Picking the best paediatric probiotic is fundamentally a matching exercise rather than a brand hunt: which strain, against which clinical reason, in which child. The trial literature in children keeps landing on two species, Lactobacillus rhamnosus GG (ATCC 53103) and Saccharomyces boulardii (CNCM I-745) — each deployed to head off or shorten the diarrhoea that trails an antibiotic course or a stomach virus. A third option, Lactobacillus reuteri DSM 17938, has a far tighter brief, restricted to colicky breastfed infants. Beyond those three names, the paediatric evidence in otherwise healthy children ranges from thin to non-existent.
Three anchor documents underpin this piece: the 2023 ESPGHAN paediatric probiotics position paper, the 2019 Cochrane review on antibiotic-associated diarrhoea in children, plus the two 2018 New England Journal of Medicine null findings on acute gastroenteritis. It also flags four paediatric populations in whom a probiotic should never be started without clinician involvement: immunocompromised children, those fitted with a central venous catheter, and infants born preterm. Brand-driven consumer write-ups almost always paper over that safety perimeter — yet in clinical practice it outweighs any individual brand recommendation.
What this article will not do is claim that a probiotic gummy somehow "boosts your child's immunity". The phrase carries no clinical definition, has no European regulatory recognition, and does not belong on any serious paediatric supplement label.
This article is for informational purposes only and is not medical advice. Always consult your child's paediatrician before starting any supplement. Do not exceed labelled dosages. Store all supplements out of children's reach in original child-resistant packaging.
For broader context on supplementation in this age group, see our coverage of children's immune-support supplements.
Talk to your child's paediatrician first: the four populations who should not use probiotics
Retail messaging around paediatric probiotics leans heavily on everyday "wellness" framing. In immunocompetent kids that framing tracks the safety data reasonably well. In four narrow populations it diverges sharply — and almost no parent picks that up from packaging or from a product page.
Do not give your child a probiotic without first talking to your paediatrician if your child:
- is immunocompromised (active cancer, post-haematopoietic stem cell transplant, primary immunodeficiency, or on high-dose immunosuppressive therapy)
- has a central venous catheter (PICU, oncology, long-term parenteral nutrition)
- was born preterm, especially at a birth weight under 1,000 g, outside a structured NICU probiotic protocol
- has short bowel syndrome, or any condition that markedly disrupts the intestinal barrier
None of those warnings is theoretical. In 2017, an European Medicines Agency safety review covering the probiotic yeast Saccharomyces boulardii counted 61 cumulative cases of fungaemia, including 10 fatal cases, almost entirely in critically ill or immunocompromised patients with indwelling central venous catheters [ema2017sb]. Paediatric intensive-care case reports also describe catheter-related S. cerevisiae fungaemia after S. boulardii dosing [atici2017]. The mechanism is not exotic: a live yeast can translocate across a compromised mucosal lining, or it can contaminate the catheter line from sachets opened in the same hospital bay.
The 2021 American Academy of Pediatrics statement on the preterm question is unambiguous. Its Fetus and Newborn Committee published a clinical report whose explicit conclusion was that "current evidence does not support the routine, universal administration of probiotics to preterm infants, particularly those with a birth weight of <1000 g" [poindexter2021]. Behind that wording is partly a product-quality concern: in the US, probiotics are regulated as supplements, not as licensed drugs, so neither strain identity nor batch-level CFU counts are guaranteed to the standard a NICU population would require. The US FDA doubled down in 2023 by issuing a public safety alert after the in-hospital death of a preterm infant from Bifidobacterium longum subsp. infantis sepsis linked to a hospital-administered probiotic product [fda2023preemie].
A separate adult signal is worth flagging. The 2008 PROPATRIA trial, reported in The Lancet, randomised n=296 patients with predicted severe acute pancreatitis to a prophylactic multi-strain probiotic blend versus placebo, and mortality more than doubled in the probiotic arm [besselink2008]. The setting was adult intensive care, not paediatric outpatient practice, but the underlying lesson stands: the label "probiotic" does not automatically translate to "safe in every clinical setting".
