From the blog

Cheese Is Low in Carbs. Why Is It So Hard to Stop Eating?

An illustrated wedge of cheese with holes

Quick answers

Does cheese have carbs?
Yes, but very few. The cheeses listed here run from about 0.1 to 2.1 grams per ounce, according to USDA data [1]. What carbohydrate remains is largely leftover milk sugar, and in one Italian cheese lactose was below the quantification limit [2].
Is cheese addictive?
The evidence does not show that it is addictive the way drugs are. In a 2015 survey study, cheese ranked 10th of 35 foods for problematic, addictive-like eating, behind pizza, chocolate and chips [9]. The study measured what people reported, not what happens in the brain.
Do casomorphins in cheese act like opioids?
Casomorphin-7 does bind opioid receptors in lab tests [4]. But a simulated-digestion study found a serving of cheese may not release enough to act as an opioid in the body, and its authors call for human trials [7].

Yes, cheese has carbs, but not many. The cheeses in the table below carry between 0.1 and 2.1 grams per ounce [1]. Whether cheese is "addictive" is a separate question, and the answer is less dramatic than the headlines.

How many carbs are in cheese?

Much of the carbohydrate in fresher cheese is lactose, the sugar in milk. Most of the lactose leaves with the whey when the curds are drained, and bacteria use up more of it early in ripening. In one study of a stretched-curd Italian cheese (Caciocavallo Silano), lactose was below the quantification limit at 30, 60 and 120 days of ripening [2]. That is one cheese from one dairy, so treat it as an example, not a rule.

Here is what USDA data show per ounce (about 28 g) for the cheeses people ask about most [1]:

| Cheese | Carbs per ounce | |---|---| | Brie, Camembert | 0.1 g | | Swiss | 0.4 g | | Cheddar | 0.6 g | | Gouda | 0.6 g | | Mozzarella (whole milk) | 0.7 g | | Blue | 0.7 g | | Parmesan (hard) | 0.9 g | | Cottage cheese (creamed) | 1.0 g | | Feta | 1.1 g | | Ricotta (part-skim) | 1.5 g | | Cream cheese | 1.6 g | | Ricotta (whole milk) | 2.1 g |

Cheddar here is the USDA's sharp, sliced entry. The generic USDA Cheddar entry is higher, at 3.4 grams per 100 g, or about 1 gram per ounce [1]. All values are USDA's carbohydrate-by-difference figures, which are calculated from the other nutrients, not measured as lactose. They were converted from 100 grams to one ounce and rounded.

Aging alone does not sort this list. Parmesan is among the longest-aged cheeses on it and sits near the top. Brie and Camembert are lowest, and whole-milk ricotta and cream cheese are highest. Cottage cheese and ricotta are often served by the cup rather than the ounce, so check the label for your serving size.

Form matters too. USDA lists grated Parmesan at 13.9 grams per 100 g, against 3.2 for the hard wedge [1]. The entry does not say why, but added ingredients are one possibility, so the label tells you more than the cheese name does.

Why does cheese feel so hard to stop eating?

Cheese is concentrated milk. It takes about 10 pounds of milk to make 1 pound of cheese [10]. When milk's casein protein is digested, some of it breaks into small fragments called casomorphins. The best studied is β-casomorphin-7 (BCM-7), which binds to μ-opioid receptors [4], the type of receptor morphine acts on.

That is where the "cheese is addictive" idea comes from. Four details keep it in proportion.

First, BCM-7 comes mainly from one variant of β-casein, called A1. Milk with only the A2 variant does not release it the same way [3]. In 26 commercial US Cheddars, BCM-7 turned up only in cheeses made from milk that carried A1, at up to about 10 milligrams per kilogram, although total casomorphins did not differ between the two cheese types [5]. It has also been found in blue cheeses, where mold enzymes help release it [6].

Second, having an opioid-like peptide in food is not the same as having an opioid effect. In a laboratory model of digestion, a serving of Cheddar made from A1 milk released BCM-7 at 36 to 63 micrograms per 100 grams of protein. The authors concluded that this may not reach the levels needed for opioid activity in the body, and said human trials were still needed [7].

Third, the wider evidence is still limited. One review of BCM-7's possible effects on gut health, fractures and obesity concluded that the evidence "is limited and warrants further pre-clinical and clinical studies" [8].

Fourth, I searched PubMed through October 2026 and could not find a human trial that tests whether casomorphins drive cheese cravings. Human trials of A1 versus A2 milk exist, but they looked at digestive outcomes. The craving idea is a hypothesis, not a finding.

What does the research on food addiction show?

The best-known study here is a 2015 paper from the University of Michigan, published in PLOS ONE [9]. In its second study, 384 adults rated 35 foods on how problematic each was for their eating. Pizza came first, followed by chocolate, chips, cookies, ice cream, French fries, a cheeseburger, soda and cake. Cheese came tenth, with a mean rating of 3.22 on a 7-point scale against 4.01 for pizza. In the first study, 120 undergraduates picked, in pairs, which food they were more likely to have problems with, and cheese was picked 16th most often of 35 [9].

