Mozzarella: the cheese trials were mostly run on cheddar

The strongest numbers on this page were not measured on mozzarella. In a 2023 meta-analysis of 7 randomized trials, 135 g a day of hard or semi-hard cheese for at least 14 days lowered fasting LDL cholesterol by 0.19 mmol/L (95 percent CI 0.12 to 0.27) compared with about 52 g a day of butter. The reviewers note that fresh mozzarella is not represented in those trials, and that swapping cheese for milk showed no benefit at all. Mozzarella itself has one lipid trial, from 1993. Twenty-six adults with total cholesterol between 5.69 and 7.24 mmol/L ate 100 g a day of part-skim mozzarella for two months, and no lipid value moved significantly. A linoleate-modified mozzarella built for that study lowered LDL by 0.38 mmol/L against the ordinary cheese, which tells you about the modified fat rather than about mozzarella. US law fixes only what the name covers: more than 52 percent and at most 60 percent moisture, with at least 45 percent milkfat in the solids. That is a legal range, so a given block can sit anywhere inside it.

The numbers, per 100 g

How much is actually in it

• US law defines mozzarella by water and fat: more than 52% and at most 60% moisture, with at least 45% milkfat in the solids, which is why USDA and CoFID numbers for fresh and low-moisture mozzarella differ so much per 100 g. — mozzarella cheese sold under that name in the United States Evidence E source
  Moisture more than 52% and not more than 60% by weight; minimum milkfat 45% by weight of the solids (21 CFR 133.155(a)(1)); low-moisture and part-skim variants are defined in 133.156 to 133.158

What your body absorbs

• Vitamin D3 baked into mozzarella on a weekly pizza was absorbed: 28,000 IU once a week for 8 weeks raised serum 25-hydroxyvitamin D by 73 nmol/L in 49 adults, against 5.1 nmol/L with 200 IU, with no adverse events. — 96 apparently healthy, ethnically diverse adults, randomized double-blind, weekly pizza for 8 weeks (n = 96) Evidence B source
  Serum 25(OH)D +5.1 ± 11 nmol/L with 200 IU weekly (n = 47, p = 0.003) and +73 ± 22 nmol/L with 28,000 IU weekly (n = 49, p < 0.0001); PTH fell in the high-dose group; no adverse events

What cooking and storage change

• Eating 120 g of cheddar a day melted rather than unmelted for 6 weeks raised total cholesterol by 0.20 mmol/L and triglycerides by 0.17 mmol/L in 162 overweight adults over 50, with no change in weight, LDL or glucose. — 162 overweight adults aged 50 or over, 43% male, 6-week parallel trial, about 40 g dairy fat per day in three forms (n = 162) Evidence B source
  Melted vs unmelted cheese: total cholesterol +0.23 ± 0.79 vs +0.02 ± 0.67 mmol/L (p = 0.008), triglycerides +0.17 ± 0.39 vs 0.00 ± 0.42 mmol/L (p = 0.016); no difference in HDL, LDL, VLDL, weight, fasting glucose or insulin

What it does to your health

• Eating 100 g a day of part-skim mozzarella for 2 months produced no statistically significant change in blood lipids in 26 adults with moderately high cholesterol, while a linoleate-modified mozzarella product lowered LDL by 0.38 mmol/L relative to it. — 26 adults (17 men, 9 women) with total cholesterol 5.69 to 7.24 mmol/L, community outpatients, 4-month crossover (n = 26) Evidence B source
  No significant lipid change on skim-milk mozzarella; modified-fat cheese lowered LDL-C by 0.28 mmol/L vs baseline (95% CI -0.14 to -0.42) and 0.38 mmol/L vs mozzarella (95% CI -0.2 to -0.70); no change in weight or calories
• In a 2023 meta-analysis of 7 randomized trials, 135 g a day of hard or semi-hard cheese for at least 14 days lowered fasting LDL cholesterol by 0.19 mmol/L relative to about 52 g of butter. Swapping cheese for milk showed no benefit. — adults in 10 RCTs of isoenergetic dairy substitution, searches to June 2022, all rated some concerns for bias (n = 7 RCTs) Evidence A source
  Total cholesterol WMD -0.24 mmol/L (95% CI -0.34 to -0.15), LDL-C -0.19 (-0.27 to -0.12), HDL-C -0.04 (-0.08 to -0.00) versus butter; cheese versus milk: no evidence of benefit (n = 2)

