The clinic deals with what a patient brings into the room. Most of what is acting on a body, however, is happening outside it — and most of that is happening in the patient's own home. Air, water, and the surfaces they live among are not background. They are dose.
Air: the single largest modifiable exposure
Fine particulate matter — particles small enough to enter the bloodstream from the lungs — has the cleanest mortality signal in environmental epidemiology. The Six Cities and ACS cohorts established the relationship; the Medicare analysis of 60.9 million people refined it. Per 10 µg/m³ of PM2.5, all-cause mortality rises by roughly 6–8%, and the curve has no apparent safe threshold. Lower is better, almost without qualification.
Australian urban background — typically 6–8 µg/m³ in Sydney and Melbourne — already exceeds the WHO 2021 annual guideline of 5 µg/m³. Bushfire seasons make it worse. The Black Summer of 2019–20 was associated with around 417 excess deaths and several thousand cardiorespiratory hospitalisations. The case for indoor filtration during smoke episodes is not aesthetic; it is mortality data.
Indoor is often worse than outdoor
Gas cooktops without externally vented range hoods can drive nitrogen dioxide peaks above WHO one-hour guidelines within minutes. US sampling has found detectable benzene above health benchmarks in some kitchens. Attached garages, scented candles, incense and inadequately ventilated bathrooms each add their own loads.
Two interventions sit at the top of the list. An externally vented rangehood used every time the cooktop is on — or a switch to induction — is the structural fix. A portable HEPA unit in the bedroom is the second; the meta-analysed trial data show roughly a 56% reduction in indoor PM2.5 and a mean drop of around 4 mmHg in systolic blood pressure. The bedroom is the most consequential room in the house for this conversation, since a person spends roughly a third of life there, breathing slowly and deeply — a point that compounds with the case made in the note on sleep.
A simple CO₂ monitor — used as a proxy for ventilation rather than as an exposure measure in itself — is among the more useful instruments a household can own. The COGfx work suggests cognitive performance falls measurably as CO₂ climbs through the 1,000–1,400 ppm range that closed bedrooms and small meeting rooms routinely reach.
Water
Municipal water in Australia is generally safe. That is not the same as optimal. Chlorination by-products are detectable in most chlorinated supplies and carry a consistent, if modest, bladder-cancer signal in pooled cohorts. The NHMRC has been refreshing the Australian Drinking Water Guidelines across 2024 and 2025, with revised PFAS, lead and plumbing-products values. PFAS contamination is documented in specific catchments — Williamtown, Oakey, Katherine — driven by historical use of defence firefighting foams.
An activated-carbon block filter handles most routine concerns at modest cost. Reverse osmosis is the rational add-on where long-chain PFAS or lead is a known issue. Bottled water is, on the evidence, a worse answer than filtration in most cases — both for environmental reasons and for the microplastic burden it carries (one widely cited 2018 study detected plastic particles in 93% of bottles sampled across eleven brands). The food side of the same dose argument is in the note on nutrition and food systems.
Plastics, lead and the things below the noise floor
Microplastics are now consistently detected in blood, placenta, lung, testis and brain tissue. The 2024 NEJM study of carotid-artery plaque — the cholesterol deposits that narrow neck arteries — reported a substantially higher rate of cardiovascular events over three years in patients with detectable micro- and nanoplastics in the plaque. The finding is suggestive rather than confirmatory, with real questions about contamination during surgery. The reasonable posture is to reduce exposure where it is cheap and uncontroversial: glass or stainless for food storage, hot liquids decanted out of plastic, bedding and mattresses chosen with material in mind.
Lead is a less fashionable conversation but a more settled one. The 2018 NHANES analysis estimated that background blood-lead exposure could account for as much as 18% of all-cause mortality in US adults, with no clean threshold below which the cardiovascular signal disappeared. Australian adults sit at a low median, but legacy paint and plumbing in older housing are worth flagging in renovation conversations.
What does not work
"Detox" regimens — juice protocols, sauna stacks, "binders" — have no convincing evidence for clearing legacy pollutants or PFAS outside established clinical chelation in acute poisoning. Alkaline water, negative-ion generators (some of which emit ozone), salt lamps, EMF-protection devices and hair mineral analysis for exposure assessment all sit in the same category: confident marketing, no credible outcome data.
The cumulative case
None of these interventions, taken alone, is dramatic. The case is cumulative. Twenty-four hours a day, seven days a week, fifty years of cleaner air and water and fewer combustion exposures is a measurably different exposure profile to the alternative. The work is to identify the highest-leverage environmental changes for a particular household and to make them once, well.
