Hand-Checking a One-Compartment Prediction

Why a Hand Calculation Helps

A model output can be numerically precise and still be wrong for a basic reason: the time may be interpreted in the wrong units, a concentration may be reported on an unexpected scale, or the calculation may use a different parameter set than the one being reviewed. A small hand calculation provides a transparent reference for a deliberately simple case. It does not validate every part of a pharmacometric model, but it can reveal whether the model's basic concentration prediction is consistent with its stated dose, volume, clearance, and elapsed time.

The check in this tutorial is for a single intravenous bolus in a one-compartment model with linear elimination. The dose is assumed to enter the compartment instantaneously, and the concentration is evaluated at a specified time after dosing. Before comparing any result, establish the dose, parameter values, time origin, units, and whether the output is a prediction rather than an observed concentration.

Definition: For an intravenous bolus dose \(D\) into a one-compartment model with volume \(V\) and clearance \(CL\), the elimination rate constant is \(k=CL/V\). Under linear, first-order elimination, the concentration \(C(t)\) at elapsed time \(t\geq 0\) is \(C(t)=(D/V)e^{-kt}\). Here \(D\) is in amount units, \(V\) is in volume units, \(CL\) is in volume per time, \(k\) is in inverse time, and \(C(t)\) is amount per volume.

Model Assumptions and the Hand Calculation

The formula applies when the model treats the dose as an instantaneous intravenous bolus, the drug is immediately and uniformly distributed in one apparent volume, and elimination is first order with constant clearance over the interval. It also assumes there is no additional input, dose, or process changing concentration during the interval being calculated. This is a model prediction, not a claim that real concentration measurements must follow the curve exactly.

The units should be compatible. If \(CL\) is reported in L/h and \(V\) in L, then \(k\) is in h\(^{-1}\), and \(t\) must be entered in hours. The ratio \(D/V\) has concentration units. The exponent \(kt\) has no units, as required for an exponential. These checks are useful before evaluating any numbers.

The relationship follows from the one-compartment mass balance. Let the amount in the compartment at time \(t\) be \(A(t)\). With first-order elimination, the rate of change in amount is proportional to the negative amount, with proportionality constant \(k\). Immediately after the bolus, the amount is \(D\). Solving this initial-value equation gives the amount remaining, and dividing by the fixed volume gives concentration.

\frac{dA(t)}{dt}=-kA(t),\qquad A(0)=D

Separate variables and integrate from time zero to \(t\). The initial condition determines the integration constant. Since \(C(t)=A(t)/V\), the result is the bolus concentration formula.

\frac{dA}{A}=-k\,dt \quad\Longrightarrow\quad \ln\!\left(\frac{A(t)}{D}\right)=-kt \quad\Longrightarrow\quad A(t)=De^{-kt} \quad\Longrightarrow\quad C(t)=\frac{D}{V}e^{-kt}

In practice, calculate \(k\), calculate the initial post-dose concentration \(C(0)=D/V\), and then multiply by the decay factor \(e^{-kt}\). For an output comparison, use the same dose, time after dose, parameter values, and concentration units as the model. If the software reports more digits than the inputs justify, compare at a sensible precision rather than treating a tiny rounding difference as an error.

Worked Example: Checking a Prediction at Three Hours

Suppose a hypothetical model uses a single 500 mg intravenous bolus, \(V=25\) L, and \(CL=5\) L/h. Calculate the concentration three hours after dosing and compare it with a hypothetical model output of 10.98 mg/L.

First, check the units and calculate the elimination rate constant:

k=\frac{CL}{V}=\frac{5\ \text{L/h}}{25\ \text{L}}=0.2000\ \text{h}^{-1}

The initial concentration is \(500\ \text{mg}/25\ \text{L}=20.00\) mg/L. At \(t=3\) h:

C(3)=\frac{500\ \text{mg}}{25\ \text{L}}e^{-(0.2000\ \text{h}^{-1})(3\ \text{h})} =20.00e^{-0.6000} =20.00(0.5488) =10.98\ \text{mg/L}

The hand result agrees with the stated output to the reported precision. The units cancel correctly in the exponent, and the output is plausible relative to the initial concentration: it is lower than 20.00 mg/L because elimination has occurred. Agreement supports this particular calculation, but it does not by itself establish that other model components or predictions are correct.

Worked Checks for Units, Timing, and Parameters

Worked Example: Detecting a Time-Unit Mismatch

For a hypothetical bolus of 400 mg, let \(V=40\) L and \(CL=8\) L/h. The model output at two hours is reported as 9.048 mg/L. Check the stated time and determine whether the output matches the hand calculation.

The rate constant and initial concentration are:

k=\frac{8\ \text{L/h}}{40\ \text{L}}=0.2000\ \text{h}^{-1}, \qquad C(0)=\frac{400\ \text{mg}}{40\ \text{L}}=10.00\ \text{mg/L}

Using two hours, as stated, gives:

C(2)=10.00e^{-(0.2000)(2)} =10.00e^{-0.4000} =10.00(0.6703) =6.703\ \text{mg/L}

This does not match 9.048 mg/L. To see whether the output suggests a timing issue, evaluate the same formula at 0.5000 h:

C(0.5000)=10.00e^{-(0.2000)(0.5000)} =10.00e^{-0.1000} =9.048\ \text{mg/L}

Thus, the reported value corresponds to 0.5000 h under these parameters, not two hours. This is a diagnostic clue, not proof of the cause: the review should check the event-time input and time units actually used by the model. The concentration should not be “corrected” by changing a parameter until the time definition has been checked.