Safety, limits and interactions

• A 1989 multistate US outbreak of 136 Salmonella javiana and 11 S. oranienburg infections was traced to mozzarella from a single plant, with an odds ratio of 7.2, even though the cheese carried as few as 0.36 organisms per 100 g. — Minnesota, Wisconsin, Michigan and New York, May to October 1989; two case-control studies (31 cases vs 60 controls; 50 cases vs 164 controls) (n = 147 cases) Evidence C source
  OR 7.2 (95% CI 1.7 to 23.2, p < 0.01) for mozzarella from plant X or shredded alongside it; most probable numbers 0.36 and 4.3 Salmonella per 100 g in unopened blocks
• In an Italian survey of 3,879 foods in 2000 to 2001, Verocytotoxin-producing E. coli O157 was found in 0.43% of minced beef samples and in none of 2,948 young dairy products, including 501 water-buffalo mozzarella samples. — 3,879 samples from 14 Italian regions, retail, plants and farms, May 2000 to September 2001, 9 veterinary public health laboratories (n = 3,879 samples) Evidence C source
  VTEC O157 in 4 of 931 minced beef (0.43%); 0 of 2,948 dairy products under 60 days ripening, of which 501 buffalo mozzarella; one E. coli O157 VT-eae+ isolate from raw cow's milk cheese
• Across 31 papers and 2,277 cheese samples worldwide, aflatoxin M1 was detected in about half of cheese samples (pooled prevalence 49.85%) and ochratoxin A in about a third (35.64%), with the margin of exposure for aflatoxin M1 below 10,000 in most countries. — 31 papers with 34 data reports, 2,277 cheese samples, 2010 to February 2023, multiple countries (n = 2,277 samples) Evidence C source
  AFM1 pooled prevalence 49.85% (95% CI 37.93 to 61.78%), range 12.18% El Salvador to 100% Serbia; OTA 35.64% (17.16 to 56.44%); MOE for AFM1 under 10,000 for adults except Pakistan and Iran

What people believe that the data does not support

• Across 5 randomized trials, about 145 g a day of hard cheese lowered LDL cholesterol by 6.5% (0.22 mmol/L) compared with butter of the same fatty acid profile, so the saturated fat in cheese does not behave like the same fat in butter. — healthy adults in 12 RCTs, 5 of them comparing hard cheese with butter at a similar P/S ratio (n = 5 RCTs) Evidence A source
  LDL-C -6.5% (-0.22 mmol/L; 95% CI -0.29 to -0.14) and HDL-C -3.9% (-0.05 mmol/L; -0.09 to -0.02) versus butter; no effect on triglycerides; versus tofu or fat-modified cheese, cheese raised LDL as the P/S ratio predicted
• Forty grams of dairy fat a day for 4 weeks as butter raised LDL cholesterol to a median 3.9 mmol/L in 19 mildly hypercholesterolaemic adults. The same fat as cheddar left it at 3.7 mmol/L, not significantly above the 3.4 mmol/L run-in. — 14 men and 5 women, mean age 56, total cholesterol 5.6 mmol/L, free-living, randomized crossover with 4-week periods (n = 19) Evidence B source
  Median LDL-C butter 3.9 (3.5 to 4.1) vs run-in 3.4 (3.0 to 4.1) mmol/L, p < 0.05; cheese 3.7 (3.3 to 3.9), not significant vs run-in; in 13 subjects with LDL over 4 mmol/L, butter 4.4 vs cheese 3.9 mmol/L (p = 0.014)