Outside those four populations, in immunocompetent kids, both Cochrane evidence and ESPGHAN guidance describe a reassuring safety profile [goldenberg2019] [szajewska2023]. The driver of risk here is the patient, not the dose. If your child sits inside any of those four buckets, the right conversation is a paediatric one, not a retail-shelf one.
What probiotics actually do (and what they don't) in a child's gut
Since 2014, the operative ISAPP consensus definition has not been revised. Probiotics, per that consensus, are "live microorganisms that, when administered in adequate amounts, confer a health benefit on the host" [hill2014]. Three pieces of that sentence do most of the analytical load: live, adequate amounts (a defined CFU dose tied to a named strain), and strain-specific health benefit. Routine paediatric probiotic marketing tends to walk past all three.
The four real mechanisms
First on the list is colonisation resistance. A live transient strain takes up mucosal binding sites and outcompetes potential pathogens for the same nutrient pool. The effect is genuinely transient: within an otherwise healthy child's gut, supplemented strains fail to take up permanent residence and become largely undetectable somewhere between a week and a month after dosing ends [zmora2018]. Hence why a probiotic's benefit manifests during the dosing window itself, rather than as some durable microbiome "reset".
Second is short-chain fatty acid production. Certain probiotic strains — chiefly those within the Bifidobacterium genus — break down dietary fibre into three SCFAs: butyrate, propionate, and acetate. Both the resulting acidified colonic environment and the direct colonocyte fuel they supply produce downstream effects. The magnitude attributable to diarrhoea duration is modest, but in antibiotic-recovery and gastroenteritis settings (discussed further down) it does become measurable.
Third is modulation of mucin and tight-junction proteins. In cell-culture and animal models, L. rhamnosus GG paired with Bifidobacterium infantis drives up mucin-2 expression and reinforces tight-junction components such as ZO-1 and occludin; the net result is a less permeable intestinal barrier. The clinical payoff in healthy children is modest. It contributes to the AAD and acute-gastroenteritis effect sizes, but it does not amount to a standalone "gut healing" claim.
Fourth on the list: direct antimicrobial output. S. boulardii secretes proteases that can cleave Clostridioides difficile toxins A and B, while L. rhamnosus GG releases bacteriocins active against certain Gram-positive pathogens. Neither effect generalises across the genus; both ride on the strain and on the dose.
What probiotics will NOT do
This is the section that paediatric probiotic marketing usually glosses over.
- They do not "boost" any child's immune system; the phrase carries no clinical definition, and EFSA, under Regulation (EC) No 1924/2006, has rejected it as a health claim [efsahealth].
- They will not permanently rewire a healthy child's gut microbial community. Whatever strain you supplement is, by and large, undetectable in stool about a month after the last dose [zmora2018].
- Atopic dermatitis, autism spectrum disorder, ADHD, and so-called "leaky gut" are not curable with probiotics. ESPGHAN 2023 has explicitly judged the evidence insufficient on several of those indications [szajewska2023].
- They are no substitute for antibiotics in a bacterial infection. Treat them as an adjunct only, never as a replacement.
Across the entirety of the EU regulatory record, the European Food Safety Authority has, so far, authorised no probiotic-specific health claims under Regulation (EC) No 1924/2006. The single live-bacteria health claim that has been approved attaches to traditional yoghurt cultures — namely Lactobacillus delbrueckii subsp. bulgaricus together with Streptococcus thermophilus — and it covers improved lactose digestion only; nothing about immunity, nothing about gut health [efsayoghurt]. The term "probiotic" is itself contested on EU labels: most member states refuse to permit it either as a nutrition claim or as a health claim. National interpretations that do permit the word cluster around Italy and Spain, with the Czech Republic, the Netherlands, Bulgaria, Denmark, France, Greece, and Poland joining more recently [ipaeurope]. The "immune support" wording printed on US-marketed probiotic packaging would not clear EU label review on the same product in most member states. For a companion read on what actually belongs in a child's daily nutrition stack, see our children's multivitamin guide.
The strain-specific evidence base: matching strain to outcome
If a single label-reading rule matters for paediatric probiotics, it is this: the trial data attach to the strain, not to the species, and certainly not to the brand. A bare "Lactobacillus rhamnosus" by itself fails the bar. What the literature actually tracks is the alphanumeric strain ID held by a recognised culture collection.
