.0, and convert through odds both times, because with 4.0 the shortcut of multiplying the chance gives an answer under one hundred percent that's still wrong.

**Sarah:** Next time, it's Lesson seven, Cannabis, Psychedelics and New Psychoactive Substances, where the same pharmacology meets the endocannabinoid system and the early warning systems that track new drugs.

**Kiffer:** Take care, everyone.

**Sarah:** See you in Lesson seven.