Worked Example: Checking Which Parameter Set Was Used

A hypothetical subject receives 600 mg. The output is intended to use individual values \(V=24\) L and \(CL=6\) L/h at two hours. A separate typical, or population, parameter set is \(V=30\) L and \(CL=6\) L/h. The software output is 13.41 mg/L. Calculate both predictions to see which parameter set it matches.

For the individual values, \(k=6/24=0.2500\) h\(^{-1}\), and \(C(0)=600/24=25.00\) mg/L. Therefore:

C_{\text{individual}}(2) =25.00e^{-(0.2500)(2)} =25.00e^{-0.5000} =25.00(0.6065) =15.16\ \text{mg/L}

For the typical values, \(k=6/30=0.2000\) h\(^{-1}\), and \(C(0)=600/30=20.00\) mg/L. Therefore:

C_{\text{typical}}(2) =20.00e^{-(0.2000)(2)} =20.00e^{-0.4000} =20.00(0.6703) =13.41\ \text{mg/L}

The output matches the typical-parameter calculation, not the calculation using the stated individual values. This comparison makes parameter selection a plausible source of the discrepancy. The reviewer should confirm which parameter values the output actually used and whether the intended comparison calls for individual or typical values.

Interpreting Discrepancies and Choosing a Comparison

A useful hand check is a controlled comparison: the mathematical calculation and the model output must refer to the same prediction. In particular, confirm that both use the same dose and dose time, the same elapsed time, the same \(V\) and \(CL\), and the same units. Also confirm whether the output is a model-predicted concentration. An observed concentration can differ from the prediction because of residual error and measurement variability; it is not the same quantity as the deterministic formula above.

For this simple case, the analytical expression is preferable as the primary reference because it is direct and transparent. A second calculation using a calculator or spreadsheet can help check arithmetic, but it is not independent if it repeats the same incorrect time or parameter inputs. More complicated models may require numerical evaluation or other checks; those approaches do not remove the need to define inputs and units precisely. A hand calculation is most informative when the example is simple enough that its assumptions and arithmetic can be inspected line by line.

A discrepancy is a reason to investigate, not a reason to assume the software is wrong. First verify dose amount and time origin; then verify time and concentration units; then check whether the parameter values and prediction type match. Rounding can cause small differences, while a large difference usually warrants checking these inputs and the model's stated bolus convention. A concentration immediately after a bolus is \(D/V\); a value at the dose time may depend on whether the output convention represents the pre-dose or post-dose state.

Common Pitfalls and Limits

Mixing time units. Using \(k\) in h\(^{-1}\) with time in minutes makes the exponent incorrect. Convert time to hours or express \(k\) in min\(^{-1}\) before calculating.
Confusing clearance with the elimination rate constant. Clearance has volume-per-time units; \(k=CL/V\) has inverse-time units. Substituting \(CL\) directly into the exponential is dimensionally wrong.
Using an inconsistent parameter set. A typical-parameter prediction need not equal one calculated with individual parameters. State which values are being checked.
Comparing different concentration scales. For example, mg/L and mg/mL differ by a factor of 1,000. Convert to a common unit before judging agreement.
Treating agreement as full validation. Matching one simple prediction checks only that calculation under its assumptions. It does not verify complex dosing histories, additional compartments, time-varying processes, or all implementation details.

A particularly misleading argument is, “The model output is close to \(D/V\), so it must be correct.” The initial concentration alone does not check elimination or elapsed time. The full comparison requires \(e^{-kt}\), with \(k\) derived from the stated clearance and volume and \(t\) expressed in compatible units.

Key takeaway: For a one-compartment intravenous bolus, calculate \(k=CL/V\) and \(C(t)=(D/V)e^{-kt}\). Compare only after matching the dose, elapsed time, parameter set, prediction convention, and units; treat agreement as a targeted check, not a general validation.

Check Your Understanding

Answer each question, then compare with the solutions below.

A hypothetical 300 mg bolus is given with \(V=15\) L and \(CL=3\) L/h. What are \(k\) and the predicted concentration at two hours?
Why is it incorrect to use a clearance of 4 L/h directly as the exponential rate constant when \(V=20\) L?
For \(D=500\) mg and \(V=25\) L, what is the immediate post-dose concentration, and what additional quantity is needed to calculate concentration one hour later?
A hand calculation gives 8.00 mg/L, while an output gives 0.00800 g/L. Are the values inconsistent? Explain.

Solutions. (1) \(k=3/15=0.2000\) h\(^{-1}\), and \(C(2)=(300/15)e^{-(0.2000)(2)}=20.00e^{-0.4000}=13.41\) mg/L. (2) Clearance is volume per time, whereas the exponential rate must be inverse time; divide by volume to obtain \(k=4/20=0.2000\) h\(^{-1}\). (3) \(C(0)=500/25=20.00\) mg/L. The elimination rate constant and elapsed time are also needed; with \(k=CL/V\), use \(C(1)=20.00e^{-k(1\text{ h})}\). (4) They are equal after conversion: \(0.00800\) g/L is \(8.00\) mg/L because one gram is 1,000 mg.