Lactobacillus rhamnosus GG (ATCC 53103), or "LGG"
By trial count alone, LGG sits at the top of the paediatric probiotic evidence base in current commerce. Two anchor documents carry that evidence base. The first: the 2019 Cochrane review on the prevention of paediatric antibiotic-associated diarrhoea. That review pooled 33 RCTs across 6,352 children, returning a relative risk of 0.46 and an NNT of 9 [goldenberg2019]. The second: the ESPGHAN 2023 position paper, which awards LGG ATCC 53103 a weak recommendation, low certainty of evidence, on two indications — namely AAD prevention at the ≥5 billion CFU/day threshold, and as an acute-gastroenteritis adjunct at ≥10 billion CFU/day across a 5- to 7-day course [szajewska2023].
The majority of LGG products are shelf-stable: properly formulated lyophilised material holds up at room temperature without needing refrigeration. Culturelle is the brand most American parents will know; across EU markets the same strain trades under a patchwork of local brand names, generally with the strain ID printed on the panel.
Saccharomyces boulardii (CNCM I-745), or "S. boulardii"
A yeast, rather than a bacterium. That distinction matters in two practical ways. First, antibacterial drugs leave it untouched, so it can be co-administered with the antibiotic without needing the 2-hour gap. Second, in compromised populations the safety signal to watch is fungaemia, not bacteraemia [ema2017sb].
S. boulardii's evidence profile sits closely alongside LGG's. Within the Cochrane synthesis it ranked second-best for AAD prevention, sharing a comparable lower-bound dose of ≥5 billion CFU/day — equivalent in mass terms to 250 to 500 mg/day [goldenberg2019]. ESPGHAN 2023 likewise issues a weak recommendation, low certainty, for S. boulardii CNCM I-745 in acute gastroenteritis, dosed at 250 to 750 mg/day across a 5- to 7-day course [szajewska2023].
On the retail shelf, US parents will most often meet this strain as Florastor or Florastor Kids; in continental Europe the same organism trades as Enterol, Yomogi or Perenterol, depending on country. Check the panel for "CNCM I-745". Strip that designation and the trial data discussed above cannot be presumed to carry over.
Lactobacillus reuteri DSM 17938
This is the strain with the narrowest indication: infant colic. The 2018 individual-patient-data meta-analysis by Sung and colleagues, published in Pediatrics, combined 4 RCTs covering 345 infants and found that L. reuteri DSM 17938 cut daily crying time in breastfed infants with colic by roughly 65 minutes per day; supplemented infants were about twice as likely as their placebo counterparts to achieve a meaningful improvement [sung2018]. No benefit has been demonstrated in formula-fed infants.
For colicky breastfed infants, ESPGHAN 2023 assigns L. reuteri DSM 17938 a conditional recommendation at roughly 10⁸ CFU/day (100 million) given over a 21- to 28-day course [szajewska2023]. The brand parents are likeliest to come across is BioGaia.
Bifidobacterium animalis subsp. lactis BB-12 and others
BB-12 carries supportive paediatric AGE data plus a clean tolerability profile, even though its overall trial volume sits well below LGG's and S. boulardii's. Multi-strain "daily gut health" blends sit in roughly the same evidence neighbourhood: small studies, mixed signals, and no major paediatric guideline body backing routine use. ESPGHAN 2023 does cite an L. rhamnosus 19070-2 plus L. reuteri DSM 12246 combination in AGE, though it grades the underlying evidence very-low-certainty [szajewska2023].
Outside LGG, S. boulardii, and L. reuteri DSM 17938, the working rule of thumb is straightforward: the marketing usually runs ahead of the trial evidence.
Probiotics after antibiotics for kids: the timing protocol that actually works
Of all the paediatric probiotic indications on the table, this is the one with the firmest evidence — and yet consumer-facing write-ups routinely mistime the dosing schedule. Antibiotic courses unsettle gut flora and cause diarrhoea in roughly 20 to 25% of children across the following days and weeks. The 2019 Cochrane analysis demonstrated that pairing a specific probiotic strain with the antibiotic course pulls that rate down from approximately 23% to 8%, corresponding to a number needed to treat of 9. Framed another way: for every nine children put on a probiotic during an antibiotic course, one episode of antibiotic-associated diarrhoea is averted [goldenberg2019].
