Does coffee raise cholesterol?

Yes, and how you brew it decides how much. A 2020 meta-analysis of 12 randomized trials found that coffee raised total cholesterol by 8.1 mg/dL (95% CI 1.5 to 15; metric: 0.21 mmol/L, 0.04 to 0.39) and LDL by 5.4 mg/dL (95% CI 1.9 to 9.3; metric: 0.14 mmol/L, 0.05 to 0.24), with no change in HDL. Those trials mixed brews. Unfiltered coffee, such as boiled, French press or Turkish, drives most of the rise, and a GRADE-rated review judged with high certainty that it raises LDL moderately to largely, by at least 7.7 mg/dL (0.20 mmol/L). Paper-filtered coffee does far less, though in two trials it was not zero.

Established. Unfiltered coffee raises LDL cholesterol. A meta-analysis of 14 trials found the rise with unfiltered coffee and very little with filtered coffee, and a GRADE-rated review of foods and LDL rated the unfiltered coffee finding high certainty.

What the trials found

All coffee, pooled

In the 12-trial meta-analysis, total cholesterol, LDL and triglycerides rose more as people drank more, in a curve that was not a straight line. The authors suggest staying under 3 cups a day. That cut-off is their suggestion, and no trial tested it as a threshold. The abstract does not give the total number of participants.

Unfiltered against filtered

An older meta-analysis of 14 trials, all from before 1999 and some at high doses, found that unfiltered coffee raised total and LDL cholesterol in step with the dose, while trials of filtered coffee showed very little increase. A later review that ranked foods by their effect on LDL, built from 37 guidelines, 108 systematic reviews and 20 randomized trials, placed unfiltered coffee among the few foods that push LDL up. The rise it rated is of the same size as the fall it rated for oils low in saturated fat, plant sterols and soluble fiber, at least 0.20 to 0.40 mmol/L.

Filtered coffee is not always neutral

Unsettled. How much paper-filtered coffee moves cholesterol is not settled. Three trials found some effect, and none of them compared blinded coffee with a coffee-free arm.

In a crossover trial of 20 healthy volunteers, 3 to 4 cups a day of paper-filtered coffee for 4 weeks raised total cholesterol by 10% to 12% and LDL by 12% to 14%. The two arms were two roasts, with no coffee-free arm, and the trial reports these as percentages without the starting level. In a randomized trial of 191 Norwegian coffee drinkers, people who had averaged 4 cups of filtered coffee a day and then stopped for 6 weeks saw total cholesterol fall by 11 mg/dL (0.28 mmol/L). That trial was unblinded. In 121 healthy adults who stopped and restarted about four cups a day of filtered coffee several times, the two coffee-free periods lowered cholesterol by 8.5 mg/dL (95% CI 5 to 12; metric: 0.22 mmol/L, 0.13 to 0.31) and 14 mg/dL (95% CI 10 to 18; metric: 0.36 mmol/L, 0.26 to 0.46). It had no separate control group.

Espresso and plunger coffee in daily life

The Tromso Study surveyed 21,083 adults aged 40 and over in Norway. Men who drank 3 to 5 cups of espresso a day had total cholesterol 6.2 mg/dL (0.16 mmol/L) higher than men who drank none (95% CI 0.07 to 0.24). For 6 or more cups a day of boiled or plunger coffee, the gap was 0.30 mmol/L in women and 0.23 mmol/L in men. This is a single snapshot, and cup sizes and espresso sources, machines and pods among them, were not standardized.

Why the brew matters

The cholesterol-raising substances are cafestol and kahweol, which a paper filter traps and a metal one lets through. In lab tests of 27 coffee samples, unfiltered coffee from Indonesia held 4.43 mg of cafestol a cup, similar to Scandinavian boiled, Turkish and French press coffee. At 5 cups a day that predicts a rise of 0.33 mmol/L in cholesterol. That rise is modeled from the cafestol content and was not measured in drinkers.

None of these trials measured heart attacks or deaths. They show a change in cholesterol, and how far that change carries through to heart disease is outside what this evidence can tell.

Who should be careful

Not for everyone. People who already have high blood lipids. In the 14-trial meta-analysis, cholesterol rose more in studies of patients with hyperlipidemia, and more again with boiled or caffeinated coffee. If your cholesterol is high, how you brew your coffee is worth raising with whoever treats you.

