Blood levels of a drug measure the concentration of that medication — and sometimes its active metabolites — circulating in your blood, serum, or plasma at a specific moment in time. This practice, known as therapeutic drug monitoring, exists to confirm that a drug is present in quantities high enough to actually work but not so high that it becomes dangerous. For someone taking lithium to manage behavioral symptoms associated with dementia, for example, a blood draw might reveal a level of 0.7 mmol/L, which falls within the accepted maintenance range of 0.6 to 0.8 mmol/L. That single number tells a clinician more about what is happening inside the body than any symptom checklist could.
The reason this matters — particularly in dementia care and brain health — is that the people most vulnerable to drug toxicity are often the same people taking multiple medications with narrow safety margins. Older adults with declining kidney or liver function process drugs differently, and a dose that works perfectly for a 45-year-old may accumulate to toxic levels in someone who is 80. Therapeutic drug monitoring provides an objective measurement rather than relying on clinical guesswork, which becomes especially important when a patient cannot reliably report their own symptoms. This article explains what peak and trough levels actually represent, why only a fraction of an oral drug reaches the bloodstream, how long it takes for medications to reach a stable concentration, which drugs require the closest monitoring, and what factors — from genetics to grapefruit juice — can throw blood levels off course.
Table of Contents
- What Do Blood Drug Levels Actually Measure and When Are They Drawn?
- Bioavailability — Why Not All of a Drug Reaches the Bloodstream
- The Steady-State Principle and Why Timing Blood Draws Matters
- Which Drugs Require Blood Level Monitoring and What the Therapeutic Ranges Look Like
- Factors That Alter Blood Drug Levels in Older Adults
- The Evolving Science of How Blood Levels Are Measured
- What Caregivers Should Understand Going Forward
- Conclusion
- Frequently Asked Questions
What Do Blood Drug Levels Actually Measure and When Are They Drawn?
Therapeutic drug monitoring captures two key data points. The peak level is the highest plasma concentration a drug reaches after administration, reflecting how quickly and completely the body absorbs it. For an intravenous infusion, this is typically drawn 15 to 30 minutes after the infusion ends. For an oral medication, it is usually measured about one hour after the dose is taken. The trough level is the opposite — the lowest concentration in plasma, drawn immediately before the next scheduled dose. Trough levels reflect how quickly the body is eliminating the drug.
The distinction between these two measurements is not academic. A patient whose vancomycin trough level falls below 10 micrograms per milliliter faces a meaningfully higher risk of treatment failure and, critically, of fostering bacterial resistance. A patient whose trough exceeds 20 micrograms per milliliter risks kidney damage. The therapeutic range for vancomycin — 10 to 20 micrograms per milliliter, with a target of 15 to 20 for complicated infections — illustrates how narrow the window can be between a drug doing its job and doing harm. One important clarification: blood, serum, or plasma specimens are appropriate for therapeutic drug monitoring. Urine drug concentrations, despite being commonly used in other contexts, do not correlate with therapeutic effect or toxicity and should never be used for this purpose. If a provider is relying on urine levels to adjust a medication dose, that is a red flag worth questioning.
Bioavailability — Why Not All of a Drug Reaches the Bloodstream
When a drug is injected directly into a vein, 100 percent of it reaches systemic circulation by definition. That is the baseline against which all other routes of administration are measured. Oral medications, however, must survive the gastrointestinal tract and then pass through the liver before reaching the rest of the body — a gauntlet known as first-pass metabolism that can destroy a substantial portion of the drug before it ever has a chance to work. The numbers are striking. Propranolol, a beta-blocker sometimes used to manage agitation in dementia, has an oral bioavailability of roughly 26 percent — meaning first-pass metabolism destroys 75 to 85 percent of each oral dose. Morphine fares only slightly better at around 30 percent, which is why the oral dose (approximately 30 milligrams) is so much larger than the intravenous dose (2 to 10 milligrams).
Digoxin tablets reach about 63 percent bioavailability, with the liquid form somewhat higher at 75 percent. Even levothyroxine, a medication many older adults take daily for thyroid function, achieves only up to 80 percent bioavailability — and only when taken on an empty stomach. However, bioavailability is not fixed for every patient. A person with significant liver disease may have reduced first-pass metabolism, paradoxically allowing more of an oral drug into circulation and pushing blood levels higher than expected. This is one reason why liver function is always relevant when interpreting drug levels, and why a standard dose may need to be reduced in someone with hepatic impairment. Sublingual nitroglycerin exploits this principle deliberately — by dissolving under the tongue and absorbing directly into the bloodstream, it bypasses first-pass metabolism entirely, which is why it works within minutes rather than the hour or more an oral tablet might require.