§ For professionals — mechanisms & evidence+
Ambient PM2.5
The mortality signal is among the cleanest in environmental epidemiology. Dockery's Six Cities cohort gave an adjusted mortality-rate ratio of 1.26 (1.08–1.47) between the most and least polluted cities 1. Pope's ACS CPS-II analysis estimated all-cause RR 1.06 per 10 µg/m³ PM2.5, with cardiopulmonary 1.09 and lung cancer 1.14 2. Di's Medicare cohort of 60.9 million beneficiaries refined this to HR 1.073 (1.071–1.075) per 10 µg/m³, with the association persisting below the US NAAQS of 12 µg/m³ 3. GBD 2021 attributes approximately 4.7 million annual deaths to ambient PM2.5 and 3.1 million to household air pollution. WHO halved its annual PM2.5 guideline from 10 to 5 µg/m³ in 2021; the NEPM Australian annual standard moves from 8 to 7 µg/m³ in 2025. The Black Summer analysis (Borchers Arriagada 2020) attributed 417 excess deaths to bushfire PM2.5 4.
NO₂, ozone, traffic-related pollution and dementia
Atkinson's 2018 pooled analysis put all-cause HR per 10 µg/m³ NO₂ at 1.02 (1.01–1.04), CV 1.03, respiratory 1.03. Turner 2016 found per-10-ppb O₃ respiratory-mortality HR 1.12 in two-pollutant models. Chen 2017 (Ontario, 2.2 million) found dementia incidence rose with proximity to major roads (HR 1.07 within 50 m), and Wilker 2023 (BMJ) reported a 17% increase in dementia risk per 2 µg/m³ annual PM2.5 in actively ascertained cohorts 5.
Indoor air
Lebel 2022 measured NO₂ peaks above the WHO one-hour guideline within minutes of gas-stove use; Kashtan/Lebel 2023 found detectable benzene above health benchmarks in some kitchens 6. Gruenwald 2023 attributed 12.7% of US current childhood asthma to gas-stove use applying Lin's 2013 meta-analysis OR 1.34 — a figure with wide CIs and acknowledged confounding limits. Darby's 2005 European pooled analysis showed an 8.4% lung-cancer risk increase per 100 Bq/m³ residential radon, largely linear with no threshold 7. Allen's COGfx study (Allen 2016 EHP) found cognitive scores 15% lower at ~945 ppm CO₂ and 50% lower at ~1,400 ppm versus ~550 ppm 8. Guan's 2020 meta-analysis of ten RCTs found portable HEPA filtration reduced indoor PM2.5 by 55.9% with mean systolic BP −3.94 mmHg (−7.00 to −0.89) 9.
Water
Villanueva 2004 — pooled bladder-cancer OR 1.44 (1.20–1.73) in men exposed to >50 µg/L THMs; Xie 2025 confirmed long-term THM exposure OR 1.59 (1.40–1.81) 10. NHMRC's 2025 ADWG update revised PFAS values to PFOA 200 ng/L, PFOS 8 ng/L, PFHxS 30 ng/L, PFBS 1000 ng/L. WHO/IARC reclassified PFOA as Group 1 and PFOS as Group 2B in 2023; Steenland's C8 review summarises evidence for kidney and testicular cancer, hypercholesterolaemia, ulcerative colitis and attenuated vaccine response 11. Activated carbon handles THMs and chlorine reasonably; reverse osmosis is the relevant technology for long-chain PFAS and lead.
Microplastics and EDCs
Marfella's 2024 NEJM carotid-plaque study reported composite MI/stroke/death HR 4.53 (2.00–10.27) over 34 months in patients with detectable plaque MNPs, with subsequent correspondence noting intra-operative contamination concerns 12. Yang's NHANES analysis found urinary BPA detectable in 95.7% of samples (2013–14). Trasande 2024 estimates >90,000 attributable US deaths annually from phthalate exposure alone. Levine's 2022 update reported a global sperm-count decline of roughly 51% from 1973 to 2018, with continuing methodological debate 13.
Lead, heavy metals, climate
Lanphear 2018 (NHANES-III) attributed an 18% population-attributable fraction of all-cause mortality to background lead exposure, with CV HR 1.70 and IHD HR 2.08 across blood-lead 1.0 → 6.7 µg/dL — no apparent threshold 14. The Lancet Countdown 2024 reported heat-related deaths in adults over 65 were 167% higher in 2023 than in 1990–99.
What does not work
Detoxification regimens, "alkaline" water, negative-ion generators (some of which emit ozone), salt lamps, EMF-protection devices and hair mineral testing for exposure assessment lack credible evidence. Clinical chelation has a defined role in acute heavy-metal poisoning, not general "burden reduction."
Australian regulatory frame
NEPM Ambient Air Quality: annual PM2.5 8 → 7 µg/m³ (2025), 24-hour 25 µg/m³. ADWG (NHMRC) revised across 2024–2025 covering PFAS, lead, manganese and plumbing-products leaching. PFAS NEMP 3.0 governs contaminated-site management. FSANZ Australian Total Diet Study tracks dietary metal, pesticide and PFAS exposure. Victoria has banned new-dwelling gas connections from 2024; ACT from late 2023; NSW under review. enHealth bushfire smoke guidance supports P2/N95 use and HEPA-filtered indoor "clean rooms."