Genes change the size of the effect. Pooled data from 26 dietary trials in 496 people found that people with the CC form of the CYP7A1 gene had a larger rise in total cholesterol from cafestol than people with the AA form, 1.01 against 27 mg/dL (0.69 mmol/L). The cafestol dose in those trials was far above what a normal cup gives, and the gene groups were not randomized.

Pregnancy is a separate caution about caffeine, not about cholesterol. The review behind the Nordic Nutrition Recommendations 2023 says pregnant women should not exceed 200 mg of caffeine a day, the level the European Food Safety Authority sets as safe for the fetus, and that applies to filtered coffee too. In that same review, high caffeine intake in pregnancy is associated with pregnancy loss, preterm birth and low birth weight, an observational link.

What expert bodies say

The scoping review written for the Nordic Nutrition Recommendations 2023 says high consumption of boiled coffee should be avoided because of its effect on blood lipids. It judges moderate coffee drinking to have no long-term adverse effects, and it notes that randomized trials show brews rich in diterpenes, such as boiled coffee, raise serum cholesterol. We found no other agency position specific to coffee and cholesterol.

How we searched

What would change this answer

• A randomized trial of espresso from modern machines and pods, with cholesterol as the outcome. What we have for espresso is one cross-sectional survey.
• A blinded trial of paper-filtered coffee against no coffee, long enough and large enough to fix the size of its effect.
• Evidence that the cholesterol rise from unfiltered coffee does or does not lead to heart disease.
• A check of trial registries. We did not search ClinicalTrials.gov for this question, so an ongoing trial may exist that we have not seen.

What is in a cup of coffee, nutrient by nutrient, is on the coffee page.

Sources

1. ev-cfch-01 · Meta-analysis or systematic review · systematic review and meta-analysis of randomized controlled trials · n = —
   The pooled results showed that coffee consumption significantly increased total cholesterol (TC) (WMD: 0.21 mmol/L, 95% CI: 0.04; 0.39, P = 0.017), triglyceride (TG) (WMD: 0.12 mmol/L, 95% CI: 0.03; 0.20, P = 0.006) and low-density lipoprotein (LDL-C) (WMD: 0.14 mmol/L, 95% CI: 0.05; 0.24, P = 0.003) while had no significant effect on high-density lipoprotein (HDL-C) (WMD: -0.01 mmol/L, 95% CI: -0.06; 0.04, P = 0.707).
   
   Who: adults in randomized trials of coffee consumption vs no or less coffee
   Effect: TC WMD 0.21 mmol/L (95% CI 0.04 to 0.39); LDL-C 5.4 mg/dL (1.9 to 9.3; metric: 0.14 mmol/L, 0.05 to 0.24); TG 11 mg/dL (2.7 to 18; metric: 0.12 mmol/L, 0.03 to 0.20); HDL-C -0.01 (-0.06 to 0.04)
   Certainty: Trials mixed filtered and unfiltered brews; participant total not given in the abstract.
   Nutr Metab Cardiovasc Dis, 2020 · checked 2026-10-07 · we read the abstract
2. ev-cfch-02 · Meta-analysis or systematic review · systematic review and meta-analysis of randomized controlled trials, dose-response analysis · n = —
   Dose-response analysis results revealed significant positive nonlinear associations between coffee consumption and the increase in TC, LDL-C, and TG levels.
   
   Who: adults in randomized coffee trials
   Effect: positive non-linear dose-response for TC, LDL-C and TG; authors suggest under 3 cups a day
   Certainty: Dose cut-off of 3 cups is the authors' suggestion, not a tested threshold.
   Nutr Metab Cardiovasc Dis, 2020 · checked 2026-10-07 · we read the abstract
3. ev-cfch-03 · Meta-analysis or systematic review · meta-analysis of randomized controlled trials · n = —
   Consumption of unfiltered, but not filtered, coffee increases serum levels of total and LDL cholesterol.
   
   Who: adults in 14 published randomized trials of coffee consumption
   Effect: dose-response for total and LDL cholesterol (p < 0.01); effect seen with unfiltered, not filtered, coffee
   Certainty: Older trials (before 1999); high doses in some; participant total not given in the abstract.
   Am J Epidemiol, 2001 · checked 2026-10-07 · we read the abstract
4. ev-cfch-04 · Meta-analysis or systematic review · meta-analysis of randomized controlled trials · n = —
   Trials using filtered coffee demonstrated very little increase in serum cholesterol.
   
