(© Prathankarnpap – stock.adobe.com)
In a nutshell
- Workplace coffee machines contain higher levels of cholesterol-raising compounds (diterpenes) than paper-filtered coffee due to their metal filters.
- Switching from machine-brewed to paper-filtered coffee at work could potentially reduce LDL cholesterol and cardiovascular risk over time.
- Not all machines are equal: liquid-model machines generally produce coffee with low diterpene content similar to paper-filtered coffee.
UPPSALA, Sweden — That coffee machine in your office break room might be doing more than just keeping you alert during the workday. According to new Swedish research, it could be pumping your body full of compounds that boost your cholesterol levels, potentially increasing your cardiovascular risk over time.
Scientists at Uppsala University and Chalmers University of Technology have discovered that workplace coffee machines—those convenient automated brewers found in offices worldwide—contain significantly higher levels of these cholesterol-elevating compounds than traditional paper-filtered coffee. The alarming finding connects your daily caffeine habit at work to heart health in ways nobody suspected.
The Cholesterol-Coffee Connection
Coffee consumption has long been associated with potential health benefits, from improved cognitive function to reduced risk of certain diseases. However, not all coffee is created equal when it comes to cardiovascular health. Since the 1980s, scientists have known that unfiltered coffee can raise LDL (low-density lipoprotein) cholesterol—the “bad” kind associated with heart disease risk.
The new research from examined coffee from 14 machines in Swedish healthcare facilities, analyzing their content of two specific compounds: cafestol and kahweol.
These natural chemicals, known as diterpenes, were identified in the 1990s as the main culprits behind unfiltered coffee’s cholesterol-raising effect. Paper filters typically trap these compounds, but metal filters found in many coffee machines allow them to pass into your cup.
The numbers tell a compelling story. Researchers found median cafestol concentrations of 176 milligrams per liter in brewing machine coffee—nearly 15 times higher than the 12 milligrams per liter found in paper-filtered brews. Kahweol showed similar patterns.
In Nordic countries like Sweden, workplace coffee isn’t just an occasional indulgence—it’s practically institutionalized. Many companies provide free coffee as a tax-deductible benefit, making those self-serve machines standard features in office rest areas. Workers might consume three or more cups daily throughout their careers, creating decades of exposure to these cholesterol-raising compounds.
Long-Term Health Impact
The researchers calculated that replacing three cups of machine coffee with paper-filtered alternatives five days weekly could reduce LDL cholesterol by 0.58 millimoles per liter. For context, that effect equals what would happen if you added about two ounces of heavy cream to each cup of paper-filtered coffee.
The long-term impact is remarkable. Based on established risk models, this reduction in LDL cholesterol could decrease heart disease risk by approximately 13% over five years. Over an entire 40-year career, the risk reduction could reach 36%—simply by changing how your coffee is filtered.
The study, published in Nutrition, Metabolism and Cardiovascular Diseases, identified three main types of workplace machines: brewing machines, liquid-model machines, and instant machines. The brewing machines, which produce coffee in 10-30 seconds by passing hot water through ground beans and a metal filter, showed the highest diterpene levels. In contrast, liquid-model machines—which mix liquid coffee concentrate with hot water—generally contained much lower levels comparable to paper-filtered coffee.
For comparison, the team also tested various home-brewing methods. French press and percolator coffees contained moderate diterpene levels (around 90 mg/L cafestol), while espresso samples varied wildly, with some containing extremely high concentrations up to 2447 mg/L.
The Power of Filtration
One striking finding was how effectively even simple filtration methods reduce these compounds. When researchers poured boiled coffee through a fabric “sock” filter (a two-layer polyester/acrylic fabric), cafestol levels dropped dramatically from 939 mg/L to just 28 mg/L—highlighting that the filtering process itself plays the crucial role.
Why do these machines produce higher diterpene levels? The metal filters commonly used in brewing machines have larger pores compared to paper filters, allowing more diterpenes to pass through. The researchers also noted considerable variation between samples from the same machines taken weeks apart, suggesting that cleaning schedules might affect filter porosity. Ironically, cleaning a metal filter might actually increase its permeability, allowing more diterpenes into your cup.
This research aligns with a large Norwegian study showing that habitual unfiltered coffee consumption is associated with higher cardiovascular mortality over 20 years. The 2023 Nordic Nutrition Recommendations specifically advise choosing filtered over unfiltered coffee for heart health. Know
For coffee lovers, there’s no need to panic or abandon your beloved beverage. Paper-filtered coffee—whether drip-brewed at home or from liquid-model workplace machines—appears perfectly safe for cardiovascular health. Even fabric filters proved remarkably effective at removing diterpenes. A single cup of machine coffee won’t harm you, but decades of daily exposure could contribute to cardiovascular risk in subtle ways.
Next time you’re reaching for that workplace coffee machine, consider what’s actually brewing in your cup. Your future self might thank you for switching to filtered options—especially if you plan to enjoy a long career fueled by coffee.
Paper Summary
Methodology
The research team collected coffee samples from 14 machines across four Swedish healthcare facilities. They selected machines from Uppsala University Hospital and Falu County Hospital (six from each), plus two from primary health care centers. For each machine, they took two samples 2-3 weeks apart, choosing the standard settings for a brewed coffee cup. The samples represented different commercial brands and were frozen before being transported for analysis.
