Generic drugs are held to a real equivalence bar. The difference is tiny.

A generic drug is required by the FDA to be the same as its brand-name original in active ingredient, strength, dosage form, route and quality, and to be "bioequivalent" — to deliver the drug into the bloodstream at the same rate and extent [s1]. That last requirement is not a slogan; it is a specific statistical test that generics must pass. And when the FDA reviewed how far generics actually differed from their brands across 2,070 studies, the average gap was under 5% [s3]. The popular worry that generics are watered-down substitutes is not supported by the standard or the data.

What "bioequivalent" actually requires

The most common misconception is that "80 to 125%" — a number that circulates among patients and even clinicians — means a generic can contain anywhere from 80% to 125% of the active drug. It does not. A generic must contain the same amount of active ingredient as the brand [s1]. The 80-to-125% window applies to something else: a statistical test of how the drug behaves in the body.

To run it, a manufacturer typically gives the generic and the brand to the same volunteers on separate occasions and measures two things in the blood over time — the peak concentration reached, and the total exposure (the area under the concentration curve). It then calculates the ratio of generic to brand for each measure. Under the FDA's guidance, the 90% confidence interval for that ratio of geometric means must fall entirely within a bioequivalence limit of, in the FDA's words, "usually 80.00" to 125.00% [s2]. Because the entire confidence interval — not just the average — has to sit inside that window, a generic whose true average drifted toward either edge would fail; the test is far stricter than the raw bounds suggest.

How different are generics in practice

The window describes what is allowed. What actually happens is a separate, answerable question, and the FDA answered it. Researchers at the agency pooled 2,070 bioequivalence studies for generic drugs approved over 12 years, from 1996 to 2007, and compared each generic against its brand [s3]. The average difference in total exposure (area under the curve) was 3.56%, and the average difference in peak concentration was 4.35% [s3]. In other words, the typical generic differed from its brand by a few percent — well inside the allowable window, and smaller than the batch-to-batch variation seen in the brand-name drugs themselves.

That does not mean the difference is zero for every product or every patient. For most drugs a few percent is clinically irrelevant. For a narrow set of medicines with a very narrow therapeutic index — where the effective dose and the toxic dose sit close together — some clinicians prefer to keep a patient on a single consistent product, brand or generic, and the FDA applies tighter bioequivalence criteria to some of these. The general reassurance and the narrow exception are both true, and it is worth stating both rather than either alone.

Why the price falls so far

Generics are cheap not because they are lesser but because their makers skip the cost the original bore. A brand-name manufacturer funds the discovery, the full clinical trial programme and the marketing, and holds patents that block competition while it recoups that. A generic maker enters only after those patents expire, proves bioequivalence rather than repeating the entire trial programme, and then competes on price with every other generic maker of the same molecule. Competition, not corner-cutting, is what drives the price down.

The scale of the result is easy to understate. Generics and biosimilars together make up about 90% of all US prescriptions but less than 18% of prescription drug spending [s4]. The savings come precisely because the product is therapeutically equivalent at a small fraction of the price — a substitution most of the health system already makes automatically at the pharmacy counter.

What the evidence supports

The claim that generics are meaningfully weaker or less reliable than brand-name drugs is not supported: they carry the same active ingredient, must pass a stringent bioequivalence test, and in the FDA's own large review differed from their brands by under 5% on average [s1] [s2] [s3]. The narrow, honest caveat is that for a small number of narrow-therapeutic-index drugs, staying on one consistent product is a reasonable preference some clinicians hold. For the overwhelming majority of prescriptions, the difference between generic and brand is measured in low single-digit percentages, and the difference in price is measured in multiples.