Plasma concentration and clinical effect do not rise and fall in lock-step — the drug must first cross into the "biophase" (the effect site: brain, synapse, receptor-rich tissue). This delay is modelled with a hypothetical, unmeasurable effect compartment linked to plasma by a first-order rate constant ke0 (Sheiner–Hull link model):
Plasma (1-compartment IV bolus): Cp(t) = Cp0 · e^(−ke·t), ke = ln2 / t½,elim
Effect-site equilibration: dCe/dt = ke0 · (Cp(t) − Ce(t)), ke0 = ln2 / t½,ke0
Pharmacodynamic response (Emax): E(t) = 100 · Ce(t) / (EC50 + Ce(t))
This 2D view renders the same equations as a physical two-tank diffusion analogy: the plasma tank's liquid level is Cp(t), the effect-site tank's level is Ce(t), and molecules physically stream through the connecting channel at a rate proportional to the instantaneous flux ke0·(Cp−Ce) — exactly the same first-order equilibration, just rendered as diffusing particles rather than a 3D trajectory. The strip chart traces Cp and Ce against time; the loop panel plots effect against Cp directly, tracing the same counterclockwise hysteresis loop as the 3D version's floor shadow.
- Cp0 — peak plasma concentration right after the bolus (sets dose/volume-of-distribution).
- Elimination half-life — how fast the plasma level itself decays.
- ke0 half-life — how sluggishly the effect site follows plasma; a fast drug (short t½,ke0) tracks plasma almost immediately, a slow one lags for tens of minutes.
- EC50 — the effect-site concentration producing 50% of maximal effect.
Because Ce lags Cp, effect is lower for a given plasma level on the way up (still equilibrating) than on the way down (effect site draining slower than plasma) — the same mechanism anesthesiologists use to explain why a propofol or opioid bolus "kicks in" minutes after infusion, and why trough-only plasma monitoring can under- or over-estimate real-time effect.