The Steady-State Principle and Why Timing Blood Draws Matters
When a patient begins a new medication, the drug does not immediately reach a stable concentration in the blood. With each dose, the level rises, the body eliminates some of it, the next dose adds more, and a gradual accumulation occurs until the rate of drug entering the body equals the rate at which it is being cleared. This equilibrium is called steady state, and it takes approximately four to five half-lives of the drug to achieve, regardless of the dose being given. Consider phenytoin, an anti-seizure medication with a half-life of roughly 22 hours. Steady state would not be reached for approximately four to five days.
Drawing a blood level on day two of therapy would produce a number that does not yet reflect where the drug concentration will ultimately settle. For drugs with much longer half-lives — some antidepressants have half-lives measured in days — reaching steady state can take weeks. This is why clinicians are taught to wait before checking levels and why early blood draws can be misleading, potentially prompting unnecessary dose changes that overshoot the target. For drugs with long half-lives where waiting four to five half-lives is clinically impractical — say, in a patient with an active infection — a loading dose can be used. A loading dose is a larger initial dose designed to push blood levels into the therapeutic range immediately, followed by smaller maintenance doses to keep them there. This approach carries its own risks, however, since the line between a therapeutic loading dose and a toxic one can be thin, especially in elderly patients with unpredictable drug clearance.
Which Drugs Require Blood Level Monitoring and What the Therapeutic Ranges Look Like
Not every medication needs its blood levels checked. Therapeutic drug monitoring is most critical for drugs with a narrow therapeutic index, meaning the margin between the dose that works and the dose that causes toxicity is dangerously small. If a drug is effective across a wide range of concentrations and relatively harmless at higher levels — acetaminophen at normal doses, for instance — routine monitoring is unnecessary. But for a drug like lithium, where the therapeutic range is 0.6 to 0.8 mmol/L for maintenance and toxicity begins around 1.5 mEq/L, the difference between benefit and harm is a matter of fractions. The drug classes most commonly requiring monitoring include anti-seizure medications such as phenytoin (therapeutic range of 10 to 20 micrograms per milliliter for total drug, or 1 to 2.5 micrograms per milliliter for the free, unbound fraction), valproic acid, and carbamazepine.
Cardiac drugs like digoxin — whose recommended range has actually been revised downward to 0.5 to 0.9 nanograms per milliliter for heart failure based on 2013 ACC/AHA guidelines — also require close surveillance. Antibiotics such as vancomycin and aminoglycosides, mood stabilizers like lithium, and immunosuppressants including tacrolimus and cyclosporine round out the list. The tradeoff in monitoring is real. Each blood draw costs money, requires a venipuncture, and demands proper timing to produce meaningful results. For patients in long-term care facilities or those with dementia who may find blood draws distressing, the frequency of monitoring becomes a clinical judgment call — balancing the safety information gained against the burden of the procedure itself. Skipping monitoring altogether, though, is almost never the right answer for narrow-index drugs.
Factors That Alter Blood Drug Levels in Older Adults
Age, liver function, kidney function, genetics, drug-drug interactions, body weight, and patient adherence all affect how a drug concentrates in the blood. In the context of dementia care, nearly every one of these factors is amplified. Kidney function declines with age, often without obvious symptoms, slowing the clearance of drugs like vancomycin and lithium and pushing blood levels upward. Liver metabolism slows as well, affecting the first-pass processing of oral medications. Drug-drug interactions deserve particular attention in this population. A person with dementia may be taking a cholinesterase inhibitor, an antidepressant, a blood pressure medication, and a proton pump inhibitor — each of which can compete for the same metabolic pathways in the liver.
Add an antibiotic prescribed for a urinary tract infection, and the entire equilibrium shifts. Pantoprazole, for example, has an oral bioavailability of about 77 percent, but if it is competing with other drugs for absorption or metabolism, the effective level of every medication in the mix may change. Pharmacogenomics — the study of how genetic variation affects drug metabolism — adds another layer of complexity that is only beginning to influence clinical practice. Some patients are ultrarapid metabolizers of certain drugs, clearing them so quickly that standard doses never reach therapeutic levels. Others are poor metabolizers, accumulating the drug to toxic concentrations on the same dose. Genetic testing for drug metabolism enzymes is available but not yet routine in most dementia care settings, which means clinicians often discover a patient’s metabolizer status only after a drug level comes back unexpectedly high or low.
