HSCI 180 · Lesson 3

How Drugs Work: Routes, Pharmacokinetics and Pharmacodynamics

Drugs and Society

Learning objectives for this lesson:

  1. Describe the main routes of administration and explain how each shapes the onset, peak, duration, reinforcing effect and risks of a drug.
  2. Interpret British Columbia’s data on the mode of consumption in drug toxicity deaths, distinguishing a route’s share of deaths from its risk per use.
  3. Explain absorption, first-pass metabolism and distribution, and calculate the amount of a dose that reaches the circulation from its bioavailability.
  4. Explain how liver enzymes, genetic variation and drug interactions change drug levels, using codeine, caffeine and grapefruit as examples.
  5. Calculate the amount of a drug remaining from its half-life and explain why repeated doses approach a steady state after four to five half-lives.
  6. Describe how drugs act at receptors, distinguish agonists, partial agonists, antagonists and allosteric modulators, and distinguish potency from efficacy on a dose-response curve.
  7. Calculate a therapeutic index and explain the limits of therapeutic indexes and safety ratios, including the effects of tolerance and cross-tolerance.
  8. Explain how expectancy, placebo responses, conditioned tolerance and the risk environment shape drug effects and drug-related harm.
  9. Integrate route, pharmacokinetics, pharmacodynamics, set and setting to analyze a real Canadian pattern of drug-related deaths.

This course was developed by Dr. Kiffer G. Card.

Lesson 3 · HSCI 180

How Drugs Work

Routes, pharmacokinetics and pharmacodynamics: a short guided walkthrough before you work through the lesson at your own pace.

Drugs and Society
Why this lesson

A shared toolkit for Lessons 4 to 7

Every later lesson on a drug class applies the ideas introduced here.

  • The route of administration sets how fast a drug arrives.
  • Pharmacokinetics describes what the body does to a drug.
  • Pharmacodynamics describes what a drug does to the body.
  • Set and setting describe how the person and the place change the effect.
The road map

Four sections, one question: what does a dose do?

1 · Routes

Oral, inhaled, intranasal and injected routes, and why speed matters.

2 · Pharmacokinetics

Bioavailability, enzymes, genes, interactions and half-life.

3 · Pharmacodynamics

Receptors, dose and response, safety margins and tolerance.

4 · Drug, set and setting

Expectancy, conditioning and the risk environment.

A Canadian thread

British Columbia's 2025 drug toxicity data

65%
of deaths involved smoking as the mode of consumption
48%
of deaths occurred in a private residence

Preliminary data from the BC Coroners Service (2026). Each section opens with a content note and support lines.

Section 1 of 5

Routes of Administration: How a Drug Reaches the Brain

⏱ Estimated reading time: 40 minutes
Section 1 of 5

Routes of Administration

How a drug reaches the brain, and why the speed of delivery matters.

Four routes

Different paths to the brain

Oral

Stomach and intestine, then the liver, then the circulation.

Smoked or inhaled

Lungs, then straight to the left heart and the brain.

Intranasal

Lining of the nose, then veins, heart and lungs.

Injected

Directly into a vein, so the whole dose enters the blood.

One drug, many routes

Nicotine by route

10–20 s
cigarette: nicotine reaches the brain
3–12 h
patch: time to peak level
44%
oral capsule: bioavailability

Laboratory values compiled in a review by Benowitz et al. (2009).

Delayed onset

Slow routes invite repeated dosing

When the first dose seems not to work, a second dose can peak on top of it.

  • Eaten or drunk cannabis: effects can begin 30 minutes to 2 hours after use (Health Canada, 2024).
  • A review of lethal toxicity listed repeated dosing among the reasons that toxicity can exceed single-dose predictions (Gable, 2004).
Speed and reinforcement

Faster delivery, stronger reinforcement

  • Rapid onset and intermittent use both appear to push the addiction process forward (Allain et al., 2015).
  • Why speed matters in the brain is still debated (Samaha & Robinson, 2005).
  • Crack and powder cocaine have similar effects; speed, size and frequency of effect tracked the difference in harm (Hatsukami & Fischman, 1996).
British Columbia, 2025

Mode of consumption in drug toxicity deaths

65%
smoking
11%
nasal
9%
injection
4%
oral

BC Coroners Service (2026), preliminary; interview study by Bonn et al. (2025). A share of deaths is not a rate per use.

Carry forward

From the route to the bloodstream

  • The route sets the path to the brain and the speed of onset.
  • Speed shapes reinforcement as well as risk.
  • Shares of deaths must be read with denominators in mind.
  • Section 2 follows the drug through absorption, metabolism and elimination.

Content note

This section discusses deaths from the toxic unregulated drug supply in British Columbia. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service (1-888-688-6677) supports people who use drugs alone, and the Hope for Wellness Help Line (1-855-242-3310) serves Indigenous people across Canada.

Introduction

A cigarette and a nicotine patch contain the same drug, and so do a cannabis joint and a cannabis cookie. The people who use them describe very different experiences, and the risks of each product differ in ways that matter for health. Much of the difference lies in the route of administration, the way a drug enters the body: by swallowing, by inhaling smoke or vapour, by absorption through the lining of the nose, or by injection. Lesson 1 introduced psychoactive substances and the spectrum of use. This lesson introduces the shared pharmacology that Lessons 4 to 7 apply to each drug class.

Pharmacology, the study of how drugs interact with living systems, has two main parts. Pharmacokinetics describes what the body does to a drug: how it is absorbed, carried around the body, chemically changed and removed. Pharmacodynamics describes what the drug does to the body: where it acts and how its effects change with dose. Sections 2 and 3 develop these two parts, and Section 4 adds the person and the setting. This first section shows that the route shapes how fast a drug reaches the brain, how high its level climbs and how long it lasts.

Learning Objectives

  • 1.1 Describe the oral, smoked or inhaled, intranasal and injected routes of administration and trace the path that a drug takes from each point of entry to the brain.
  • 1.2 Compare the speed of onset, the peak and the duration of a drug's effect across routes, using published data on nicotine, cannabis and diazepam.
  • 1.3 Explain why rapid delivery of a drug to the brain is associated with stronger reinforcement and a greater potential for addiction.
  • 1.4 Describe how routes differ in their risks, including rapid peaks, repeated dosing when onset is delayed, and infection from shared equipment.
  • 2.1 Interpret British Columbia's data on the mode of consumption in drug toxicity deaths, including what those data can and cannot show about the risk of each route.

1.1 Four Main Routes into the Body

A psychoactive drug has to reach the brain to change mood, perception or behaviour, and almost every route gets it there through the bloodstream. Routes differ in which tissues the drug crosses first, whether it passes through the liver before reaching the general circulation, and how far the blood must travel before it is pumped to the brain. Figure 3.1 traces the simplified path for each of the four routes that this section compares.

Brain via arterial blood Oral Stomach and intestine Liver (first pass) Heart Smoked or inhaled Lungs (large surface area) Left side of the heart, then straight into arterial blood Intranasal Lining of the nose Veins Heart and lungs Injected (intravenous) Vein Heart and lungs, then arterial blood
Figure 3.1. Simplified paths from four routes of administration to the brain. A swallowed drug passes through the liver before it reaches the general circulation; an inhaled drug travels from the lungs to the left side of the heart and then into arterial blood; intranasal and intravenous drugs enter veins and return through the heart and lungs before reaching the brain. The diagram is a teaching simplification drawn from Benowitz et al. (2009), Del Pozo et al. (2026) and Susa et al. (2023).

The four tabs below describe each route with examples from published pharmacology and Canadian sources. The differences in speed that the tabs report follow from the differences in path that Figure 3.1 shows.

A swallowed drug is absorbed through the stomach and small intestine and carried by the portal vein to the liver before it reaches the rest of the body. Absorption is slow and variable, and the liver can chemically change part of the dose before it reaches the brain (Section 2 calls this first-pass metabolism). Health Canada (2024) advises that the effects of cannabis that is eaten or drunk can begin within 30 minutes to 2 hours. Oral diazepam, a benzodiazepine, begins to act within 15 to 60 minutes (Dhaliwal et al., 2023).

Smoking or vaping heats a drug so that it can be inhaled as smoke or vapour. The lungs offer a very large surface for absorption, and blood leaving them goes straight to the left side of the heart and into the arteries that supply the brain. After a puff on a cigarette, nicotine reaches the brain in 10 to 20 seconds, faster than with an intravenous infusion (Benowitz et al., 2009), and the effects of smoked or vaped cannabis can be felt within seconds to minutes (Health Canada, 2024). Inhalation also bypasses the first pass through the liver (Del Pozo et al., 2026). In British Columbia, most people who died of drug toxicity in 2025 had smoked the drugs involved (BC Coroners Service, 2026).

Intranasal use, also called insufflation or snorting, deposits a drug on the moist lining of the nose, where it is absorbed into small veins. The route avoids the first pass through the liver, and medications given this way are usually absorbed within 30 minutes (Del Pozo et al., 2026). A nicotine nasal spray reaches its peak level in venous blood 11 to 18 minutes after use (Benowitz et al., 2009). Powder cocaine can be taken intranasally (Hatsukami & Fischman, 1996), and naloxone, which reverses opioid effects, can be given as a nasal spray (Jordan et al., 2024).

An intravenous (IV) injection delivers the whole dose into a vein, so the fraction that reaches the circulation is 100 percent (Price & Patel, 2023), and the effect begins quickly: intravenous diazepam acts within 1 to 3 minutes, compared with 15 to 60 minutes when it is swallowed (Dhaliwal et al., 2023). Intramuscular injection (into a muscle) and subcutaneous injection (under the skin) release a drug more slowly. The Public Health Agency of Canada (2025) lists sharing equipment for preparing and consuming drugs, including needles, syringes and cookers, as a way that hepatitis C spreads. HSCI 212 develops infections among people who use drugs.

Medicine uses several other routes to control how fast and how completely a drug is absorbed. The flip cards below describe three of them.

Sublingual (under the tongue)Click to explore
Transdermal (through the skin)Click to explore
Intramuscular and subcutaneousClick to explore

The paragraphs above have traced the main routes and three others used in medicine. The next part turns to the time course of a drug's level in the blood, where the differences between routes become measurable.

1.2 Speed of Onset, Peak and Duration

Pharmacologists describe the time course of a drug with a concentration-time curve, a graph of the amount of drug in a given volume of blood plotted against the time since it was taken. Three features of the curve matter most. The onset is how quickly the level rises enough to produce an effect. The peak is the highest level reached and the time at which it occurs, often written Tmax. The duration is how long the level stays high enough to produce an effect. A route that delivers the drug quickly usually produces an early, high peak and a short duration; a slow route produces a later, lower and longer-lasting curve.

Nicotine is a well-documented example of one drug delivered by many routes. Table 3.1 summarizes the time to peak concentration and the bioavailability (the share of the dose that reaches the general circulation) of nicotine products, as compiled by Benowitz et al. (2009).

Table 3.1. Time to peak nicotine concentration in venous blood and bioavailability for nicotine taken by different routes, from single-dose laboratory studies compiled by Benowitz et al. (2009).

Product and routeTime to peak in venous bloodBioavailability
Cigarette, smoked over 5 minutes5 to 8 minutes (3 to 5 in arterial blood)80 to 90 percent of inhaled nicotine
Intravenous infusion over 30 minutes30 minutes (end of infusion)100 percent
Nasal spray11 to 18 minutes60 to 80 percent
Gum, chewed for 30 minutes30 minutes55 to 78 percent
Oral capsule, swallowed90 minutes44 percent
Transdermal patch3 to 12 hours, by product68 to 100 percent

Three patterns stand out in Table 3.1. The cigarette produces the fastest rise, and nicotine from a puff reaches the brain sooner than it does from an intravenous infusion, because blood from the lungs goes directly to the left side of the heart (Benowitz et al., 2009). The patch is the slowest, taking hours to reach its peak. The swallowed capsule delivers the smallest share of its dose, 44 percent, because the liver breaks down much of the nicotine before it reaches the general circulation. Figure 3.2 shows the general shapes that these patterns produce.

Time since the drug was taken (hours, illustrative) Concentration in blood 02468 Smoked or intravenous Intranasal Oral Transdermal patch
Figure 3.2. Illustrative concentration-time curves for four routes, drawn from a one-compartment model with the same elimination rate, absorption at three different speeds and, for the patch, steady release. The shapes reflect the patterns reported for nicotine in Table 3.1; the times and levels are illustrative, and the curves describe no specific product.

A slow route has its own hazard. When onset is delayed, a person may conclude that the first dose is not working and take more, so that two doses peak together. Gable (2004) listed repeated administration among the reasons that toxicity exceeds single-dose predictions, noting that after oral MDMA the concentration in blood does not peak for about 2 hours. Cannabis edibles raise the same timing problem (Health Canada, 2024). Section 2 shows how repeated doses accumulate.

The methylphenidate box below shows that the speed of delivery can change what a drug is used for.

One drug, two time courses: methylphenidate

Methylphenidate is a stimulant prescribed for attention-deficit/hyperactivity disorder. Volkow and Swanson (2003) reviewed brain imaging and clinical studies and concluded that its reinforcing effects are associated with rapid changes in blood concentration, as produced by intravenous injection or insufflation, while its therapeutic effects are associated with slowly rising concentrations, as produced by swallowing a tablet. They argued that clinical use, with low oral doses adjusted by a clinician, limits non-medical use. The same molecule can produce a treatment or a reinforcing experience depending on how quickly it arrives. Lesson 6 develops prescription stimulants.

Interactive 3.1 asks you to apply the timing data above. For each example, choose the onset band that the cited source reports, then check your choices.

Interactive 3.1
Onset sorter: how quickly does each route act?

Choose one band for each example, then select Check my choices. The feedback names the source for each answer.

Nicotine from a puff on a cigarette reaches the brain
Effects of cannabis eaten in a baked product begin
Diazepam injected into a vein begins to act
Diazepam swallowed as a tablet begins to act
A nicotine nasal spray reaches its peak in venous blood
A nicotine patch reaches its peak level in blood

The next part examines why the speed of the rise matters for reinforcement and addiction.

1.3 Why Speed Matters for Reinforcement and Addiction

Psychologists use the term reinforcement for any consequence that makes a behaviour more likely to be repeated. A drug that produces a pleasant effect within seconds of an action, such as a puff or an injection, links the action and its consequence very closely in time. Lesson 8 develops how the brain's reward and learning systems respond to such links.

