The Bioavailability Formula and How to Calculate It

Bioavailability is calculated by comparing the amount of a drug that reaches systemic circulation after a given route of administration against a reference, usually an intravenous dose. The core formula for absolute bioavailability is F = (AUC_oral / AUC_IV) × (Dose_IV / Dose_oral), where AUC stands for the area under the plasma concentration-time curve. That ratio captures how much of an oral dose actually makes it into the bloodstream in usable form, but the real story is in the layers beneath that fraction, from gut-wall metabolism to transporter proteins to the physical properties of the drug itself.

The Core Formula for Absolute Bioavailability

When a drug is injected directly into a vein, all of it enters the bloodstream. That makes intravenous delivery the gold standard, assigned a bioavailability of 1.0 (or 100%). For any other route, you measure how much drug shows up in the blood over time and compare it to that intravenous benchmark. The formula looks like this:

F = (AUC_oral ÷ AUC_IV) × (Dose_IV ÷ Dose_oral)

The dose correction matters because studies often give different amounts by different routes. If you gave 100 mg orally but only 50 mg intravenously, you need to account for that imbalance before the AUC comparison means anything. This calculation assumes the drug behaves proportionally across doses, a condition known as linear pharmacokinetics. When that assumption holds, doubling the dose roughly doubles the blood levels, and the formula works cleanly. When it does not hold, the math gets considerably more complicated.

One practical wrinkle is that many drugs have long half-lives, meaning they linger in the body for days or even weeks. Measuring the full AUC in those cases would require following blood levels for an impractically long time. Researchers have developed workarounds using truncated AUC measurements and regression methods to estimate bioavailability without waiting for the drug to fully clear the body.¹

What Determines the Fraction That Gets Through

The single number F obscures the fact that oral bioavailability is actually the product of three sequential barriers. Each one removes some fraction of the drug before it reaches general circulation. The relationship is:

F = Fa × Fg × Fh

Fa is the fraction absorbed across the intestinal wall. Fg is the fraction that escapes metabolism in the gut wall itself. Fh is the fraction that survives its first pass through the liver.² A drug could be well absorbed (high Fa) but still have poor bioavailability if the liver aggressively metabolizes it on that first pass (low Fh). Researchers have developed methods to tease these components apart, including studies with cyclosporine that separately measured how much the gut wall and the liver each contributed to the drug’s overall losses.³

More recently, lab-on-a-chip systems that mimic the human gut and liver have been used to estimate each of these fractions without giving the drug to a person at all. These microphysiological systems combine living human gut and liver tissue with mathematical models to predict oral bioavailability and its component parts.⁴

Some drugs add a fourth twist: enterohepatic recycling. The liver excretes a portion of the drug into bile, which dumps it back into the intestines, where it can be reabsorbed. This cycling effectively inflates the amount of drug in the bloodstream beyond what a simple three-barrier model would predict. The math for systemic availability in these cases has to account for the reabsorption fraction and the hepatic extraction ratio together, producing a more complex expression.⁵

How AUC Is Actually Measured

The area under the concentration-time curve sounds abstract, but it comes from a concrete and somewhat tedious process. Volunteers receive a dose, and blood samples are drawn at scheduled intervals afterward, sometimes extending over days. In a bioequivalence study of clonidine, for example, blood was drawn at 18 time points stretching from 15 minutes to 192 hours after dosing.⁶ Each sample is processed and analyzed, typically by liquid chromatography coupled with mass spectrometry, to determine the drug concentration at that moment.

Once you have a series of concentration-time data points, you need to estimate the total area underneath the curve they trace. The standard approach uses the trapezoidal rule: you connect adjacent data points with straight lines and sum the areas of the resulting trapezoids. During the elimination phase, when concentrations are declining, a log-trapezoidal variant is preferred because drug levels tend to fall in a curve rather than a straight line. The choice between arithmetic and geometric means at each time point depends on whether the data follow a normal or lognormal distribution, with pharmacokinetic data commonly being lognormal.⁷

Two versions of AUC appear frequently. AUC from time zero to the last measurable concentration (often written AUC₀₋ₜ) captures only the measured portion. AUC extrapolated to infinity (AUC₀₋∞) adds an estimated tail based on the drug’s terminal elimination rate, giving a more complete picture. Bioequivalence studies report both, along with the peak concentration and the time it takes to reach that peak.⁸

Absolute Versus Relative Bioavailability

The formula described above gives absolute bioavailability because it compares an oral dose against an intravenous reference. But many studies instead measure relative bioavailability, which compares two non-intravenous formulations against each other. The math is the same, except you substitute the reference formulation’s AUC where the IV values would go. This is exactly what happens in generic drug approval: the generic tablet is the test product, and the brand-name tablet is the reference. If their AUC ratio falls within an acceptable window, the generic is considered bioequivalent.

