Titration: The Start-Low-Go-Slow Principle

Why the start-low-go-slow principle isn't just a catchphrase

Roughly 46% of adverse drug reactions recorded in Australian hospitals are dose-related. That single figure, drawn from the Australian Commission on Safety and Quality in Health Care's published medication safety data, sits in my head every time I read someone asking why they can't just take "a normal amount" of something from day one. The answer is titration, and the start-low-go-slow principle behind it deserves to be understood properly, not just repeated as a slogan.

Titration, in a pharmacological sense, means adjusting the dose of a substance incrementally over time until the desired response is reached, the point at which adverse effects appear, or some pre-defined endpoint. The word itself comes from analytical chemistry: titrating a solution meant adding a reagent drop by drop until a reaction occurred. The medical borrowing is apt. You're looking for a threshold, not a target written on a packet.

A 1998 Australian Prescriber article on opioid titration in palliative care uses almost identical language to guidelines published in 2023. That is a useful reminder. The underlying pharmacology hasn't changed. What changes is the clinical context, the substance, and the individual.

What titration actually involves

The core structure is simple. A starting dose is chosen, typically low relative to the anticipated therapeutic range. After a defined interval (which varies considerably by substance and context), the dose is either held at the same level, increased modestly, or occasionally reduced. This continues until the prescriber and patient agree an appropriate response has been reached, or until tolerability becomes the limiting factor.

Each step is a small experiment. The prescriber changes one thing, waits, and observes. Changing several variables at once, such as the dose, the timing and the formulation, makes it very hard to tell which change produced which effect. That is one of the most common mistakes in any dose-adjustment process, and a slow, single-variable approach is designed to avoid it.

The idea also depends on the shape of the dose-response relationship. For many substances the curve is not a straight line. A small increase at one point may add little, while the same increase at another point may shift the effect considerably. Small steps give the clinician a chance to see where on the curve a person actually sits, instead of guessing.

Intervals, pharmacokinetics and half-life

Intervals matter. A substance with a half-life of a few hours behaves very differently from one that accumulates in fatty tissue over days. Pharmacokinetics, the study of how a body absorbs, distributes, metabolises, and excretes a compound, underpins every titration schedule. You cannot sensibly titrate without knowing roughly how long a substance persists, where it goes, and how it exits.

The half-life is the time it takes for the concentration of a substance in the body to fall by half. It gives a rough guide to how long to wait before judging the effect of a given step. If the next increase arrives before the previous dose has settled, the prescriber is effectively reading a moving target. The related concept is covered in more detail in Half-Life and Dosing Frequency (concept).

Two related concepts often come up here. Bioavailability, the proportion of a dose that reaches systemic circulation, varies by route of administration, formulation, and individual physiology. And first-pass metabolism can substantially reduce effective bioavailability for orally administered substances, because a significant fraction is metabolised by the liver before reaching the bloodstream. Both factors mean that a 10 mg oral dose and a 10 mg inhaled dose of the same compound are not equivalent. Titration schedules account for this, or should. For a fuller treatment, see Bioavailability and the First-Pass Effect.

Titration and the endocannabinoid system

Cannabinoid-based medicines are a particularly instructive case study for titration, and not just because they're relatively new to the regulated prescription landscape in Australia. The endocannabinoid system is highly individual. Receptor density, baseline endocannabinoid tone, and genetic variation in metabolic enzymes (especially CYP2C9 and CYP3A4, which process many cannabinoids) differ considerably between people. There is no standard dose that will produce a predictable response across all individuals. The wider physiology is outlined in The Endocannabinoid System in Whole-Body Physiology.

Tetrahydrocannabinol (THC) illustrates this particularly clearly. THC binds with high affinity to CB1 receptors, which are densely expressed in the brain and central nervous system. Its psychoactive effects are dose-dependent and highly variable between individuals. Someone with lower CB1 receptor density, or with faster hepatic metabolism of THC to its active metabolite 11-hydroxy-THC, will have a qualitatively different experience at the same nominal dose than someone with slower metabolism or higher receptor density. Starting low allows a person, under medical supervision, to identify where their individual threshold sits before reaching doses that may produce unwanted effects.

Cannabidiol (CBD) has a different pharmacological profile. It does not bind directly to CB1 or CB2 receptors with high affinity, and its dose-response curve appears more complex. Some research (including work published in Frontiers in Pharmacology) describes a biphasic response for certain endpoints, meaning effects may differ at lower versus higher doses. This further underlines why titration, not a fixed dose, is the framework prescribers use.

