Why Diagnostic Errors Matter
A drug concentration is evidence about exposure at a particular time, not a complete description of treatment success, toxicity, or the dose a patient should receive. A result can be analytically sound yet misleading if it is compared with the wrong target, collected at an unexpected time, or treated as a complete explanation of a clinical event. In therapeutic drug monitoring (TDM), interpretation therefore depends on the measurement and on the circumstances that give it meaning.
The earlier material on interpreting concentrations relative to a target, unexpected-result triage, repeat concentrations, and trough, peak, and AUC interpretation provides the relevant vocabulary. This tutorial focuses on diagnostic errors that can still arise when a number looks straightforward. The practical sequence is to verify what was measured and when, identify the applicable clinical target, compare like with like, and interpret the result alongside the patient's response and other relevant information.
Error 1: Treating a Reference Range as a Universal Target
A reference range is not a guarantee that every value inside it is effective and safe for every patient, nor does a value just outside it automatically require a dose change. Its interpretation depends on the drug, indication, sampling time, assay, and the clinical purpose of monitoring. The clinically relevant target may differ with the treatment goal and patient context. A laboratory interval can help frame interpretation, but it does not replace the target selected for the individual clinical question.
Before comparing a result with a range, establish what the range represents and whether it applies to this indication and sample type. Confirm that it refers to the same sampling phase, such as a trough, as the sample being interpreted. A range for one sampling time cannot be applied uncritically to a sample collected at another time. Also ask whether the patient is achieving the intended clinical response and whether there are signs of toxicity. Those observations inform interpretation; they do not make the concentration irrelevant.
Worked Example: A Result Above the Patient's Documented Goal
Consider a wholly hypothetical drug Q. A laboratory report displays a general reference interval of 5 to 15 mg/L. For a particular patient and indication, the treating team has documented a goal of 8 to 12 mg/L for a correctly timed trough. A trough result is 13.2 mg/L. What can be concluded?
Step 1: Check comparability. Assume the sample was collected immediately before the scheduled dose, the patient took the prescribed regimen, and the assay and units match those used for the documented goal. Under these assumptions, it is reasonable to compare 13.2 mg/L with the patient-specific goal.
Step 2: Make the comparison accurately. The result is \(13.2-12=1.2\) mg/L above the upper end of the documented goal. It is nevertheless within the laboratory's general 5 to 15 mg/L interval.
Step 3: State the conclusion narrowly. The result is above this patient's documented trough goal, but that fact alone does not establish toxicity or dictate a particular dose change. The response, adverse effects, dosing history, and clinical reason for the goal still matter. Calling the result simply “therapeutic” because it is inside the general interval would ignore the patient-specific goal; calling it proof of harm would overstate what the measurement shows.
Error 2: Ignoring Dose-to-Sample Timing
A measured concentration is meaningful only in relation to the time of collection and the dosing history. “A concentration of 12 mg/L” is incomplete if it is not known whether the sample was collected before a dose, shortly after a dose, or at another point in the interval. A value intended to represent a trough cannot be interpreted as a trough if it was collected substantially earlier. Missed or delayed doses, extra doses, and dose-time recording errors can also make the apparent sampling time misleading.
A simple model demonstrates why this matters. Suppose, as an approximation, that absorption is complete, there is no further dose during the period considered, and elimination is first order with a constant rate constant \(k\). Then concentration falls exponentially over that period. This is an illustrative calculation, not a substitute for the appropriate PK model: the assumptions may not hold during absorption, with nonlinear kinetics, or when the dosing history is uncertain.
Here \(C(t)\) is concentration at time \(t\), \(\Delta t\) is elapsed time, and \(t_{1/2}\) is the elimination half-life under the stated model. Because the exponential factor is less than 1 when \(k\) and \(\Delta t\) are positive, a later sample is predicted to have a lower concentration in this elimination-only phase.
Worked Example: An Early Sample Mistaken for a Trough
In a hypothetical case, a drug is dosed every 24 hours. A sample is collected 8 hours after the dose and measures 12.0 mg/L. For illustration, assume the drug has a 12-hour half-life, absorption is complete by hour 8, elimination is first order, and no dose is given before hour 24. Estimate the concentration at hour 24 under this simplified model.