   Who: adults in randomized trials of filtered coffee
   Effect: very little increase in serum cholesterol with filtered coffee
   Certainty: Size of the filtered-coffee effect not given in the abstract.
   Am J Epidemiol, 2001 · checked 2026-10-07 · we read the abstract
5. ev-cfch-05 · Meta-analysis or systematic review · meta-analysis of randomized controlled trials, subgroup analysis · n = —
   Increases in serum lipids were greater in studies of patients with hyperlipidemia and in trials of caffeinated or boiled coffee.
   
   Who: patients with hyperlipidemia and healthy adults in coffee trials
   Effect: larger increases in serum lipids in hyperlipidemic patients and with boiled or caffeinated coffee
   Certainty: Subgroup comparison across trials; people with high cholesterol are the group to be careful.
   Am J Epidemiol, 2001 · checked 2026-10-07 · we read the abstract
6. ev-cfch-06 · Meta-analysis or systematic review · systematic review of guidelines, systematic reviews and RCTs (RCTs for coffee) with GRADE · n = —
   Unfiltered coffee caused a moderate to large increase.
   
   Who: adults in randomized trials of coffee lasting at least 13 days
   Effect: unfiltered coffee: moderate to large LDL increase (moderate = 7.7 to 15 mg/dL (0.20 to 0.40 mmol/L)); high-certainty evidence
   Certainty: High certainty by GRADE for the unfiltered coffee finding.
   Nutr Metab Cardiovasc Dis, 2021 · checked 2026-10-07 · we read the abstract
7. ev-cfch-07 · Randomized controlled trial(s) · randomized crossover trial · n = 20
   Both roasts increased plasma total cholesterol, low-density lipoprotein-cholesterol, and soluble vascular cell adhesion molecule-1 (sVCAM-1) concentrations (10%, 12%, and 18% for MLR; 12%, 14%, and 14% for MR, respectively) (P < 0.05).
   
   Who: healthy volunteers drinking 3 or 4 cups (150 mL) of medium light or medium roast paper-filtered coffee per day for 4 weeks
   Effect: total cholesterol +10% (MLR) and +12% (MR); LDL-C +12% and +14% (P < 0.05)
   Certainty: Very small trial; no no-coffee arm in the crossover between roasts; shows filtered coffee is not always neutral.
   Nutrition, 2013 · checked 2026-10-07 · we read the abstract
8. ev-cfch-08 · Randomized controlled trial(s) · unblinded randomized controlled trial · n = 191
   Abstention from coffee for 6 wk was associated with a decrease in the tHcy concentration of 1.08 micromol/L and a decrease in the total cholesterol concentration of 0.28 mmol/L in participants who had been drinking on average 4 cups of filtered coffee daily for the past year.
   
   Who: healthy non-smoking coffee drinkers aged 24 to 69, mostly drinking caffeinated filtered coffee; no coffee, 1-3 cups or 4+ cups a day for 6 weeks
   Effect: total cholesterol -11 mg/dL (-0.28 mmol/L) after 6 weeks without coffee (habitual 4 cups/day filtered)
   Certainty: Unblinded; Scandinavian brewing and cup sizes.
   Am J Clin Nutr, 2001 · checked 2026-10-07 · we read the abstract
9. ev-cfch-09 · Observational data · population-based cross-sectional study · n = 21,083
   Men drinking 3–5 cups of espresso per day had 0.16 mmol/L (95% CI 0.07 to 0.24) higher S-TC level than men drinking 0 cups of espresso per day.
   
   Who: residents of Tromso, Norway, aged 40 years and older (Tromso7, 2015-2016)
   Effect: men 3-5 cups espresso/day: +0.16 mmol/L (95% CI 0.07 to 0.24) vs 0 cups
   Certainty: Cross-sectional; cup size and espresso source (machines, pods) not standardized.
   Open Heart, 2022, 9(1):e001946 (Svatun et al.) · checked 2026-10-07 · we read the fulltext
10. ev-cfch-10 · Observational data · population-based cross-sectional study · n = 21,083
    From the lowest (0 cups) to the highest (≥6 cups) consumption category, mean S-TC concentrations increased by 0.30 mmol/L (95% CI 0.13 to 0.48) in women and 0.23 mmol/L (95% CI 0.08 to 0.38) in men compared with participants drinking 0 cups of boiled/plunger coffee per day.
    
