Air quality: effects & evidence—what is actually supported
Air quality affects health through pollutants that vary in composition depending on source and setting. The most robust evidence for specific health outcomes comes mainly from studies on indoor air from biomass smoke and an increased COPD risk. For schools, biomonitoring, and models, there are sometimes hints, but often clean causal studies are missing.
What air quality affects in health—the plausible mechanisms
Indoor and outdoor air can influence tissues through inflammation, oxidative stress, and direct irritant effects. The key point is not “air is bad,” but how much pollutant reaches you and for how long (exposure level and exposure duration). This explains why studies vary strongly depending on whether they assess room air under real-life conditions, outdoor air measurements, or biomarkers.
Indoors, the pollutant mixture depends heavily on the source: when cooking with biomass, you typically generate a combination of fine particulate matter and gaseous combustion products. Biologically plausible is that particles and gases chronically burden the airways: they can trigger inflammatory processes, alter immune responses in the respiratory tract, and thereby lead over time to structural and functional changes. COPD is an example of a disease that typically develops over longer periods, so it responds not only to “days with poor values,” but to repeated or sustained exposure patterns.
Another issue is the measurement strategy. Many studies assess air pollutants through direct measurements (e.g., concentrations in the room) or through biomonitoring (e.g., via biomarkers). Both approaches have strengths but also limitations: direct measurements can distinguish what is in the air from what you actually inhale. Biomonitoring can integrate exposure, but its transferability to specific sources and pollutants is not always clear. That is exactly why comparability between studies is often limited.
Especially in biomonitoring research, the literature also emphasizes that methodological gaps can make interpretation difficult. For example, (Bouchriti et al., 2026, PMID 42133163) describes in a systematic review on air quality monitoring in North Africa the dominance of certain research directions (e.g., lichen and heavy metal studies) and at the same time substantial methodological weaknesses. For the question “what does this mean for health?”, this heterogeneity is relevant.
If you want to understand how strongly studies are causal versus only correlational, it also helps to look at: Bias: effects & evidence—what is supported and what is not.
What the best evidence really provides: COPD and indoor exposure
The most convincing evidence in your evidence set relates to COPD risk in connection with indoor air pollution from biomass cooking fuels. The data are not about “day-to-day values” but about a long-term pathway: COPD develops over time, so exposure patterns are central. Important caveat: these findings are mostly observational data, but they are summarized in a meta-analysis.
The meta-analysis Pathak et al., 2020, PMID 30754998 (“Risk of COPD due to indoor air pollution from biomass cooking fuel: a systematic review and meta-analysis”) is the key element in this selection, because it systematically combines multiple individual studies, improving the strength of the overall conclusion compared with any single study. Meta-analyses are particularly helpful when individual studies are small or heterogeneous, making it harder to see a clear picture.
For quantification: In this article draft, the specific effect sizes (e.g., “relative risk per exposure level”) are not provided as numbers. If you need a number for reader interest (e.g., per a specific increase in exposure), you would need to extract the effect sizes directly from the meta-analysis. In the provided set of studies, no further detail is included. Therefore, I cannot report any claimed percentage or RR figure here without citing the corresponding numbers from the original paper.
What you can infer from the nature of the evidence: COPD is a chronic condition that develops over years. That aligns with the biologically plausible mechanism (long-lasting exposure leads to persistent inflammation and airway damage). This is exactly why “how long” and “how often” matter—not just “what was the single measurement.”
Practical takeaway: If biomass cooking is the relevant exposure source, the most effective strategy in general is to reduce exposure (e.g., switching fuels or improving the venting of smoke). Supplements or “protection through dietary supplements” cannot meaningfully replace this exposure source, because the cause (smoke/particles) continues to act. More on lifestyle levers below.
Evidence hierarchy: meta-analysis, review, modeling study—what it means for causality
If you want to assess causality, an evidence hierarchy helps: meta-analyses from systematic reviews are usually closer to the best overall statement than single observational studies. Modeling studies can make relationships plausible, but they do not directly show what happens in real populations. Cross-sectional studies tend to show patterns of behavior more than health effects.
In a meta-analysis like (Pathak et al., 2020, PMID 30754998), results from multiple studies are combined. Even if the individual studies are often not randomized, pooling can reduce heterogeneity and random fluctuations. Still, without randomized exposure control, causality is never as “hard” as it would be in a true exposure RCT—which often does not exist for natural real-world conditions.