Beyond workplace machines, the researchers prepared various coffee types at home for comparison. They made drip-brewed coffee using five common brands and two different paper filters (regular and perforated “aroma” filters), percolator coffee, French press coffee, and boiled coffee. They even tested an improvised filtration method by pouring boiled coffee through a fabric “sock” made of two-layer polyester/acrylic material. Four espresso samples were also collected from cafeterias and a laboratory workplace.
All samples underwent analysis using liquid chromatography-mass spectrometry to measure cafestol and kahweol concentrations with high precision. The analytical method showed excellent repeatability, with low coefficient of variation values across multiple quality control samples.
Results
The findings revealed stark differences between brewing methods. Of the 14 workplace machines, 11 were classified as brewing machines and 3 as liquid-model machines. None were instant machines (which typically use freeze-dried, previously filtered coffee).
Brewing machines produced coffee with median cafestol and kahweol concentrations of 176 mg/L and 142 mg/L, respectively—far exceeding the levels in paper-filtered coffee (12 mg/L cafestol, 8 mg/L kahweol). Only one brewing machine sample had a cafestol concentration below 100 mg/L. Liquid-model machines generally produced coffee with low diterpene concentrations similar to paper-filtered coffee, except for one outlier sample with unusually high levels. When excluding this outlier, the average cafestol concentration was just 5.9 mg/L—comparable to paper-filtered coffee.
Boiled coffee contained the highest diterpene levels (939 mg/L cafestol, 678 mg/L kahweol), but passing it through the fabric filter reduced these by about 97% (to 28 mg/L cafestol). French press and percolator methods showed moderate levels around 90 mg/L cafestol. Espresso samples showed extreme variation, with cafestol concentrations ranging from 36 mg/L to 2447 mg/L. This wide range suggests that espresso preparation methods significantly affect diterpene content.
The researchers calculated that replacing three cups of brewing machine coffee with paper-filtered coffee five days weekly could reduce LDL cholesterol by 0.58 mmol/L. Based on established cardiovascular risk models, this could translate to a 13% reduction in atherosclerotic cardiovascular disease risk over five years, or a 36% reduction over a 40-year working life.
Limitations
The study had several important constraints worth noting. With only 14 workplace machines analyzed, the sample size was relatively small. The researchers also lacked detailed information about each machine’s design specifications, particularly regarding filter characteristics, water pressure and temperature, contact time, and coffee grinding and roasting parameters.
While the analytical method showed high precision, the researchers acknowledged that their measured diterpene concentrations were higher than in some previous studies. This could potentially lead to overestimating the effects on LDL cholesterol and cardiovascular risk. Most importantly, the study calculated theoretical effects on cholesterol levels and cardiovascular risk based on previously established relationships between diterpenes and LDL cholesterol. These calculations remain theoretical until directly confirmed through randomized controlled trials measuring actual cholesterol changes in people drinking different types of workplace coffee.
The researchers also noted substantial variation between samples from the same machines taken weeks apart, with an average difference of 58% for cafestol and 65% for kahweol. This suggests that machine operation conditions significantly affect diterpene levels, adding another layer of complexity to assessing real-world exposure.
Key Takeaways and Implications
This research highlights how everyday workplace habits might affect long-term health in unexpected ways. Most workplace brewing machines produced coffee with significantly higher diterpene concentrations than paper-filtered coffee, potentially impacting cholesterol levels over time.
The filtration process emerged as the critical factor determining diterpene content. Paper filters effectively removed these compounds, while metal filters allowed more to pass through. Even simple fabric filtration dramatically reduced diterpene levels in boiled coffee, suggesting that filter type matters more than brewing method.
For coffee drinkers, this doesn’t mean abandoning workplace coffee entirely. Liquid-model machines generally produced coffee with low diterpene levels similar to paper-filtered coffee. The researchers emphasized that “thoroughly filtered coffee seems like the preferable choice for cardiovascular health, accordingly, filtered coffee should be preferred, also in workplace settings.” Future research directions include investigating how different machine factors influence diterpene levels, conducting interventional studies to directly measure cholesterol effects, and performing observational studies comparing cardiovascular outcomes between employees at workplaces with different coffee preparation methods.
For now, health-conscious coffee lovers might want to consider their workplace brewing method—especially if they plan to enjoy a long career fueled by multiple cups daily.
Funding and Disclosures
The researchers explicitly stated that this study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. They acknowledged contributions from Matti Marklund and Ulf Holmbäck, as well as Katarina Eriksson for creating the illustrations used in the paper.
Publication Information
This research paper, “Cafestol and kahweol concentrations in workplace machine coffee compared with conventional brewing methods,” was authored by Erik Orrje, Rikard Fristedt, Fredrik Rosqvist, Rikard Landberg, and David Iggman. It was received on November 6, 2024, accepted in revised form on February 19, 2025, and published online on February 20, 2025, in the journal Nutrition, Metabolism and Cardiovascular Diseases. The paper is available at https://doi.org/10.1016/j.numecd.2025.103933 and is published as an open access article under the CC BY license.
