Penicillin and Alcohol: What the Labelling Says, and What It Leaves Unsaid

The NHS gives a one line answer for each of the two penicillins it is asked about most. Phenoxymethylpenicillin: “Yes, you can drink alcohol with phenoxymethylpenicillin.” Flucloxacillin: “Yes, you can drink alcohol while taking flucloxacillin.” No interaction between alcohol and this family is recognised in UK guidance.

That is the whole of the interaction answer. The rest of this page is what the licensed labels actually contain, why the belief refuses to die, and the one penicillin where something worth knowing has nothing to do with drinking.

Penicillin is a family, not a drug

Worth checking the box, because “penicillin” covers several different prescriptions.

Phenoxymethylpenicillin, usually written as penicillin V, is the one the NHS describes as treating “ear, chest, throat and skin infections”, and it is also taken long term to prevent infection in people with sickle cell disease, people who have had their spleen removed, and people who have had rheumatic fever or chorea. Flucloxacillin is the penicillin used against staphylococcal infections, most familiarly of the skin.

Amoxicillin is a penicillin too, and the most prescribed of the three. We have written amoxicillin and alcohol separately, including co-amoxiclav, so this page covers the class and the two penicillins that page does not.

Where alcohol appears in the labelling, which is almost nowhere

We searched all four documents mechanically. The three occurrences in the flucloxacillin leaflet sit inside one sentence naming dehydrated alcohol, isopropyl alcohol and butyl alcohol as constituents of the capsule printing ink. In the penicillin V documents the word does not occur at all.

That is a searched negative, and worth being exact about. Neither label states that alcohol is safe. Both are silent. Their interactions sections are not short: the penicillin V label names bacteriostatic antibiotics, probenecid, methotrexate, oral contraceptives, warfarin and typhoid vaccine among others, and the flucloxacillin label names probenecid, methotrexate, sugammadex, warfarin, voriconazole and paracetamol. Alcohol is absent from a long list rather than missing off a short one, which our reading is what you would expect for a substance with no recognised interaction. It is still not a positive finding of safety.

The nearest thing to one is in the research. The 2020 systematic review Fact versus Fiction in Antimicrobial Agents and Chemotherapy concludes that “alcohol may be consumed with penicillins”, and reports that alcohol slows the rate at which amoxicillin is absorbed while leaving the extent of absorption unchanged. Its verdict on the field is that the evidence behind these beliefs is “poor and controversial”.

One freshness point: the NHS phenoxymethylpenicillin pages carry a review date of 26 January 2022 and were due for review by 26 January 2025, so they are overdue. The flucloxacillin pages were reviewed in June 2025.

Why everyone believes otherwise, and why we are not repeating the usual story

The account you will find on most pages covering this is that penicillin was so scarce during the Second World War that it was recovered from patients’ urine and reused, that beer diluted the supply, and that commanding officers banned drinking as a result.

We went looking for the source. It traces to a personal recollection, reported second hand in a short 2008 BMJ piece by Lwanga and colleagues titled “Do antibiotics and alcohol mix? The beliefs of genitourinary clinic attendees”. We could not obtain the full text, so we are not reproducing that paper’s survey figures and we are not presenting the wartime account as verified history. Our observation is that a story repeated this confidently ought to have firmer footing than we could find.

The belief is correct about a small number of drugs. The NHS antibiotics interactions page names metronidazole and tinidazole as the two to avoid alcohol with completely, describing sickness, stomach pain, hot flushes, palpitations, headaches and dizziness. It separately flags linezolid, affected by fermented drinks, and doxycycline, which it says may be less effective in people with a history of drinking heavily. No penicillin appears on any of those lists. We have covered metronidazole and alcohol separately, because that is the drug people are usually thinking of.

The flucloxacillin liver point, which is real and is not about drinking

Flucloxacillin causes liver injury. The UK label contraindicates it in anyone “with a previous history of flucloxacillin associated jaundice/hepatic dysfunction”, and says it “should be used with caution in patients with evidence of hepatic dysfunction, patients >50 years or patients with underlying disease all of whom are at increased risk of hepatic reactions”.

The part that matters to a reader is the timing. The label says “the onset of these hepatic effects may be delayed for up to two months post-treatment”, and that the reactions have sometimes lasted for months. The NHS side effects page repeats it and names the signs: pale poo with dark pee, and yellowing of the whites of the eyes or the skin, less obvious on black or brown skin.

Wing and colleagues used UK primary care records covering 861,962 people prescribed flucloxacillin between 2000 and 2012. Laboratory confirmed liver injury in the 1 to 45 days after a prescription ran at 8.47 per 100,000, against 1.62 per 100,000 for a comparison antibiotic, rising with repeat prescribing and age to 110.57 per 100,000 among people over 70 on consecutive courses. LiverTox describes the injury as mostly cholestatic, often starting after therapy has stopped, and strongly associated with carrying the HLA-B*57:01 allele.

Now the half that cuts against us. Alcohol is not implicated. The Wing study measured alcohol intake and adjusted for it as a confounder, and does not report drinking as a driver of the risk. The LiverTox chapter covering flucloxacillin does not contain the word alcohol anywhere, and the chapter covering penicillin G and V uses it once, in a case report, only to record that the patient had no history of alcohol abuse.