Who this works differently for

• In pooled data from 197 middle-aged adults, 120 g of cheese a day for 6 weeks lowered total and LDL cholesterol compared with the same fat, protein and calcium given as butter, caseinate and a supplement. Women responded to the cheese matrix more favourably than men. — 197 adults (41.6% male) from two parallel-arm RCTs, cheese n = 104, deconstructed cheese n = 93, about 40 g fat per day for 6 weeks (n = 197) Evidence B source
  Cheese lowered total and LDL cholesterol versus deconstructed cheese in the whole group; no sex x treatment interaction, but within women deconstructed cheese raised total and LDL while cheese lowered them
• Pooling 16 studies, about 1.2% of mozzarella samples at the end of manufacture carried Listeria monocytogenes, at levels below the limit of quantification, typically from cross-contamination after the stretching step. — 16 studies (13 published, 2 grey literature, 10 years of one Italian laboratory database) of mozzarella at the end of manufacturing (n = 16 studies) Evidence C source
  Pooled prevalence 1.2% (95% CI 0.0 to 4.2%), I2 = 78%; concentrations below the limit of quantification
• In a 2023 cross-sectional study of 365 Italian rheumatoid arthritis patients, eating half a pizza more than once a week was associated with about 70% lower disease activity, and mozzarella and olive oil were the pizza components linked to the benefit. — 365 rheumatoid arthritis patients in Italy, tertiles of pizza-related food intake, DAS28-CRP and SDAI (n = 365) Evidence C source
  About 70% reduction in disease activity overall and 80% in the more severe stratum for half a pizza over once a week vs twice a month or less; beta -0.70 for DAS28-CRP, -3.6 for SDAI

What is still unknown

• Across 5 randomized trials, vitamin D3-fortified cheese raised serum 25-hydroxyvitamin D by a non-significant 16.8 nmol/L, against 18.3 nmol/L for fortified milk and 26.2 nmol/L for fortified yoghurt, and the cheese estimate turned significant only after dropping one trial. — 4,965 participants in 35 RCTs of vitamin D-fortified dairy versus unfortified dairy, searches to October 2024 (n = 35 RCTs) Evidence A source
  Cheese D3: MD 16.78 nmol/L (95% CI -3.61 to 37.16, I2 99%, n = 5); leave-one-out 24.13 (4.69 to 43.58); milk D3 18.31 (13.30 to 23.33); yoghurt D3 26.22 (18.67 to 33.77)

