Antibiotic resistance
The Drug That Worked Too Well
Every time you don't finish a course of antibiotics, you are not being cautious — you may be running a tiny, unintentional experiment in accelerated evolution.
The Idea
Antibiotics don't create resistant bacteria — they select for them. The distinction matters. Resistance mutations arise randomly, all the time, in bacterial populations going about their business. What antibiotics do is act as a filter: susceptible bacteria die, resistant ones survive and reproduce. The drug doesn't cause the problem; it reveals and amplifies a problem that was always latent in the population. This is Darwinian selection operating at speeds that make human evolution look glacial. Bacteria can divide every twenty minutes. A single bacterium can become a billion in under twelve hours. In that time, a resistance gene can spread not just through reproduction but through a process called horizontal gene transfer — bacteria literally passing genetic material to unrelated neighbours, the way you might hand someone a USB drive. This is why antibiotic resistance doesn't stay in one place or one species. It moves. What makes the situation genuinely alarming is not that bacteria are becoming resistant to some antibiotics — that was always going to happen — but the rate at which it is occurring relative to the rate at which we are developing new drugs. The pipeline of novel antibiotics has been nearly dry for decades, partly because pharmaceutical economics make them unattractive to develop: a drug you take for a week generates far less revenue than one you take for life.
In the World
In 2016, a woman in Nevada died from an infection caused by Klebsiella pneumoniae — a bacterial strain resistant to all 26 antibiotics available in the United States at the time. There was nothing left to try. She had been hospitalised in India earlier that year, and the bacteria had almost certainly been acquired there, in a healthcare environment where antibiotic use is heavy and regulation uneven. Her case was a glimpse of what epidemiologists had long described as a post-antibiotic era — not a distant apocalyptic scenario but a present reality for an unlucky few. The bacterium that killed her carried a gene called NDM-1, New Delhi Metallo-beta-lactamase, first identified in a Swedish patient of Indian origin in 2008. Within three years, NDM-1-carrying bacteria had been detected on six continents. The gene travels in mobile genetic elements that can hop between bacterial species, and it codes for an enzyme that disables an entire class of antibiotics — carbapenems — that doctors treat as a last resort. The Nevada case didn't make global headlines in the way a new virus would. Resistant bacteria kill quietly, incrementally, often appearing in the record as a secondary complication. Globally, it is estimated that infections caused by drug-resistant organisms now claim more lives each year than malaria and HIV combined — though the number is contested, it is not small.
Why It Matters
This is one of those issues where individual behaviour and systemic failure are genuinely intertwined, which makes it hard to know where to direct concern. Finishing a course of antibiotics matters, but the far larger drivers are agricultural overuse — roughly two-thirds of all antibiotics consumed globally are given to livestock, often not to treat illness but to promote growth — and the economic structures that have left antibiotic development financially unappealing for decades. Knowing this should sharpen your instincts in a few ways. It's worth being more deliberately sceptical when a doctor reaches for an antibiotic prescription for a viral infection, where it will achieve nothing except selecting for resistant bacteria in your gut microbiome. And it's worth understanding that this is fundamentally a collective action problem — the costs of misuse are diffuse and slow, the benefits immediate and personal — which means it almost certainly requires policy solutions, not just behaviour change. The post-antibiotic future is not inevitable. But it requires treating this as the slow-moving emergency it actually is.
A Question to Ponder
If the economic model that funds drug development actively disincentivises the creation of the medicines we most urgently need, what would a better model actually look like — and who would have to give something up for it to exist?
Get a new one of these every morning.
Start learning with Thinkable