First calculate the elimination rate constant:
The elapsed time from hour 8 to hour 24 is \(16\) hours. Substitution gives:
Under these assumptions, the early sample of 12.0 mg/L corresponds to an estimated pre-dose concentration of about 4.76 mg/L, not 12.0 mg/L. This estimate is not a measured trough and should not be treated as one. Its purpose is to show that timing can materially change the interpretation. In a real case, verify the dose and sample times and use an appropriate model and monitoring plan; do not correct a mistimed result using this simple calculation without justification.
Error 3: Overreading an Isolated Result
One result can be an important signal, but it does not by itself establish that exposure has changed persistently or reveal why it changed. A difference between two measurements is interpretable only after checking whether they are comparable: dose and regimen, time since dose, steady-state status where relevant, assay, units, and clinical circumstances. Analytical uncertainty and within-patient variability also mean that small differences may not represent a meaningful change in underlying exposure.
A useful first step is to distinguish the observation from the inference. “The measured concentration is higher than the two prior results” is a statement about the available data. “Clearance has fallen,” “adherence has improved,” or “the patient is now toxic” are explanations or clinical conclusions that require additional evidence. If the result is unexpected, the appropriate response may include checking sample metadata, reviewing the dosing history, considering analytical uncertainty, assessing the patient, and deciding whether a repeat measurement is warranted. A repeat is useful only if its timing and purpose are clear.
Worked Example: A Single Higher Result
Suppose a hypothetical patient has two earlier, correctly timed trough results of 9.8 mg/L and 10.2 mg/L. A new result is 16.0 mg/L. Assume all three results use the same assay and units, were collected at the same point in the dosing interval, and correspond to the same prescribed regimen. Calculate how the new result compares with the earlier measurements, then state what can and cannot be inferred.
The mean of the two earlier results is:
The new result is \(16.0-10.0=6.0\) mg/L above that earlier mean. Relative to the mean, the increase is:
Thus, the new measurement is 6.0 mg/L, or 60%, above the mean of the two earlier measurements. That is a substantial observed difference in this hypothetical series and merits attention. It does not, by itself, identify the cause or prove that the patient's typical exposure has permanently changed. Check the dose and sample records, relevant clinical changes, assay context, and symptoms. Whether and when to repeat the concentration depends on the clinical question and the result of those checks.
Error 4: Attributing Causation to Concentration Alone
A concentration can support an explanation of an outcome, but a concentration and an outcome observed together do not prove that the concentration caused it. The measured value may be related to a causal factor, may reflect a change that occurred alongside the outcome, or may be one part of a more complicated clinical picture. A concentration within a stated target does not rule out toxicity or inadequate response; a concentration outside it does not prove that the result caused a symptom or treatment failure.
The correct interpretation keeps several questions separate: Is the result valid and appropriately timed? Does it suggest exposure that is higher or lower than intended? Is the observed clinical effect plausibly related to the drug, and what other explanations should be considered? The concentration contributes evidence to the clinical assessment. It does not replace evaluation of the patient's condition, other medicines, illness, organ function, or other relevant factors. In particular, avoid turning “occurred after” into “was caused by.”
Worked Example: Symptoms Despite an In-Range Result
A hypothetical patient taking drug R develops new dizziness. A correctly timed concentration is 14.0 mg/L, within the documented goal of 10 to 15 mg/L. The timing and assay are verified, but the cause of the dizziness is not known. What does the result establish?
The result establishes that the measured concentration at that sampling time is 14.0 mg/L and, assuming the documented goal applies, lies within that goal. It does not establish that drug R caused the dizziness, and it does not exclude a drug contribution merely because the concentration is in range. The clinician should assess the symptom and relevant alternative explanations, review dosing and concomitant medicines, and interpret the concentration with the full clinical picture. A dose change should not be justified by the concentration alone without considering the treatment goal and clinical findings.
The same caution applies in the opposite direction. If a concentration is above a goal and a symptom is present, their coexistence raises a question worth assessing; it is not, on its own, proof that the elevated concentration caused the symptom.
