Air Pollution and Health: The Boundaries of Evidence on Exposure Risk, Combustion Pollution, “Detoxification,” and Air-Cleaning Products

Air Pollution and Health: How to Interpret Risks, Protective Measures, and Product Claims

Bottom line

Air pollution does increase health risks. The most reliable course is to reduce pollution at its source and reduce actual exposure. No food, supplement, or air-cleaning device should be claimed to prevent disease solely on the basis of cell, animal, or surrogate-endpoint studies. PMID: 29279932

The harm from air pollution is more than simply “dirty air”

Short-term exposure to PM2.5 and ozone is associated with a higher risk of death among older adults. This association was still observed at concentrations below the air-quality standards in effect at the time, so whether public-health standards are sufficiently stringent merits continued review. PMID: 29279932

Research on PM2.5 and lung cancer also cannot be reduced to the claim that “pollution directly creates all mutations.” A more precise account is that PM2.5 can promote the progression of lung cells with pre-existing EGFR driver mutations into lung adenocarcinoma. The same study did not show that, at the same concentration, people in Taiwan are genetically more susceptible to cancer than people in other countries. Claims of cross-national differences by a specific multiple are an overextension. PMID: 37020004

Climate change is a further upstream determinant of health. Human-caused greenhouse-gas emissions drive modern climate change, and systematic evidence supports the disease, injury, mortality, and mental-health burdens caused by high temperatures and extreme events. PMID: 36211706

Exercise is beneficial, but location and exposure conditions alter the net benefit

Walking does not have identical effects in every environment. A specific London study found that the cardiopulmonary benefits of walking were attenuated when older adults walked along a heavily traffic-polluted road. This finding supports avoiding highly polluted routes, but it cannot be generalized to mean that no one should exercise whenever air quality is poor. PMID: 29221643

The Taipei commuting study measured a reduction in heart rate variability, not a slowing of the heartbeat itself. Walking commuters had the highest PM2.5 exposure and also showed a larger effect on heart rate variability. News reports must preserve the correct name of this physiological measure. PMID: 25638696

Ordinary parks or wooded areas at night should not be conflated with carbon-dioxide poisoning caused by dry-ice accidents or confined spaces. The direction of the current literature supports Professor Lin’s rebuttal of claims about poisoning in ordinary outdoor settings, but the evidence did not directly measure every nighttime woodland scenario. PMID: 28378268

As for whether there is one best time of day to exercise that applies to everyone, the literature has been updated. Studies of exercise timing now include physiological endpoints, so it is no longer accurate to say that there is no scientific evidence at all. Even so, no single best time has been established for every person and every health outcome. In practice, the priority remains exercising safely and consistently. PMID: 42220558

Incense and joss-paper burning are avoidable sources of combustion pollution

Burning joss paper emits particulate matter, polycyclic aromatic hydrocarbons, and water-soluble heavy metals. Environmental research has also observed contributions to black carbon and deterioration in air quality. Direct measurements therefore support treating joss-paper burning as an avoidable source of pollution. PMID: 40490634

The extrapolation of evidence must still have limits. Air pollution in general is associated with major health risks, but this does not mean that studies of joss paper have individually proven that it causes lung cancer, heart disease, or stroke. Conclusions about air pollution in general cannot substitute for evidence of a causal effect of joss paper on specific diseases. PMID: 41941397

Burning incense likewise produces particulate matter, volatile organic compounds, polycyclic aromatic hydrocarbons, and various gases. Measurements in homes show that combustion raises indoor PM10 and particle-bound polycyclic aromatic hydrocarbons even with ventilation. Whether an incense product is marketed as natural or lead-free therefore does not establish that exposure to its combustion emissions is harmless. PMID: 12756013

A study of older adults found that indoor incense burning was associated with poorer cognitive performance and lower functional brain connectivity. Because exposure was based on self-reported observational data, however, this association cannot be presented as proof that incense causes dementia or causal brain damage. PMID: 32341386

Another cohort study found that burning incense at home every day for more than 40 years was associated with a higher risk of chronic limb-threatening ischemia among middle-aged and older adults. This is an adjusted long-term association; a single study still does not establish causality. PMID: 39853267