The supporting trial evidence rests on two strains: Lactobacillus rhamnosus GG (ATCC 53103) at ≥5 billion CFU/day, plus Saccharomyces boulardii (CNCM I-745) at 250 to 500 mg/day. Layered together, Cochrane and ESPGHAN 2023 produce a workable four-part protocol.
When to start
Begin the probiotic on day one of antibiotics, the same day the prescription starts. The most frequent error is waiting until after the antibiotic course has finished. Diarrhoea risk peaks during the course itself and in the days immediately after; the probiotic only works if live cells are circulating in the gut throughout that window of disruption [goldenberg2019].
How to space the doses
With L. rhamnosus GG, or with any other live bacterial probiotic, space the probiotic and the antibiotic by a minimum of 2 hours. Most antibacterial drugs will knock out a meaningful share of any co-ingested bacterial probiotic. S. boulardii sidesteps the issue because it is a yeast rather than a bacterium, so the 2-hour gap is unnecessary against an antibacterial. It does remain vulnerable to antifungals such as fluconazole or nystatin, so the same 2-hour interval applies whenever a child is on both.
How long to continue
Continue the probiotic for 1 to 2 weeks after the antibiotic course ends [goldenberg2019]. The microbiota take longer than the antibiotic schedule to bounce back, and diarrhoea risk does not switch off the moment the last pill is swallowed.
Which strain to pick
In most paediatric cases, the LGG-versus-S. boulardii decision comes down to practicality more than to evidence. The Cochrane review puts their effect sizes in broadly the same neighbourhood. The yeast option, S. boulardii, can sit alongside the antibiotic with no scheduling logistics. The bacterial option, L. rhamnosus GG, brings decades of outpatient paediatric safety data in immunocompetent children. Either is a defensible pick.
What does not work is a "10-strain blend" boasting only a combined CFU number and no per-strain dosing. The Cochrane evidence attaches to specific strains at specific CFU floors, not to a bulk bacterial load.
Probiotics for acute gastroenteritis (vomiting and diarrhoea): an honest read of the evidence
Of all the indications, this one has seen the biggest evidence shift in the past decade, and consumer pages have largely failed to update.
Up to roughly 2015, the pooled meta-analytic picture suggested that LGG and S. boulardii shaved approximately a day off acute-gastroenteritis duration; ESPGHAN 2023 carries that recommendation forward at a weak grade with low certainty of evidence [szajewska2023]. The trials underpinning that read dosed either ≥10 billion CFU/day of LGG ATCC 53103 or 250 to 750 mg/day of S. boulardii CNCM I-745, across courses of 5 to 7 days.
What shifted in 2018 was the publication of two large, rigorously conducted RCTs in the New England Journal of Medicine. In the Schnadower trial (n=971), LGG ATCC 53103 was pitted against placebo in US paediatric emergency departments among children aged 3 months to 4 years with acute gastroenteritis [schnadower2018]. In the Freedman trial (n=886), a Lactobacillus rhamnosus/L. helveticus blend was tested against placebo in Canadian paediatric emergency departments in children aged 3 to 47 months [freedman2018]. Neither trial detected a significant benefit on any clinically meaningful endpoint: diarrhoea duration, vomiting duration, treatment failure, or caregiver work-days lost.
An honest reading is that, in well-nourished children presenting early to well-resourced healthcare systems, the gastroenteritis effect from probiotics is probably smaller than the older literature suggested. A meaningful effect may still surface in malnourished populations or in lower-resource settings — which is where the original Cochrane signal ran strongest. ESPGHAN 2023 has acknowledged both NEJM trials and grades the overall certainty of evidence as low [szajewska2023].
For a parent at home the clinical bottom line stands: oral rehydration solution is the primary intervention. A probiotic is an add-on that may, in some children, trim diarrhoea duration by roughly a day. It is not a treatment for the underlying gastroenteritis, and it does not replace medical assessment when a child presents with dehydration signs, blood in stool, persistent fever, or generally worsening illness.