The bottom line

Mozzarella gets melted more than any other cheese, and the one trial of melting used cheddar. For 6 weeks, 162 overweight adults aged 50 or over ate about 40 g of dairy fat a day. Melted cheese raised total cholesterol by 0.20 mmol/L and triglycerides by 0.17 mmol/L compared with the same cheese unmelted. Weight, HDL, LDL, fasting glucose and insulin did not differ between the two forms. Nobody has run that comparison on mozzarella, so treat the melting result as a question rather than an answer for the cheese on your pizza. Vitamin D, at least, survives the oven. Baked into mozzarella on a weekly pizza, 28,000 IU once a week for 8 weeks raised serum 25-hydroxyvitamin D by 73 nmol/L in 49 adults, against 5.1 nmol/L on 200 IU weekly, with no adverse events recorded. That was cheese fortified for the trial, and ordinary mozzarella carries little vitamin D. Pooled across 5 trials, vitamin D3-fortified cheese raised 25-hydroxyvitamin D by 16.8 nmol/L, which did not reach significance, and heterogeneity between those trials was extreme at I2 99 percent. The safety cards are about a fresh, high-moisture cheese. Pooling 16 studies of mozzarella at the end of manufacture, 1.2 percent of samples carried Listeria monocytogenes, at levels below the limit of quantification and usually picked up after the stretching step. The authors of that review name pregnant, older and immunocompromised people as the ones the finding is for, and it counts bacteria in samples rather than illness in people. In 1989, an outbreak of 136 Salmonella javiana and 11 S. oranienburg infections across four US states was traced to mozzarella from a single plant. The odds ratio was 7.2 (95 percent CI 1.7 to 23.2) and unopened blocks held as few as 0.36 organisms per 100 g. One plant in one year proves nothing about the cheese counter today. What it does show is that a barely detectable level in a nationally shipped cheese can still make people ill in several states. Mould toxins are a cheese-wide question. Across 31 papers and 2,277 cheese samples of every kind, aflatoxin M1 was detected in 49.85 percent and ochratoxin A in 35.64 percent, counting any detectable level as presence, with no mozzarella figure separated out. An Italian survey of 2000 to 2001 found no Verocytotoxin-producing E. coli O157 in 2,948 young dairy products, 501 of them water-buffalo mozzarella. That was one supply chain a quarter century ago, and absence in a survey is not a guarantee. We have no card on mozzarella in children or alongside any medicine, so this page says nothing about either. The belief worth correcting is that the saturated fat in cheese acts like the saturated fat in butter. Across 5 randomized trials, about 145 g a day of hard cheese lowered LDL cholesterol by 6.5 percent (0.22 mmol/L) against butter of a similar fatty acid profile, and HDL fell by 3.9 percent alongside it. Those trials used hard cheese, mostly cheddar or Gouda, and the reviewers say plainly that they cannot tell whether calcium, particular fatty acids or the structure of the cheese explains the gap. Butter is what flatters cheese here. In the same review, when cheese was compared with tofu or with a fat-modified cheese instead, it raised LDL cholesterol exactly as its ratio of polyunsaturated to saturated fat predicted. A crossover in 19 mildly hypercholesterolaemic adults points the same way. Forty grams of dairy fat a day as butter raised median LDL to 3.9 mmol/L over 4 weeks, while the same fat as matured cheddar left it at 3.7, which was not significantly above the 3.4 mmol/L run-in. None of that is a reason to eat more cheese. It is a reason to stop reading the saturated fat line as the whole story. Pooled data from two trials in 197 middle-aged adults make the same point from the other side. Cheese at 120 g a day for 6 weeks lowered total and LDL cholesterol against the same fat, protein and calcium served apart as butter, caseinate and a supplement. Within women, the taken-apart version raised total and LDL cholesterol while the cheese lowered them, though the sex by treatment interaction was not significant and the cheese was cheddar again. One last association belongs to mozzarella. Among 365 Italian people with rheumatoid arthritis, eating half a pizza more than once a week went with about 70 percent lower disease activity, and mozzarella and olive oil were the components linked to it. That study is cross-sectional and self-reported, from one country, and its own authors ask for cohort studies before anyone believes it.

