Antibiotics: When They Help and When They Harm

Almost everyone has had the conversation. You have been coughing for five days, you feel wretched, and someone suggests an antibiotic will sort it out. Sometimes they are right. Far more often, the medicine will do nothing for the illness and something unwanted to you. Understanding why turns a frustrating "no" into a decision that makes sense.

What an antibiotic actually does#

Antibiotics attack structures and processes that bacteria have and human cells do not. Penicillins and cephalosporins break the bacterial cell wall, so the cell bursts under its own internal pressure. Macrolides and tetracyclines jam the bacterial ribosome, which is built differently from ours, so protein production stops. Quinolones interfere with the enzymes bacteria use to unwind and copy their DNA.

Viruses have none of these targets. A virus is genetic material in a protein coat that hijacks your own cells to reproduce. There is no bacterial cell wall to break and no bacterial ribosome to jam. An antibiotic given for a viral illness is not a weak treatment. It is no treatment at all.

This is not a matter of dose or strength. It is a matter of the drug having nothing to bind to.

What is usually viral, and what is usually bacterial#

Notice what this table is not saying. It is not saying antibiotics are bad. In the bottom rows they are among the most life-saving medicines ever developed. The problem is that the top rows account for most of the prescriptions.

The color of your mucus proves nothing#

This deserves its own section because it is the single most common reason people ask for antibiotics.

Green and yellow mucus gets its color from myeloperoxidase, a green-tinted enzyme inside your neutrophils. When those white cells arrive to fight an infection and then break down, they release it. That happens in viral infections just as reliably as in bacterial ones. In fact, most people with an ordinary cold will produce colored mucus somewhere around day three to five, and it clears again as they recover.

What does carry information: how long you have been ill, whether you improved and then clearly got worse again, whether you have a high persistent fever, and whether you are struggling to breathe.

What antibiotics cost you personally#

The argument against unnecessary antibiotics is often framed as a public duty. That is true, but it undersells the personal side.

Side effects are common, not rare. Roughly one in five to one in ten people get diarrhea, nausea or thrush. Allergic reactions range from a mild rash to anaphylaxis. Quinolones carry warnings about tendon rupture and nerve damage; several regulators have restricted their use for minor infections for exactly this reason.

Your microbiome takes a real hit. A single course measurably reduces gut bacterial diversity, and studies tracking recovery show that some species do not return for six to twelve months, with a few not recovering at all. That gut community influences digestion, immune signaling and how you absorb nutrients.

Clostridioides difficile. When antibiotics clear the competing bacteria, C. difficile can overgrow and cause severe, sometimes dangerous colitis. Risk is highest in older adults, in hospital, and after broad-spectrum agents such as clindamycin, cephalosporins and quinolones.

You may be carrying resistance for months afterwards. Resistant bacteria colonise your gut and skin after a course and can persist for up to a year. If you then get a real infection, your own bacteria may be harder to treat.

Resistance: how it actually happens#

Bacteria reproduce roughly every twenty minutes, and they make copying errors. In any large population, a few individuals will by chance carry a change that helps them survive a given antibiotic. A modified target site, a pump that expels the drug, or an enzyme that chops it apart.

Antibiotic exposure does not create these mutants. It removes their competition. The susceptible bacteria die, the resistant ones inherit the space and multiply. Worse, bacteria trade resistance genes directly with each other on small loops of DNA called plasmids, including across species. A harmless gut bacterium can hand a resistance gene to a dangerous one.

The consequences are already measurable. WHO describes antimicrobial resistance as one of the top global public health threats, and large modeling studies attribute well over a million deaths each year directly to bacterial resistance, with several million more associated with it. Carbapenem-resistant Klebsiella, extensively drug-resistant typhoid, and multidrug-resistant tuberculosis are not future problems. They are current ward realities.

The South Asia and Gulf picture#

This is where honesty is more useful than diplomacy.