The Evolving Science of How Blood Levels Are Measured
The traditional approach to therapeutic drug monitoring relies on trough levels — a single snapshot taken at the drug’s lowest point. But a growing body of research suggests that area-under-the-curve monitoring, which captures the total drug exposure over an entire dosing interval, may be more accurate for certain medications. A 2025 study published in the British Journal of Clinical Pharmacology examined whether AUC monitoring for tacrolimus after kidney transplant is superior to traditional trough monitoring, reflecting an ongoing debate about which method better predicts clinical outcomes.
For caregivers and families navigating dementia care, this shift matters because it may change how often and when blood is drawn. AUC monitoring typically requires multiple samples over several hours, which is more burdensome but potentially more informative. As this methodology matures, it may eventually trickle down from transplant medicine into other areas — including the management of anti-seizure drugs and mood stabilizers commonly used in patients with dementia.
What Caregivers Should Understand Going Forward
The practical takeaway for families managing medications in dementia care is that blood drug levels are not just numbers on a lab printout — they are safety signals. A level in the therapeutic range does not guarantee the drug is working, but a level outside that range almost certainly means something needs to change. As pharmacogenomic testing becomes more accessible and AUC-based monitoring methods continue to develop, the precision of drug dosing will improve.
Until then, the most important step a caregiver can take is ensuring that blood draws happen at the correct time relative to dosing, that all medications — including over-the-counter supplements — are reported to the prescribing clinician, and that any sudden changes in behavior, alertness, or appetite trigger a conversation about whether drug levels should be checked. The blood test itself is simple. Interpreting it correctly, in the context of a specific patient’s age, organ function, and medication list, is where the real clinical skill lies.
Conclusion
Therapeutic drug monitoring measures the concentration of a medication in blood at a precise moment, providing clinicians with objective data about whether a drug is present in quantities sufficient to be effective but not so high as to cause harm. This is especially critical in dementia care, where patients often take multiple narrow-index medications, have declining organ function that affects drug clearance, and may not be able to articulate symptoms of toxicity themselves. Understanding the basics — that blood levels depend on bioavailability, timing relative to steady state, liver and kidney function, and interactions with other drugs — helps caregivers ask better questions and recognize when something may be going wrong.
The science behind drug monitoring continues to evolve, with newer approaches like AUC monitoring and pharmacogenomic testing promising greater precision. But the fundamentals remain unchanged: the right drug at the right blood level in the right patient is the foundation of safe prescribing. For families navigating the complexity of dementia medications, insisting on regular monitoring of narrow-index drugs is one of the most concrete, actionable steps available to protect someone who cannot always protect themselves.
Frequently Asked Questions
How long after starting a new medication should blood levels be checked?
Blood levels should generally not be drawn until the drug has reached steady state, which takes approximately four to five half-lives. For a drug with a 24-hour half-life, this means waiting about five days. Drawing levels too early can produce misleadingly low numbers that prompt unnecessary dose increases.
Does it matter what time of day a blood sample is drawn for drug monitoring?
Yes, timing is critical. Trough levels must be drawn immediately before the next scheduled dose, and peak levels are drawn at specific intervals after administration — typically 15 to 30 minutes after an IV infusion or about one hour after an oral dose. A mistimed blood draw can produce a number that does not accurately reflect the drug’s behavior in the body.
Why does kidney function affect drug blood levels?
Many drugs are eliminated primarily through the kidneys. As kidney function declines — a common and often undetected change in aging — the body clears these drugs more slowly, causing them to accumulate in the blood. This is why drugs like vancomycin and lithium require more frequent monitoring in older adults and often need dose reductions.
Can food or supplements affect blood drug levels?
Absolutely. Levothyroxine, for example, achieves up to 80 percent bioavailability on an empty stomach but significantly less when taken with food, calcium, or iron supplements. Grapefruit juice inhibits enzymes in the gut wall that break down certain drugs, effectively increasing their bioavailability. All supplements and dietary habits should be disclosed to prescribers.
What is the difference between total and free drug levels?
Total drug level measures all of the drug in the blood, including the portion bound to proteins like albumin. Free drug level measures only the unbound fraction, which is the portion that is pharmacologically active. For phenytoin, the total therapeutic range is 10 to 20 micrograms per milliliter, but the free level range is 1 to 2.5 micrograms per milliliter. In patients with low albumin — common in older and malnourished individuals — total levels may appear normal while free levels are dangerously high.
