Bioequivalence Analysis (Two One-Sided Tests)
Also known as: TOST Procedure, Average Bioequivalence, BE Analysis, Biyoeşdeğerlik Analizi
Bioequivalence Analysis is a regulatory-grade statistical framework used to determine whether a test drug formulation (generic or reformulated) delivers the active ingredient to the systemic circulation at a rate and extent comparable to a reference product. Introduced by Donald J. Schuirmann in 1987, the method operationalizes equivalence through the Two One-Sided Tests (TOST) procedure, replacing the ambiguous absence-of-difference paradigm with an explicit equivalence margin evaluated on log-transformed pharmacokinetic endpoints such as AUC and C_max.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Bioequivalence Analysis is indicated whenever a generic drug, reformulated product, or post-approval manufacturing change must demonstrate therapeutic interchangeability with a reference listed drug. It assumes a two-period or replicate crossover design with sufficient washout, log-normally distributed PK endpoints, and homogeneous within-subject variance. It is not appropriate for narrow therapeutic index drugs without adjusted limits, for biologics requiring immunogenicity assessment, or when between-subject rather than within-subject designs must be used. Alternatives include reference-scaled average bioequivalence for highly variable drugs.
Strengths & limitations
- Directly controls the consumer risk: the Type I error rate (declaring equivalence falsely) is maintained at alpha by construction.
- The 90% CI criterion is intuitively interpretable and directly auditable by regulatory agencies without access to raw data.
- Works well with the standard two-period crossover design, minimizing subject burden and between-subject confounding.
- Mathematically rigorous equivalence framework that has replaced the scientifically flawed absence-of-significance approach.
- Fixed 80–125% limits may be too wide for narrow therapeutic index drugs and too narrow for highly variable drugs without regulatory adjustment.
- Requires a crossover design; when crossover is infeasible (e.g., irreversible endpoints), power drops sharply in parallel-group designs.
- Sample size is sensitive to within-subject variability; underestimating CV at the planning stage leads to underpowered studies.
- Addresses only average bioequivalence; individual or population bioequivalence may be required in certain regulatory contexts.
Frequently asked
Why is a 90% confidence interval used rather than the conventional 95%?
The TOST procedure involves two one-sided tests each conducted at the 5% significance level. By duality of tests and confidence intervals, rejecting both one-sided nulls at alpha = 0.05 is exactly equivalent to showing the 90% CI falls within the equivalence limits. Using a 95% CI would correspond to alpha = 0.025, imposing a stricter and unnecessary standard.
What does the 80–125% rule mean in practice?
It means the geometric mean ratio of the test to the reference product — for AUC and C_max — must have its entire 90% confidence interval contained within 0.80 to 1.25 on the ratio scale. Note that 1/1.25 = 0.80, so the interval is symmetric on the log scale, reflecting equal tolerance for underdosing and overdosing relative to the reference.
Can Bioequivalence Analysis be applied to highly variable drugs?
Standard TOST with fixed 80–125% limits has low power for highly variable drugs (within-subject CV > 30%). Regulatory agencies permit reference-scaled average bioequivalence for such drugs, where the acceptance limits widen proportionally to the reference product's own variability, provided a replicate crossover design is used and the widened limits stay within 69.84–143.19%.
Sources
- Schuirmann, D. J. (1987). A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. Journal of Pharmacokinetics and Biopharmaceutics, 15(6), 657–680. DOI: 10.1007/BF01068419 ↗
How to cite this page
ScholarGate. (2026, June 2). Bioequivalence Analysis (Two One-Sided Tests). ScholarGate. https://scholargate.app/en/pharmacometrics/bioequivalence-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Equivalence Test (TOST)Statistics↔ compare
- Pharmacokinetic Compartment ModelPharmacometrics↔ compare