How clean is Australian background air, really?+
Sydney and Melbourne urban background PM2.5 typically sits in the 6–8 µg/m³ range — comfortably above the WHO 2021 annual guideline of 5 µg/m³ and within the NEPM standard of 8 µg/m³ (tightening to 7 µg/m³ in 2025). The Medicare cohort (Di 2017) showed mortality continues to rise with exposure well below these values, with no apparent threshold. Bushfire seasons make the dose substantially worse: the 2019–20 Black Summer was associated with 417 excess deaths and roughly 3,150 cardiorespiratory hospitalisations.
Is indoor air really worse than outdoor air?+
Often, yes. Gas cooktops without externally vented range hoods can drive NO₂ peaks above WHO one-hour guidelines within minutes, and US sampling has found detectable benzene above health benchmarks in some kitchens. CO₂ used as a ventilation proxy is one of the more useful instruments a household can own — cognitive performance scores in controlled office studies fell roughly 15% at ~945 ppm and 50% at ~1,400 ppm relative to ~550 ppm.
Do HEPA air purifiers actually help?+
Yes — the meta-analysed RCT data (Guan 2020) shows portable HEPA units reduce indoor PM2.5 by roughly 56%, with mean systolic blood pressure reductions of around 4 mmHg. The bedroom is the most consequential room: a person spends roughly a third of life there, lungs working at the slowest, deepest tidal volume. During bushfire smoke episodes, enHealth guidance supports HEPA-filtered indoor 'clean rooms' and P2/N95 use outdoors.
Is Australian tap water safe to drink?+
Municipal supplies are generally compliant with the Australian Drinking Water Guidelines, refreshed by NHMRC across 2024–2025. Trihalomethane disinfection by-products are detectable in most chlorinated supplies and carry a consistent bladder-cancer signal in pooled cohorts (Villanueva 2004; Xie 2025). PFAS values were revised in 2025 — PFOA 200 ng/L, PFOS 8 ng/L. Activated-carbon block filtration handles routine THM concerns at modest cost; reverse osmosis is the rational add-on in PFAS-impacted catchments such as Williamtown, Oakey or Katherine.
Is bottled water better than filtered tap water?+
On the available evidence, no. Mason 2018 found microplastic particles in 93% of bottles sampled across eleven brands. A point-of-use carbon block filter, replaced on schedule, is the simpler and better default — and reverse osmosis where PFAS is a known issue.
How concerned should I be about microplastics?+
Detectable burdens in blood, placenta, lung, testis and brain tissue are now consistently reported. Marfella's 2024 NEJM paper found that micro- and nanoplastics in carotid plaque were associated with a fourfold higher composite cardiovascular event rate over 34 months — hypothesis-generating rather than confirmatory, with real concerns about intra-operative contamination and unmeasured confounding. The precautionary posture — glass and stainless for food and hot liquids — is reasonable rather than evidence-based for hard endpoints.
- 1. Dockery DW, et al. An association between air pollution and mortality in six US cities. NEJM. 1993.
- 2. Pope CA, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution (ACS CPS-II). JAMA. 2002.
- 3. Di Q, et al. Air Pollution and Mortality in the Medicare Population. NEJM. 2017.
- 4. Borchers Arriagada N, et al. Unprecedented smoke-related health burden associated with the 2019–20 bushfires in eastern Australia. MJA. 2020.
- 5. Wilker EH, et al. Ambient air pollution and clinical dementia: systematic review and meta-analysis. BMJ. 2023.
- 6. Lebel ED, et al. Methane and NOx Emissions from Natural Gas Stoves, Cooktops, and Ovens. Environ Sci Technol. 2022.
- 7. Darby S, et al. Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies. BMJ. 2005.
- 8. Allen JG, et al. Associations of Cognitive Function Scores with CO2, Ventilation, and VOC Exposures (COGfx). Environ Health Perspect. 2016.
- 9. Guan T, et al. Effects of indoor air filtration intervention on cardiovascular health: meta-analysis. Hypertension. 2020.
- 10. Villanueva CM, et al. Disinfection byproducts and bladder cancer: a pooled analysis. Epidemiology. 2004.
- 11. Steenland K, Fletcher T, Stein CR, et al. Review: evolution of evidence on PFOA and health following the assessments of the C8 Science Panel. Environ Int. 2020.
- 12. Marfella R, et al. Microplastics and Nanoplastics in Atheromas and Cardiovascular Events. NEJM. 2024.
- 13. Levine H, et al. Temporal trends in sperm count: a systematic review and meta-regression analysis of samples collected globally in the 20th and 21st centuries. Hum Reprod Update. 2023.
- 14. Lanphear BP, et al. Low-level lead exposure and mortality in US adults: a population-based cohort study. Lancet Public Health. 2018.
- 15. National Health and Medical Research Council. Australian Drinking Water Guidelines (updates 2024–2025).
- 16. World Health Organization. Global Air Quality Guidelines. 2021.