A Practical Diagnostic Sequence
A consistent sequence helps prevent these errors without turning every unexpected result into an elaborate investigation. The checks should be proportionate to the clinical urgency and the uncertainty in the result.
Confirm the drug and analyte, units, assay context, and sample identity; check for collection or contamination concerns when indicated.
Record the actual recent doses and the actual collection time. Establish whether the sample is the intended trough, peak, AUC-related sample, or another planned measurement.
Identify the target for the indication and sampling phase. Do not substitute a general laboratory interval for a documented patient-specific goal without checking that it is applicable.
Review previous results only after checking whether dose, timing, assay, and relevant clinical conditions are sufficiently comparable. Separate observed differences from proposed explanations.
Assess response, adverse effects, and other relevant clinical information. Decide whether the result supports action, further evaluation, or a repeat measurement, and state the remaining uncertainty.
This sequence is not a claim that every result needs every possible check. A clearly mistimed sample, for example, may already explain why a trough comparison is invalid. Conversely, an urgent clinical problem should not be left unevaluated while waiting for a repeat concentration. The aim is to make the reasoning explicit and to avoid conclusions that the available evidence cannot support.
Comparison, Limitations, and Common Pitfalls
A fixed reference interval offers a quick common comparison, but it can mislead when the indication, sampling phase, or patient-specific goal differs. A patient-specific target is more relevant when it has been deliberately selected for the clinical objective, but it still requires correct sample timing and clinical interpretation. A single concentration is faster to obtain and interpret than a trend, but it carries less evidence about persistence. Serial comparable results help characterize change, yet a trend still does not prove its cause. Clinical assessment provides context that a concentration cannot, while clinical impressions alone may miss an exposure problem. These approaches complement rather than replace one another.
The simple exponential calculation above also has clear limits: it assumes a specific elimination phase and a constant first-order rate, and it does not account for ongoing absorption, additional doses, nonlinear kinetics, or uncertainty in the recorded times. More generally, a concentration-based inference depends on the model and assumptions used to connect the sample with exposure. If those assumptions are doubtful, report what was measured and what is uncertain rather than presenting a derived estimate as fact.
- Wrong reasoning: “The concentration is inside the laboratory range, so it must be the correct target.” Correction: Verify that the interval applies to the indication and sampling phase, and consider the documented patient-specific goal.
- Wrong reasoning: “This is the trough because it was drawn on the dosing day.” Correction: A trough is defined by its timing relative to the next dose; verify actual dose and collection times.
- Wrong reasoning: “One high result proves a persistent exposure change.” Correction: Check comparability, uncertainty, and possible explanations before making that inference.
- Wrong reasoning: “The symptom appeared while the concentration was high, so the concentration caused it.” Correction: Treat the concentration as evidence to integrate with clinical assessment, not as proof of causation.
Check Your Understanding
Answer each question, then compare with the brief solutions below.
- A hypothetical patient has a correctly timed result of 7.0 mg/L. The laboratory's general interval is 5 to 15 mg/L, but the documented goal for the patient's indication is 8 to 12 mg/L. What is the most accurate comparison?
- In the simplified elimination model, a concentration is measured at 10 hours after a dose. Is a sample collected at 24 hours 14 hours later or 34 hours later? What additional condition is needed to call it a pre-dose sample?
- Two comparable earlier results are 9.8 mg/L and 10.2 mg/L, and a new result is 16.0 mg/L. Calculate the new result's difference from the earlier mean.
- A patient has a symptom while a concentration lies within the documented target. Does this establish or exclude drug causation? State one appropriate next interpretive step.
Solutions. (1) The result is within the general interval but 1.0 mg/L below the patient's documented lower goal of 8 mg/L. Neither comparison alone determines efficacy or a dose change; clinical response and context matter. (2) The interval is \(24-10=14\) hours. It is a pre-dose sample only if the next scheduled dose is due at hour 24 and the sample was collected before that dose. (3) The earlier mean is \((9.8+10.2)/2=10.0\) mg/L, so the new result is \(16.0-10.0=6.0\) mg/L higher. This difference does not identify its cause. (4) The result neither establishes nor excludes causation. Assess the symptom and other clinical information, and review the dose and sampling context.