Allain et al. (2015) reviewed clinical studies of stimulant use and experiments in animals and concluded that rapid onset and intermittent use both appear to push the addiction process forward. Why speed matters in the brain is still debated, and research on the question continues. Samaha and Robinson (2005) noted that it was then widely accepted that the more rapidly a drug reaches the brain, the greater its potential for addiction, which might explain why cocaine and nicotine are more addictive when smoked than when taken by other routes, but they described the evidence for the usual explanation, that rapidly delivered drugs produce more euphoria or more reinforcement, as not compelling. They proposed that rapid delivery more readily produces lasting changes in the brain circuits involved in motivation, changes that appear in behaviour as sensitization (a growing response to the drug and its cues with repeated use) and contribute to the compulsive pursuit of drugs. Lesson 8 develops the incentive sensitization theory on which their proposal draws. For nicotine, Benowitz et al. (2009) concluded that the rapid rise in nicotine levels during smoking lets a person adjust their intake puff by puff and makes smoking the most reinforcing and dependence-producing form of nicotine. Case 3.1 describes a 1996 review that compared two forms of the same drug.

Case 3.1: Crack and powder cocaine

Powder cocaine (cocaine hydrochloride) is snorted or dissolved and injected; crack cocaine (cocaine base) is smoked. Hatsukami and Fischman (1996) reviewed twenty years of studies comparing the two forms and found that the physiological and psychoactive effects of cocaine were similar in either form. Smoking cocaine base, or injecting cocaine hydrochloride, was associated with a greater propensity for dependence and more severe consequences than intranasal use. The variables that mattered were the immediacy, duration and magnitude of the effect and the frequency and amount used; the chemical form did not account for the difference. The authors therefore judged excessive the United States federal sentencing guidelines of the time, under which 100 times as much powder as crack triggered the same mandatory minimum penalty. Lesson 11 develops this sentencing disparity.

Case 3.1 shows that route, and the speed of onset that comes with it, can matter more than the chemical form in which a drug is sold. The next part turns from reinforcement to the immediate risks of each route.

1.4 Routes and Risk

Each route brings its own risks, and the risks do not line up neatly with speed. Fast routes produce their highest level within seconds or minutes, which leaves little time between taking a dose and experiencing its strongest effect; for opioids, that effect includes slowed breathing, which Lesson 5 develops. Slow routes delay the effect and can lead to repeated dosing. Injection adds the risk of infection, and the Public Health Agency of Canada (2025) also lists shared pipes and straws among the equipment that can transmit hepatitis C. Case 3.2 describes how the routes used in British Columbia have changed during the toxic drug crisis.

2.1 Mode of Consumption in British Columbia's Drug Toxicity Deaths

Case 3.2: From injecting to smoking in British Columbia

In preliminary data released on 19 February 2026, the BC Coroners Service reported that 1,826 people died of unregulated drug toxicity in British Columbia in 2025, an approximate 21 percent decrease from 2,315 deaths in 2024. Smoking was the most common mode of consumption (65 percent of deaths), followed by nasal insufflation (11 percent), injection (9 percent) and oral consumption (4 percent) (BC Coroners Service, 2026). These percentages sum to 89 percent; the release does not describe the remainder.

Bonn et al. (2025) interviewed 31 people who smoke unregulated drugs in Vancouver, recruited in early 2023 through the Vancouver Area Network of Drug Users (VANDU) and two cohort studies. Participants described the move from injecting to smoking as a way to protect their survival and wellbeing in an unpredictable drug supply, citing problems with injection, greater control over how much they consumed, and the ease and discretion of smoking. The authors interpreted the shift as a form of resilience.

Case 3.2 raises a question of interpretation. The figure of 65 percent describes how deaths were distributed across routes in 2025. On its own, it cannot show that smoking is more dangerous per occasion of use than injecting, because a route's share of deaths depends both on the risk of each occasion and on how many people use that route. Bonn et al. (2025) describe a transition from injecting to smoking among people who use unregulated drugs. If smoking has become the most common way of using them, smoking will account for most deaths even if each occasion of smoking carries the same risk as an occasion of injecting, or less. A comparison of risk per use needs a denominator: the number of people, or occasions of use, involving each route. HSCI 402 develops the surveillance of drug-related harms.

Try it 3.1 practises the arithmetic and reasoning that Case 3.2 requires.

Try it 3.1: Reading the British Columbia figures

Use the figures in Case 3.2 to answer three questions. (1) About how many of the 1,826 deaths in 2025 involved smoking, and about how many involved injection? (2) Check the reported decrease of about 21 percent from 2024 to 2025. (3) Name the additional information you would need before concluding that smoking is riskier, or less risky, per occasion than injecting.

Check your reasoning for Try it 3.1v

(1) Smoking: 0.65 × 1,826 ≈ 1,187 deaths. Injection: 0.09 × 1,826 ≈ 164 deaths, so smoking accounted for about seven times as many deaths (65 ÷ 9 ≈ 7.2).

(2) (2,315 − 1,826) ÷ 2,315 = 489 ÷ 2,315 ≈ 0.211, or about 21 percent.

(3) You would need the number of people, or occasions of use, involving each route in the same population and period, so that deaths could be expressed as a rate for each route. You would also want to know whether people who smoke and people who inject differ in tolerance, the drugs they use or whether they use alone (Section 4).

This section has shown that the route of administration determines the path a drug takes to the brain and the speed, height and duration of its effect; that rapid delivery strengthens reinforcement; and that each route carries distinct risks. Section 2 follows the drug after it enters the bloodstream: why only part of a swallowed dose reaches the circulation, why people process the same drug differently, and how long a drug stays in the body.

Key Takeaways

  • 1.1 The four main routes are oral, smoked or inhaled, intranasal and injected. They differ in which tissues the drug crosses first, whether it passes through the liver before reaching the general circulation, and how far the blood must travel before it is pumped to the brain; blood from the lungs goes straight to the left side of the heart and then to the brain.
  • 1.2 The concentration-time curve describes a drug's onset, peak and duration. Nicotine from a cigarette reaches the brain in 10 to 20 seconds and peaks in venous blood within 5 to 8 minutes, while a patch peaks after 3 to 12 hours; inhaled cannabis acts within minutes and eaten cannabis within 30 minutes to 2 hours, and oral diazepam acts more slowly than injected diazepam.
  • 1.3 Rapid delivery to the brain, as with smoking or injection, appears to strengthen reinforcement and the potential for addiction, although why speed matters is still debated. Crack and powder cocaine have similar effects, and the route and speed of onset matter more than the chemical form; methylphenidate taken slowly by mouth is a treatment, while rapid delivery makes it reinforcing.
  • 1.4 Each route carries distinct risks. Fast routes leave little time between a dose and its strongest effect, slow routes invite repeated dosing when a first dose seems not to work, and injection, along with shared pipes and straws, can transmit infections such as hepatitis C.
  • 2.1 In British Columbia in 2025, smoking accounted for 65 percent of unregulated drug toxicity deaths and injection for 9 percent. A route's share of deaths reflects how many people use it as well as its risk per occasion, so comparing the risk of routes needs a denominator: the number of people, or occasions of use, involving each route.

Reflection

A hypothetical provincial public health team is preparing a plain-language explainer on why the same drug can behave differently depending on how it is taken. They give you these facts. Nicotine from a cigarette reaches the brain in 10 to 20 seconds and peaks in venous blood within 5 to 8 minutes, while a nicotine patch takes 3 to 12 hours to reach its peak (Benowitz et al., 2009). The effects of inhaled cannabis begin within seconds to minutes, and those of eaten cannabis within 30 minutes to 2 hours (Health Canada, 2024). In British Columbia in 2025, 65 percent of unregulated drug toxicity deaths involved smoking and 9 percent involved injection (BC Coroners Service, 2026). In 150 to 250 words, explain (a) why a faster route is associated with stronger reinforcement, (b) why a slower route carries its own risk, and (c) why the British Columbia percentages alone cannot show which route is riskier per use.

Model answer

(a) A faster route produces a faster rise in the drug’s level in the brain. Nicotine from a cigarette arrives within 10 to 20 seconds, so the effect follows the act of smoking almost immediately. Benowitz et al. (2009) concluded that this rapid rise makes smoking the most reinforcing and dependence-producing form of nicotine, and Samaha and Robinson (2005) proposed, in an explanation that is still debated, that rapid delivery also more readily produces lasting changes in the brain’s motivation circuits. A patch, which takes hours to peak, delivers nicotine without that rapid rise.

(b) A slow route has a different hazard. Eaten cannabis can take up to 2 hours to act, so a person may conclude that the first dose has failed and take another. The two doses can then peak together, the repeated-dosing problem that Gable (2004) described for oral MDMA.

(c) The British Columbia figures describe how deaths were distributed across routes in one year. A route’s share of deaths depends on how many people use it as well as on the risk of each occasion. If most people who used unregulated drugs in 2025 smoked them, smoking will account for most deaths even if each occasion of smoking carries the same risk as injecting, or less. A fair comparison would need the number of people, or occasions of use, for each route.

Minimum 20 characters required.

✓ Reflection saved
Knowledge Check: this section

Question 1: Why does nicotine from a puff on a cigarette reach the brain sooner than nicotine from an intravenous infusion, according to Benowitz et al. (2009)?

Inhaled nicotine is absorbed in the lungs, and blood leaving the lungs goes directly to the left side of the heart and into the arteries that supply the brain, so it arrives in 10 to 20 seconds. Dose does not explain the speed: in the studies cited, a cigarette delivered about 2 mg of nicotine and the intravenous infusion about 5 mg. Inhalation also bypasses the liver on the first pass, so the liver plays no role in the speed.

Question 2: Hatsukami and Fischman (1996) compared crack cocaine and powder cocaine. Which conclusion did they reach?

The review found that the physiological and psychoactive effects of cocaine were similar in either form, and that smoking or injecting was associated with more dependence and harm than intranasal use because of the immediacy, magnitude and frequency of effect. The tempting answer, that crack is chemically stronger, is the claim the review rejected.

Question 3: In British Columbia in 2025, smoking accounted for 65 percent of unregulated drug toxicity deaths and injection for 9 percent. What can be concluded from these figures alone?

The figures describe how deaths were distributed across routes. A route’s share of deaths depends on how many people use it as well as on the risk per occasion, so comparing risk requires denominators. The seven-to-one ratio of deaths is real, but reading it as a ratio of risk per use ignores how common each route is.

Question 4: According to Volkow and Swanson (2003), what distinguishes the therapeutic effects of methylphenidate from its reinforcing effects?

Volkow and Swanson linked reinforcing effects to rapid changes in blood concentration, as with injection or insufflation, and therapeutic effects to slowly rising concentrations from oral doses. The same molecule acts on the same target in both cases, so the difference lies in the time course, which is why the first option is wrong.
Section 2 of 5

Pharmacokinetics: What the Body Does to a Drug

⏱ Estimated reading time: 45 minutes
Section 2 of 5

Pharmacokinetics

What the body does to a drug: absorption, distribution, metabolism and elimination.

ADME

Four processes after a dose

Absorption

Entry into the blood.

Distribution

Movement to tissues and the brain.

Metabolism

Transformation by liver enzymes.

Elimination

Removal, mainly by the kidneys.

Bioavailability

How much of a dose arrives

Equation 3.1
\[ \text{amount absorbed} = D \times F \]

Oral morphine: \( 10\ \text{mg} \times 0.35 = 3.5\ \text{mg} \); the range between people is about 1.5 to 6.4 mg (Murphy et al., 2025). Intravenous: all 10 mg.

Enzymes and interactions

Inhibitors raise levels; inducers lower them

Inhibition

Grapefruit blocks CYP3A4 in the gut wall; a 2013 review counted more than 85 interacting drugs, a number that was rising (Bailey et al., 2013). CYP3A4 inhibitors can raise fentanyl levels.

Induction

Tobacco smoke induces CYP1A2 (Zevin & Benowitz, 1999): caffeine's half-life is up to 50 percent shorter in people who smoke (Evans et al., 2024).

Genes and codeine

One dose, very different amounts of morphine

  • Poor metabolizers: little morphine, poor pain relief. Ultrarapid metabolizers: more morphine, higher risk (Crews et al., 2021).
  • 2006: the Toronto case report (Koren et al., 2006). 2013: Health Canada advises against codeine under age 12.
  • 2020: a review argues that the case is implausible (Zipursky & Juurlink, 2020). 2026: expression of concern in The Lancet (2026).
Half-life

Falling levels and steady state

Equation 3.2
\[ A_t = D \times F \times \left(\tfrac{1}{2}\right)^{t / t_{1/2}} \]
  • Remaining after 1, 2, 3, 4 and 5 half-lives: 50, 25, 12.5, 6.25 and about 3 percent.
  • Repeated doses approach steady state after about four to five half-lives (Hallare & Gerriets, 2025).
Carry forward

From amounts to actions

  • Bioavailability and first-pass metabolism decide how much arrives.
  • Enzymes, genes and interactions explain why people differ.
  • Half-life governs elimination and accumulation.
  • Section 3 asks what the drug does at its target.

Content note

This section discusses the death of an infant attributed to opioid toxicity, and the risk of overdose when drug levels accumulate. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service (1-888-688-6677) supports people who use drugs alone, and the Hope for Wellness Help Line (1-855-242-3310) serves Indigenous people across Canada.

Introduction

An afternoon coffee can disturb that night's sleep, more for some people than for others. Both observations concern pharmacokinetics, the time course of a drug in the body, often summarized by the acronym ADME: absorption (entry into the blood), distribution (movement to the tissues), metabolism (chemical transformation, mainly by the liver) and elimination (removal from the body). Figure 3.3 sets out the four processes in order.

Dose taken Absorption Distribution Metabolism Elimination By one of theroutes inSection 1 Drug enters theblood; the sharethat arrives intactis thebioavailability Blood carries itto tissues, thebrain and, innursing parents,breast milk Liver enzymeschange it, oftento inactive forms,sometimes toactive ones Mainly thekidneys removeit; the half-lifedescribesthe speed
Figure 3.3. The four processes of pharmacokinetics (ADME) that follow a dose. The processes overlap in time: elimination begins while absorption is still under way. Drawn for this lesson from Price and Patel (2023), Susa et al. (2023) and Vuckovic (2025a).

The rest of this section takes the processes in turn and then shows how they combine when a drug is taken repeatedly. The worked examples use hypothetical doses; the bioavailability and half-life values they use come from the cited sources.