One other thing in that label deserves a drinker’s attention, and it concerns paracetamol rather than alcohol. It warns of high anion gap metabolic acidosis when flucloxacillin and paracetamol are taken together, calls it serious and in need of urgent treatment, and says it arises “particularly in certain groups of patients at risk, e.g. patients with severe renal impairment, sepsis or malnutrition”. Alcohol is named nowhere in that warning. Our extension, rather than the label’s finding, is that heavy drinking is a well recognised route into malnutrition, and that paracetamol is what people reach for when they feel rough. Ask a pharmacist, not a web page.

Our reading is that the useful instruction is not “do not drink on flucloxacillin” but this: learn the jaundice signs, and know they can arrive weeks after the course finished and you stopped thinking about the antibiotic. Someone who drinks, and whose eyes turn yellow a month later, has an obvious explanation to hand. It may be the wrong one.

What drinking actually does when you have an infection

None of this is an interaction. All of it is more useful than the interaction question.

The overlap in symptoms is the practical part. Flucloxacillin’s common side effects are feeling or being sick, diarrhoea, bloating and indigestion; penicillin V’s are feeling sick and diarrhoea. Alcohol produces all of them. Drink during a course, feel awful, and you cannot tell which did it. The version that matters is where you decide the antibiotic disagrees with you and abandon a course you needed.

Drinking is separately associated with acquiring infections. A 2018 systematic review and meta-analysis in BMJ Open by Simou, Britton and Leonardi-Bee pooled fourteen studies and found an 83% higher risk of community acquired pneumonia among people who drank, or drank more, than among those who drank nothing or less, with a dose response of roughly 8% for each 10 to 20 grams more alcohol a day. The authors report substantial heterogeneity between the studies, attributable in part to continent, confounder adjustment and how pneumonia was diagnosed, so read it as a direction rather than a figure. It is a finding about catching pneumonia, not about drinking while being treated for one, and we are not stretching it into “a drink will make your chest infection last longer”. No source we found shows that drinking during a penicillin course causes an infection to fail to clear.

The dullest risk is the biggest. Penicillin V and flucloxacillin both have to be taken on an empty stomach and more than once a day. Courses depend on remembering, and heavy drinking nights are not. Read the leaflet in your own box, and finish the course.

Years rather than days, which nobody has written about

Everything above assumes a short course. Penicillin V is also prescribed for years at a time as prophylaxis, for people with sickle cell disease, people without a functioning spleen, and people with a history of rheumatic fever or chorea.

Nothing we read addresses what regular drinking over that timescale means for that group. The label is silent, and the NHS page is written for somebody with a sore throat who wants to know about Saturday. Our observation is that this is a genuine gap rather than reassurance. Anyone on penicillin V for years has a serious underlying reason for it, and the question about their drinking belongs with the clinician managing that condition. If that is you, ask them.

If the hard part was the week without drinking

Most people reading this wanted a yes or no before an evening out. A smaller number arrived because going a few days without a drink turned out to be harder than expected. Finding a short break genuinely difficult is a recognised clinical signal rather than a character question, and a course of antibiotics is often what surfaces it. The early signs of alcohol dependence and what a functioning drinker looks like are more useful reading than another page about penicillin. If you drink heavily every day, do not stop abruptly on your own, because unmanaged alcohol withdrawal can be dangerous. Speak to a clinician first.

The Orchid Recovery is a residential addiction and mental health treatment centre in Hang Dong District, Chiang Mai, Thailand, for international English speaking adults. If drinking has stopped being something you can put down, our pages on alcohol addiction treatment and medically supervised alcohol detox set out what that involves, our clinical team is listed here, and you can talk to us. Almost everyone reading this came for the antibiotic answer and needs none of it, which is the right outcome.

We have also written a general page on drinking while taking antibiotics, and one on pregabalin and alcohol, where the interaction is genuine and serious rather than a folk belief.

Sources

• NHS, About phenoxymethylpenicillin
• NHS, Common questions about phenoxymethylpenicillin
• NHS, Common questions about flucloxacillin
• NHS, Side effects of flucloxacillin
• NHS, Who can and cannot take flucloxacillin
• NHS, Antibiotics: interactions
• electronic Medicines Compendium, Phenoxymethylpenicillin 250mg Film-coated Tablets, Summary of Product Characteristics
• electronic Medicines Compendium, Phenoxymethylpenicillin 250mg Film-coated Tablets, Patient Information Leaflet
• electronic Medicines Compendium, Flucloxacillin 500 mg capsules, hard, Summary of Product Characteristics
• electronic Medicines Compendium, Flucloxacillin 500 mg capsules, hard, Patient Information Leaflet
• LiverTox, Penicillins (2nd Generation), NCBI Bookshelf
• LiverTox, Penicillin G and V, NCBI Bookshelf
• Wing K et al, Quantification of the risk of liver injury associated with flucloxacillin: a UK population-based cohort study, Journal of Antimicrobial Chemotherapy 2017
• Mergenhagen KA et al, Fact versus Fiction: a Review of the Evidence behind Alcohol and Antibiotic Interactions, Antimicrobial Agents and Chemotherapy 2020
• Simou E, Britton J, Leonardi-Bee J, Alcohol and the risk of pneumonia: a systematic review and meta-analysis, BMJ Open 2018
• Lwanga J, Mears A, Bingham JS, Bradbeer CS, Do antibiotics and alcohol mix? The beliefs of genitourinary clinic attendees, BMJ 2008;337:a2885