For the omeprazole study mentioned earlier, the 90% confidence intervals for the AUC ratios between two enteric formulations all fell within the 80–125% range commonly used as the bioequivalence criterion, meaning the two formulations delivered the drug to the bloodstream in statistically similar amounts.⁸ Relative bioavailability is more common in practice than absolute bioavailability because many drugs are never given intravenously, making the absolute comparison impossible.

Why the Same Drug Varies From Person to Person

A bioavailability number published in a drug’s label represents an average across a study population, but individual values can spread widely around that average. Several biological factors drive this variability.

One of the most important is genetics. Variations in the genes coding for drug-metabolizing enzymes and transporter proteins can dramatically change how fast a person processes a drug. Someone who carries a version of a metabolizing enzyme that works unusually fast may break down the drug in the gut wall or liver before much of it reaches the bloodstream, producing low bioavailability. Someone with a slow variant of the same enzyme may get much higher blood levels from the same dose.⁹

P-glycoprotein, a transporter protein embedded in the intestinal lining, actively pumps certain drugs back out of the gut wall before they can be absorbed. It acts as a gatekeeper, reducing oral bioavailability for drugs it recognizes. Studies in humans have shown that intestinal P-glycoprotein is particularly important for the bioavailability of the immunosuppressant cyclosporine.¹⁰ The impact of P-glycoprotein depends on how easily the drug can cross membranes on its own: for compounds with moderate passive permeability, P-glycoprotein significantly limits absorption, but for highly permeable drugs, passive diffusion overwhelms the efflux pump and the drug gets through regardless.¹¹

Food adds yet another variable. Eating can change stomach pH, slow gastric emptying, increase blood flow to the gut, and alter the composition of gut bacteria, all of which can push a drug’s absorption up or down depending on its properties.¹² This is why some medications specify whether to take them with or without food: the instruction is not arbitrary, it reflects measured differences in bioavailability.

The Drug’s Own Properties Matter Too

Before you even consider what happens inside the body, the drug’s solubility and permeability set boundaries on what is possible. The Biopharmaceutics Classification System groups drugs into four classes based on how well they dissolve in water and how easily they cross the intestinal membrane.¹³ A drug that is both highly soluble and highly permeable will generally have good bioavailability. A drug that is poorly soluble and poorly permeable faces an uphill battle no matter what formulation tricks you apply.

A related system, the Biopharmaceutics Drug Disposition Classification System, uses the same solubility criterion but swaps permeability for a measure based on how the drug is metabolized and eliminated. The two systems serve different purposes: one focuses on how the drug gets in, the other on how it behaves once it is there.¹⁴ For practical bioavailability questions, the original classification based on solubility and permeability is the one you encounter most.

When the Formula Breaks Down

The standard bioavailability formula assumes that doubling the dose doubles the blood levels. For many drugs, this linear relationship holds well enough. But some drugs are substrates for both efflux transporters and metabolizing enzymes in the gut wall, and those systems can become saturated at higher doses. When that happens, the relationship between dose and AUC becomes nonlinear, and the simple ratio-based formula can mislead.

Simulations of drugs that are transported by influx or efflux mechanisms in the intestine have shown nonlinear dose dependence even within normal therapeutic ranges.¹⁵ One modeling study illustrated how this plays out with a drug that is both a P-glycoprotein and a CYP3A4 substrate: at low doses, P-glycoprotein actively pumps the drug out and actually shields it from gut-wall metabolism. At moderate doses, as P-glycoprotein saturates, more drug encounters the metabolizing enzyme and extraction increases. At the highest doses, the enzyme itself saturates, and bioavailability jumps again.¹⁶ The result is a zigzag pattern where bioavailability changes unpredictably with dose, making a single F value essentially meaningless for those compounds.

Formulation Strategies That Change the Numbers

When a drug’s intrinsic properties make bioavailability poor, formulation scientists have several ways to fight back. Lipid-based delivery systems are among the most successful. Oils and surfactants can dissolve water-insoluble drugs, disperse them in an aqueous medium, and help them cross the gut barrier. Technologies like nanoemulsions and self-emulsifying systems have been designed specifically to improve absorption of poorly soluble compounds.¹⁷

Curcumin, the yellow compound in turmeric, is a well-known example of why formulation matters. In its natural form, curcumin has notoriously poor oral bioavailability. Researchers loaded it into solid lipid nanoparticles with PEGylated surface coatings, which allowed the compound to permeate the intestinal lining rapidly. The result was a more than 12-fold increase in bioavailability compared to curcumin in plain solution.¹⁸ That kind of dramatic improvement shows how much of the bioavailability equation is under the formulator’s control, not fixed by the drug’s chemistry alone.