Learning Objectives

  • 3.1 Define absorption and bioavailability, calculate the amount of a dose that reaches the general circulation from its bioavailability, and explain first-pass metabolism.
  • 3.2 Describe distribution, including the factors that govern how much of a drug reaches each tissue, the blood-brain barrier and passage into breast milk.
  • 4.1 Explain how liver enzymes transform drugs, including prodrugs such as codeine, and how enzyme inhibition and induction change drug levels.
  • 4.2 Explain how genetic variation in the CYP2D6 enzyme changes the response to codeine, and evaluate the evidence behind the Toronto codeine case.
  • 5.1 Explain first-order elimination and calculate the amount of a drug remaining after a given time from its half-life.
  • 5.2 Explain why repeated doses approach a steady state after about four to five half-lives, and why a long half-life allows slow accumulation.

3.1 Absorption and Bioavailability

Absorption is the movement of a drug from its site of administration into the blood. Bioavailability is the extent to which a drug reaches its intended destination unchanged, usually expressed as the fraction of a dose that reaches the general circulation (Price & Patel, 2023). A drug injected into a vein has a bioavailability of 100 percent. A swallowed drug usually has less, because absorption from the intestine is incomplete, because food and other drugs can interfere, and because of first-pass metabolism, the breakdown of part of a swallowed dose by the liver before it reaches the rest of the body (Price & Patel, 2023). The oral bioavailability of morphine is about 35 percent, with a range of 15 to 64 percent between people (Murphy et al., 2025), and that of swallowed nicotine is 44 percent (Benowitz et al., 2009). Nitroglycerin and buprenorphine lose so much of a swallowed dose to the liver that both are given under the tongue (Del Pozo et al., 2026; Kumar et al., 2024). Equation 3.1 expresses the amount that reaches the circulation, and Worked Example 3.1 applies it.

Amount reaching the general circulation
amount absorbed = D × FEq 3.1
The amount that reaches the general circulation equals the dose taken (D) multiplied by the bioavailability (F), written as a fraction between 0 and 1.

Worked Example 3.1: Oral and intravenous morphine

Take a hypothetical 10 mg dose of morphine. Swallowed, with a typical bioavailability of 0.35, it delivers 10 × 0.35 = 3.5 mg to the circulation; across the reported range of 0.15 to 0.64, it delivers 1.5 to 6.4 mg, a fourfold difference between people. Injected intravenously, it delivers 10 mg. Matching the intravenous amount by mouth would take about 10 ÷ 0.35 ≈ 29 mg. This is why oral and intravenous doses of the same drug differ, and why one oral dose can affect two people differently.

Once a drug is in the circulation, the next question is where it goes.

3.2 Distribution: Where the Drug Goes

Distribution is the movement of a drug from the blood into the tissues. How much reaches each tissue depends on its blood supply, on how easily the drug crosses cell membranes, and on how much of the drug is bound to proteins in the blood, which hold it inactive (Price & Patel, 2023). Psychoactive drugs must also cross the blood-brain barrier, the tightly joined cells lining the brain's blood vessels that limit which substances enter the brain. Drugs that dissolve readily in fat cross it more easily: heroin is more fat-soluble than morphine, enters the central nervous system rapidly, and is then converted to 6-monoacetylmorphine and morphine, both of which act at opioid receptors (Oelhaf & Azadfard, 2025). Distribution can also carry a drug to another person. Codeine and its products, including morphine, pass into breast milk, typically in small, dose-dependent amounts (Crews et al., 2021). The discussion of codeine and CYP2D6 later in this section returns to this point.

4.1 Metabolism, Enzymes and Interactions

Metabolism is the chemical conversion of a drug into other compounds, called metabolites. Most of it happens in the liver. Phase I reactions change the drug's structure, mostly through the cytochrome P450 family of enzymes, which have names such as CYP2D6 and CYP3A4; phase II reactions attach another molecule, which usually makes the drug inactive and water-soluble so that the kidneys can remove it (Susa et al., 2023). Metabolism usually inactivates a drug, but it can also convert an inactive prodrug into an active one (Susa et al., 2023). Codeine is a prodrug: CYP2D6 converts part of each dose to morphine, which produces most of its pain relief (Crews et al., 2021).

Anything that changes enzyme activity changes drug levels, which is the basis of a pharmacokinetic interaction. An enzyme inhibitor slows an enzyme, so the drugs it metabolizes build up; an enzyme inducer increases an enzyme's activity, so its drugs are removed faster and reach lower levels. Table 3.2 gives examples for five enzymes that act on psychoactive drugs.

Table 3.2. Five liver enzymes, examples of the psychoactive drugs they metabolize, and factors reported to change their activity.

EnzymeDrugs it metabolizesFactors that change its activity
CYP2D6Codeine, converted to morphine (Crews et al., 2021)Highly variable genes (4.2 Genetic Variation: Codeine and CYP2D6)
CYP3A4Fentanyl (Ramos-Matos et al., 2023); methadone, with CYP2B6 (Durrani & Bansal, 2024)Inhibited by grapefruit (Bailey et al., 2013) and by some antibiotics, antifungals and HIV drugs (Ramos-Matos et al., 2023)
CYP1A2Caffeine (Evans et al., 2024)Induced by chemicals in tobacco smoke (Zevin & Benowitz, 1999)
CYP2A6Nicotine (Benowitz et al., 2009)Genes, diet, age, sex, pregnancy and other drugs (Benowitz et al., 2009)
CYP2E1Alcohol, with alcohol dehydrogenase (Jones, 2010)Increased by heavy, prolonged drinking (Jones, 2010)

The effects in Table 3.2 can be large. In a 2013 review, researchers in London, Ontario, counted more than 85 drugs known or predicted to interact with grapefruit, 43 with possible serious effects, and reported that the number with serious effects had been rising as new drugs came to market; because grapefruit inactivates CYP3A4 in the gut wall until new enzyme is made, one serving still produced 25 percent of its maximum effect 24 hours later (Bailey et al., 2013). Inhibitors of CYP3A4 raise fentanyl levels and can deepen opioid-induced respiratory depression (Ramos-Matos et al., 2023), and inhibitors of methadone's enzymes can raise its levels enough to cause fatal respiratory depression (Durrani & Bansal, 2024). In the other direction, chemicals in tobacco smoke induce CYP1A2 (Zevin & Benowitz, 1999), and caffeine's half-life is up to 50 percent shorter in people who smoke (Evans et al., 2024). In people with heavy, long-term alcohol use who are going through detoxification, increased CYP2E1 activity can raise the rate at which alcohol is eliminated to 25 to 35 mg per 100 mL of blood per hour, compared with 10 to 15 after drinking on an empty stomach (Jones, 2010). Interactions in which two drugs add to each other's effects at their sites of action are pharmacodynamic; Lesson 4 develops them for alcohol, benzodiazepines and opioids. The next part turns to another source of difference between people: their genes.

4.2 Genetic Variation: Codeine and CYP2D6

The gene for CYP2D6 exists in more than 130 named versions, and some people carry extra copies of it or lack it entirely (Crews et al., 2021). A person's combination of versions predicts a metabolizer phenotype, a category describing how much enzyme activity they have. Table 3.3 shows what each phenotype means for codeine, using the guideline of the Clinical Pharmacogenetics Implementation Consortium (CPIC), an international group that translates genetic test results into prescribing advice.

Table 3.3. CYP2D6 metabolizer phenotypes, their predicted frequency across world populations (Gaedigk et al., 2017, using the phenotype definitions of the time), and the implications and recommendations for codeine (Crews et al., 2021).

PhenotypePredicted frequencyEffect on codeineCPIC recommendation
Ultrarapid1 to 21 percentMore morphine; higher risk of toxicityAvoid codeine
Normal67 to 90 percentExpected morphineUsual dosing
Intermediate0.4 to 11 percentLess morphineUsual dosing; another opioid if relief is poor
Poor0.4 to 5.4 percentMuch less morphine; poor pain reliefAvoid codeine

The guideline has since reclassified one group of genotypes, those whose combined activity is half that of the most common normal genotype, from normal to intermediate metabolizers, so under current definitions the intermediate group is larger and the normal group smaller than Table 3.3 shows (Crews et al., 2021). Figure 3.4 illustrates why the same dose of codeine can be ineffective for one person and dangerous for another. In healthy volunteers, ultrarapid metabolizers had a 45 percent higher median exposure to morphine than normal metabolizers after the same codeine dose (Crews et al., 2021).

Codeine (prodrug) CYP2D6 in the liver Poor metabolizer little morphine Weak pain relief Normal metabolizer expected morphine Expected pain relief Ultrarapid metabolizer more morphine Higher risk of toxicity
Figure 3.4. How the CYP2D6 phenotype changes the conversion of codeine to morphine. Bar lengths are illustrative; the implications follow the CPIC guideline (Crews et al., 2021).

A widely publicized report of harm attributed to this mechanism comes from Toronto (Zipursky & Juurlink, 2020). Case 3.3 describes it, and its history shows how regulators act on rare events and what happens when the evidence behind a warning is questioned.

Case 3.3: The Toronto codeine case and the evidence behind a warning

In 2006, Koren and colleagues at the Hospital for Sick Children in Toronto reported in The Lancet the death of a 13-day-old breastfed infant from opioid toxicity. The mother had taken codeine while breastfeeding and carried an extra copy of the CYP2D6 gene, and the authors concluded that she had converted codeine to morphine unusually quickly and passed large amounts to the infant in breast milk (Koren et al., 2006; Zipursky & Juurlink, 2020). Health Canada warned in 2008 that nursing infants of ultrarapid metabolizers were at risk. In June 2013, after reviewing rare cases of serious harm and death in children who had taken codeine themselves, it advised against codeine for children under 12 (Health Canada, 2013), and in July 2016 it contraindicated codeine for people under 18 for pain after tonsil or adenoid removal (Health Canada, 2016).

The report was questioned from early on. Bateman et al. (2008) asked whether breast milk could deliver a fatal dose. Zipursky and Juurlink (2020) argued that the explanation was implausible: very little opioid passes into breast milk whatever the mother's genotype, dangerous accumulation in a newborn requires severely impaired kidney function, and the infant's post-mortem blood contained a markedly high concentration of codeine itself. In February 2026, The Lancet published an expression of concern after receiving new allegations of falsified toxicological data, authorship problems and ethical concerns, and referred the matter to the hospital for investigation (The Lancet, 2026; Retraction Watch, 2026). As of October 2026, the report's PubMed record lists the expression of concern and no retraction.

Case 3.3 separates two questions. The first is whether the mechanism is real: pharmacokinetic studies show higher morphine exposure in ultrarapid metabolizers, case reports in children describe serious toxicity after standard doses, and the 2021 CPIC guideline advises against codeine for ultrarapid metabolizers (Crews et al., 2021). The second is whether one widely cited case report is reliable, and there the evidence has weakened considerably. Zipursky and Juurlink (2020) also noted a cost of the warnings: when codeine is avoided after childbirth, stronger opioids with greater potential for addiction may be prescribed in its place. The speed of metabolism matters for other drugs too: people who smoke tend to smoke fewer cigarettes when a drug inhibits their CYP2A6, and people who metabolize nicotine rapidly appear to be more dependent and to have more difficulty quitting (Benowitz et al., 2009). Lesson 9 develops the genetics of addiction. The next part turns to the last of the four processes.

5.1 Elimination and Half-Life

Elimination is the removal of a drug by metabolism and excretion, mainly through the kidneys (Vuckovic, 2025a). For most drugs it is first-order: a constant fraction of the drug is removed in each unit of time. A few substances follow zero-order elimination, in which a constant amount is removed per unit of time whatever the level (Vuckovic, 2025a). Alcohol is a familiar example, because the main enzyme that breaks it down is working at full capacity after the first few drinks (Jones, 2010); Lesson 4 applies this to blood alcohol concentration.

First-order elimination is summarized by the half-life, the time it takes for a drug's concentration to fall by half. After one half-life 50 percent remains, after two 25 percent, after three 12.5 percent, after four 6.25 percent and after five about 3 percent; about 90 percent has been eliminated after roughly 3.3 half-lives (Hallare & Gerriets, 2025). Table 3.4 lists half-lives for substances that later lessons discuss.

Table 3.4. Average or typical adult half-lives for selected substances. Individual values vary widely.

SubstanceHalf-lifeNotes and source
Naloxone30 to 80 minutes after injectionShorter than methadone, fentanyl and buprenorphine (Jordan et al., 2024)
NicotineAbout 2 hours(Benowitz et al., 2009)
MorphineAbout 2 hoursAfter intravenous dosing (Murphy et al., 2025)
Fentanyl3 to 7 hours(Ramos-Matos et al., 2023)
CaffeineAbout 5 hoursUp to 50 percent shorter in people who smoke; up to 15 hours late in pregnancy (Evans et al., 2024)
BuprenorphineAbout 38 hours(Kumar et al., 2024)
DiazepamAbout 46 hoursIts active metabolite lasts about 100 hours (Dhaliwal et al., 2023)
MethadoneAbout 8 to 60 hours(Durrani & Bansal, 2024)

Equation 3.2 gives the amount remaining a given time after a single dose, assuming immediate absorption and first-order elimination; Worked Example 3.2 applies it.

Amount remaining after time t
At = D × F × (½)t ÷ t½Eq 3.2
The amount remaining (At) equals the dose (D) times the bioavailability (F) times one half raised to the power of the time elapsed (t) divided by the half-life (t½).

Worked Example 3.2: Afternoon caffeine

A person drinks a coffee containing a hypothetical 100 mg of caffeine at 8 a.m.; assume all of it is absorbed (F = 1). With the average half-life of 5 hours, at 6 p.m. (10 hours, or two half-lives) 100 × (½)2 = 25 mg remains. For a person who smokes, with a half-life as short as 2.5 hours, 10 hours is four half-lives and about 6 mg remains. Late in pregnancy, with a half-life of up to 15 hours, 100 × (½)10 ÷ 15 ≈ 63 mg remains, ten times as much as in the second case.

Calculator 3.1 performs the calculation in Equation 3.2. Start with the worked example, then change the half-life or the time.

Many drugs are taken repeatedly, and the next part shows what happens when a new dose arrives before the last one has gone.

5.2 Repeated Dosing and Steady State

When a new dose arrives before the last one has been eliminated, the doses add up and the level rises. It does not rise forever: because first-order elimination removes more drug when the level is higher, the level approaches a steady state, in which the amount eliminated between doses equals the amount taken. Steady state is reached after approximately four to five half-lives, whatever the size of the dose (Hallare & Gerriets, 2025).

