Benefits
Lactose digestion compared with milk
In 15 adults who digest lactose poorly, test meals each holding 20 g of lactose were compared. Breath hydrogen, a marker of undigested lactose, was lower after plain kefir than after milk (87 versus 224 ppm x h), flavored kefir gave an intermediate result, and kefirs and yogurts cut rated flatulence by 54% to 71% versus milk. Kefir still contains lactose.
Stool frequency and bowel satisfaction
In 20 adults with functional constipation who drank 500 mL of kefir a day for 4 weeks, stool frequency, stool consistency and bowel satisfaction scores improved from baseline and laxative use fell, while straining did not change significantly. There was no control group, so these before and after changes cannot be separated from placebo effects or time.
Digestive comfort during antibiotic courses
In 125 children aged 1 to 5 taking antibiotics, a commercial kefir drink did not reduce diarrhea compared with a heat-killed placebo (18% versus 21.9%), and no secondary outcome differed. In 82 adults on a 14-day antibiotic regimen for H. pylori, those also given kefir reported fewer and milder side effects than those given milk with placebo.
Fasting blood sugar and insulin markers
Two pooled analyses of small trials agree only in part. A review of 24 intervention studies found lower fasting glucose (by 8.46 mg/dL) and HOMA-IR, an insulin resistance score. A review of six randomized trials found lower fasting insulin and HOMA-IR but no change in fasting glucose or HbA1c. These are blood markers from small trials.
Cholesterol and blood pressure compared with milk
In 75 women with overweight or obesity, extra kefir and extra low-fat milk lowered LDL cholesterol by similar amounts versus a control diet, with no kefir versus milk difference. In 62 adults with metabolic syndrome, kefir raised apolipoprotein A1 versus milk, while blood pressure fell with both. A pooled analysis of seven trials found no blood pressure change.
Gut bacteria composition
In a 6-week placebo-controlled trial in children with ADHD, 125 mL a day of kefir did not change overall gut microbial diversity compared with UHT milk, but several Bifidobacterium species rose. In critically ill adults given kefir, kefir species were found in stool at varying levels. These are microbiome measures, not health outcomes.
Bone density alongside calcium carbonate
In adults with osteoporosis (40 of 69 recruited completed) taking 1,500 mg of calcium carbonate a day, 1,600 mg a day of freeze-dried kefir powder or unfermented milk was added for 6 months. Spine and hip bone density did not differ significantly between groups; the authors reported links between early bone turnover changes and hip density gains within the kefir group.
Mechanism of action
Lactose use during fermentation
Kefir microbes ferment part of the milk's lactose, and Kluyveromyces yeasts make up most of the lactose-using yeasts in kefir. In the lactose test, servings were matched at 20 g of lactose and plain kefir still produced less breath hydrogen than milk, so lower lactose per serving does not explain the result, and the reason was not tested.
Live microbes passing through the gut
Kefir carries live bacteria and yeasts at counts that vary by product; one trial drink was estimated at about 50 billion live microorganisms per 125 mL. In critically ill adults, kefir species were later found in stool at varying levels, which points to passage through the gut, but lasting colonization was not shown.
Kefiran and milk peptides (lab and animal work)
Fermentation produces kefiran, a polysaccharide made mainly by Lactobacillus kefiranofaciens, and peptides released from milk proteins. Lab and animal studies link these compounds to cholesterol, blood pressure enzyme (ACE) and immune effects, but human trials comparing kefir with plain milk did not show matching differences in cholesterol or blood pressure.
Clinical trials
Randomized block feeding study; test meals each holding 20 g of lactose as 2% milk, plain or raspberry kefir, or plain or raspberry yogurt, after a 12-hour fast; breath hydrogen and symptoms were monitored hourly for 8 hours (Hertzler et al. 2003, J Am Diet Assoc)
15 healthy adults with lactose maldigestion.
Breath hydrogen area under the curve was 224 ppm x h after milk versus 87 after plain kefir (P<.001); flavored kefir was intermediate (156). Kefirs and yogurts reduced rated flatulence severity by 54% to 71% relative to milk, and abdominal pain and diarrhea were negligible with every meal. This was a single-meal test, not a long-term trial.
Open, uncontrolled before and after pilot study; 500 mL a day of a probiotic kefir beverage for 4 weeks after a baseline period, with stool diaries and colonic transit studies; no control group (Turan et al. 2014, Turk J Gastroenterol)
20 adults with functional constipation (Rome II), 10 with normal and 10 with slow colonic transit.