Probiotics for toddlers (1 to 3) and preschoolers (3 to 5): what changes from school-age
Two variables shift as the age band drops: the appropriate dose falls, and the available formats narrow.
For toddlers between roughly 12 months and three years, the typical AAD-prevention course runs at 1 to 5 billion CFU/day of L. rhamnosus GG, or alternatively 250 mg/day of S. boulardii CNCM I-745, paired with the antibiotic prescription. In an infant or young toddler dosed with L. reuteri DSM 17938 for stubborn functional GI symptoms, the working dose typically sits at 10⁸ CFU/day (100 million). Every one of these schedules is condition-specific and is agreed with a paediatrician; none functions as a default "daily probiotic" prescription. For what it is worth, the ESPGHAN 2023 paper did not endorse routine probiotic dosing for paediatric functional constipation, citing thin evidence [szajewska2023].
Format is the more pressing question. The AAP and AAPD both advise against gummy and chewable supplements in under-4s on choking-risk grounds, and AAPD additionally singles out the cariogenic sugar load carried by gummy supplements [aapd2022]. Typical paediatric probiotic gummies pack 2 to 4 grams of added sugar per serving, usually with food-dye colouring and occasionally with a sugar coating on top. The AAPD added-sugar ceiling for school-age children sits at under 25 grams per day. Two probiotic gummies, a flavoured yoghurt, and a juice box at lunchtime burns through most of that allowance before dinner is served.
For under-4s, the formats that fit are liquid drops (a fit for infants and young toddlers) or stick-pack powder stirred through a small portion of cold, non-acidic food — typically yoghurt or apple sauce. Temperature counts as much as pH: heat kills the live cells, and so does an acidic environment. Stirring the powder into hot porridge or fruit juice cuts the delivered live count substantially.
If your child has trouble with appetite or selective eating that is making everyday nutrition hard, the broader question is probably not a probiotic; for a related angle on age-banded dietary support, see our companion piece on fiber for kids and constipation.
Probiotic formats compared: drops, powder, chewables, gummies, refrigerated vs shelf-stable
Format matters more than brand. The same strain can be packaged in five different ways, and the choice changes what arrives alive in the gut and what arrives accompanied by 4 grams of added sugar.
Drops and powder
The right choice for infants and toddlers. Liquid drops dose easily by the millilitre; powder stick-packs mix into a small amount of cold yoghurt or apple sauce. Both formats avoid the choking and dental concerns that apply to gummies and chewables in the under-4 group. Storage instructions matter. Some drops (particularly L. reuteri DSM 17938 formulations) require refrigeration; others are shelf-stable.
Chewables and capsules
Suitable from preschool age onwards (age 4 and up, in line with AAP choking-prevention guidance). Capsules sidestep the sugar question outright. Chewables usually pack less sugar than gummies, more than capsules. Both deliver well-defined CFU doses, with negligible interference from the delivery matrix.
Gummies: the honest critique
Gummies dominate the consumer-market shelf for paediatric probiotics, and they carry three structural problems that consumer-facing pages routinely play down.
First, the gummy matrix is hostile to CFU stability across shelf life. Live cells inside a gummy are exposed to moisture, oxygen and temperature swings that a properly lyophilised powder largely shields its cells from. The International Probiotics Association best-practice document calls this out plainly [ipa2017]. A gummy labelled "5 billion CFU at manufacture" can deliver materially fewer cells by the expiry date, and the label rarely flags the distinction.
Second, sugar load. Two to four grams of added sugar per serving, multiplied out across a daily-dosing schedule, lands squarely inside the cariogenic-load conversation that AAPD has been pushing for years [aapd2022]. In a school-age child, two gummies a day amounts to as much as 8 g of added sugar, somewhere around a third of AAPD's daily cap.
Third is the appearance-as-candy problem. Bright, fruit-flavoured, jelly-textured supplements register as sweets to a young child, full stop. For decades, iron-containing supplements have headed the list of fatal paediatric supplement poisonings in the US for precisely this reason. Probiotics lack iron's acute toxicity, but the over-consumption mechanism is identical. Store all gummy supplements out of children's reach in original child-resistant packaging.
