Probiotics: what they actually do, what dose, and whether they're worth it

Probiotics are one of the few supplements with serious research behind them — but only for a handful of narrowly defined situations, not for vague "gut health". Most people take them for a month, feel nothing, and write them off as a scam. Usually they took the wrong strain, at too low a dose, for a problem probiotics don't solve anyway.

What they actually are

A probiotic is a live microorganism that produces a measurable effect when taken in sufficient quantity. The dose is measured in CFU (colony forming units), and labels typically list somewhere between 1 and 50 billion CFU per capsule.

What manufacturers don't emphasise: the effect belongs to the strain, not the genus. "Lactobacillus" on a label tells you nothing useful. What matters is the full designation — for example Lactobacillus rhamnosus GG or Saccharomyces boulardii CNCM I-745. If the box lists only genus and species with no strain code, you have no way to check whether it was ever studied.

Two products with the same genus and the same CFU count can behave completely differently. The strain is the active ingredient, not "probiotic" as a category.

Where the evidence actually sits

Reasonably solid

• Antibiotic-associated diarrhoea. The strongest case. Large reviews show incidence dropping from roughly 18 to 8 cases per 100 people. The doses used were 5–40 billion CFU per day, started on day one of the antibiotic course and continued 7–14 days after the last tablet.
• Acute infectious diarrhoea. Older data suggested about 24 hours shorter duration, but two large 2018 trials in children found no benefit at all. Realistically, the effect is smaller than was claimed for years.
• Irritable bowel syndrome. Moderate effect on bloating and pain. Roughly speaking, about 1 in 7 people gets a clear improvement they wouldn't have got from placebo.

Weak or nothing

• Fat loss. Study differences come in under 1 kg over 8–12 weeks. That's inside the noise of your bathroom scale.
• "Immune support". The best you see is a cold that's 1–2 days shorter. Not nothing, but not a reason to buy capsules forever.
• Strength, hypertrophy, recovery. No evidence a probiotic on its own moves performance. If 2 g of protein per kilo and 7–8 hours of sleep aren't in place, a capsule fixes nothing.
• Bloating in a healthy person with no diagnosis. More common causes: 4 coffees a day, 300 g of vegetables added overnight, or sweeteners like maltitol and sorbitol.

Dose and how to take it

• General use: 1–10 billion CFU per day. More is not automatically better.
• Alongside antibiotics: 10–40 billion CFU per day, spaced 2–3 hours apart from the antibiotic dose.
• The yeast S. boulardii: 250–500 mg per day (roughly 5–10 billion CFU). Its advantage is that antibiotics don't kill it.
• Timing: with a meal or 20–30 minutes before one — food buffers stomach acid and more organisms survive the trip.
• Label check: look for CFU count at end of shelf life, not "at time of manufacture". The gap can be tenfold.

Test them like this instead of taking them blindly for a year:

1. Write down 3 symptoms that bother you and rate each from 0 to 10.
2. Take the same product, same dose, every day for 4 weeks (up to 8 for IBS).
3. Don't change anything else major in your diet during those weeks.
4. Re-rate. If nothing dropped by at least 2 points, stop — that's 15–25 euros a month going nowhere.

Food instead of capsules

Fermented food delivers live cultures plus protein, calcium and fibre. It isn't standardised, but it's cheaper and fits into normal eating.

• Kefir: 200–250 ml per day
• Yoghurt with live cultures: 150–200 g
• Unpasteurised sauerkraut or kimchi: 50–100 g
• Tempeh: 100 g

One caveat: anything pasteurised after fermentation contains zero live bacteria. Canned sauerkraut and "yoghurt" desserts with 12 g of sugar per 100 g don't count. In a 2021 trial, people eating around 6 servings of fermented food per day for 10 weeks showed greater microbiome diversity and lower inflammatory markers than the group that only increased fibre.

Fibre probably matters more

If you're choosing where to put the effort, fibre feeds the bacteria you already have. Target 25–30 g per day for women and 30–38 g for men; average intake sits around 15–20 g.

• 100 g cooked beans or chickpeas: about 7 g
• 50 g rolled oats: about 5 g
• 1 pear with the skin: about 5 g
• 30 g almonds: about 3.5 g

Increase by 5 g per week, with 2–2.5 litres of fluid. Jumping from 15 to 35 g overnight buys you three days of bloating and the wrong conclusion that you "can't tolerate legumes".

Side effects

In healthy people, the usual complaint is gas and bloating for the first 3–7 days. If it lasts beyond 2 weeks or gets worse, stop. Where small intestinal bacterial overgrowth (SIBO) is suspected, probiotics can make things worse rather than better.

When to see a doctor

• Blood in the stool, or black tarry stools
• Unintentional loss of more than 5% of body weight in 6 months (at 80 kg that's 4 kg)
• Diarrhoea lasting more than 14 days, or diarrhoea that wakes you at night
• Fever above 38.5 °C with severe cramping, especially during or after antibiotic treatment
• A weakened immune system — chemotherapy, immunosuppressants, HIV, a central venous catheter, severe acute pancreatitis, recent bowel surgery. In these situations probiotics should not be taken without medical advice; cases of bacteraemia and fungal infection have been reported, and one trial in severe pancreatitis found higher mortality in the probiotic group.

The short version

• The strain with its full code is what matters, not the genus and not the biggest number on the box.
• The clearest benefit is antibiotic-associated diarrhoea: 10–40 billion CFU daily, spaced 2–3 hours from the antibiotic, continued 7–14 days after the course.
• For fat loss, strength and "immunity" the effect is negligible — don't buy them for that.
• Give them 4 weeks with symptom scores before and after; no clear change means stop.
• 25–38 g of fibre a day and 200 ml of kefir are cheaper and probably more useful than a capsule.