Compared with baseline, stool frequency (p<0.001), stool consistency (p=0.014), laxative use (p=0.031) and bowel satisfaction scores (p<0.001) improved, and colonic transit sped up in the slow transit group (p=0.013). Straining did not change significantly (p=0.18). Without a control group, placebo effects and time cannot be ruled out.
Double-blind, randomized, placebo-controlled trial with allocation concealment; a commercially available kefir drink or a heat-killed matching placebo during antibiotic treatment, with 14 days of follow-up (Merenstein et al. 2009, Arch Pediatr Adolesc Med)
125 children aged 1 to 5 years taking antibiotics, recruited in primary care in the Washington, DC area.
The primary outcome, diarrhea during follow-up, did not differ: 18% with kefir versus 21.9% with placebo (relative risk 0.82, 95% CI 0.54 to 1.43). No secondary outcome differed. The authors concluded that kefir did not prevent antibiotic-associated diarrhea.
Randomized, double-blind trial; 14 days of lansoprazole, amoxicillin and clarithromycin twice daily plus either 250 mL of kefir twice daily or 250 mL of milk containing placebo; groups were unequal in size (Bekar et al. 2011, J Med Food)
82 adults with dyspepsia and H. pylori infection confirmed by urea breath test; 46 received kefir and 36 placebo.
Eradication on breath testing at day 45 was 78.2% with kefir versus 50.0% with placebo (P = .026), and side effects of the antibiotic regimen were less frequent and less severe with kefir. Kefir was tested only as an addition to prescription antibiotics in this single trial.
Randomized, double-blind, placebo-controlled parallel trial; 1,600 mg a day of freeze-dried kefir-fermented milk or 1,600 mg of unfermented milk, both with 1,500 mg of calcium carbonate a day; registered NCT02361372 after the study; funded by Taiwan government grants (Tu et al. 2015, PLoS One)
69 adults with osteoporosis were recruited and 40 completed and were analyzed (24 kefir, 16 control); people using bisphosphonates or other bone drugs for more than 4 months were excluded.
In the comparison the authors labeled intention-to-treat, which included only the 40 completers, spine, femoral neck and total hip bone density at 6 months did not differ significantly between groups, nor did bone turnover markers at any time point. Within the kefir group, the authors reported that early turnover changes were associated with greater 6-month hip density gains, and parathyroid hormone rose with kefir but fell in the control group.
Randomized, open-label parallel trial; 180 mL a day of probiotic kefir or unfermented milk for 12 weeks; registered NCT03966846 after the study ended, with lipid profile, glycemic profile and gut microbiota as primary outcomes (Bellikci-Koyu et al. 2022, Nutr Res)
62 adults with metabolic syndrome in Turkey.
Apolipoprotein A1 rose 3.4% with kefir and fell 2.4% with milk (P = .03). LDL cholesterol and apolipoprotein B fell from baseline with kefir only in participants with LDL above 130 mg/dL and not compared with milk. Systolic and diastolic blood pressure fell from baseline with both drinks. No other between-group lipid or glucose difference was reported in the abstract.
Single-center, three-arm, parallel randomized controlled trial; a weight-maintenance diet with 2 servings a day of low-fat dairy (control) or with 2 extra servings a day of low-fat milk or of kefir, for 8 weeks (Fathi et al. 2017, J Clin Lipidol)
75 Iranian women aged 25 to 45 with overweight or obesity.
Compared with control, kefir lowered total cholesterol (-10.4 mg/dL), LDL cholesterol (-9.7 mg/dL) and non-HDL cholesterol (-11.5 mg/dL), all P < .05. The milk group showed similar results, and kefir and milk did not differ from each other.
Randomized, double-blind, placebo-controlled trial; 125 mL a day of organic cow's milk kefir (estimated at about 50 billion live microorganisms per serving) or UHT cow's milk for 6 weeks; registered NCT05155696; funded by The Waterloo Foundation, with kefir supplied at reduced cost by its maker (Lawrence et al. 2025, BMC Psychiatry)
UK children aged 8 to 13 with ADHD; 80 were recruited, 27 withdrew (18 from the kefir group, 9 from the milk group) and only the 53 who completed were analyzed.
The primary outcome, ADHD symptom severity, did not differ between groups. Actigraphy showed fewer minutes awake during the down period with kefir (p = 0.04), but children on kefir self-reported more sleep problems (p = 0.02). Overall gut microbial diversity did not change, while several Bifidobacterium species rose with kefir.