A simple sequence shows why. Suppose each dose produces a peak of 1 unit and a dose is taken every half-life. Just before the second dose, 0.5 unit remains, so the second peak is 1.5. Before the third dose 0.75 remains and the peak becomes 1.75, then 1.875, then about 1.94, approaching 2. A longer gap between doses produces a lower steady state and a shorter gap a higher one, but the time needed to approach it depends mainly on the half-life.

Long half-lives therefore create a specific risk: a drug with a half-life of a day or more keeps accumulating for days after each dose increase. Durrani and Bansal (2024) advise caution with increases in methadone dose early in treatment, noting that its peak respiratory depression occurs later and lasts longer than its peak pain relief, and Gable (2004) noted that seemingly modest daily doses of methadone can become life-threatening as the drug gradually accumulates. Short half-lives create the opposite problem: naloxone can wear off while a longer-acting opioid is still active, so people who respond to it are monitored for several hours afterward, and for longer after a long-acting opioid (Jordan et al., 2024). Lesson 5 develops this point. Simulator 3.1 lets you explore these patterns; its levels are multiples of the single-dose peak, with instant absorption and first-order elimination, so it is illustrative and describes no particular drug.

Simulator 3.1
Half-life and repeated dosing

Set the half-life, the interval between doses and the number of doses. The chart shows the level after each dose, the dashed lines show the steady-state peak and trough, and the summary reports how long the troughs take to reach 90 percent of their steady-state level. An optional threshold line marks an illustrative level above which harm would occur.

Try it 3.2 uses Simulator 3.1 to find the time to steady state.

Try it 3.2: Finding the time to steady state

(1) Choose the preset in which the interval equals the half-life (12 hours, 8 doses). Record the steady-state peak and the time for the troughs to reach 90 percent of steady state. (2) Keep the half-life at 12 hours and change the interval to 24 hours, then to 6 hours with 14 doses, and record the same values. (3) Choose the long half-life preset with a threshold of 2.5. At which dose does the level first cross the threshold?

Check your results for Try it 3.2v

(1) The steady-state peak is 2.0 times the single-dose peak, and the troughs reach 90 percent of steady state 48 hours after the first dose, four half-lives.

(2) With a 24-hour interval the steady-state peak is about 1.33; with a 6-hour interval it is about 3.41. The time to 90 percent stays similar, 48 and 42 hours (about 3.5 to 4 half-lives). The interval changes the level reached; the half-life sets how long it takes.

(3) With a 40-hour half-life and daily doses, the steady-state peak is about 2.94 and the level first rises above 2.5 at the fifth dose, 96 hours after the first. The same dose that stayed well below the threshold on the first day exceeds it on the fifth.

This section has followed a drug through absorption, distribution, metabolism and elimination, and shown how bioavailability, enzymes, genes, interactions and half-life determine how much drug acts and for how long. Section 3 turns to what the drug does to the body: the receptors on which drugs act and the relationship between dose and effect.

Key Takeaways

  • 3.1 Bioavailability is the fraction of a dose that reaches the general circulation: 100 percent for an intravenous dose, and less for a swallowed drug because of incomplete absorption and first-pass metabolism in the liver. The amount absorbed equals the dose multiplied by the bioavailability, so 10 mg of oral morphine at a bioavailability of 0.35 delivers about 3.5 mg.
  • 3.2 Distribution carries a drug from the blood to the tissues, depending on blood supply, how easily the drug crosses membranes and how much is bound to blood proteins. Fat-soluble drugs such as heroin cross the blood-brain barrier more easily, and some drugs, including codeine and its products, pass into breast milk.
  • 4.1 Liver enzymes, mostly of the cytochrome P450 family, usually inactivate drugs but can convert a prodrug such as codeine into an active drug. Inhibitors raise drug levels, as grapefruit does for drugs metabolized by CYP3A4, and inducers lower them, as tobacco smoke does for caffeine through CYP1A2.
  • 4.2 Variation in the CYP2D6 gene predicts poor, intermediate, normal and ultrarapid metabolizer phenotypes, which convert different amounts of codeine to morphine. The mechanism is supported by pharmacokinetic studies and case reports, but the Toronto codeine case shows that the evidence behind a single widely cited report deserves separate scrutiny.
  • 5.1 Most drugs follow first-order elimination, summarized by the half-life: after one half-life 50 percent remains and after five about 3 percent. The amount remaining equals the amount absorbed multiplied by one half raised to the number of half-lives that have passed, so the half-life strongly controls how much of a drug is left after a fixed time.
  • 5.2 Repeated doses add up until the level approaches a steady state, in which the amount eliminated between doses equals the amount taken, after about four to five half-lives. A drug with a long half-life, such as methadone, keeps accumulating for days after each dose increase.

Reflection

In 2006, a case report from Toronto attributed the death of a breastfed 13-day-old infant to morphine formed from the mother’s codeine by an unusually active CYP2D6 enzyme (an ultrarapid metabolizer). Health Canada warned in 2008 about nursing infants of ultrarapid metabolizers, and in 2013, after reviewing rare cases of serious harm in children who had taken codeine themselves, advised against codeine for children under 12. In 2020, Zipursky and Juurlink argued that the Toronto explanation was implausible, and in February 2026 The Lancet issued an expression of concern about the report. Separately, pharmacokinetic studies show that ultrarapid metabolizers have about 45 percent higher median morphine exposure than normal metabolizers after the same codeine dose (Crews et al., 2021). In 150 to 250 words, explain the pharmacokinetic mechanism that would make codeine riskier for ultrarapid metabolizers, and argue whether the advice against codeine for young children should stand if the Toronto report proves unreliable. Consider the other evidence available and one possible unintended consequence of the warnings.

Model answer

Codeine is a prodrug. The liver enzyme CYP2D6 converts part of each dose into morphine, which produces most of codeine’s effect. A person with extra copies of the CYP2D6 gene converts codeine faster and more completely, so the same dose produces more morphine. Studies in healthy volunteers found about 45 percent higher median morphine exposure in ultrarapid metabolizers, which is why the 2021 CPIC guideline advises them to avoid codeine.

I think the advice for young children should stand, because it does not rest on the Toronto report alone. Health Canada’s 2013 advice followed a review of cases in children who had taken codeine themselves, which is a different route of exposure from breast milk, and the pharmacokinetic studies support the mechanism independently. The Toronto report bears most directly on the warnings for breastfeeding, and its weakening, together with the 2026 expression of concern, is a reason to re-examine that specific warning. I would also weigh the unintended consequence that Zipursky and Juurlink (2020) identified: if codeine is avoided after childbirth, stronger opioids with greater potential for addiction may be prescribed in its place. A good review would ask what the evidence shows for each population separately and what the alternatives to codeine cost.

Minimum 20 characters required.

✓ Reflection saved
Knowledge Check: this section

Question 1: A hypothetical 20 mg dose of a drug with an oral bioavailability of 0.25 is swallowed. How much reaches the general circulation?

By Equation 3.1, the amount absorbed is the dose multiplied by the bioavailability: 20 × 0.25 = 5 mg. The answer of 15 mg is the amount lost to incomplete absorption and first-pass metabolism, which is the most common slip.

Question 2: Why does the 2021 CPIC guideline advise avoiding codeine for both CYP2D6 ultrarapid metabolizers and poor metabolizers?

Codeine is a prodrug converted to morphine by CYP2D6. Ultrarapid metabolizers form more morphine and risk toxicity; poor metabolizers form little and get poor pain relief. The last option reverses the two groups, which is the usual error.

Question 3: Grapefruit raises the blood levels of some swallowed drugs. Which mechanism did Bailey et al. (2013) describe?

Grapefruit inactivates CYP3A4 in the intestinal wall until new enzyme is made, so more of a swallowed dose survives into the circulation. Enzyme induction would have the opposite effect for most drugs, lowering their levels.

Question 4: A drug with a half-life of 6 hours is taken at the same dose every 6 hours. About how long will its level take to approach steady state?

Steady state is approached after about four to five half-lives whatever the dose, here 24 to 30 hours. The level does not rise forever, because first-order elimination removes more drug as the level rises until elimination between doses equals each dose.
Section 3 of 5

Pharmacodynamics: What a Drug Does to the Body

⏱ Estimated reading time: 45 minutes
Section 3 of 5

Pharmacodynamics

What a drug does to the body: receptors, dose and response, safety margins and tolerance.

Receptors

Binding is one thing; activating is another

Affinity

How strongly a drug binds to its receptor.

Efficacy

How large a response the drug produces once bound.

Caffeine binds and blocks adenosine receptors (Evans et al., 2024).

Four kinds of action

Agonists, partial agonists, antagonists and modulators

Agonist

Morphine, fentanyl

Partial agonist

Buprenorphine

Antagonist

Naloxone, caffeine

Allosteric modulator

Diazepam at the GABA-A receptor

Dose and response

Potency is not the same as efficacy

  • Potency: the dose needed for a given effect (the EC50).
  • Efficacy: the maximum effect the drug can produce.
  • Fentanyl is typically 50 to 100 times more potent than morphine (Ramos-Matos et al., 2023).
  • Buprenorphine, a partial agonist, has a ceiling on respiratory depression (Kumar et al., 2024).
Margin of safety

Therapeutic index and safety ratio

Equation 3.3
\[ \text{TI} = \frac{\text{LD}_{50}}{\text{ED}_{50}} \]

One analysis of published reports estimated safety ratios of about 6 for intravenous heroin, 10 for alcohol and about 1,000 for LSD (Gable, 2004). Such ratios omit curve slope, combinations, accumulation, tolerance and uncertainty about the dose.

Unknown doses and uneven tolerance

When the curves move

Variable supply

Drug checking in Victoria, 2020 to 2021: fentanyl in 96 percent of opioid samples, at widely varying concentrations (Larnder et al., 2022).

Uneven tolerance

Reviews describe tolerance to pain relief developing faster than tolerance to slowed breathing (Dumas & Pollack, 2008; Farahbakhsh & Siciliano, 2023).

Carry forward

From the receptor to the person

  • Drugs act as agonists, partial agonists, antagonists or modulators.
  • Potency and efficacy are different properties.
  • Safety margins hide slope, combinations and uncertainty about the dose.
  • Section 4 shows how set and setting move the curves.

Content note

This section discusses lethal doses and overdose deaths in the unregulated drug supply. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service (1-888-688-6677) supports people who use drugs alone, and the Hope for Wellness Help Line (1-855-242-3310) serves Indigenous people across Canada.

Introduction

Section 2 followed a drug through the body as a matter of amounts and times. Pharmacodynamics asks what the drug does once it arrives: which molecules in the body it acts on, what response it produces, and how that response changes with dose and with repeated use. A cup of coffee illustrates the questions. Caffeine acts mainly by blocking the receptors for adenosine, a chemical messenger, and its blockade of one adenosine receptor subtype is especially responsible for its effect on wakefulness (Evans et al., 2024). The size of that effect depends on how much caffeine reaches the receptors, and the effects of many drugs weaken with repeated use. The rest of this section develops these ideas in turn: receptors, types of drug action, dose-response curves, margins of safety and tolerance.

Learning Objectives

  • 6.1 Define pharmacodynamics, receptor and ligand, distinguish affinity from efficacy, and name the main receptor targets of the drug classes that Lessons 4 to 7 examine.
  • 6.2 Distinguish full agonists, partial agonists, antagonists and allosteric modulators.
  • 6.3 Read a dose-response curve and distinguish potency from efficacy, using opioids as the example.
  • 7.1 Calculate a therapeutic index from effective and lethal dose curves, and explain what the safety ratios compiled by Gable (2004) show and leave out.
  • 7.2 Explain tolerance and cross-tolerance, including pharmacokinetic, pharmacodynamic and learned tolerance, and why tolerance to different effects develops at different rates.

6.1 Receptors and Ligands

Cells communicate through chemical messengers, such as neurotransmitters, that bind to receptors: proteins on or in cells that respond to a particular messenger by changing the cell's activity. Any molecule that binds to a receptor is called a ligand. The body's own messengers are endogenous ligands; the endorphins, for example, are endogenous ligands of the opioid receptors, and Section 4 describes evidence that they contribute to placebo pain relief (Levine et al., 1978). Most psychoactive drugs work by acting on the receptors or other proteins that the body's own messengers use.

A common analogy describes a receptor as a lock and a ligand as a key. The analogy captures how selectively ligands bind. It also hides a distinction that matters for drugs. Affinity is the strength of the bond between a drug and its receptor; efficacy is the size of the response that the drug produces once bound (Vuckovic, 2025b). A key can fit a lock tightly and still fail to turn it. Not every drug acts at a receptor: some act on transporters, the proteins that clear messengers away after they are released. Methylphenidate, for example, increases the amount of dopamine outside nerve cells (Volkow & Swanson, 2003). Table 3.5 lists the targets that this lesson uses as examples and the later lessons that develop them.

Table 3.5. Selected drug targets, the body's own messengers that act on them, and example drugs with their type of action.

TargetBody's own messengerExample drugs and their actionLesson
Mu-opioid receptorEndorphins and other endogenous opioidsMorphine and fentanyl (agonists); buprenorphine (partial agonist); naloxone (competitive antagonist)5
GABA-A receptorGABA, the main inhibitory neurotransmitterDiazepam binds a separate site and increases how often the receptor's channel opens4
Adenosine receptorsAdenosineCaffeine blocks all four subtypes (antagonist)6
Serotonin 2A (5-HT2A) receptorSerotoninClassic psychedelics (agonists)7
Dopamine signallingDopamineMethylphenidate increases dopamine outside nerve cells6 and 8

The entries in Table 3.5 come from Murphy et al. (2025), Ramos-Matos et al. (2023), Kumar et al. (2024) and Jordan et al. (2024) for the opioids; Dhaliwal et al. (2023) and Canver et al. (2024) for GABA; Evans et al. (2024) for caffeine; Carhart-Harris et al. (2018) for psychedelics; and Volkow and Swanson (2003) for methylphenidate. Lessons 6 and 7 add the nicotinic acetylcholine receptors and the cannabinoid CB1 receptor. 3-D Model 3.1 shows the first two targets in Table 3.5 as their measured molecular structures: the mu-opioid receptor passes its signal to a separate protein inside the cell, called a G protein, while the GABA-A receptor forms a pore through the cell membrane that opens to let chloride ions (charged particles) flow into the cell. The next part explains the kinds of action listed in the table.

3-D Model 3.1. The model places the mu-opioid receptor, with fentanyl bound and its G protein attached, beside the GABA-A receptor, with GABA and diazepam bound, in one membrane at the same scale. Atomic structures are from the Protein Data Bank (PDB 8EF5, Zhuang et al., 2022; PDB 6X3X, Kim et al., 2020), and the membrane and the movement of ions are schematic.

