The Pharmacology of Coming Off an Antidepressant Slowly

The Pharmacology of Coming Off an Antidepressant Slowly

Coming off an antidepressant is harder at the end of the taper than at the beginning because the drug's effect on the brain does not track its dose in a straight line. Selective serotonin reuptake inhibitors bind the serotonin transporter along a hyperbolic curve, so the last few milligrams do far more pharmacological work than the first few. Cutting a dose by the same number of milligrams at each step therefore delivers progressively larger jumps in effect as the dose falls, which is the mechanistic case for reducing by shrinking proportions rather than fixed amounts. The receptor-occupancy math behind this is well established; the clinical question of how slow a taper actually needs to be is still being answered.

Why occupancy, not dose, is the quantity that matters

An antidepressant does not act in proportion to the number printed on the tablet. It acts in proportion to how much of its target it engages. For SSRIs the relevant target is the serotonin transporter, and the fraction of transporters a given dose blocks is what positron emission tomography studies actually measure. That fraction, called occupancy, rises steeply at the lowest doses and then flattens. In their 2019 Personal View in The Lancet Psychiatry, Mark Horowitz and David Taylor assembled these imaging data and showed that the dose-occupancy relationship follows a Michaelis-Menten curve, the same saturating hyperbola that describes an enzyme approaching its maximum rate.

The practical consequence is counterintuitive. Near a standard dose the transporter is already close to saturated, so removing a substantial amount of drug changes occupancy only a little. Near zero the curve is almost vertical, so removing a tiny amount changes occupancy a great deal. A milligram matters far more at the bottom of the range than at the top.

The hyperbola, in numbers

Horowitz and Taylor's synthesis puts figures on this. About 20 mg of citalopram occupies roughly 80% of serotonin transporters. Drop to 5 mg and occupancy is still near 60%; even 1 mg holds something over 20%. So the move from 20 mg to 10 mg, which looks like halving the dose, subtracts only a few percentage points of occupancy, while the move from 5 mg to zero crosses most of the biologically active range.

To make each step feel the same to the brain, the paper models reductions that remove an equal slice of occupancy rather than an equal slice of dose. Holding the occupancy change constant means the dose decrements have to shrink as the dose falls, so the schedule descends in ever-smaller steps toward very small final doses rather than in fixed amounts. That curve-shaped, shrinking pattern is what the phrase hyperbolic tapering names. This is an illustration of the pharmacology, not a prescription; any change to a real regimen is a clinical decision.

What the tapering-rate evidence now shows

Mechanism is a hypothesis until outcomes test it, and the outcome data have started to arrive. A 2023 prospective cohort in Therapeutic Advances in Psychopharmacology followed 608 people tapering their antidepressants in small, graduated daily steps, collecting nearly 33,000 daily withdrawal ratings over trajectories of 28 to 140 days. Withdrawal was limited overall and, tellingly, inverse to the rate of taper: slower reductions produced less withdrawal than faster ones. Younger age and female sex were associated with more withdrawal, which the authors read as an argument for individualized, monitored tapering rather than a single fixed algorithm.

A large controlled synthesis arrived in 2025, when a network meta-analysis in The Lancet Psychiatry pooled 76 randomized trials and 17,379 adults with remitted depression or anxiety. Gradual tapering combined with psychological support prevented relapse about as well as staying on the drug, while fast tapering of four weeks or less performed no better than stopping abruptly. The direction of the cohort and the trial evidence agrees: slower is gentler.

What the evidence does not yet establish

Direction is not the same as precision, and the honest account has to widen here. The 2025 network meta-analysis measured relapse, not withdrawal symptoms, and its authors noted that withdrawal and quality-of-life data were largely missing and that the certainty for the psychotherapy comparisons was low. The 2023 cohort was naturalistic rather than randomized, so it cannot cleanly separate the taper shape from the expectations of the people who chose it.

The underlying withdrawal literature is shakier still. A 2025 appraisal and reanalysis found that in most efficacy trials withdrawal was captured only through spontaneously reported side effects rather than systematic checklists, that prior antidepressant exposure averaged under six months, and that only a handful of studies measured withdrawal properly at all. The people most likely to struggle, long-term users stopping after years, are the least represented in the trials.

So the strong claim, that a specifically hyperbolic schedule outperforms every other slow taper for those patients, has mechanistic plausibility and supportive but incomplete outcome data, not settled proof. That gap is the useful place to stand. The receptor math is real and explains why equal dose cuts do not produce equal effects. The clinical corollary, that tapers should generally be slow, flexible, and monitored, is well supported in direction, while the exact curve, the right duration, and the size of the benefit for long-term users remain questions current trials were not built to answer. This article is educational and is not medical advice.