A practical ranking for paediatric probiotic delivery: drops or powder for infants and toddlers; capsules or chewables for preschool and up; gummies only when no other format is tolerated.
Refrigerated vs shelf-stable
A perennial parental question, usually framed as "are refrigerated probiotics better for kids?". The honest answer turns on two variables: the strain in play, and whether the entire supply chain actually honoured the storage requirement.
A subset of strains — L. rhamnosus GG and S. boulardii in particular — hold up well as shelf-stable lyophilised products and reliably hit the labelled CFU through to expiry at room temperature. Several Bifidobacterium species are, by contrast, thermally fussier and do better when refrigerated.
The metric worth reading is CFU at expiry, not CFU at manufacture. A shelf-stable formulation that holds its labelled CFU through the printed expiry beats a refrigerated formulation that spent a couple of hours baking inside a courier van in 30 °C summer heat. If you do buy refrigerated, source it from a retailer with a proper cold-chain workflow, and get it into the fridge the moment it lands.
Probiotic dosage for kids by age band
Every dose listed in the table is anchored to a specific strain and a specific indication. No generic "daily probiotic dose for kids" actually exists. A figure that worked in one trial against one indication cannot be presumed to carry across to a different strain or to a different indication.
| Age band | Strain | Indication | Dose | Duration |
|---|---|---|---|---|
| Infant (0 to 12 mo, breastfed, paediatrician-directed only) | L. reuteri DSM 17938 | Colic | 10⁸ CFU/day (100 million) | 21 to 28 days |
| Toddler (1 to 3 yr) | L. rhamnosus GG ATCC 53103 | AAD prevention | 1 to 5 × 10⁹ CFU/day | During antibiotic + 1 to 2 weeks after |
| Toddler (1 to 3 yr) | S. boulardii CNCM I-745 | AAD prevention | 250 mg/day | During antibiotic + 1 to 2 weeks after |
| Preschool (3 to 5 yr) | L. rhamnosus GG ATCC 53103 | AAD prevention | 5 × 10⁹ CFU/day (5 billion) | During antibiotic + 1 to 2 weeks after |
| Preschool (3 to 5 yr) | S. boulardii CNCM I-745 | AAD prevention | 250 to 500 mg/day | During antibiotic + 1 to 2 weeks after |
| School-age (6 to 12 yr) | L. rhamnosus GG ATCC 53103 | AAD prevention | 5 to 10 × 10⁹ CFU/day | During antibiotic + 1 to 2 weeks after |
| School-age (6 to 12 yr) | L. rhamnosus GG ATCC 53103 | Acute gastroenteritis adjunct | ≥10¹⁰ CFU/day | 5 to 7 days |
| School-age (6 to 12 yr) | S. boulardii CNCM I-745 | Acute gastroenteritis adjunct | 250 to 750 mg/day | 5 to 7 days |
| Teen (13 to 17 yr) | L. rhamnosus GG ATCC 53103 | AAD prevention | 10 to 40 × 10⁹ CFU/day | During antibiotic + 1 to 2 weeks after |
The doses above derive from the Cochrane evidence [goldenberg2019] combined with the ESPGHAN 2023 position paper [szajewska2023]. Within the Cochrane analysis, the AAD-prevention dose response inflects at the "≥5 billion CFU/day" mark; below that floor, the protective effect did not show up reliably [goldenberg2019].
One general rule runs across the entire table: do not exceed labelled doses, and resist the urge to "stack" multiple probiotic products in pursuit of a bigger combined CFU number. The figures listed above are clinical doses for defined indications, not floors that must be cleared in order for the supplement to "really work". Past the indication-specific floor, the evidence for incremental benefit is inconsistent, and the bill rises linearly with the dose.
What to look for on the label (and the red flags to walk away from)
The label tells you more than the brand name on the front does. A defensible paediatric probiotic label carries five things.
The specific strain designation, written out as genus + species + alphanumeric code. "Lactobacillus rhamnosus GG (ATCC 53103)" passes the bar; a bare "Lactobacillus rhamnosus" does not. "Saccharomyces boulardii CNCM I-745" passes the bar; a bare "S. boulardii" does not. Strip the alphanumeric code and the published trial data cannot reliably be assumed to apply to the actual product in the bottle.
