Fanning the coals
Clearing the stem.
18+Reference resource. Nicotine is addictive. Tobacco use harms your health.
Clearing the stem.
A non-smoker in a hookah lounge breathes air with carbon monoxide and PM2.5 particles that exceed WHO and EPA limits several times over — water does not filter the smoke. Below are specific measurements from venues in several countries: where the instruments stood, how long the measurement lasted, and what the result was compared against.
In Virginia, USA, from March 4 to May 27, 2011, researchers surveyed 28 venues: 17 hookah lounges, 5 restaurants that allowed cigarette smoking, and 6 fully non-smoking restaurants. Average PM2.5 in the smoking sections of hookah lounges was 374 µg/m³, in the non-smoking sections of those same lounges — 123 µg/m³, in the smoking sections of cigarette-permitting restaurants — 119 µg/m³, and in non-smoking restaurants — 9 µg/m³. Even the formally "non-smoking" section of a hookah lounge turned out dirtier than a smoking section where cigarettes were allowed.
In Baltimore, Maryland, USA, in 2011–2012, across 7 hookah bars over 9 peak-hour visits on Friday and Saturday nights, average CO came to 18 ppm (range 2–115) — the EPA's 8-hour standard of 9 ppm was exceeded on 4 of 6 visits. Average PM2.5 was 712 µg/m³. The venues were poorly ventilated: indoor PM2.5 was 1.3–24.1 times higher than outdoors.
In Hong Kong, from February to April 2018, researchers compared 8 hookah bars with 6 ordinary bars without hookah (measurements lasted about two hours during peak evening hours). Median PM2.5 in hookah bars was 332.1 µg/m³ on weekdays and 1001 µg/m³ on weekends, versus 36.7 and 27.9 µg/m³ in ordinary bars; CO was 22.5 and 27.9 ppm versus 0.8 and 0.71 ppm. The WHO/EPA 24-hour PM2.5 standard (12–15 µg/m³) was exceeded 22 to 67 times over.
In Iran, the numbers diverge even within a single city: in Ardabil, a sample of 60 venues found average PM2.5 of 303.3 ± 201.9 µg/m³, while a different sample of 50 venues in the same city found 624 ± 327 µg/m³. These are different research groups with different measurement-point selection methods; which figure is closer to reality for any given room can't be determined from the published abstracts. The gap isn't an order of magnitude, but it's several-fold — and both figures exceed the standards many times over.
The key question is what's actually burning. A chemical analysis of three "herbal" nicotine-free hookah mixtures (Hammal et al., Tobacco Control, 2013) found a spread in heavy metals: one product contained elevated levels of lead, chromium, nickel, and arsenic. When these mixtures were burned in a smoking machine, the toxic combustion products were equal to or higher than those from tobacco mixtures, and across 6 surveyed venues, average PM2.5 and CO were significantly higher than in casinos, where cigarette smoking is permitted. The authors' conclusion: carbon monoxide and polycyclic aromatic hydrocarbons are generated primarily by the burning coal, not the smoking mixture itself — giving up tobacco alone does not remove the source.
The CDC (USA), in its official fact sheet, states this directly: secondhand hookah smoke is dangerous for non-smokers in part because it contains smoke not just from the tobacco but also from the heat source — the coal. In a controlled chamber experiment at the American University of Beirut, a single hookah session (Nakhla tobacco, Three Kings charcoal) released 2269 mg of CO into sidestream smoke, versus 65.5 mg from a Marlboro Red cigarette — a ratio of roughly 35:1; for total PAHs, roughly 4:1. The authors estimated that an hour-long hookah session creates a toxic load equivalent to the presence of 2–10 people smoking cigarettes.
Separately, there's a press release from the maker of the OOKA electric, coal-free device (April 2025): a laboratory hired by the manufacturer itself claimed zero CO, benzene, toluene, and benzo[a]pyrene content, and a 96% reduction in aldehydes compared with a traditional hookah. The technology here has no coal at all — but the source has a stake in the outcome, and no independent replication of these figures has been found. This claim should be treated with the same caution as any study commissioned by the maker of the product being tested.
In Istanbul, Moscow, and Cairo, in 2013–2014, researchers surveyed 276 employees from 46 hookah venues, 73 of whom were non-smokers. For these non-smoking staff, median levels were: exhaled CO — 1.67 ppm, urine cotinine — 1.1 µg/g creatinine, salivary cotinine — 5.5 ng/mL, hair nicotine — 0.95 ng/mg, urinary NNAL (a metabolite of the carcinogen NNK) — 1.48 pg/mg creatinine. Every additional 8 hours of weekly work was linked to a 1.68-fold rise in urinary cotinine and a 1.22-fold rise in hair nicotine; staff who lit the coals themselves saw salivary cotinine rise 2.83-fold.