Sources

1. mozz-c1-13 · regulation
   The minimum milkfat content is 45 percent by weight of the solids, and the moisture content is more than 52 percent but not more than 60 percent by weight as determined by the methods described in § 133.5.
   21 CFR 133.155 Mozzarella cheese and scamorza cheese, eCFR, current · checked 2026-09-23
2. mozz-c1-07 · randomized controlled trial
   Serum 25-hydroxyvitamin D increased by 5.1 ± 11 nmol/L in the low-dose group (n = 47; P = 0.003), and by 73 ± 22 nmol/L in the high-dose group (n = 49; P < 0.0001). None of the subjects in either group developed any adverse events during the supplementation protocol.
   Can J Diet Pract Res, 2015 · checked 2026-09-23
3. mozz-c1-05 · randomized controlled trial
   Melted cheese increased total cholesterol concentrations by 0.20 ± 0.15 mmol L-1 and triglyceride concentrations by 0.17 ± 0.08 mmol L-1 compared to unmelted cheese. No significant differences were observed between the cheese forms for change in HDL, LDL or VLDL cholesterol.
   Food Funct, 2024 · checked 2026-09-23
4. mozz-c1-01 · randomized crossover trial
   No differences in weight or in the amount or type of calories consumed were found during the study. No statistically significant changes in lipid values resulted from consumption of mozzarella cheese.
   Ann Intern Med, 1993 · checked 2026-09-23
5. mozz-c1-03 · meta-analysis of RCTs
   Pooled analyses of 7 RCTs showed a reduction in fasting total cholesterol, LDL-C and HDL-C concentrations after ≥14 d mean daily intake of 135 g cheese (weighted mean difference [WMD]: -0.24 mmol/L; 95% confidence interval (CI): -0.34, -0.15; I2 = 59.8%, WMD: -0.19 mmol/L; 95% CI: -0.27, -0.12; I2 = 42.8%, and WMD: -0.04 mmol/L; 95% CI: -0.08, -0.00; I2 = 58.6%, respectively) relative to ∼52 g/d butter.
   Adv Nutr, 2023 · checked 2026-09-23
6. mozz-c1-10 · case-control outbreak investigation
   Cases were more likely than controls to have consumed mozzarella cheese manufactured at a single cheese plant (plant X) or cheese that had been shredded at processing plants that also shredded cheese manufactured at plant X (odds ratio [OR], 7.2; 95% confidence interval [CI], 1.7 to 23.2; P < .01).
   JAMA, 1992 · checked 2026-09-23
7. mozz-c1-11 · prevalence survey
   VTEC O157 was isolated from four (0.43%) of the minced beef samples, collected in four different regions and during different months, but was not detected in any of the dairy products.
   Int J Food Microbiol, 2004 · checked 2026-09-23
8. mozz-c1-12 · meta-analysis of prevalence studies
   The pooled prevalence of AFM1 was 49.85 %, 95 %CI (37.93-61.78 %)...The pooled prevalence of OTA was 35.64 %, 95 %CI (17.16-56.44 %).
   Rev Environ Health, 2024 · checked 2026-09-23
9. mozz-c1-02 · meta-analysis of RCTs
   Compared with butter intake, cheese intake (weighted mean difference: 145.0 g/d) reduced low-density lipoprotein cholesterol (LDL-C) by 6.5% (-0.22 mmol/l; 95%CI: -0.29 to -0.14) and high-density lipoprotein cholesterol (HDL-C) by 3.9% (-0.05 mmol/l; 95%CI: -0.09 to -0.02) but had no effect on triglycerides.
   Nutr Rev, 2015 · checked 2026-09-23
10. mozz-c1-04 · randomized crossover trial
    A total of 40 g dairy fat eaten daily for 4 weeks as butter, but not as cheese, raised total and LDL cholesterol significantly compared with a diet containing significantly less saturated fat.
    Eur J Clin Nutr, 2005 · checked 2026-09-23
11. mozz-c1-06 · pooled analysis of randomized controlled trials
    In the group as a whole, cheese was found to lower total and LDL cholesterol, compared to deconstructed cheese. No sex × treatment interactions were observed although, within-sex analysis found men and women to respond differently.
    Atherosclerosis, 2025 · checked 2026-09-23
12. mozz-c1-09 · meta-analysis of prevalence studies
    In total, 16 studies were analyzed using random-effects meta-analysis, and the pooled prevalence of L. monocytogenes in mozzarella was estimated at 1.2% (95% CI: 0.0-4.2) although a high heterogeneity among studies was observed (I2 = 78%).
    J Dairy Sci, 2025 · checked 2026-09-23
13. mozz-c1-14 · cross-sectional study
    Among the pizza-related food items/groups, mozzarella cheese and olive oil showed beneficial effects, especially in the more severe stratum. Future cohort studies are needed to confirm this beneficial effect of pizza and related food items/groups on RA disease activity.
    Nutrients, 2023 · checked 2026-09-23
14. mozz-c1-08 · meta-analysis of RCTs
    but insignificantly increased by vitamin D3-fortified cheese (n = 5, MD: 16.78 nmol/L, 95% CI: -3.61-37.16, I2 = 99%). However, the results of vitamin D3-fortified cheese became significant when leave-one-out analysis was performed by omitting one RCT (MD: 24.13 nmol/L, 95% CI: 4.69-43.58, I2 = 90%).
    Nutrients, 2025 · checked 2026-09-23