Diet can be studied, but “detoxification” claims must not be stretched without limit

A diet rich in antioxidants may mitigate some harm from air pollution, but the current signal is limited. Findings on antioxidant supplements are uncertain and contradictory, so “may help” cannot be rewritten as proven protection that offsets air pollution. PMID: 29716592

Oral GSH is not necessarily devoid of biological or clinical effects, making the absolute claim that supplementation “must be injected” too strong. On the other hand, a small trial also did not prove that oral GSH can counteract PM2.5. PMID: 33740389

NAC is a precursor of GSH, so the biochemical rationale is valid. Its effects, however, depend on baseline status and the population studied, and results in healthy people are inconsistent. This does not justify a supplementation recommendation for everyone. PMID: 42232577

There is a research basis for the attention given to broccoli sprouts: three-day-old broccoli sprouts may contain approximately 10 to 100 times as much glucoraphanin as mature broccoli. PMID: 9294217

A human challenge study found that after short-term consumption of a specific broccoli-sprout extract, nasal cells showed a reduced response to diesel exhaust particles. This was a result involving a specific intervention and a surrogate biomarker; it cannot be equated directly with prevention of long-term disease. PMID: 24287881

Another study observed increased urinary excretion of metabolites of specific pollutants, but it did not measure long-term clinical outcomes such as lung cancer, cardiopulmonary disease, or death. It is therefore accurate to say that an “excretion marker increased,” but not that the intervention has been proven to protect people from the long-term harms of air pollution. PMID: 24913818

Cooking may affect myrosinase activity and the efficiency of conversion to sulforaphane. Yet the statement that broccoli sprouts “must be eaten raw or they are completely useless” is also overly absolute: some bioavailability remains without added exogenous myrosinase, and gut bacteria may also participate in the conversion. PMID: 41692762

The chlorella study supports only the limited conclusions that breast-milk dioxin concentrations were lower in the pregnancy-intervention group and that transfer through breast milk might be reduced. It did not show that dioxins can be eliminated from the entire body of an adult in the general population, nor does it provide evidence for a detoxification treatment after exposure to tear gas. PMID: 17472477

Evidence cannot leap from cells and animals to endorsement in humans

The paper promoted as proof of the efficacy of Pyunkang-Hwan studied rat and cell models. It can support only a preclinical hypothesis, not independently prove treatment efficacy for human rhinitis, asthma, emphysema, bronchiectasis, or pulmonary fibrosis. This search also found no identifiable human clinical trial supporting those therapeutic claims. PMID: 29933536

The original TW01 study likewise examined only in vitro cellular markers; it did not test effects on human respiration, immunity, bowel movements, or so-called constitutional health. The literature has been updated, and a mouse study is now also available, so the historical statement that “there is only one paper” is no longer valid. Even with this scientific progress, however, the available research is still not a human clinical trial. PMID: 40077701

Air-cleaning devices must be judged by real exposure and health endpoints

Evidence on blue-light titanium-dioxide photocatalysis comes mainly from reactor, materials, and laboratory studies. It cannot directly prove that any commercially available air purifier can inactivate COVID-19 or improve users’ health. PMID: 35935800

The toxicity and carcinogenicity of titanium dioxide must be interpreted in light of dose, particle size, and exposure context. Animal findings or occupational inhalation risks cannot be treated as proof that ordinary contact with a consumer product inevitably causes cancer. Statements about human evidence and classifications should likewise not go beyond the literature obtained. PMID: 35565075

Prototype nanofiber window screens can capture PM2.5 with high efficiency while retaining some light transmission and ventilation. Prototype results, however, cannot endorse the household performance of every commercial brand sold in Taiwan. Uniformity in mass production and performance after actual installation are questions at different levels. PMID: 31792730

For wearable negative-ion air cleaners, no identifiable studies currently establish effects on real PM2.5 exposure or health endpoints, and there are insufficient data to show that these devices release enough ozone to cause respiratory harm. Indoor negative-ion devices may remove some particles, but reliable evidence of health benefit is lacking. “Can process particles” and “can protect health” must therefore be evaluated separately. PMID: 30274196