How Australian regulation frames dose management

In Australia, medicinal cannabis products containing THC are listed under Schedule 8 (Controlled Drug) of the Poisons Standard, whilst CBD above certain concentrations sits under Schedule 4 (Prescription Only). Access is via prescription, and depending on the product and indication, through either the TGA's Special Access Scheme or an Authorised Prescriber arrangement. These regulatory frameworks exist partly because dose management of these substances requires clinical oversight, which is, in itself, an argument for titration being a medical conversation rather than a self-directed one.

The scheduling logic is worth a closer look if you're new to it. Schedules in the Poisons Standard group substances by the level of control considered appropriate, and the distinction is explained in Schedule 4 vs Schedule 8: What They Mean. The regulator behind the system is covered in The Therapeutic Goods Administration (TGA) Explained, and the TGA website remains the primary reference for current requirements.

Clinical oversight also gives a safety net. A prescriber can recognise an adverse drug reaction early, distinguish it from an unrelated symptom, and adjust the plan. Someone adjusting a dose alone has none of that context, and I'd argue it is the strongest practical reason the "just try a bit more" instinct is a poor guide.

Formulation and its effect on titration

The form a substance takes affects how it should be titrated, sometimes dramatically. Oils and tinctures typically allow for fine-grained dose adjustments. Capsules and tablets come in fixed amounts, which can make small incremental changes harder. Inhalation, where absorption is rapid and peak concentration is reached within minutes, produces a faster onset but also a faster offset, which changes the practical meaning of "steady state" compared to oral administration.

For cannabinoids specifically, the distinction between full-spectrum, broad-spectrum, and isolate formulations is relevant here. Full-spectrum products contain a range of cannabinoids and terpenes alongside the primary compound, and the interactions between these components may alter the overall dose-response in ways that aren't fully characterised. These interactions are sometimes discussed under the heading of the entourage effect, though honestly that term carries more commercial baggage than clinical evidence at this stage. Titrating with a consistent, standardised extract is generally considered more reliable than titrating with variable batches, for exactly that reason.

The practical point is consistency. If the product changes mid-schedule, the earlier steps tell you less about the later ones. Batch variation, a switch of route, or a change from one formulation type to another can each reset the question the titration was trying to answer.

Titration in other contexts: peptides and beyond

The start-low-go-slow principle applies well beyond cannabinoids. GLP-1 receptor agonists, a class of compounds that includes several prescription medicines for type 2 diabetes and weight management, are almost always initiated at low doses and increased on a structured schedule. The published prescribing information for these compounds includes specific titration steps precisely because gastrointestinal adverse effects are dose-dependent and substantially more common if uptitration is rapid.

Peptides add another layer. Their persistence in the body can differ widely depending on structure and modification, which is why peptide half-life matters when a schedule is designed. Approved products have their steps set out through the evidence and review behind the product's registration, and the process by which that evidence is built is described in How Clinical Trials Move from Lab to Patient.

A note on unapproved peptide products marketed outside the prescription framework: the TGA has been explicit that supplying peptides for performance or anti-ageing purposes without appropriate approval is illegal in Australia, and enforcement activity has increased in this space. Any discussion of titration for such products is therefore separate from any question of whether a person should be using them at all, and for unapproved peptides, the TGA's position is that they should not.

Individual variation: the reason titration isn't optional

I'd push back gently on the framing I sometimes see that treats titration as excessive caution, a kind of bureaucratic hesitancy when the "right" dose is obvious. It's not a bureaucratic habit. It's a direct acknowledgement that human pharmacology is variable in ways that population-level studies don't fully capture.

Age, body composition, liver function, co-administered medicines, genetic polymorphisms in metabolic enzymes, and the state of a person's own endocannabinoid tone (including baseline levels of compounds like anandamide) all influence how a given dose behaves in a given person at a given time. None of these factors are visible on a label. Titration is how a prescriber, working with the person in front of them, finds the relevant dose for that individual, rather than applying a population average and hoping it fits.

Evidence quality matters here too. Trial averages describe groups, and a group result can hide wide differences between the individuals inside it. That is not a flaw in the research so much as a limit on what an average can tell anyone about a single person, and it is why a careful schedule keeps checking against the individual's actual response.

The short version is: population pharmacology tells you where to start. Titration tells you where to stop.

Sources

• Guidance documents for medicinal cannabis; Therapeutic Goods Administration (TGA)
• An Update on Safety and Side Effects of Cannabidiol: A Review of Clinical Data and Relevant Animal Studies, NCBI / Cannabis and Cannabinoid Research
• Medication Safety, Australian Commission on Safety and Quality in Health Care
• Medicinal cannabis: pharmacology and prescribing; Australian Prescriber

Tom Whitfield, Health librarian, references & fact-checking