    Who: residents of Tromso aged 40 and older
    Effect: boiled/plunger 6+ cups vs 0: women +0.30 mmol/L (0.13 to 0.48); men +0.23 mmol/L (0.08 to 0.38)
    Certainty: Cross-sectional; agrees in size with older trials of boiled coffee.
    Open Heart, 2022, 9(1):e001946 (Svatun et al.) · checked 2026-10-07 · we read the fulltext
11. ev-cfch-11 · Observational data · survey of coffee samples plus population cross-sectional study · n = 27 samples
    Unfiltered coffee from Indonesia contained an amount of cafestol (4.43 mg/cup) similar to Scandinavian boiled, Turkish and French press coffee with substantial predicted increases in serum cholesterol (0.33 mmol/l) and triglycerides (0.20 mmol/l) concentrations for consumption of 5 cups per day.
    
    Who: coffee samples from shops in Singapore, Indonesia and India
    Effect: cafestol 4.43 mg/cup unfiltered vs 0.09 mg/cup paper or sock filtered; predicted +0.33 mmol/L cholesterol at 5 cups/day
    Certainty: Predicted cholesterol change is modeled from cafestol content, not measured in drinkers.
    Nutr J, 2011 · checked 2026-10-07 · we read the abstract
12. ev-cfch-12 · Observational data · pooled analysis of controlled dietary trials by genotype · n = 496
    Upon intake of cafestol, AA-subjects had a smaller increase in plasma total cholesterol than CC-subjects (0.69 +/- 0.10 vs. 1.01 +/- 0.10 mmol/L; P = 0.028).
    
    Who: normolipidemic adults in controlled dietary trials, cafestol about 57 mg/day
    Effect: total cholesterol rise with cafestol: AA genotype 0.69 vs CC genotype 39 mg/dL (1.01 mmol/L) (P = 0.028)
    Certainty: Genotype comparison is not randomized; cafestol dose far above a normal cup.
    J Nutr, 2004 · checked 2026-10-07 · we read the abstract
13. ev-cfch-13 · Position of an expert body · scoping review for food-based dietary guidelines · n = —
    High consumption of boiled coffee should be avoided due to negative effect on lipid profile.
    
    Who: adults in the Nordic and Baltic countries
    Effect: advice: avoid high consumption of boiled coffee
    Certainty: Guideline basis document; moderate filtered coffee is judged to have no long-term harm.
    Food Nutr Res, 2024 · checked 2026-10-07 · we read the abstract
14. ev-cfch-14 · Position of an expert body · scoping review for food-based dietary guidelines · n = —
    Pregnant women should not exceed the recommended daily dose of caffeine intake of 200 mg set by the European Food Safety Authority as a safe level for the fetus.
    
    Who: pregnant women
    Effect: caffeine at most 200 mg/day in pregnancy
    Certainty: Limit concerns caffeine, so it also applies to filtered coffee.
    Food Nutr Res, 2024 · checked 2026-10-07 · we read the abstract
15. cof-dr-01 · Meta-analysis or systematic review · meta-analysis of randomized controlled clinical trials of coffee consumption identified from the English-language literature to December 1998, with data abstracted independently by two reviewers and pooled with a random-effects model weighted by inverse variance · n = 14 trials
    Consumption of unfiltered, but not filtered, coffee increases serum levels of total and LDL cholesterol.
    
    Who: 14 published trials of coffee consumption
    Effect: a dose-response relation was found between coffee consumption and both total and LDL cholesterol (p<0.01); increases were larger in people with hyperlipidemia and in trials of caffeinated or boiled coffee; trials using filtered coffee showed very little increase
    Certainty: this meta-analysis is old, and the trials behind it are older, but nothing has overturned it because the mechanism is simple: the diterpenes cafestol and kahweol are trapped by a paper filter and pass through a metal one. It says nothing about heart attacks, only about cholesterol. French press, moka pot, Turkish and boiled coffee are all unfiltered
    Am J Epidemiol, 2001 · checked 2026-09-22 · we read the abstract
16. cof-dr-02 · Meta-analysis or systematic review · systematic review of the accumulated evidence from guidelines, systematic reviews and meta-analyses of randomized controlled trials of at least 13 days, three databases searched from inception to June 2019 plus a guideline database, with GRADE strength of evidence · n = 165 documents
    With high evidence, foods high in unsaturated and low in saturated and trans fatty acids (e.g. rapeseed/canola oil), with added plant sterols/stanols, and high in soluble fiber (e.g. oats, barley, and psyllium) caused at least moderate (i.e. 0.20-0.40 mmol/L) reductions in LDL cholesterol. Unfiltered coffee caused a moderate to large increase.
    