Excipients, the supposedly inactive ingredients in a pill, can also have outsized effects. A famous episode in the late 1960s in Australia demonstrated this starkly. The manufacturer of phenytoin sodium capsules changed one excipient, replacing calcium sulfate with lactose. Calcium sulfate had been interacting with phenytoin and reducing its absorption. When that interaction disappeared, patients who had been stable on their doses suddenly developed toxic blood levels of the drug, causing an intoxication outbreak.¹⁹²⁰ That incident became a landmark case in pharmaceutical science and helped drive the modern requirement for bioequivalence testing whenever a formulation changes.

Bioavailability in Nutrition

The concept of bioavailability extends well beyond pharmaceuticals. In nutrition, it describes how much of a vitamin or mineral from food actually gets absorbed and used by the body. The principles overlap with drug bioavailability, but the factors at play are sometimes different.

Fat-soluble vitamins like A, D, E, and K are absorbed more efficiently when consumed with dietary fat. Certain vitamin forms are inherently more bioavailable than others: calcifediol is absorbed better than standard cholecalciferol (both forms of vitamin D), and methylfolate outperforms folic acid. On the other side, plant-based foods often have reduced micronutrient bioavailability because minerals get trapped in cellular structures or bound to compounds like phytate and fiber that block absorption.²¹

There is no AUC-based formula for nutritional bioavailability in the way there is for drugs. Instead, researchers use a mix of techniques: balance studies that compare intake to excretion, isotope labeling to track absorption of a specific nutrient, and blood-level measurements after a test meal. The concept is the same as in pharmacology, but the tools are adapted to the messier reality of food.

How Bioavailability Is Predicted Without Human Trials

Running a full pharmacokinetic study in humans is expensive and time-consuming. Computational models called physiologically based pharmacokinetic (PBPK) models have become a standard tool for estimating bioavailability earlier in drug development. These models build a virtual human body, with compartments for the gut, liver, kidneys, and other organs, and simulate how a drug moves through each one based on laboratory measurements of properties like solubility, permeability, and metabolic stability.²² The goal is to predict what will happen in a person before actually dosing anyone.²³

A related approach combines in vitro lab data with in silico PBPK modeling to estimate the hepatic component of bioavailability specifically, predicting how much drug the liver will remove on the first pass.²⁴ These tools are not perfect, but they have become good enough that regulatory agencies now accept PBPK modeling data as part of drug applications, particularly for predicting the effects of food, age, organ impairment, and drug interactions on bioavailability.

The Regulatory Meaning of Bioequivalence

When a generic drug company wants to bring a product to market, it does not need to repeat the original clinical trials. It needs to show that its formulation delivers the same amount of drug to the bloodstream as the brand-name version. This comparison is a relative bioavailability study, and the statistical framework around it is called bioequivalence testing.

The general rule is that the 90% confidence interval for the ratio of AUC values (and peak concentrations) between the generic and the reference must fall within 80–125%. That window sounds wide, but it is applied to the confidence interval rather than the point estimate, which in practice keeps products closer together than the bounds suggest. Where things get tricky is with “highly variable drugs,” compounds whose blood levels bounce around a lot from one dose to the next even in the same person. Regulatory agencies have developed special procedures for these drugs, but the details differ: the FDA and the European Medicines Agency use different statistical constants, and the FDA’s approach creates a discontinuity in the acceptable limits at a within-subject variation of about 30%. The EMA’s approach avoids that discontinuity but produces slightly higher error rates in certain ranges.²⁵

Another regulatory tool, in vitro-in vivo correlation, aims to link a drug’s dissolution behavior in a lab flask to its absorption in a person. When a strong correlation is established, the dissolution test can substitute for a human study, saving time and reducing unnecessary testing.²⁶ The foundational framework for deciding when such correlations are likely to work relies on the same drug classification system based on solubility and permeability, where high-solubility, high-permeability drugs are the best candidates for dissolution-based waivers.²⁷

Beyond the Oral Route

Although the oral route dominates the bioavailability conversation, every route of administration has its own bioavailability profile. Intravenous is 100% by definition. Sublingual delivery (under the tongue) avoids first-pass liver metabolism but the amount absorbed varies by drug and formulation. Rectal, nasal, pulmonary, and transdermal routes each have distinct absorption characteristics, and researchers continue developing non-invasive alternatives for drugs that traditionally require injection, such as heparin.²⁸

Comparative pharmacokinetic studies sometimes test several routes head to head. A study comparing sublingual tablets, suppositories, vaporized delivery, and an approved oromucosal spray for cannabis products found that the vaporizer provided the fastest absorption and the highest peak blood concentration, while the sublingual and rectal formulations delivered somewhat less drug than the oromucosal product.²⁹ Each route sidesteps some barriers and introduces others. The bioavailability formula remains the same regardless of route: you still compare the AUC of the test route against a reference, whether that reference is intravenous or another formulation.