Higher PM2.5 levels were associated with a nearly 90% increased risk of combined impaired fasting glucose and impaired glucose tolerance (IFG+IGT).
Lower greenness (NDVI) was linked to higher risk of incident prediabetes and reduced remission of isolated impaired fasting glucose.
Environmental mixtures (higher air pollution, temperature variation, impervious surfaces, and light at night, lower NDVI) increased the risk of all prediabetes phenotypes.
No statistically significant associations were found between these exposures and progression from prediabetes to type 2 diabetes.
Study at a Glance
Design: Prospective cohort study (KORA, 2006–2022)
Population: 1,618 adults without diabetes at baseline; 367/420 with prediabetes assessed for remission/progression
Exposures: PM2.5, PM2.5–10, PM10; light at night; NDVI (greenness); imperviousness; air temperature variation
Primary Outcomes: Incidence of isolated IFG, isolated IGT, combined IFG+IGT; remission to normal glucose tolerance; progression to type 2 diabetes
Statistical Approach: Complementary log-log regression (per IQR increase); quantile g-computation for mixtures
Key Findings: PM2.5 and lower NDVI associated with increased risk of IFG+IGT; lower NDVI and higher LAN associated with reduced remission of iIFG
Why This Study Matters
Prediabetes—a state of elevated blood glucose not yet meeting diabetes criteria—affects hundreds of millions of people worldwide and often progresses to type 2 diabetes. Although environmental factors such as air pollution and the built environment have been linked to the risk of diabetes, their role in the development, remission, and progression of distinct prediabetes phenotypes has remained poorly understood. This study from the KORA cohort in southern Germany provides some of the first evidence that long-term exposure to particulate matter, lack of green space, and other environmental features may influence not only the onset of prediabetes but also the likelihood of reverting to normal glucose tolerance.
How the Study Was Conducted
Researchers analyzed data from the KORA (Cooperative Health Research in the Region of Augsburg) cohort, following participants between 2006 and 2022. Glucose tolerance status was defined according to the 1999/2006 WHO criteria, classifying participants into normal glucose tolerance (NGT), isolated impaired fasting glucose (iIFG), isolated impaired glucose tolerance (iIGT), combined IFG+IGT, or type 2 diabetes.
Environmental exposures were assessed annually at the residential level and included fine particulate matter (PM2.5, PM2.5–10, PM10), light at night (LAN), the Normalized Difference Vegetation Index (NDVI) as a measure of greenness, percentage of impervious surface (IMP, reflecting built-up area), and air temperature variation (Tsd, the SD of daily temperature).
The primary analysis used complementary log-log regression models to estimate hazard ratios (HRs) per interquartile range (IQR) increase in each exposure. A separate quantile g-computation model examined the joint effect of the environmental mixture. Analyses were adjusted for age, sex, body mass index, smoking, education, and other potential confounders.
What the Researchers Found
Among 1,618 participants without diabetes at baseline, 370 developed prediabetes during follow-up: 124 with iIFG, 199 with iIGT, and 47 with combined IFG+IGT. Of those with prediabetes (excluding participants on glucose-lowering medication), 136 regressed to NGT and 133 progressed to type 2 diabetes.
Exposure to higher levels of PM2.5 was associated with increased risk of IFG+IGT (HR 1.89, 95% CI 1.07–3.31 per IQR increase). Lower greenness (lower NDVI) was also linked to a higher risk of IFG+IGT (HR 1.59, 95% CI 1.03–2.46). Associations with iIGT were consistent but weaker; no significant associations were observed for iIFG alone.
When considering the environmental mixture (higher air pollution, Tsd, IMP, LAN, and lower NDVI), the combined exposure was associated with increased risks of all three prediabetes phenotypes—iIFG, iIGT, and IFG+IGT—suggesting that co-exposure may be more harmful than single agents.
Importantly, the environment also appeared to influence remission of prediabetes. Lower NDVI (HR 0.52, 95% CI 0.28–0.96), higher LAN (HR 0.22, 95% CI 0.08–0.62), and higher IMP were associated with lower likelihood of remission from iIFG to NGT. No statistically significant associations were found between these exposures and the progression from prediabetes to type 2 diabetes.
What the Findings May Mean
The results suggest that environmental exposures may affect glucose metabolism in a phenotype-specific manner. The stronger links with phenotypes involving impaired glucose tolerance (iIGT and IFG+IGT) raise the possibility that air pollution and lack of greenness particularly impair postprandial glucose regulation, possibly through inflammatory or oxidative stress pathways. The finding that lower greenness and higher light at night were associated with reduced remission of isolated IFG indicates that environmental factors might also hinder the body's ability to reverse early glucose abnormalities.
Because this is an observational study, causal conclusions cannot be drawn. However, the coherence of the findings across multiple exposures and outcomes adds weight to the hypothesis that modifying the environment—for example, by reducing air pollution or increasing access to green spaces—could help prevent prediabetes and promote remission. The lack of association with progression to type 2 diabetes may reflect limited statistical power, competing risks, or a true null effect, and requires further investigation.
Strengths and Limitations
Strengths include the prospective design with up to 16 years of follow-up, systematic characterization of glucose tolerance by OGTT, a wide range of environmental exposures, and the use of both single-exposure and mixture analyses. However, several limitations must be considered. The study was conducted in a single region (Augsburg, Germany), which may limit generalizability. Residual confounding by unmeasured factors such as diet, physical activity, noise, or social deprivation is possible. Environmental exposure estimates were based on residential address and may not capture personal exposure or indoor environments. The sample size for some phenotype subgroups was modest, leading to wide confidence intervals and the possibility of chance findings. Finally, as an observational study, the results do not demonstrate causation.
Implications for Practice and Research
For clinicians, these findings underscore the potential importance of the physical environment in the development and course of prediabetes. While individual management remains focused on diet, exercise, and pharmacotherapy, the study adds to a growing body of evidence that population-level interventions—such as clean air policies, urban greening, and reduced light pollution—could complement clinical approaches to diabetes prevention.
Future research should aim to replicate these findings in diverse populations, incorporate more detailed personal exposure assessment, and explore the biological mechanisms linking the environment to specific prediabetes phenotypes. Randomized or quasi-experimental studies of environmental interventions would be valuable to test causality.
Funding, Disclosures, and Registration
The authors' funding sources and potential conflicts of interest are described in the original publication. The KORA study is supported by the Helmholtz Zentrum München and the German Federal Ministry of Education and Research.
References
Xi Y, Thorand B, Schneider A, et al. Association of long-term exposure to air pollution, built environment and air temperature with the incidence of prediabetes phenotypes, and prediabetes remission and progression to type 2 diabetes: a longitudinal study in the KORA cohort. Diabetologia. 2026. PMID: 42502136. https://pubmed.ncbi.nlm.nih.gov/42502136/
