Antibiotics that once worked reliably now fail against some infections. The mechanism is evolutionary rather than mysterious, and it operates on timescales measured in years rather than centuries.
Selection happens with every course
A bacterial population is not uniform. Some cells carry variations that let them survive a drug that kills the rest, and those survivors reproduce into the space the others vacated.
This occurs whenever an antibiotic is used, including when it is used correctly. Appropriate use slows the process considerably but does not eliminate the selective pressure.
Because bacterial generations are short and populations enormous, the arithmetic favours the survivors quickly, which is why resistance emerges within years of a drug entering wide use.
Resistance genes move between bacteria
Bacteria do not only inherit traits from a parent cell. They exchange loops of genetic material directly, including with unrelated species sharing the same environment.
A resistance gene arising in a harmless gut organism can therefore transfer to a pathogen without either lineage having been exposed to the drug in a clinical setting.
This horizontal transfer is why resistance spreads far faster than ordinary inheritance would allow, and why it appears in organisms and places with no obvious connection to treatment.
The mechanisms are varied and combinable
Some bacteria produce enzymes that break the drug down, others alter the molecular target so the drug no longer binds, and others pump the compound out before it acts.
A single organism can carry several mechanisms at once, and genes for different mechanisms frequently travel together on the same mobile element.
The practical result is that exposure to one drug can select for resistance to several, which narrows the remaining options faster than the treatment history alone suggests.
Agricultural and environmental use adds pressure
Antibiotics are used in animal production in many countries, and residues plus resistant organisms enter water and soil through waste streams.
Manufacturing effluent contributes where discharge is poorly controlled, creating environments with sustained low-level exposure that is ideal for selecting resistant populations.
Regulation of these uses varies substantially between jurisdictions and has been tightening in several, but the environmental reservoir already established does not disappear when use stops.
The development pipeline is structurally weak
New antibiotics are commercially awkward. A successful one is deliberately held in reserve to preserve its effectiveness, which limits the revenue that would justify developing it.
Most other medicines earn more the more they are used, so investment has flowed elsewhere, and several classes in current use are decades old in origin.
Alternative funding models that pay for availability rather than volume are being trialled in some countries. Anyone with a specific infection needs a clinician, not general information, to decide treatment.