Patrons' condition changes too: in Tampa, Florida, USA, personal CO dosimeters on 166 hookah bar patrons recorded a rise from 6.5 ppm to 58.2 ppm after a visit — a 795% increase over baseline.
In New York, from 2015 to 2019, researchers surveyed 57 non-smoking households that used hookah, e-cigarettes, or conventional cigarettes, measuring PM2.5, black carbon, and CO in the room where smoking occurred and in an adjacent room without smoking. Hookah use significantly raised PM2.5 not only in the smoking room but in the adjacent one as well — pollution spread into neighboring rooms; e-cigarette use raised PM2.5 only in the room where vaping happened. Hookah also produced higher CO than cigarettes within the same room.
Researchers in the US surveyed 24 households with a child aged 5 or younger: 5 non-smoking families and 19 hookah-smoking families (8 smoking daily, 11 weekly or less often). Over three visits within a week, they measured air and surfaces in the child's bedroom and took urine samples from the child. In daily hookah-smoking households, nicotine in the air and on surfaces of the child's bedroom was significantly higher than in non-smoking households. Children from daily-smoking households had significantly elevated urinary cotinine, the carcinogen metabolite NNAL, and the acrolein metabolite 3-HPMA; children from occasional-smoking households had elevated cotinine and NNAL. This was the first study to show hookah-specific toxicant uptake biomarkers in children — not symptoms, but traces of the substances in the body.
In North Carolina, USA, in September 2015 and September 2016, researchers surveyed 6 hookah venues, all of which formally had a ventilation system in place at both measurement points. Median CO was 46.2 ppm in 2015 and 64.6 ppm in 2016; median PM2.5 was 379.3 µg/m³ and 484.0 µg/m³. There was no statistically significant improvement between the two measurements (p=0.173 for CO, p=0.116 for PM2.5). The authors' conclusion: the mere presence of ventilation does not bring pollution down to a safe level.
In Ardabil, Iran, in a separate sample of 60 venues, natural ventilation significantly affected indoor air quality, while fans mainly helped where natural ventilation was lacking — the effect depended on the building's layout. How long airing out a room takes to bring hookah smoke concentration specifically down to a safe level in a residential room has not been measured in the available studies.
No. The CDC states this plainly: "Even after passing through water, the smoke produced by a hookah contains high levels of toxic compounds, including carbon monoxide, heavy metals, and cancer-causing chemicals." Water cools the smoke and some of the larger particles settle out, but carbon monoxide, heavy metals, and carcinogens pass through the bowl with practically no loss — the lab measurements above confirm this too.
Precision matters here. No clinical studies — of asthma flare-ups, of emergency room visits — specifically linked to secondhand hookah smoke in children or in people with asthma and COPD have been found. There is only the biomarker study in children described above: it shows that hookah smoke toxicants enter a child's body and turn up in urine, but it does not measure clinical outcomes — flare-ups, hospitalizations, reduced lung function. Filling in this picture with data on secondhand cigarette smoke would mean attributing to hookah something that hasn't been studied for hookah specifically — so what's stated here is only what was actually measured.
Does the water in a hookah actually filter out harmful substances?
No: according to the CDC, carbon monoxide, heavy metals, and carcinogens pass through the water with practically no loss.
Do nicotine-free mixtures make secondhand smoke safer?
No: a study of three "herbal" mixtures found that toxic combustion products from burning them were equal to or higher than from tobacco mixtures, and CO and PAHs are mostly generated by the burning coal.
Does ventilation at a venue help?
Based on the North Carolina measurements — not fully: at all 6 venues with claimed ventilation, no statistically significant drop in CO or PM2.5 was recorded between measurements.
Is there data on the harm of secondhand hookah smoke specifically to children?
There is biomarker data: children in hookah-smoking households show elevated cotinine and carcinogen metabolites in urine. No clinical studies of disease flare-ups from this smoke in children have been found.
How much dirtier is hookah lounge air than air where only cigarettes are smoked?
Per the 2011 Virginia data, even the non-smoking section of a hookah lounge was dirtier than the smoking section of a cigarette-permitting restaurant: 123 µg/m³ PM2.5 versus 119 µg/m³.
Related topics on this site: on carbon monoxide and ventilation, see the article on safety; on what science says about hookah's effects on health and how to minimize risks when smoking, see the corresponding materials; common misconceptions are covered in the top 10 hookah myths; on what actually happens with milk, juice, alcohol, and ice in the bowl, see the separate material; on how a hookah at home differs from a hookah at a venue, see the same piece.