6.2 Agonists, Partial Agonists, Antagonists and Modulators

Drugs that act at receptors fall into a few types according to their efficacy. An agonist binds to a receptor and produces a biological effect. A partial agonist binds and produces a weaker maximal effect than a full agonist, however much is given. An antagonist binds and blocks the receptor's activity without activating it (Vuckovic, 2025b). A fourth type, the allosteric modulator, binds to a different site on the receptor and changes how the receptor responds to its natural messenger. Figure 3.5 shows the four types schematically.

Full agonistPartial agonistAntagonistAllosteric modulator ResponseResponseResponseResponse Binds and fullyactivates Binds; responsehas a ceiling Binds, blocks;no response Separate site;boosts messenger
Figure 3.5. Schematic of four types of drug action at a receptor. Circles and the square represent ligands in the binding site; the triangle represents an allosteric modulator at a separate site. Bars show the response relative to a full agonist and are illustrative. Based on the definitions in Vuckovic (2025b) and Dhaliwal et al. (2023).

The flip cards below give an example of each type from Table 3.5; each reappears in a later lesson.

Agonist: morphine and fentanylClick to explore
Partial agonist: buprenorphineClick to explore
Competitive antagonist: naloxoneClick to explore
Allosteric modulator: diazepamClick to explore

3-D Model 3.2 lets you step through a dopamine synapse, the junction where one nerve cell passes a chemical signal to the next, and then compare the synapse with and without four kinds of drug: an agonist, an antagonist, a reuptake inhibitor (which blocks the transporter that carries dopamine back into the sending cell) and a releaser (which makes that transporter push dopamine out of the cell).

3-D Model 3.2. The model follows one signal across a dopamine synapse, through release, binding, reuptake and breakdown, and then shows how an agonist, an antagonist, a reuptake inhibitor and a releaser change it, with a toggle that compares the synapse with and without each drug. The model is schematic, and sizes, numbers of molecules and speeds are not to scale.

The paragraphs above have classified drugs by the kind of action they have at a receptor. The next part adds the quantity of drug, which determines how large that action is.

6.3 Dose-Response Curves, Potency and Efficacy

A dose-response curve plots the size of a drug's effect against its dose. Kenny et al. (2024) distinguish two forms. The continuous form, for one person or tissue, answers the question "how much?": the effect grows with the dose until it reaches a maximum. The quantal form, for a population, answers "yes or no?": it plots the percentage of people who show a defined effect at each dose. Both forms give rise to a median dose. The median effective dose (ED50) is the dose that produces a specified effect in 50 percent of a population (Kenny et al., 2024), and on a continuous curve the concentration that produces half the maximum response is called the EC50 (Vuckovic, 2025b).

Two properties of a drug can be read from its curve. Potency is the amount of drug needed to produce a given effect, measured by the EC50: a more potent drug produces the same effect at a smaller dose. Efficacy, as defined under 6.1 Receptors and Ligands, is the maximum response the drug can produce, often written Emax (Vuckovic, 2025b). The two are independent, and Figure 3.6 shows three illustrative curves that separate them. The curves use a logarithmic dose axis, on which each labelled step multiplies the dose by ten; on this axis they take an S shape.

0255075100 0.11101001000 Dose (logarithmic scale, illustrative units) Effect (percent of maximum) A: high potency B: low potency C: partial agonist
Figure 3.6. Illustrative dose-response curves. Dots mark the dose that produces half of each drug's own maximum effect. A and B are full agonists with the same efficacy that differ in potency by a factor of 50; C is a partial agonist with lower efficacy. The curves are drawn from a simple mathematical model and describe no specific drug.

Opioids show why the distinction matters. Fentanyl is typically 50 to 100 times more potent than morphine (Ramos-Matos et al., 2023): a much smaller amount produces the same effect, the relationship between curves A and B. Potency describes the amount needed, so with a highly potent drug a small error in the amount taken produces a large change in effect. Buprenorphine resembles curve C: as a partial agonist, its effect on breathing levels off at higher doses (Kumar et al., 2024). The next part turns from desired effects to harmful ones.

7.1 Therapeutic Index and Margin of Safety

Every drug has a dose-response curve for its desired effect and another for its harmful effects, and the distance between them describes its margin of safety. The therapeutic index (TI) compares the two medians. In human pharmacology it is usually the median toxic dose (TD50) divided by the ED50; in animal studies it is often calculated with the median lethal dose (LD50), the dose that kills half the animals tested (Kenny et al., 2024). Drugs with a narrow therapeutic index, such as warfarin and lithium, need close monitoring, and Kenny et al. (2024) report a United States Food and Drug Administration definition of a narrow therapeutic ratio as an LD50 less than twice the ED50. Diazepam, in contrast, has a broad therapeutic index when taken alone (Dhaliwal et al., 2023). Equation 3.3 gives the calculation, and Worked Example 3.3 applies it.

Therapeutic index
TI = LD50 (or TD50) ÷ ED50Eq 3.3
The therapeutic index equals the dose that kills (or harms) half of a population divided by the dose that produces the desired effect in half of the population.

Worked Example 3.3: A therapeutic index from illustrative curves

A hypothetical drug produces its desired effect in half of a test population at 10 units (ED50) and is lethal to half at 100 units (LD50), so TI = 100 ÷ 10 = 10. The index uses only the two midpoints. If both curves are steep (a Hill slope of 3 in Simulator 3.2), the dose that produces the desired effect in 95 percent of the population, about 27 units, is lethal to about 2 percent. If both curves are shallow (a slope of 1.5), the same TI of 10 hides a much worse overlap: reaching 95 percent of people requires about 71 units, which is lethal to about 38 percent.

Gable (2004) made a related point about curve shape: of two drugs with the same median lethal dose, the one with a flat curve may kill some people at much lower doses, and the one with a steep curve is safer in most applications unless people are unwilling or unable to measure doses accurately, a condition he noted frequently occurs in social situations. Simulator 3.2 lets you compute a therapeutic index from illustrative curves and see how slope and tolerance change the overlap.

Simulator 3.2
Dose-response curves and the therapeutic index

The teal curve shows the percentage of a hypothetical population that has the desired effect at each dose; the red curve shows the percentage that dies. Read the ED50 and LD50 where each curve crosses 50 percent, divide to find the therapeutic index, and compare your answer with the summary. The tolerance controls multiply the ED50 or the LD50, as repeated use would. All values are illustrative.

In an analysis of published reports, Gable (2004) applied the same logic to drugs used outside medicine. From about 3,000 records, he estimated for 20 substances a safety ratio: the usual lethal dose divided by the dose most commonly used for non-medical purposes, for a 70 kg adult without tolerance. Table 3.6 lists a selection of his ratios, which are the estimates of one 2004 analysis.

Table 3.6. Safety ratios (estimated lethal dose divided by usual non-medical dose) for selected substances, from Gable (2004). Lower numbers indicate a smaller margin. Ratios marked uncertain were flagged by Gable because they were extrapolated or rested on very few cases.

Substance (route)Safety ratioSubstance (route)Safety ratio
Heroin (intravenous)6Codeine (oral)20
GHB (oral)8Methadone (oral)20
Alcohol (oral)10Ketamine (intranasal)38, uncertain
Methamphetamine (oral)10Nitrous oxide (inhaled)more than 150
Cocaine (intranasal)15LSD (oral)1,000, uncertain
MDMA (oral)16Cannabis (smoked)more than 1,000, uncertain

The ratios in Table 3.6 differ by more than a factor of 100. Intravenous heroin had the smallest margin and several hallucinogens the largest, and most published cases of lethal toxicity involved a second substance, most often alcohol (Gable, 2004). Gable was explicit about what the ratios leave out, and the box below summarizes his cautions.

What a safety ratio leaves out

  • Combining substances raises the probability of harm substantially; deaths increased when alcohol was taken with a benzodiazepine despite the benzodiazepine's wide therapeutic index.
  • Repeated doses can accumulate, for example when a person believes a slow-acting first dose has failed, or when a long half-life lets methadone build up.
  • The ratios describe acute toxicity only; direct acute deaths from alcohol or nicotine are rare, yet alcohol and tobacco use are among the leading long-term causes of preventable death.
  • The ratios assume no tolerance; when tolerance to a drug's reinforcing effects grows faster than tolerance to its physiological (metabolic) effects, the ratio narrows.
  • Environment, expectation and stress alter drug reactions (Section 4 develops this).
  • Serious non-lethal harms, such as impaired performance, are not counted.

Source: Gable (2004).

The ratios also assume that the person knows the dose. In the unregulated supply, that assumption fails. Case 3.4 describes the evidence from drug checking in British Columbia.

Case 3.4: The unknown dose in the unregulated opioid supply

Larnder et al. (2022) analyzed 454 samples that people brought to a drug checking service in Victoria, British Columbia, between November 2020 and July 2021, describing them as opioids. Fentanyl was found in 96 percent of the samples, at a median concentration of 9 percent with a standard deviation of 7 percentage points, and nearly 20 percent of results fell outside the control limits used to judge whether a process is stable. More than half of the samples contained an additional, unexpected active drug, most often etizolam (43 percent). The authors concluded that the overdose crisis was better explained by the general variability of the market than by occasional bad batches.

In the terms of this section, a person buying from such a market cannot know where a dose falls on the dose-response curve. With a drug as potent as fentanyl, a difference of a few percentage points in concentration can move the same amount of powder from the desired effect toward the lethal curve. Lesson 5 develops the composition of the toxic supply, and HSCI 402 develops drug checking services. The next part adds the final factor that moves the curves: tolerance.

7.2 Tolerance and Cross-Tolerance

Tolerance is a decrease in a drug's effect following repeated or prolonged use (Dumas & Pollack, 2008). A person with tolerance needs a larger dose to produce the same effect, which on a dose-response curve appears as a shift to the right. Dumas and Pollack (2008) distinguish innate tolerance, a person's inborn level of response, from acquired tolerance, which develops with repeated exposure. They classify acquired tolerance by mechanism, as the accordion below summarizes.

Pharmacokinetic tolerancev

The body removes the drug faster, so less reaches its target. Increased CYP2E1 activity in people with heavy, long-term alcohol use, which speeds the clearance of alcohol (Jones, 2010), is an example (Section 2, 4.1 Metabolism, Enzymes and Interactions).

Pharmacodynamic tolerancev

The receptor system becomes less responsive, so the same amount of drug at the target produces a smaller effect (Dumas & Pollack, 2008). Lesson 8 develops the neuroadaptations behind this kind of tolerance.

Learned tolerance: behavioural and conditionedv

Behavioural tolerance develops when a person learns to function despite a drug's effects. Conditioned tolerance follows Pavlovian principles, in which cues associated with drug use come to trigger responses that counteract the drug (Dumas & Pollack, 2008). Section 4 develops conditioned tolerance.

Acute tolerancev

Tolerance can develop within a single episode of use. After intranasal cocaine, euphoria at first rises with the blood concentration, then declines even though the concentration stays the same or rises (Dumas & Pollack, 2008).

Cross-tolerancev

Tolerance to one drug can reduce the response to other drugs of the same class. Cross-tolerance is the principle behind methadone treatment for people who use heroin: tolerance from long-term heroin use reduces methadone's euphoric and side effects. Cross-tolerance between opioids is often incomplete, so switching from one opioid to another can restore some effect (Dumas & Pollack, 2008).

Tolerance does not develop at the same rate for every effect of a drug. With long-term opioid use, tolerance to pain relief develops rapidly, while tolerance to respiratory depression, nausea and slowed digestion develops slowly (Dumas & Pollack, 2008), and a short review by Farahbakhsh and Siciliano (2023) describes the same contrast between pain relief and breathing. When the curve for the desired effect moves right faster than the curve for the harmful effect, a person who raises the dose to restore the desired effect moves closer to the harmful curve, which is the narrowing that Gable (2004) warned about. The reverse problem arises when tolerance is lost after a period without use, which Lesson 5 develops for overdose after release from prison. Try it 3.3 asks you to quantify uneven tolerance.

Try it 3.3: When tolerance is uneven

In Simulator 3.2, choose the preset Uneven tolerance (ED50 10 units and LD50 100 units before tolerance, steepness 3). (1) With both tolerance controls at 1, record the therapeutic index and the percentage of the population that dies at a dose of 40 units. (2) Set tolerance to the desired effect to 4 and tolerance to the lethal effect to 1.4. What are the new ED50, LD50 and therapeutic index? (3) What percentage now has the desired effect at 40 units, and what percentage dies?

Check your results for Try it 3.3v

(1) The therapeutic index is 100 ÷ 10 = 10, and about 6 percent of the population dies at 40 units.

(2) The ED50 becomes 10 × 4 = 40 units and the LD50 becomes 100 × 1.4 = 140 units, so the therapeutic index falls to 140 ÷ 40 = 3.5.

(3) At 40 units, only half the population now has the desired effect, and about 2 percent dies. Raising the dose to 60 units to restore the desired effect for about three-quarters of the population raises deaths to about 7 percent. The margin has narrowed even though tolerance protects against some of the harm.

This section has shown that drugs act on receptors and other targets as agonists, partial agonists, antagonists or modulators; that dose-response curves separate potency from efficacy; that the therapeutic index and safety ratio summarize the margin of safety but omit curve slope, combinations, accumulation and uncertainty about the dose; and that tolerance shifts the curves at different rates. Section 4 shows that the person and the setting move these curves as well.

Key Takeaways

  • 6.1 Pharmacodynamics describes what a drug does to the body. Drugs act as ligands at receptors and other targets used by the body's own messengers, such as the mu-opioid, GABA-A and adenosine receptors; affinity is how strongly a drug binds, and efficacy is how large a response it produces once bound.
  • 6.2 Agonists activate receptors, partial agonists such as buprenorphine activate them less fully, antagonists such as naloxone bind without activating and block other ligands, and allosteric modulators such as diazepam bind a separate site and change the response to the natural messenger.
  • 6.3 A dose-response curve separates potency, the amount of drug needed for a given effect, from efficacy, the maximum effect. Fentanyl is 50 to 100 times more potent than morphine, so small errors in amount produce large changes in effect.
  • 7.1 The therapeutic index divides the median lethal or toxic dose by the median effective dose, and Gable's safety ratios apply the same logic to non-medical use. Both omit the slopes of the curves, combinations of drugs, accumulation, tolerance, long-term harms and uncertainty about the dose, which in the unregulated supply can be large.
  • 7.2 Tolerance is a decrease in a drug's effect with repeated use, and it can be pharmacokinetic, pharmacodynamic, learned or acute; cross-tolerance extends it to drugs of the same class. Tolerance to pain relief develops faster than tolerance to slowed breathing, which narrows the margin of safety.