    Who: 37 guidelines, 108 systematic reviews and 20 randomized controlled trials
    Effect: with high evidence, foods high in unsaturated fat, with added plant sterols or stanols, and high in soluble fiber lowered LDL by at least 7.7 to 15 mg/dL (0.20 to 0.40 mmol/L), while unfiltered coffee caused a moderate to large increase; soy protein, tomatoes, flaxseed and almonds caused small reductions; with moderate evidence avocado and turmeric caused moderate to large reductions
    Certainty: this places coffee on a scale rather than in isolation, which is the useful part: the increase from unfiltered coffee is the same size as the decrease from the foods people are told to eat for cholesterol. High evidence here refers to consistency across reviews, not to hard outcomes. Nobody has shown that the cholesterol rise from unfiltered coffee translates into events
    Nutr Metab Cardiovasc Dis, 2021 · checked 2026-09-22 · we read the abstract
17. cof-dr-03 · Position of an expert body · scoping review drawing on qualified systematic reviews from the World Cancer Research Fund, the US Dietary Guidelines Advisory Committee, the European Food Safety Authority and the World Health Organization, plus an additional search, prepared as a basis for food-based dietary guidelines · n = —
    The long-term favorable effects of coffee consumption are related to reduced risk of endometrial and liver cancer, type 2 diabetes, and cardiovascular deaths. However, results from randomized controlled trials (RCTs) suggest that coffee brews that are rich in diterpenes, such as boiled coffee, increase serum cholesterol concentrations. High caffeine intake in pregnancy is associated with higher risk of pregnancy loss, preterm birth, and low birth weight.
    
    Who: published systematic reviews on coffee, tea, sugar-sweetened beverages and low-calorie sweetened beverages
    Effect: moderate coffee and tea consumption do not have long-term adverse health effects; favorable long-term associations for coffee cover endometrial cancer, liver cancer, type 2 diabetes and cardiovascular death; randomized trials show diterpene-rich brews such as boiled coffee raise serum cholesterol; high caffeine intake in pregnancy is associated with higher risk of pregnancy loss, preterm birth and low birth weight
    Certainty: a scoping review of other people's reviews carries the weakest evidence level, and the favorable coffee associations all come from observational data where people who drink coffee differ from those who do not. The pregnancy statement is the one to act on, and the word is associated: no trial randomizes pregnant women to caffeine
    Food Nutr Res, 2024 · checked 2026-09-22 · we read the abstract
18. cof-dr-05 · Observational data · prospective controlled study with four consecutive periods, two of three weeks of total coffee abstention alternating with two of four weeks drinking 20 fl oz (600 mL) of filter-brewed coffee a day, in a free-living population, measuring total and HDL cholesterol, triglycerides, lipoprotein(a), blood pressure and heart rate · n = 121
    The two coffee abstention periods were associated with a decline in serum cholesterol of 0.22 mmol/l (95% CI -0.31, -0.13) and 0.36 mmol/l (95% CI -0.46, -0.26), respectively.
    
    Who: 121 healthy nonsmoking men and women aged 29 to 65
    Effect: the two abstention periods lowered serum cholesterol by 8.5 mg/dL (95% CI -12 to -5; metric: 0.22 mmol/L, -0.31 to -0.13) and 14 mg/dL (-18 to -10; metric: 0.36 mmol/L, -0.46 to -0.26); 20 fl oz (600 mL) a day of filtered coffee raised it by 9.7 mg/dL (5.8 to 14; metric: 0.25 mmol/L, 0.15 to 0.36) and 0.15 mmol/L (0.04 to 0.26)
    Certainty: this is a within-person design without a control group, so seasonal drift and diet changes ride along with the coffee, though repeating the switch four times guards against that. 20 fl oz (600 mL) is about four cups. The effect is smaller than with unfiltered coffee and it is not zero, which is the correction this card adds to the two above
    Eur J Clin Nutr, 2003 · checked 2026-09-22 · we read the abstract