Systematic reviews—even if they do not include RCTs—can provide important insights by structuring the available data. This also applies to indoor settings and schools: (Ferraz et al., 2026, PMID 42058141) brings together studies on indoor air quality in schools focusing on the microbiome and its relationship to particle exposure and chemical pollutants. Such reviews are useful to show that associations can exist, but they do not automatically deliver a clear cause-effect number.
Modeling studies are a different category. (Gu et al., 2027, PMID 42184814) uses modeling to estimate potential effects of land use and vegetation changes on air quality, health, and costs. This can help compare policy options or evaluate scenarios. But: a modeling study does not replace the real interplay of exposure, behavior, adaptation, and individual vulnerability. For causality, direct observational or intervention evidence is missing.
Cross-sectional studies often answer the question “what people do” rather than “what works for health.” (Siddiqui et al., 2026, PMID 42074417) looks at knowledge, awareness, and practices regarding indoor air quality among students; this is methodologically useful, but it is not a direct health endpoint.
If you want to go deeper into “effect vs. study design,” it may also help: Understanding effect size: effects & evidence of 1–2 levers.
Lifestyle levers before supplements: reduce exposure before taking anything
The best “intervention” is usually not a supplement but reducing exposure: less smoke/particles in the indoor air, better exhaust, targeted ventilation, and source management. For biomass cooking, this is particularly relevant because the strongest evidence in your set focuses on exactly that (indoor exposure from biomass). Accordingly, priority should go to the underlying cause.
If you identify a source like biomass smoke, the problem is often so specific that you can make faster progress with practical measures than with nutrition supplements. This also holds when you consider chemical/biological protective mechanisms: as long as smoke reaches the airways daily or regularly, the pollutants keep acting. In the evidence on COPD from biomass cooking (Pathak et al., 2020, PMID 30754998), this exact relationship is the central argument—and it is explained through exposure.
What is often practical and useful?
- Improve smoke exhaust/ventilation: vent smoke directly outdoors as much as possible, rather than distributing it indoors.
- Switch fuel if possible: fossil or cleaner alternatives typically reduce emissions substantially (the magnitude depends on context and technology; you would need matching study values for each case).
- Ignition and cooking routines: spend less time during the most heavily burdened phases, improve positioning, and reduce “smoke buildup.”
- Source management indoors: in daily life it is not only cooking processes; other particle sources may also be relevant. The principle remains: reduce sources first.
In schools and other indoor settings, exposure composition can be more complex (e.g., microbiome + particles + chemicals). (Ferraz et al., 2026, PMID 42058141) systematically discusses indoor air quality in schools with this focus. That supports taking a broad approach (ventilation, particle reduction, hygiene concept) without expecting that a single “biohacking product” replaces the cause.
If you want to track exposure, biomonitoring can help—but methods can distort interpretation. (Bouchriti et al., 2026, PMID 42133163) emphasizes methodological gaps. That means: if you invest in measurement, pay attention to validity and comparability; otherwise you may measure “the wrong thing” or be unable to translate results into health risk estimates reliably.
Evidence landscape at a glance: indoors, biomonitoring, models, and technology
The research landscape is broad: from reviews on schools (with a focus on microbiome and pollutants) to biomonitoring reviews, and to modeling for land use change and technical indoor sensing. What is common across all of them: associations are often studied, but causal evidence with health endpoints and controlled exposure is less frequently established.
For indoor settings in schools, (Ferraz et al., 2026, PMID 42058141) compiles studies on indoor air quality with regard to the microbiome and its relationship to particle and pollutant patterns. This is relevant because schools have dense, recurring exposures, making chronic exposure-pathway issues likely—however, the direct proof that “this intervention reduces this disease risk” is usually harder.
In biomonitoring, (Bouchriti et al., 2026, PMID 42133163) conveys a methodological double message: there are many studies (e.g., using lichens as biomonitors), but there are substantial methodological gaps. For you as a reader, this means: trends in monitoring approaches may provide hints, but translating them into health effects is not automatically robust.
Modeling studies complement the picture when real experiments are missing. (Gu et al., 2027, PMID 42184814) assesses impacts of land-use changes on air quality, health, and costs. Scenario planning is useful to compare “what happens if…” situations. But these studies do not directly establish individual health effects in a controlled population.
