What Are Probiotics?
Probiotics are, per the WHO definition, “live microorganisms which, when administered in adequate amounts, confer a health benefit on the host.” The most important genera in supplements: Lactobacillus, Bifidobacterium, and the yeast Saccharomyces boulardii.
The human gut microbiome contains approximately 38 trillion bacteria—roughly as many as human body cells. This community influences digestion, immune defense, inflammation regulation, metabolism, and—via the gut-brain axis—mental health. Approximately 90% of the body’s serotonin is produced in the gut, illustrating the tight link between microbiome and nervous system.
The Central Problem: Strain Specificity
The most important principle in probiotics:
What is true for one strain is not automatically true for another—even within the same genus.
Lactobacillus rhamnosus GG has strong evidence for antibiotic-associated diarrhea. Lactobacillus acidophilus NCFM for IBS. Saccharomyces boulardii CNCM I-745 for traveler’s diarrhea. This does not mean that any Lactobacillus-containing yogurt or cheap combination probiotic has these effects.
EFSA rejected all generic probiotic health claims for precisely this reason—not because specific strains don’t work, but because strain, product, and indication must be jointly substantiated.
What Does the Evidence Show?
Clearly Established (Strain-Specific)
Antibiotic-associated diarrhea (AAD): The strongest evidence area. The Hempel et al. meta-analysis (63 RCTs, n=11,811) shows RR 0.58—a 42% risk reduction. NNT: approximately 13. Strongest evidence: L. rhamnosus GG and S. boulardii CNCM I-745. These strains are also studied for Clostridioides difficile-associated diarrhea, with moderate evidence.
Irritable bowel syndrome (IBS): Ford et al. meta-analysis (43 RCTs): significant improvement in global IBS symptoms and abdominal pain, NNT ~7. No single strain clearly superior; multi-species preparations are well-studied. Effects are moderate but reproducible.
Traveler’s diarrhea: S. boulardii and L. rhamnosus GG show preventive effects in RCTs. Effect size varies by destination and exposure.
Well-Studied but More Moderate Evidence
Immune modulation: Multiple RCTs show shorter cold duration (−1 to −2 days) with specific strains. Effect size small to moderate but clinically plausible.
Lactose intolerance: Fermented products with S. thermophilus and L. bulgaricus improve lactose tolerance through bacterial lactase activity—one of the most mechanistically direct effects.
Vaginal microbiome: Lactobacillus-dominated vaginal microbiome associated with lower infection risk; oral probiotic transfer to the vaginal environment is possible but not consistently demonstrated.
Biologically Plausible but Early
Gut-brain axis and mood: Smaller RCTs show reductions in anxiety and depression scores with specific “psychobiotic” strains. Results promising, but study sizes often <100 participants, short durations. Not suitable for therapeutic use.
Healthy aging and microbiome diversity: Centenarians have a characteristically different microbiome than younger elderly: more diversity, more short-chain fatty acid producers (Akkermansia muciniphila, Faecalibacterium prausnitzii), fewer inflammatory strains. The causal direction is unknown—whether the healthy microbiome causes healthy aging or results from it remains open.
Akkermansia muciniphila is particularly promising: associated in observational studies with healthy weight, better insulin sensitivity, and lower inflammation. Pasteurized Akkermansia supplements are on the market, but clinical evidence remains limited.
Not Proven
- General “gut health” without specific indication
- Lifespan extension in humans
- Cancer prevention
Aging and the Microbiome
The microbiome changes profoundly with age: diversity decreases, opportunistic pathobionts increase, short-chain fatty acid producers (butyrate producers) decline. This is associated with chronic low-grade inflammation (inflammaging).
Whether targeted probiotic supplementation can slow this process is an active research area. Initial studies with specific strains (e.g., Bifidobacterium longum) show improvements in inflammatory markers in older adults—but long-term data are absent.
Quality Criteria When Buying
- Full strain designation (genus + species + strain identifier, e.g., L. rhamnosus GG or L. rhamnosus ATCC 53103)
- CFU count (colony-forming units)—minimum 10⁹ (1 billion) CFU per daily dose at expiration date (not at manufacture)
- Stability: Refrigerated storage or microencapsulated protection; taking with food improves gastric survival
- Gastric survival: Acid-resistant capsule or taken with food—without protection, many strains lose >99% of viability during gastric passage
- Indication match: Choose the strain for which evidence supports your specific indication—not the product with the most strains
Safety
Short-term use in healthy adults is considered likely safe. Common side effects (bloating, changes in stool consistency) are usually transient.
Warnings:
- In immunosuppressed patients, active cancer therapy, or critical illness: increased risk—individual cases of sepsis with probiotic strains documented
- After bariatric surgery or with intestinal catheters: use with caution
- S. boulardii is a yeast: possible interaction with fluconazole (azole antifungals)
For healthy individuals without relevant conditions: standard short-term use (4–12 weeks) is well-tolerated.
Honest Limitations
- Strain specificity makes blanket assessments difficult: the evidence for the strongest strain does not transfer to the entire category
- Many supplements do not contain the viable CFU count stated on the label—independent quality testing (e.g., ConsumerLab) recommended
- Long-term effects (>12 months) and optimal dosing duration are insufficiently studied for most strains
- The microbiome is highly individual: the same intervention study shows very different response rates depending on baseline microbiome composition