The CFU count guaranteed at expiry, not at the time of manufacture. A common phrasing on gummy and capsule labels is "10 billion CFU at time of manufacture". After six months on a kitchen shelf, the actual delivered dose can drop substantially below that label figure. Better brands print "guaranteed through expiry" instead, and they stand by it.
A third-party testing seal. On US-marketed products, the recognised marks include USP Verified, NSF/ANSI 173, and ConsumerLab. On the EU side, the IPA Best-Practice mark serves as a serviceable proxy. Independent testing is what actually verifies strain identity, the declared CFU count, and freedom from contaminants. By default, dietary supplements guarantee none of those.
Storage instructions that line up with how the product was actually displayed. A refrigerate-required strain sitting on a non-refrigerated supermarket shelf is one to walk past.
Format matched to the child's age. Drops or powder for infants and the toddler band; capsules or chewables from preschool age upward.
Label red flags to walk past include: a "proprietary blend" disclosing only a combined CFU count (per-strain doses unverifiable); claims of "boosts immunity" or "supports immune health" (rejected as health claims in the EU and aspirational in the US); a "probiotic complex of 12+ strains" with no per-strain CFU breakdown; an absent expiration-CFU guarantee; refrigerated strains being sold from ambient shelves; and any product positioned for a clinical condition such as an eczema cure, autism intervention or ADHD support. Each major paediatric guideline body has concluded the evidence is insufficient for those claims [szajewska2023] [thomas2010].
Red flags that mean stop the probiotic and call a paediatrician
In most healthy children, tolerability across the indications above is good. A handful of presentations call for halting the probiotic and getting medical advice.
- Blood in stool, blood in vomit, or black tarry stools. These require paediatric assessment in their own right, irrespective of whether a probiotic is on board.
- Persistent fever above 38.5 °C lasting three days or more; or any high fever at all in an infant younger than 3 months.
- Signs of dehydration: dry mouth, reduced wet nappies, lethargy, sunken eyes, no tears when crying.
- Worsening abdominal pain, distension, or refusal to drink.
- Hives, facial or tongue swelling, or difficulty breathing: the picture of an allergic reaction to something in the product matrix.
- A new diagnosis of immunocompromise, placement of a central venous catheter, or transfer to intensive care. Stop the probiotic and make sure the medical team is told it had been on board. The contraindications flagged at the top of this guide apply continuously, not only at the moment a probiotic is first started.
None of the items above implicate the probiotic itself as the cause of the symptom. The point is to make sure that a deteriorating clinical picture is not being blamed on the probiotic when it actually needs medical evaluation.
For a companion read on the other immune-support category most often confused with probiotics, see our elderberry for kids safety guide.
Frequently asked questions about probiotics for kids
Are probiotics safe for kids?
In immunocompetent children outside the contraindicated populations, the indications outlined above carry a reassuring safety record across both the 2019 Cochrane review and the ESPGHAN 2023 position paper [goldenberg2019] [szajewska2023]. Without medical oversight, however, probiotics are unsafe in four groups: immunocompromised children, kids with central venous catheters, preterm infants outside a structured NICU protocol, and children living with short bowel syndrome. EMA documented 61 cumulative Saccharomyces boulardii fungaemia cases across those populations, ten of them fatal [ema2017sb]. Have that conversation with your child's paediatrician before initiating any probiotic.
How long should I give my child probiotics after antibiotics?
Begin the probiotic on day one of antibiotics; for live bacterial strains, leave at least 2 hours between the probiotic dose and the antibiotic dose; and keep dosing for 1 to 2 weeks once the antibiotic course is finished [goldenberg2019]. The gut microbiota take longer than the antibiotic course itself to recover, and diarrhoea risk persists past the final pill, not just up to it.
What is the best probiotic strain for kids?
Two paediatric strains anchor the strongest evidence for both AAD prevention and the shortening of acute-gastroenteritis episodes. On the bacterial side is Lactobacillus rhamnosus GG (ATCC 53103), at 5 billion CFU per day or above. On the yeast side is Saccharomyces boulardii CNCM I-745, at 250 to 500 mg per day [goldenberg2019] [szajewska2023]. For colic in a breastfed infant, the evidence-backed strain is Lactobacillus reuteri DSM 17938 at 10⁸ CFU per day [sung2018]. No single strain claims a universal "best" award. The right pick is the one that maps to the indication you are treating.