There are also contributions that are indirectly relevant—e.g., how forests might behave under future stressors. (Lassiter et al., 2026, PMID 42089018) examines the benefits and potential disadvantages of forests under future stressors. This can play a role in public health decision-making related to air quality, but it is not a substitute for direct health endpoint data.
On the technical side, researchers are also working on unobtrusive detection of indoor activities by combining multisensors. (Protopsaltis et al., 2026, PMID 42122579) addresses “Unobtrusive Human Activity Recognition Using Multivariate Indoor Air Quality Sensing.” In the future, this could help capture exposure timing more accurately—but at first it is a technology and methodology focus, not direct evidence of health effects.
Mandatory table: what evidence provides (and what it does not)
| Topic | Type of study (from your list) | What it most likely supports |
|---|---|---|
| Biomass cooking → COPD risk | Meta-analysis (Pathak et al., 2020, PMID 30754998) | Association between indoor exposure from biomass fuels and COPD risk over longer time periods |
| Indoor air in schools | Systematic review (Ferraz et al., 2026, PMID 42058141) | Relationship patterns among microbiome, particles, and chemical pollutants in school settings (no direct causal proof in this form) |
| Biomonitoring methods & trends | Systematic review (Bouchriti et al., 2026, PMID 42133163) | Which biomonitoring approaches dominated and where methodological gaps exist |
| Land use → air quality/health | Modeling study (Gu et al., 2027, PMID 42184814) | Scenario-based estimation of possible effects on air quality, health, and costs (no direct real-world intervention proof) |
| Indoor activity detection | Technical study/sensing (Protopsaltis et al., 2026, PMID 42122579) | Feasibility of using multisensors to detect activities (no health endpoint as focus) |
What remains unclear—and how to interpret the evidence correctly
Many findings are not designed as randomized exposure studies, so causal claims are often limited or indirect. Modeling and methodological studies are valuable for hypotheses and planning, but they do not replace measuring exposure in everyday life—and certainly not health endpoints in real settings. Cross-sectional data also frequently provide behavior-related information rather than health effects.
A core interpretation problem is: air quality is not just one variable, but a bundle of pollutant types, concentrations, exposure times, and individual factors (e.g., pre-existing respiratory conditions, ventilation behavior, room size, cooking practices). When studies use different measurement approaches (direct measurements vs. biomonitoring), additional uncertainties arise. (Bouchriti et al., 2026, PMID 42133163) explicitly highlights methodological gaps in biomonitoring—meaning: even if measurement data exist, translating them into exposure-related health effects is not automatically stable.
Even when using systematic reviews: they summarize what is available. That means if primary studies are mostly observational designs, overall evidence quality remains limited by individual-study limitations. For COPD and biomass cooking, the overall statement is stronger than for many other air-quality questions—but it is not the same evidence level as a true randomized exposure intervention.
In schools, another layer applies: even if microbiome and pollutant patterns co-occur, it remains unclear what proportion of that is causally responsible for health endpoints. (Ferraz et al., 2026, PMID 42058141) documents these relationships as the state of research, but that does not automatically imply that each intervention will have the same practical effect.
Knowledge and behavior are also not the same as health. (Siddiqui et al., 2026, PMID 42074417) shows practices and awareness—this can help design intervention approaches, but it does not answer whether this would reduce, for example, COPD or acute respiratory symptoms.
In the end, the correct interpretation is: if there is a specific exposure source (e.g., biomass smoke), then reducing exposure is the most important lever. Supplements can at best complement, but they do not replace avoiding the cause. This priority is consistent with the stronger evidence in (Pathak et al., 2020, PMID 30754998).
What you should take away
- The strongest evidence in this list concerns indoor air from biomass cooking and an increased COPD risk (Pathak et al., 2020, PMID 30754998).
- Causality is often limited: many studies are not randomized; modeling and cross-sectional data provide more hints than hard evidence.
- Lifestyle levers first: reduce exposure (reduce sources, improve smoke exhaust/ventilation) is the most practical, evidence-close approach.
- Schools, biomonitoring, and technology provide valuable puzzle pieces, but translating them into specific health outcomes is often not yet cleanly quantified (Ferraz et al., 2026, PMID 42058141; Bouchriti et al., 2026, PMID 42133163; Gu et al., 2027, PMID 42184814; Protopsaltis et al., 2026, PMID 42122579).