Reflection

A hypothetical drug has a median effective dose (ED50) of 10 units and a median lethal dose (LD50) of 100 units in a population without tolerance. After months of daily use, tolerance develops unevenly: the ED50 for the desired effect rises fourfold, to 40 units, while the LD50 rises only to 140 units. In 150 to 250 words, (a) calculate the therapeutic index before and after tolerance, (b) explain what the change means for a person who raises their dose to keep the desired effect, and (c) connect your answer to what Dumas and Pollack (2008) report about opioid tolerance and to one other limitation of safety ratios noted by Gable (2004).

Model answer

(a) Before tolerance, the therapeutic index is 100 ÷ 10 = 10. After tolerance it is 140 ÷ 40 = 3.5, so the margin between the effective and lethal medians has shrunk to about a third of its original size.

(b) A person who raises the dose to restore the desired effect is moving along a curve that has shifted four times to the right, while the lethal curve has shifted only 1.4 times. Each increase therefore brings them proportionally closer to the dose that is lethal for half the population. With the curves in Simulator 3.2, a dose of 40 units now gives the desired effect to only half the population, and raising it to 60 units to restore the effect for about three-quarters raises the lethal fraction to about 7 percent.

(c) Dumas and Pollack (2008) report this pattern for opioids: tolerance to pain relief develops rapidly, and tolerance to respiratory depression develops slowly. Gable (2004) warned that safety ratios narrow in exactly this situation, and he noted another limitation that makes it worse: his ratios assume a single substance, yet most lethal cases involved a second drug, most often alcohol. A person with uneven tolerance who also combines drugs has less margin than any published ratio suggests.

Minimum 20 characters required.

✓ Reflection saved
Knowledge Check: this section

Question 1: Fentanyl is typically 50 to 100 times more potent than morphine. What does this statement mean?

Potency is the amount of drug needed for a given effect, so a more potent drug produces the same effect at a smaller dose. A larger maximum effect would be greater efficacy, a different property, which is why the second option is wrong.

Question 2: Buprenorphine is a partial agonist with high affinity for the mu-opioid receptor. Which consequence follows?

A partial agonist produces a weaker maximal effect, so buprenorphine has a ceiling on respiratory depression; because it binds tightly, it can displace a full agonist and trigger withdrawal (Kumar et al., 2024). Blocking without activation describes an antagonist such as naloxone.

Question 3: A hypothetical drug has an ED50 of 20 mg/kg and an LD50 of 300 mg/kg in animal studies. What is its therapeutic index, and what does the index leave out?

The therapeutic index is the LD50 divided by the ED50: 300 ÷ 20 = 15. It compares the medians only, so two drugs with the same index can differ greatly in how much their curves overlap. It does not account for tolerance or combinations, which is why the first option is wrong.

Question 4: With long-term opioid use, tolerance to pain relief develops rapidly and tolerance to respiratory depression develops slowly. What does this imply?

When the curve for the desired effect shifts right faster than the curve for the harmful effect, a person who raises the dose to restore the desired effect moves closer to harm, and the margin narrows. The third option reverses this: the therapeutic index shrinks when tolerance is uneven, so larger doses become more dangerous.
Section 4 of 5

Drug, Set and Setting: Expectancy, Conditioning and Context

⏱ Estimated reading time: 40 minutes
Section 4 of 5

Drug, Set and Setting

Expectancy, conditioning and the risk environment.

The framework

Drug, set and setting

Drug

Dose, route, potency, half-life, receptor action.

Set

Expectations, intentions, mood, past learning.

Setting

Place, people, culture, laws and markets.

Leary et al. (1964) and Zinberg (1984), as described by Hartogsohn (2017).

Expectancy and placebo

Belief changes the effect

Balanced placebo design

Told alcohol or tonic, crossed with given alcohol or tonic (Marlatt et al., 1973).

Placebo pain relief

Naloxone raised pain only in placebo responders (Levine et al., 1978). Later research links placebo pain relief to opioid and other systems (Zamfira et al., 2026).

Conditioned tolerance

Tolerance can belong to a place

Familiar setting

Drug-associated cues trigger a compensatory response, and more rats survived a high dose of heroin.

Unfamiliar setting

Without the familiar cues, rats with the same drug history were less likely to survive the same dose.

Experiment: Siegel et al. (1982). Support from survivors' reports in people and causal tests in rats (Farahbakhsh & Siciliano, 2023).

Risk environment

Setting includes place, people, markets and policy

  • Harm arises from interactions between individuals and environments (Rhodes, 2009).
  • 48 percent of 2025 deaths in British Columbia occurred in a private residence; 20 percent outdoors (BC Coroners Service, 2026).
  • As of 2026, National Overdose Response Service: 1-888-688-6677.
Bringing it together

Five questions for any drug

  • How did it enter the body? (route)
  • How much reached the target, and for how long? (pharmacokinetics)
  • What does it do there? (pharmacodynamics)
  • How has experience changed the response? (tolerance)
  • Who, where and under what conditions? (set and setting)
Next steps

Reflection and final assessment

  • Work through the reading, the sorter and the task card below.
  • Complete the reflection and the knowledge check for this section.
  • Continue to the final assessment, which draws on all four sections.
  • Lesson 4 applies these tools to alcohol and other depressants.

Content note

This section discusses overdose deaths, including deaths from the toxic unregulated drug supply in British Columbia. If you would like support, the 9-8-8 Suicide Crisis Helpline is available by call or text at 9-8-8, the National Overdose Response Service (1-888-688-6677) supports people who use drugs alone, and the Hope for Wellness Help Line (1-855-242-3310) serves Indigenous people across Canada.

Introduction

Sections 1 to 3 treated drug effects as products of chemistry and physiology. The same drug at the same dose can nonetheless produce very different experiences in different people and places. In 1953 the sociologist Howard Becker described how novice cannabis users learned, through their peer group, to recognize the drug's effects and to interpret them as pleasant (Hartogsohn, 2017). Researchers have documented wide differences between cultures in how people behave when drunk (Room, 2001). This section explains these observations with three ideas: the framework of drug, set and setting; expectancy and placebo responses; and conditioned tolerance. It then widens the idea of setting to the social and policy conditions that shape harm, and ends by applying the whole lesson to a Canadian case.

Learning Objectives

  • 8.1 Explain the drug, set and setting framework and trace its history from early psychedelic research to Norman Zinberg's study of controlled drug use.
  • 8.2 Describe how expectancy and placebo responses shape drug effects, using the balanced placebo design and research on placebo pain relief.
  • 8.3 Explain conditioned tolerance and how the absence of familiar drug-associated cues can contribute to overdose.
  • 8.4 Apply the risk environment framework to explain how physical, social, economic and policy settings shape drug-related harm.
  • 9.1 Integrate route, pharmacokinetics, pharmacodynamics, set and setting to analyze a real Canadian pattern of drug-related deaths.

8.1 The Drug, Set and Setting Framework

The framework holds that a drug experience is shaped by three interacting factors: the drug itself, with its pharmacological properties; the set, meaning the person's mindset, including expectations, intentions, mood and personality; and the setting, the physical and social environment in which use occurs (Hartogsohn, 2017). Figure 3.7 shows the three factors and some of the elements within each.

Drug dose, route,potency, half-life,receptor action Set expectations, mood,intentions,past learning Setting place, people, culture,laws and markets Experienceand risk
Figure 3.7. The drug, set and setting framework. The drug experience and its risks arise from the three factors acting together. The elements listed in each circle are examples drawn from this lesson; the framework follows Zinberg (1984) as described by Hartogsohn (2017).

The ideas are older than the phrase. Hartogsohn (2017) traces them to nineteenth-century European drug explorers, and credits Timothy Leary with introducing the phrase "set and setting", first published in a paper presented in 1961 and popularized in The Psychedelic Experience (1964), which he wrote with Ralph Metzner and Richard Alpert. Norman Zinberg of Harvard University extended the framework from psychedelics to drug use in general in his 1984 book Drug, Set, and Setting: The Basis for Controlled Intoxicant Use, which studied people who used intoxicants in a controlled way, following rules about their use (Hartogsohn, 2017). Zinberg argued that controlled use operates under social controls: informal rituals and sanctions that define appropriate use, condemn compulsive use and restrict use to settings that support safer experiences (Kleiman & Hawdon, 2011). Lesson 9 examines related evidence on why many people stop or moderate their use without treatment.

Hartogsohn (2017) describes the concept as fundamental to psychedelic research and notes that it has also been applied to alcohol, heroin and stimulants. In an opinion piece, Carhart-Harris et al. (2018) argued that classic psychedelics, which act as agonists at serotonin 2A receptors (Table 3.5), are exceptionally sensitive to context, and reported findings from their group that contextual factors predicted the quality of the acute experience, which in turn predicted later mental health outcomes. Volkow and Swanson (2003) made a parallel point about methylphenidate (Section 1, 1.2 Speed of Onset, Peak and Duration): its effects differ between the rituals of non-medical use and the demands of clinical use. Lesson 7 develops psychedelics, and HSCI 485 develops their therapeutic use. The next part examines one element of set that experiments can test directly: what a person expects a drug to do.

8.2 Expectancy and Placebo Responses

Expectancy is what a person anticipates a drug will do. A placebo is an inactive substance presented as active, and a placebo response is a change in experience or physiology produced by the expectation and context of treatment. Separating the effects of a drug from the effects of expecting it requires a particular design. Marlatt, Demming and Reid (1973) introduced the balanced placebo design, which crosses what participants are told they are drinking with what they actually drink (White et al., 2011). Table 3.7 shows its four groups.

Table 3.7. The four groups of the balanced placebo design for alcohol (Marlatt et al., 1973, as described by Miller, 1995, and White et al., 2011).

Given an alcoholic drinkGiven tonic water only
Told the drink contains alcoholDrug and expectancyExpectancy only
Told the drink contains tonic onlyDrug onlyNeither

Comparing the rows isolates the effect of expectancy; comparing the columns isolates the effect of alcohol itself. Marlatt and colleagues used the design to test "loss of control", the idea that alcohol itself triggers uncontrolled drinking in people with alcohol dependence. In both participants with alcohol dependence and social drinkers, the amount people drank during a disguised taste-rating task followed their belief that the beverage contained alcohol, whatever it actually contained, and the effect was larger in the participants with alcohol dependence (Miller, 1995; White et al., 2011). Expectancies are also cultural. Reviewing research since MacAndrew and Edgerton's Drunken Comportment (1969), Room (2001) described societies in which intoxication is treated as a "time out" from ordinary rules, and noted Canadian court decisions that treated extreme intoxication as potentially "akin to automatism."

Placebo responses can work through the same biological systems that drugs use. Case 3.5 describes a classic 1978 experiment that pointed to this.

Case 3.5: Placebo pain relief and the body's own opioids

Levine, Gordon and Fields (1978) studied patients after removal of impacted wisdom teeth. Three and four hours after surgery, patients received naloxone, the opioid antagonist introduced in Section 3, or a placebo, under randomized, double-blind conditions. Patients given naloxone reported more pain. Among patients who first received a placebo, some reported less or unchanged pain (placebo responders) and others reported more. Naloxone given afterward increased pain only in the placebo responders. The authors concluded that the release of endorphins, the body's own opioids, mediates placebo pain relief in this setting.

Case 3.5 links set to pharmacodynamics: an expectation of relief appears to have released endogenous ligands at opioid receptors, and blocking those receptors removed the relief. A review by Zamfira et al. (2026) links placebo pain relief to the opioid, dopamine and cannabinoid systems and describes current explanations as fragmented, so research on how placebo responses work is still evolving. The next part describes a second way in which learning changes a drug's effect, this time by opposing it.

8.3 Conditioned Tolerance and Overdose

Section 3 (7.2 Tolerance and Cross-Tolerance) listed conditioned tolerance as a form of learned tolerance. Its explanation comes from Pavlovian conditioning, in which a cue that reliably comes before an event begins to trigger a response in anticipation of that event. Shepard Siegel of McMaster University in Hamilton, Ontario, applied this idea to drugs. When a drug disturbs the body's normal state, the body responds with corrective, homeostatic changes. Siegel (1999) argued that the places, objects and routines that regularly accompany drug use become cues that trigger these corrective responses before the drug acts. The anticipatory responses offset part of the drug's effect, contributing to tolerance; when the cues are absent, the offset is missing and the same dose has a larger effect. Siegel described such failures of tolerance as one source of "enigmatic" overdoses. Figure 3.8 summarizes the mechanism.

Familiarsetting Cues predictthe drug Body begins acompensatoryresponse Smaller net effect(tolerance) Unfamiliarsetting No familiarcues No anticipatoryresponse Larger net effect;higher overdoserisk
Figure 3.8. Conditioned tolerance. The same dose has a smaller effect in a setting whose cues predict the drug, because the body begins a compensatory response in advance; in an unfamiliar setting that response is absent. Drawn for this lesson from Siegel (1999) and Dumas and Pollack (2008).

Case 3.6 describes a classic 1982 experiment that tested this idea in animals.

Case 3.6: Siegel's experiment on setting and overdose

Siegel, Hinson, Krank and McCully (1982) gave rats a history of heroin injections in one distinctive environment. The rats then received a high, potentially lethal dose of heroin, some in the environment associated with the drug and others, with an identical drug history, in an alternative environment. Rats that received the high dose in the presence of the familiar drug-associated cues were more likely to survive (Siegel et al., 1982; Siegel & Ellsworth, 1986). In a later account of the experiment, Siegel (2017) described mortality as twice as high in the rats that received their last dose in the unfamiliar environment. He also wrote that independent studies in New Jersey and Spain found that most survivors of heroin overdose reported having used the drug in a novel or unusual environment.

The evidence in Case 3.6 has limits. A short review by Farahbakhsh and Siciliano (2023) describes support from causal tests in rats and from overdose survivors' reports in people: the decisive experiments were in animals, and the human evidence comes from survivors' recollections. In British Columbia's unregulated supply of the 2020s, the variability in potency and the combinations of drugs described in Case 3.4 are major causes of overdose, so conditioned tolerance is best treated as one contributor among several. Its practical message is that tolerance is partly a property of a person in a place. The next part widens the idea of setting beyond the immediate surroundings.