In much of the world, antibiotics are effectively available on request at a pharmacy counter, and even where prescription rules are strict, leftover courses sit in family medicine cabinets and get shared. In North America the pressure works differently: patients expect a prescription to leave the visit with, and clinicians short on time often write one for an illness that was always going to be a virus.

Several forces feed it, and none of them are stupidity:

• Seeing a doctor costs money and a day of lost wages. A pharmacist costs neither.
• Distances are large and clinics are crowded.
• There is a genuine cultural expectation that an illness deserves a medicine.
• Injections and "strong" antibiotics are widely believed to work better than tablets or rest.
• Courses get shortened because money is short, so partial treatment is common.
• Antibiotics are used heavily in poultry and livestock for growth and disease prevention, and resistant organizms move into people through food, water and the environment.

The result is predictable everywhere: resistance rates climb wherever antibiotics flow freely. The world has already seen outbreaks of common infections that stopped responding to almost every oral option. That is not a distant problem. Resistant organizms travel on planes.

If you live somewhere antibiotics are easy to buy, the most useful individual thing you can do is stop treating them as a household remedy for fever. Fever is a symptom, not a diagnosis.

What genuinely helps a viral illness#

Nothing here is exciting, which is part of why antibiotics stay popular.

• Fluids, rest, and accepting that recovery takes the time it takes.
• Paracetamol or ibuprofen for fever and aches, at the doses on the packet, if you have no reason to avoid them.
• Saline nasal rinses and steam for congestion.
• Honey for cough in anyone over one year old. The evidence for this is better than for most cough syrups.
• Salt water gargles for a sore throat.
• Stopping smoking, even for the duration of the illness.

The question I get asked most#

"Can you just give me something strong so it clears faster?"

I understand the request completely. Someone has a wedding, a shift they cannot miss, an exam, or three days of unpaid leave they cannot afford. They are not being unreasonable. They are being practical.

What I have learned is that refusing without explaining fails. If I simply say no, the person goes to a pharmacy and buys the antibiotic anyway, and now they have taken it without anyone monitoring them. What works better is telling them exactly what to expect: this cough will peak around day four, it will still be there at day ten, and here are the three specific things that would make me change my mind and want to see you again.

The other thing I say often, and mean: the strongest antibiotic in the world does nothing for a virus, and the request for a "stronger" one usually means the illness has simply not run its course yet. People accept a timeline far more readily than they accept a refusal.

At work#

In occupational health this plays out at scale. In a plant with hundreds of workers on rotating shifts, a respiratory virus moves through a shared canteen and locker room in about ten days. What I see is a wave of workers who have each self-medicated with a different leftover antibiotic, because a sick day costs them money and they want to shorten the illness.

Two things reduce this more than any lecture. First, sick leave that does not financially punish people. Presenteeism spreads the virus and drives self-medication at the same time. Second, making the occupational health room the easy option, so a worker can be assessed in fifteen minutes without losing pay. Handwashing facilities, ventilation in break areas, and annual flu vaccination for shift workers reduce the number of infections in the first place, which is the only intervention that reliably reduces antibiotic use.

When you genuinely should seek care#

Some infections need antibiotics quickly, and delay is the real danger. Seek medical assessment promptly for a fever with shivering rigors, breathlessness, confusion, a stiff neck, severe one-sided throat pain with difficulty swallowing saliva, spreading skin redness, symptoms that clearly worsen after an initial improvement, or any illness in a person who is pregnant, immunosuppressed, receiving chemotherapy, or has had their spleen removed. In young infants and frail older adults, the usual signs are muted and the threshold for review should be much lower.

The bottom line#

Antibiotics are among the greatest achievements in medicine and they are being spent carelessly on illnesses they cannot touch. Colds, flu and most sore throats and coughs are viral, and no antibiotic will shorten them. Every unnecessary course carries a real personal cost in side effects and gut disruption, plus a shared cost that eventually returns as an infection nobody can treat. Use them when a doctor judges a bacterial infection likely, take the course as prescribed, and let the viral illnesses take their unglamorous course.