Two points in that paper are worth knowing. The authors classed cheese as unprocessed, with a glycemic load of zero and 9 grams of fat in the study's standard portion. Nine of the ten foods at the top of the second ranking were processed, and cheese was the exception. Across all 35 foods, though, processing was a large predictor of a food being rated as problematic, and fat and glycemic load were large predictors in a separate model [9]. The men in the study also rated unprocessed foods such as cheese as somewhat more problematic than the women did, a small effect [9].

The study measured what people said about their own eating. It did not measure brain activity or casomorphin levels, so it cannot tell us why cheese ranked where it did.

What to do with this

If you are eating low carb, natural cheese fits. It is low in carbohydrate, and the numbers above are the reason.

If cheese is the food you cannot stop eating, I would not blame a peptide. The research has not shown that. Cheese is calorie-dense, fatty and salty. One thing worth trying is putting an ounce or two on a plate before you start, instead of eating from the block or the bag. If a sharp, strongly flavored cheese satisfies you with less, notice that. Both are suggestions, not findings.

Check labels on shredded, processed and fresh cheeses. Added ingredients vary by brand, and the carbohydrate numbers above apply to the plain cheeses listed.

This article is educational and is not medical advice. If you manage a condition that affects what you eat, such as diabetes, kidney disease or high blood pressure, talk to your own clinician about how much cheese fits your plan.

If you want the whole root-cause picture of women's health, the first chapter of Bad Medicine Blues is free. Read a free chapter.

Sources

  1. U.S. Department of Agriculture, Agricultural Research Service. FoodData Central, SR Legacy. Carbohydrate by difference, per 100 g: Brie (FDC 172177), Camembert (172178), Swiss (171251), Cheddar, sharp, sliced (170899), Mozzarella, whole milk (170845), Blue (172175), Gouda (171241), Parmesan, hard (170848), Parmesan, grated (171247), Cheddar, generic (173414), Cottage cheese, creamed (172179), Ricotta, part skim milk (171248), Ricotta, whole milk (170851), Feta (173420), Cream cheese (173418). fdc.nal.usda.gov
  2. Tardiolo G, Di Salvo E, et al. Ripening-associated changes in fatty acid composition and nutritional indices in Caciocavallo Silano PDO cheese. Foods. 2025;14(9):1566. doi.org/10.3390/foods14091566
  3. Giribaldi M, Lamberti C, et al. A2 milk and BCM-7 peptide as emerging parameters of milk quality. Front Nutr. 2022;9:842375. doi.org/10.3389/fnut.2022.842375
  4. Ali AH, Hachem M, et al. β-casomorphin-7, a review of occurrence, identification, techno-functionality, and effects on human health. npj Sci Food. 2026;10(1). doi.org/10.1038/s41538-026-00762-2
  5. Daniloski D, Ibáñez RA, et al. Comparison of the proteolysis and occurrence of a range of β-casomorphin peptides in commercial United States Cheddar cheeses manufactured from A1/A2 and A2/A2 β-casein milks. J Dairy Sci. 2026;109(8):7832-7843. doi.org/10.3168/jds.2025-28235
  6. Caira S, Picariello G, et al. Impact of β-casein variants on the formation of β-casomorphins in blue cheeses. Food Res Int. 2025;201:115556. doi.org/10.1016/j.foodres.2024.115556
  7. Daniloski D, McCarthy NA, et al. Release of β-casomorphin-7 in A1/A1 and A2/A2 dairy systems, insights from simulated semi-dynamic gastrointestinal digestion. Food Chem. 2026;506:148157. doi.org/10.1016/j.foodchem.2026.148157
  8. Aslam H, Ruusunen A, et al. Unravelled facets of milk derived opioid peptides, a focus on gut physiology, fractures and obesity. Int J Food Sci Nutr. 2020;71(1):36-49. doi.org/10.1080/09637486.2019.1614540
  9. Schulte EM, Avena NM, Gearhardt AN. Which foods may be addictive? The roles of processing, fat content, and glycemic load. PLoS One. 2015;10(2):e0117959. doi.org/10.1371/journal.pone.0117959
  10. Oregon State University Extension, Food Hero. Cheese infographic. foodhero.org/sites/default/files/shopping-storage/cheese_infographic_10.24.18.pdf

← All posts

Free — no purchase needed

Get the Companion Workbook, free.

Join the reader list and I will send you the Companion Workbook and Quick-Start Checklist. Simple tools to turn what you are reading into daily practice, before you spend a penny on anything.

You'll get the workbook and checklist right away, then an occasional grounded note about root-cause health. No spam, no selling your address, and every email has a one-click unsubscribe.