8.4 Social Settings and the Risk Environment

Setting, in Zinberg's sense, includes the social norms around use. Public health research has widened it further. Tim Rhodes's risk environment framework treats drug-related harm as contingent on social context, arising from interactions between individuals and the environments in which they use drugs, so that reducing harm involves the environment as well as the individual (Rhodes, 2009). The tabs below sort examples from this lesson into physical, social, and economic and policy settings.

In British Columbia in 2025, 48 percent of unregulated drug deaths occurred in a private residence and 20 percent outdoors (BC Coroners Service, 2026). A person who uses in a private residence may have no one present to respond if they lose consciousness. As of October 2026, the National Overdose Response Service (1-888-688-6677) supports people who use drugs alone, and HSCI 402 develops services that change the physical setting of use, such as supervised consumption sites.

Peers teach novice users how to interpret a drug's effects (Hartogsohn, 2017), cultures shape how intoxicated people behave (Room, 2001), and controlled users follow informal rituals and sanctions (Kleiman & Hawdon, 2011). The shift from injecting to smoking in Vancouver was also shaped by social considerations, including the ease and discretion of smoking (Bonn et al., 2025).

An unregulated market supplies drugs of unknown and variable composition, as Case 3.4 showed (Larnder et al., 2022), so the drug itself becomes part of the risk environment. Clinical settings, in contrast, supply known doses under supervision (Volkow & Swanson, 2003). Lesson 11 examines how prohibition, policing and drug markets shape these conditions, and Lesson 12 examines decriminalization and legal regulation.

Interactive 3.2 asks you to classify examples from this lesson as drug, set or setting factors. Some factors involve more than one; choose the one that the example mainly describes.

Interactive 3.2
Drug, set or setting?

Choose one factor for each example, then select Check my choices.

The fentanyl concentration in a sample sold as an opioid differs from the previous sample
A person believes that the drink in front of them contains alcohol
A person uses alone in a private residence
Caffeine blocks adenosine receptors
A participant in a psilocybin study arrives anxious and unsure why they enrolled
A society treats drunkenness as a time out from ordinary rules
A student takes prescribed methylphenidate in a classroom that demands quiet, focused attention

The risk environment framework shows that setting includes the place of use, the people and norms around it, and the economic and policy conditions that determine what the drug contains. The final part of this section brings the lesson's four sections together.

9.1 Bringing the Lesson Together

The four sections of this lesson provide a sequence of questions that can be asked about any drug-related event. Table 3.8 lists them with the concepts that answer each question and an example from the lesson.

Table 3.8. Five questions for analyzing a drug effect or drug-related harm, the concepts from this lesson that answer them, and an example of each.

QuestionConceptsExample from this lesson
How did the drug enter the body?Route, onset, peak, reinforcementNicotine from a cigarette reaches the brain in 10 to 20 seconds
How much reached the target, and for how long?Bioavailability, metabolism, half-life, steady stateCaffeine's half-life is up to 50 percent shorter in people who smoke
What does it do at the target?Agonist, antagonist, potency, therapeutic indexFentanyl is 50 to 100 times more potent than morphine
How has experience changed the response?Tolerance, cross-tolerance, conditioned toleranceOpioid tolerance to pain relief develops faster than to respiratory depression
Who, where and under what conditions?Set, setting, risk environment48 percent of 2025 deaths in British Columbia occurred in private residences

The task card below applies all five questions to a real Canadian pattern of harm.

Try it 3.4: Applying this lesson to British Columbia's 2025 drug toxicity deaths

The BC Coroners Service (2026) reported, in preliminary data, 1,826 unregulated drug toxicity deaths in 2025, about 21 percent fewer than the 2,315 in 2024. Smoking was the most common mode of consumption (65 percent), followed by nasal insufflation (11 percent), injection (9 percent) and oral use (4 percent). Forty-eight percent of deaths occurred in a private residence and 20 percent outdoors. Among deaths with expedited toxicology testing, fentanyl was detected in 69 percent, fluorofentanyl (a fentanyl analogue) in 54 percent, cocaine in 53 percent and methamphetamine in 52 percent. Drug checking in Victoria found fentanyl concentrations in opioid samples with a median of 9 percent and a standard deviation of 7 percentage points (Larnder et al., 2022).

Using these figures, write a short analysis that answers the five questions in Table 3.8. (1) What does smoking's share of deaths show, and what can it not show? (2) Why does a rapid route matter for a highly potent drug? (3) Using potency, the dose-response curve and the safety ratio, explain why variable fentanyl concentration raises risk whatever the route. (4) How could tolerance, and its loss, change a person's risk? (5) How would the risk environment framework interpret the share of deaths in private residences? Finish by naming two pieces of information that the figures do not provide and that you would need to strengthen your analysis.

Points a strong analysis would includev

(1) Smoking accounted for about 1,187 deaths and injection about 164, but without the number of people or occasions of use by route, the figures cannot show which route is riskier per use (Section 1, 2.1 Mode of Consumption in British Columbia's Drug Toxicity Deaths). (2) Inhalation delivers drugs to the brain rapidly, so the peak effect, including slowed breathing with opioids, arrives with little time to respond (Section 1, 1.2 and 1.4). (3) Fentanyl's high potency means that small differences in amount produce large differences in effect; with concentrations varying around a median of 9 percent, a person cannot know where a dose falls on the dose-response curve, and an opioid with a narrow safety ratio leaves little margin (Section 3, 6.3 and 7.1). The frequent detection of more than one drug echoes Gable's finding that most lethal cases involve combinations. (4) Tolerance to pain relief and euphoria develops faster than tolerance to respiratory depression, which narrows the margin; tolerance can be partly specific to familiar settings; and tolerance falls after a period without use (Section 3, 7.2; Section 4, 8.3; and Lesson 5). (5) Deaths in private residences suggest that many people used where no one could respond; the risk environment framework directs attention to housing, stigma, policing and the availability of services that change the setting of use (Section 4, 8.4). Useful additional information would include the number of people who use each route, whether each person was alone, their recent tolerance, and the full toxicology for each death.

This section has shown that drug effects depend on the person's expectations and learning and on the physical, social and policy settings of use, as well as on the drug itself. Together with Sections 1 to 3, it gives you a set of questions to ask about any drug, which Lessons 4 to 7 apply to depressants, opioids, stimulants, cannabis and psychedelics. Complete the reflection and the knowledge check below, then continue to the final assessment, which draws on all four sections.

Key Takeaways

  • 8.1 The drug, set and setting framework holds that a drug experience arises from the drug, the person's mindset and the physical and social environment acting together. The ideas trace from early psychedelic research to Zinberg's study of controlled drug use, and Hartogsohn (2017) describes the concept as fundamental to psychedelic research.
  • 8.2 Expectancy can shape drug effects: the balanced placebo design separates the effect of believing a drink contains alcohol from the effect of alcohol itself. Placebo pain relief can work through the body's own opioids, as a 1978 experiment suggested when naloxone removed the relief in placebo responders.
  • 8.3 In conditioned tolerance, cues that predict a drug trigger a compensatory response that reduces its effect. Without familiar cues the same dose can be more dangerous, although the decisive evidence comes from animal experiments and conditioned tolerance is one contributor among several to overdose.
  • 8.4 The risk environment framework treats drug-related harm as arising from physical, social, economic and policy settings. Using alone in a private residence, the norms of a peer group and an unregulated market of unknown composition all change the risk of the same dose.
  • 9.1 Five questions can be asked of any drug-related event: how the drug entered the body, how much reached the target and for how long, what it does at the target, how experience has changed the response, and who used it, where and under what conditions. Applied to British Columbia's 2025 drug toxicity deaths, they draw on every section of the lesson.

Reflection

A hypothetical classmate says: “Drug effects are just chemistry. The same dose of the same drug does the same thing to everyone.” Using at least three of these studies from Section 4, write a 150 to 250 word response that explains how set and setting change drug effects and what this implies for preventing harm: the balanced placebo design for alcohol (Marlatt et al., 1973), in which belief about a drink’s content was separated from its actual content; the study in which naloxone increased pain only in patients whose pain had been relieved by a placebo (Levine et al., 1978); the experiment in which rats given a large heroin dose in a setting not paired with heroin were less likely to survive than rats dosed in the familiar setting, a death rate Siegel later described as about twice as high (Siegel et al., 1982); and research arguing that psychedelic experiences are especially sensitive to context (Carhart-Harris et al., 2018).

Model answer

Chemistry matters, but these studies indicate that the person and the place can change what a dose does. In the balanced placebo design, Marlatt and colleagues separated what people believed they were drinking from what they actually drank, and the amount people drank followed their belief that the drink contained alcohol. Expectancy is part of the effect.

Expectation can also act through the body’s own chemistry. Levine et al. (1978) found that naloxone increased pain only in patients whose pain had been relieved by a placebo, which suggests that the placebo relief depended on endorphins acting at opioid receptors.

Learning shapes tolerance as well. In Siegel et al. (1982), rats given a large heroin dose in an unfamiliar setting were less likely to survive than rats given the same dose where they usually received heroin, and Siegel later described the death rate as about twice as high. The familiar cues appear to trigger compensatory responses that offset part of the drug’s effect, so the same dose may be more dangerous in a new place, although this evidence comes mainly from animal experiments and survivors’ reports.

For prevention, this means that advice and services cannot focus only on the substance. People need to know that tolerance can be weaker in unfamiliar settings, that expectations shape experiences, and that the setting of use, including whether someone else is present, affects whether harm becomes fatal.

Minimum 20 characters required.

✓ Reflection saved
Knowledge Check: this section

Question 1: In the balanced placebo design used by Marlatt, Demming and Reid (1973), how is the effect of expectancy isolated?

The design has four groups formed by crossing the instruction (told alcohol or tonic) with the beverage (alcohol or tonic). Comparing instructions within each beverage isolates expectancy. Varying only the amount of alcohol would measure a dose-response relationship and leave expectancy unchanged.

Question 2: Levine, Gordon and Fields (1978) gave naloxone to patients after dental surgery. Why was their finding important?

Naloxone, an opioid antagonist, increased pain only in patients whose pain had been relieved by a placebo, which suggests that placebo relief depended on endorphins. Naloxone is not a pain reliever; it blocks opioid receptors.

Question 3: In Siegel et al. (1982), rats given a large heroin dose in a setting not paired with heroin were less likely to survive than rats dosed in the familiar setting; Siegel later described the death rate as about twice as high (Siegel, 2017). Which explanation fits the findings?

According to Siegel’s conditioning account, cues that regularly precede the drug trigger anticipatory compensatory responses, which contribute to tolerance. Without those cues the offset is missing and the same dose has a larger effect. Both groups had the same drug history, so differences in total heroin cannot explain the result.

Question 4: What does the risk environment framework of Rhodes (2009) add to the study of drug-related harm?

Rhodes describes harm as contingent on social context, arising from interactions between individuals and the environments in which they use drugs. Ranking drugs by the ratio of lethal to usual doses describes Gable’s safety ratio, a different tool.
Section 5 of 5

Final Assessment

⏱ Estimated time: 25 minutes

Bringing It All Together

This lesson has followed a drug from the moment it enters the body to the moment its effect is felt, and has shown that each step changes what a dose does. The route of administration sets how fast a drug reaches the brain and how high its level climbs; rapid routes such as smoking deliver nicotine to the brain in seconds and, for nicotine, produce the most reinforcing form of use, while slow routes such as edibles invite repeated dosing. Pharmacokinetics determines how much of the drug reaches its target and for how long: bioavailability and first-pass metabolism reduce what arrives, liver enzymes and their genes make people differ, interactions raise or lower levels, and the half-life governs both the fall in a drug’s level and its accumulation with repeated doses.

Pharmacodynamics describes what the drug does at its target. Agonists, partial agonists, antagonists and allosteric modulators act in different ways, and dose-response curves separate potency from efficacy. The therapeutic index and the safety ratios compiled by Gable (2004) summarize the margin between wanted and harmful effects, but they omit the slopes of the curves, combinations of drugs, accumulation, tolerance and uncertainty about the dose, all of which matter in the unregulated supply. Tolerance shifts the curves, and it shifts them at different rates for different effects.

Finally, the person and the setting change drug effects as well. Expectancy and placebo responses can act through the body’s own chemistry, conditioned tolerance can make the same dose more dangerous in an unfamiliar place, on evidence that comes mainly from animal experiments and survivors’ reports, and the risk environment of housing, markets and policy shapes who is harmed. British Columbia’s 2025 data on drug toxicity deaths show all of these factors at work together.

Key Takeaways from this lesson

  1. The route of administration determines the path a drug takes to the brain, the speed of onset, the height of the peak and the duration of effect. Rapid delivery to the brain, as with smoking or injection, appears to strengthen reinforcement and the potential for addiction, although why speed matters is still debated, while slow routes can lead to repeated dosing.
  2. A route’s share of deaths in surveillance data reflects how many people use it as well as its risk per occasion, so risk comparisons need denominators.
  3. Bioavailability is the fraction of a dose that reaches the general circulation; first-pass metabolism in the liver reduces it for swallowed drugs, and distribution then carries the drug to the tissues and across the blood-brain barrier.
  4. Liver enzymes, their genes and interacting substances change drug levels, as codeine and CYP2D6, caffeine and smoking, and grapefruit and CYP3A4 show.
  5. The half-life describes how quickly a drug is eliminated, and repeated doses approach a steady state after about four to five half-lives, so long half-lives allow slow accumulation.
  6. Agonists activate receptors, partial agonists activate them less fully, antagonists block them and allosteric modulators change the response to the natural messenger. Potency is the amount of drug needed for an effect and efficacy is the maximum effect; a highly potent drug turns small errors in amount into large changes in effect.
  7. Therapeutic indexes and safety ratios summarize a margin of safety but omit curve slope, combinations, accumulation, tolerance and uncertainty about the dose. Tolerance shifts the curves at different rates for different effects, which can narrow the margin.
  8. Expectancy, conditioned tolerance and the physical, social and policy setting of use can change both the effects and the risks of the same dose.
  9. Asking how a drug entered the body, how much reached its target and for how long, what it did there, how experience changed the response, and who used it, where and under what conditions brings route, pharmacokinetics, pharmacodynamics, set and setting together, as British Columbia’s 2025 drug toxicity deaths show.

Core Concepts Reviewed

Section 1: routes of administration, onset, peak and duration, the concentration-time curve, reinforcement and the speed of delivery, and the interpretation of deaths by mode of consumption.

Section 2: absorption, bioavailability and first-pass metabolism, distribution and the blood-brain barrier, metabolism, prodrugs and cytochrome P450 enzymes, genetic variation in CYP2D6, drug interactions, half-life and steady state.