Can toddlers take probiotics?
Yes. In the 1- to 3-year band, probiotics can be deployed for tightly defined indications — AAD prevention during an antibiotic course being the lead example, dosed lower than for older children. The typical range comes in at 1 to 5 billion CFU per day of L. rhamnosus GG, or at 250 mg per day of S. boulardii CNCM I-745. Format is the binding constraint at this age: drops, plus stick-pack powder mixed into cold yoghurt or apple sauce, are the workable options. Gummies and chewables are off the menu under age 4, on combined choking-risk and added-sugar grounds [aapd2022].
Are probiotic gummies as good as refrigerated probiotics for kids?
Generally no, on two grounds. CFU stability inside the gummy matrix erodes across shelf life, so the dose the bottle actually delivers at month six can sit well below the label number [ipa2017]. Gummies also typically pack 2 to 4 grams of added sugar per serving, which the American Academy of Pediatric Dentistry has flagged as cariogenic [aapd2022]. Whether refrigerated outperforms shelf-stable then comes down to which strain is in the bottle. L. rhamnosus GG, for one, holds up well in shelf-stable lyophilised form. The label metric worth reading is CFU guaranteed through expiry, not CFU at the time of manufacture.
Do healthy children need a daily probiotic?
No major paediatric guideline body endorses routinely dosing healthy children with a daily probiotic. The AAP's 2010 clinical report on paediatric probiotics concluded the evidence did not support widespread routine dosing, and the ESPGHAN 2023 position paper limits its endorsements to defined clinical indications [thomas2010] [szajewska2023]. Across the entire EU regulatory record, EFSA has authorised no probiotic-specific health claims under Regulation (EC) No 1924/2006 [efsahealth]. The most coherent framing in healthy kids is to treat probiotics as indication-specific tools, not as a daily supplement category.
What probiotics should kids avoid?
The "what to avoid" question breaks into two layers. By population: absent medical supervision, probiotics are off-limits in immunocompromised children, those carrying a central venous catheter, preterm infants outside an established NICU protocol, and children with short bowel syndrome [ema2017sb] [poindexter2021] [fda2023preemie]. By product: avoid proprietary multi-strain blends that report only a pooled CFU count; avoid anything marketed as a treatment for a clinical condition (eczema, autism, ADHD) where paediatric guidelines have judged the evidence insufficient; avoid products lacking third-party testing; and avoid refrigerated strains being sold off a non-refrigerated supermarket shelf.
The bottom line on the best probiotics for kids
An honest answer to "what is the best probiotic for kids" is that the answer tracks the indication. The two strains with the strongest paediatric trial track record are Lactobacillus rhamnosus GG (ATCC 53103) at 5 billion CFU per day or higher, alongside Saccharomyces boulardii CNCM I-745 at 250 to 500 mg per day. Their clinical role: cutting antibiotic-associated diarrhoea (NNT of 9) and shortening acute-gastroenteritis episodes as an adjunct to oral rehydration [goldenberg2019] [szajewska2023]. Begin them with the first antibiotic dose; if the strain is bacterial, leave a 2-hour gap between probiotic and antibiotic; and continue dosing for another 1 to 2 weeks once the antibiotic course is over. Lactobacillus reuteri DSM 17938 has a narrow, real place in breastfed colicky infants.
What the literature does not back is daily probiotic dosing in healthy children for the purposes of "boosting immunity", preventing or treating eczema, supporting attention or behaviour, or substituting for medical care when a child shows dehydration, persistent fever or bloody diarrhoea. No paediatric guideline body, either European or American, endorses any of those uses.
The biggest safety story that consumer-facing pages tend to miss is not a product story but a population story. Without medical supervision, probiotics are not safe in immunocompromised children, in kids carrying a central venous catheter, or in preterm infants outside a structured NICU protocol. If your child falls into any of those buckets, the appropriate next move is a paediatric appointment, not a retail purchase. For broader continuing reading across the wider category, the children's immune-support hub assembles the evidence-based primers in one location.