Section 3: receptors and ligands, affinity and efficacy, agonists, partial agonists, antagonists and allosteric modulators, dose-response curves, potency, the therapeutic index and safety ratio, and tolerance and cross-tolerance.

Section 4: drug, set and setting, expectancy and placebo responses, the balanced placebo design, conditioned tolerance and overdose, and the risk environment framework.

The final reflection asks you to apply all four sections of the lesson to a single familiar drug.

Reflection

Use caffeine to show how the four sections of this lesson fit together. Caffeine is usually swallowed in drinks; it is metabolized mainly by the liver enzyme CYP1A2; its average adult half-life is about 5 hours, up to 50 percent shorter in people who smoke and up to 15 hours late in pregnancy; and it acts by blocking adenosine receptors (Evans et al., 2024). In 200 to 300 words, explain (a) what its usual route implies for the speed of its effect compared with an inhaled drug, (b) how much of a hypothetical 100 mg dose, fully absorbed, would remain 10 hours later for a typical adult, a person who smokes (half-life 2.5 hours) and a person late in pregnancy (half-life 15 hours), (c) what type of drug action caffeine has at its receptor and what that means, and (d) how a study could test whether expectancy contributes to the alertness people report after coffee.

Model answer

(a) Caffeine is usually swallowed, so it must be absorbed through the stomach and intestine and pass through the liver before reaching the general circulation. Its effect therefore builds more slowly than that of an inhaled drug, which reaches the brain from the lungs within seconds to minutes. A slow rise also tends to make a drug less reinforcing than the same drug delivered rapidly.

(b) For a typical adult, 10 hours is two half-lives, so 100 × 0.25 = 25 mg remains. For a person who smokes, smoking induces CYP1A2 and the half-life may fall to 2.5 hours, so 10 hours is four half-lives and about 6 mg remains. Late in pregnancy, with a half-life of 15 hours, 100 × (½)10 ÷ 15 ≈ 63 mg remains, ten times as much as for the person who smokes. Repeated cups would also accumulate toward a steady state after four to five half-lives, which takes much longer when the half-life is long.

(c) Caffeine is an antagonist: it binds adenosine receptors without activating them and blocks adenosine, the body’s own messenger, from acting there. Its effect therefore depends on how much adenosine signalling is under way.

(d) A balanced placebo design would cross what participants are told (caffeinated or decaffeinated coffee) with what they actually receive, in four groups. Comparing people told “caffeinated” with people told “decaffeinated” within each beverage would isolate the effect of expectancy, and comparing the beverages would isolate the effect of caffeine itself.

Minimum 30 characters required.

✓ Reflection saved

Final Knowledge Assessment

Final Assessment, this lesson: How Drugs Work: Routes, Pharmacokinetics and Pharmacodynamics (15 Questions)

Question 1: A person swallows a hypothetical 50 mg dose of a drug with an oral bioavailability of 0.4 and a half-life of 4 hours. Assuming immediate absorption, about how much remains in the body 8 hours later?

The amount absorbed is 50 × 0.4 = 20 mg. Eight hours is two half-lives, so 20 × 0.25 = 5 mg remains. The answer of 25 mg applies two half-lives to the whole dose and ignores bioavailability.

Question 2: Which pairing of a route with a pharmacokinetic consequence is correct?

A swallowed drug travels from the intestine to the liver through the portal vein, so part of it can be metabolized before reaching the general circulation. Intravenous, inhaled and sublingual routes all bypass this first pass.

Question 3: Why can the same oral dose of codeine be ineffective for one person and toxic for another?

Codeine is a prodrug. Poor metabolizers form little morphine and get little relief; ultrarapid metabolizers form more morphine and risk toxicity. The variation lies in the activity of the enzyme, which is why the first option, about receptors, is wrong.

Question 4: Which statement correctly distinguishes pharmacokinetics from pharmacodynamics?

Pharmacokinetics covers absorption, distribution, metabolism and elimination; pharmacodynamics covers drug action at receptors and the relation between dose and effect. The first option reverses the two.

Question 5: A person with long-term heroin use starts methadone treatment and experiences little euphoria from methadone. Which concept best explains this?

Dumas and Pollack (2008) describe cross-tolerance, in which tolerance to one drug reduces the response to others in the same class, as the principle behind methadone treatment. Methadone is an agonist, so competitive antagonism does not apply.

Question 6: Injected naloxone has a half-life of 30 to 80 minutes, shorter than that of methadone or fentanyl. Why are people who respond to naloxone monitored for hours?

This combines Sections 2 and 3: naloxone is a competitive antagonist with high affinity, but its shorter half-life means that its blocking effect can end before the opioid has been eliminated (Jordan et al., 2024). The first option is wrong because naloxone is a pure antagonist with no agonist activity.

Question 7: Two hypothetical drugs both have a therapeutic index of 10, but one has much shallower dose-response curves. Which statement is correct?

The index compares only the median doses. With shallow curves, the dose that produces the desired effect in 95 percent of people overlaps much more with the lethal curve (about 38 percent lethal in Worked Example 3.3, compared with about 2 percent for steep curves). Equal indexes do not mean equal safety.

Question 8: Gable (2004) estimated safety ratios of 6 for intravenous heroin and 10 for alcohol. Which limitation did he himself note?

Gable noted that most published lethal cases involved a co-intoxicant, most often alcohol, and that his ratios assume a single substance in a person without tolerance. The ratios describe acute toxicity only, so they leave out long-term harms.

Question 9: Smoking, intranasal use and intravenous injection share which pharmacokinetic feature?

None of these routes passes the drug through the portal vein to the liver first. Only intravenous injection guarantees a bioavailability of 100 percent; inhaled and intranasal drugs are absorbed incompletely, as Table 3.1 shows for nicotine.

Question 10: A person has always used an opioid in the same room and then takes the same dose in an unfamiliar place. According to the conditioning account, why might the effect be stronger?

Siegel (1999) argued that cues that regularly accompany use trigger anticipatory compensatory responses that contribute to tolerance; in a new place the cues and the responses are absent. Genotype does not change with the environment.

Question 11: Which example describes an allosteric modulator?

An allosteric modulator binds a site distinct from the messenger’s site and changes the receptor’s response; diazepam increases how often the GABA-A channel opens. Naloxone and caffeine are antagonists, and buprenorphine is a partial agonist.

Question 12: A nicotine patch takes 3 to 12 hours to reach its peak, while nicotine from a cigarette reaches the brain in 10 to 20 seconds. Which statement is best supported by the lesson?

Benowitz et al. (2009) concluded that the rapid rise during smoking makes it the most reinforcing and dependence-producing form of nicotine, consistent with Samaha and Robinson (2005). Duration of exposure does not determine reinforcement in the way that speed does, which rules out the first option.

Question 13: Which statement correctly matches a substance with a factor that changes its half-life?

Zevin and Benowitz (1999) describe the induction of CYP1A2 by chemicals in tobacco smoke, and caffeine’s half-life is up to 50 percent shorter in people who smoke (Evans et al., 2024). Alcohol follows roughly zero-order elimination after the first few drinks, and diazepam has a long half-life and an active metabolite lasting about 100 hours.

Question 14: Which statement about the Toronto codeine case report (Koren et al., 2006) is accurate as of October 2026?

The Lancet published an expression of concern in February 2026 after new allegations about the toxicological data, and the PubMed record lists no retraction. The report was questioned in a 2008 letter and in a 2020 review, but it was not retracted in 2008.

Question 15: Which combination of concepts best explains why variable fentanyl concentration in the unregulated supply raises the risk of overdose?

Fentanyl is 50 to 100 times more potent than morphine, and Larnder et al. (2022) found concentrations varying around a median of 9 percent, so a person cannot know where a dose falls on the dose-response curve. Fentanyl is an agonist without the ceiling effect of a partial agonist.
✦ Complete the final reflection above before submitting

Congratulations!

You have successfully completed this lesson: How Drugs Work: Routes, Pharmacokinetics and Pharmacodynamics.

You can now explain how the route of administration, absorption, metabolism, elimination and receptor action together determine what a dose of a drug does, calculate bioavailable doses, amounts remaining after a given time and therapeutic indexes, and explain why expectancy, learning and setting change drug effects and harms.

Lesson 4 applies this pharmacology to alcohol and other depressants: how alcohol acts on the brain and is metabolized, how blood alcohol concentration is estimated, the uses and risks of benzodiazepines and related drugs, and why combining depressants with opioids suppresses breathing.

Continue to Lesson 4 →
Reference

Glossary: Key Terms, People & Frameworks

📚 Reference page, available throughout the lesson

Definitions of the pharmacological terms, frameworks and researchers introduced in this lesson, in the order of the sections where they first appear.

Core Concepts
Route of administration The way a drug enters the body, such as swallowing, smoking or vaping, intranasal use or injection. The route shapes how fast the drug reaches the brain and how long it acts.
Concentration-time curve A graph of the amount of drug in the blood against the time since it was taken, used to describe onset, peak and duration.
Reinforcement Any consequence that makes a behaviour more likely to be repeated. Drugs that reach the brain rapidly tend to reinforce their use more strongly.
Pharmacokinetics What the body does to a drug: its absorption, distribution, metabolism and elimination over time.
Bioavailability The fraction of a dose that reaches the general circulation unchanged. It is 100 percent for intravenous injection and usually lower for swallowed drugs.
First-pass metabolism The breakdown of part of a swallowed dose by the liver before it reaches the rest of the body, because blood from the intestine passes through the liver first.
Blood-brain barrier The tightly joined cells lining the brain’s blood vessels, which limit which substances enter the brain. Fat-soluble drugs such as heroin cross it readily.
Prodrug An inactive or weakly active compound that the body converts into an active drug. Codeine is converted to morphine by the enzyme CYP2D6.
Cytochrome P450 enzymes A family of liver enzymes, with names such as CYP2D6 and CYP3A4, that carry out most of the first phase of drug metabolism. Genetic variation divides people into poor, intermediate, normal and ultrarapid metabolizers.
Enzyme inhibition and induction Inhibition slows an enzyme so that the drugs it metabolizes build up (grapefruit and CYP3A4); induction increases its activity so that they are removed faster (smoking and CYP1A2).
First-order and zero-order elimination In first-order elimination a constant fraction of the drug is removed per unit of time; in zero-order elimination, as with alcohol after the first few drinks, a constant amount is removed.
Half-life The time it takes for the concentration of a drug to fall by half. After four to five half-lives, 94 to 97 percent of a drug has been eliminated.
Steady state The condition reached with repeated dosing when the amount eliminated between doses equals the amount taken, approached after about four to five half-lives.
Pharmacodynamics What a drug does to the body: where it acts, the response it produces and how the response changes with dose and repeated use.
Receptor and ligand A receptor is a protein that responds to a particular chemical messenger by changing a cell’s activity; a ligand is any molecule that binds to it, including the body’s own messengers and drugs.
Affinity and efficacy Affinity is the strength of the bond between a drug and its receptor; efficacy is the size of the response the drug produces once bound.
Agonist A drug that binds to a receptor and activates it, producing a biological effect. Morphine is an agonist at the mu-opioid receptor.
Partial agonist A drug that binds to a receptor and produces a weaker maximal effect than a full agonist. Buprenorphine has a ceiling effect on respiratory depression.
Antagonist A drug that binds to a receptor and blocks its activity without activating it. Naloxone is a competitive antagonist at opioid receptors; caffeine blocks adenosine receptors.
Allosteric modulator A drug that binds to a site on a receptor separate from the messenger’s site and changes the receptor’s response. Diazepam enhances the action of GABA in this way.
Dose-response curve A graph of a drug’s effect against its dose, either for one person (how much effect) or for a population (the percentage who respond), from which the median effective dose (ED50) is read.
Potency The amount of drug needed to produce a given effect. Fentanyl is 50 to 100 times more potent than morphine, so much less is needed for the same effect.
Therapeutic index The median toxic or lethal dose divided by the median effective dose (TD50 or LD50 divided by ED50). A narrow index means a small margin of safety.
Tolerance A decrease in a drug’s effect after repeated or prolonged use, which can be pharmacokinetic, pharmacodynamic or learned, and can develop at different rates for different effects.
Cross-tolerance A reduced response to one drug that results from tolerance to another drug of the same class, the principle behind methadone treatment for people who use heroin.
Conditioned tolerance Tolerance produced by learning, in which cues that regularly accompany drug use trigger compensatory responses that offset part of the drug’s effect.
Expectancy and placebo response Expectancy is what a person anticipates a drug will do; a placebo response is a change produced by the expectation and context of treatment.
Frameworks & Tools
ADME The four processes of pharmacokinetics: absorption, distribution, metabolism and elimination.
Safety ratio Gable’s (2004) ratio of the usual lethal dose of a substance to the dose most commonly used for non-medical purposes, ranging from 6 for intravenous heroin to 1,000 or more for LSD and smoked cannabis.
CPIC guidelines Guidelines of the Clinical Pharmacogenetics Implementation Consortium that translate genetic test results into prescribing advice, such as avoiding codeine for CYP2D6 ultrarapid and poor metabolizers.
Drug, set and setting The framework that explains a drug experience as the product of the drug’s pharmacology, the person’s mindset and the physical and social environment of use.
Balanced placebo design An experimental design that crosses what participants are told they receive with what they actually receive, separating the effects of expectancy from those of the drug.
Risk environment framework Tim Rhodes’s framework, which treats drug-related harm as arising from interactions between individuals and the physical, social, economic and policy environments of use.
Key People
Anne-Noël Samaha and Terry Robinson Researchers who in 2005 reviewed why rapid delivery of drugs to the brain promotes addiction and proposed, in an explanation that is still debated, that it more readily produces lasting changes in motivation circuits.
Robert Gable Researcher who in 2004 compared the acute lethal toxicity of 20 psychoactive substances using safety ratios and set out the limits of such comparisons.
Shepard Siegel Psychologist at McMaster University in Hamilton, Ontario, whose research on Pavlovian conditioning, mainly in rats, indicated that drug-associated cues contribute to tolerance, and who proposed that their absence contributes to some overdoses in unfamiliar settings.
Norman Zinberg Harvard researcher whose 1984 book Drug, Set, and Setting extended the framework from psychedelics to drug use in general, through the study of controlled intoxicant use and its social controls.
Timothy Leary Psychologist credited with introducing the phrase “set and setting” in 1961 and popularizing it in The Psychedelic Experience (1964), written with Ralph Metzner and Richard Alpert.
G. Alan Marlatt Psychologist who, with Barbara Demming and John Reid, introduced the balanced placebo design for alcohol research in 1973.
No matching entries. Try a different search term.