Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.
Skip to main content
Download PDF
Download ePub

The independent role of fine particulate matter and genetic liability on cognition in older adults

Annals of General Psychiatry volume 24, Article number: 20 (2025) Cite this article

Abstract

Background

Genetic susceptibility to mental health and cognitive traits, as well as air pollution, significantly impact cognition. The interplay between polygenic liability and fine particulate matter (PM2.5) remains unclear due to the limited number of large-scale studies in Asia. This study utilized the Taiwan Biobank, a nationwide community-based database, to investigate the main and modified effect of PM2.5 on individuals’ polygenic susceptibility in cognition.

Methods

Polygenic risk score (PRS) for cognitive performance (CP PRS), Alzheimer’s disease (AD PRS), schizophrenia (SCZ PRS), and major depression (MDD PRS) were computed representing genetic susceptibility for an individual. APOE genotype was classified into E3/E3, E3/E4, and E4/E4. The five-year average concentration of PM2.5 from satellite images was used for defining environmental exposure. Cognitive performance was evaluated via the Mini-Mental State Examination (MMSE) score. The association between personal genetic susceptibility, PM2.5, and cognitive performance was examined using multilevel linear regression with the adjustment of age, sex, batch effect, and population stratification effect. The gene-environment synergism was examined with the inclusion of product term of PM2.5 and PRS in the multivariate model.

Results

Our analyses included 25,593 participants from 164 townships. Participants exposed to higher PM2.5 concentrations had a lower MMSE score (Beta=-0.0830 corresponding to a 1 µg/m3 increase in PM2.5 concentration, 95% CI, -0.0973 to -0.0688, p-value < 0.0001). After controlling for PM2.5 concentration, CP PRS (Beta = 0.1729, 95% CI, 0.1470 to 0.1988, p-value < 0.0001), SCZ PRS (Beta=-0.0632, 95% CI, -0.0891 to -0.0374, p-value < 0.0001), and AD PRS (Beta=-0.0321, 95% CI, -0.0580 to -0.0062, p-value = 0.0153) were associated with MMSE score. After further examination of gene-environment synergism, no interaction effect was identified, indicating different mechanism of PM2.5 and genetic liability to influence cognitive performance.

Conclusions

Human polygenic loading and PM2.5 may impact cognition via an independent pathway. A prevention strategy targeting air pollution reduction may effectively improve the cognitive performance. Multiple exposures and their influences on the long-term change of cognition were required in future research.

Background

Population aging is the most pervasive demographic trend worldwide. Aging impairs human cognitive ability, resulting in socioeconomic and health burdens [1, 2]. A decline in cognitive performance often presents before the onset of several psychotic disorders [3, 4]. Therefore, understanding the determinant factors associated with cognition may help generate public health prevention strategies and mitigate the detrimental effects of cognitive ability in older populations.

An increasing number of studies targeting on the influence of air pollution on human mental health, including depression, anxiety and cognitive impairment [5,6,7]. Exposure to air pollutants was postulated to harm the central nervous system and affect psychological functioning by direct toxicity or inducing neuro-inflammation [5, 8]. Fine particulate matter (PM2.5) has been found to have a negative effect on general cognition among adults > 40 years [1]. A meta-analysis presented a significant decrease in general cognition for the increase of PM2.5 concentration [1].

General cognitive ability is attributed to genetic factors, with a high heritability of 50–70% [9,10,11]. Polygenic risk score (PRS), combining the effects of trait-susceptible genetic variants identified in a genome-wide association study (GWAS), was utilized recently to comprehensively evaluate the genetic liability of a phenotype. A higher cognitive polygenic capacity was significantly correlated with better cognition in a population of European ancestry [11, 12]. Genetic correlations between general cognition and psychotic syndromes was also reported [13, 14]. A high PRS for schizophrenia or major depression (MDD) was associated with poor general cognitive ability and spatial visualization [3, 15,16,17,18]. The Apolipoprotein E (APOE) locus was known associated with Alzheimer’s disease (AD) and cognitive decline, with a higher risk for epsilon 4 (E4) allele [11, 12, 19]. In addition, the PRS for AD combining common variants except for APOE was associated with reduced hippocampal volume and cognitive impairment in late life [20, 21].

The interplay of individual genetic effects and environmental factors in a complex phenotype was important [16]. However, the literature is still limited. Two small studies conducted in China have examined interactions between MDD PRS and air pollution for cognitive impairment through the alteration of brain function via DNA methylation or the influence of stress-related corticol network function [8, 22, 23]. A child study revealed the antagonistic effect of APOE status, AD PRS and air pollution through the change of brain structural morphology [19]. In addition, large-scale genetic studies were mainly conducted in Western countries, resulting in a knowledge gap regarding the role of polygenic susceptibility, air pollutants, and their potential interaction on cognition in Asian populations.

This study aimed to utilize the Taiwan Biobank (TWB), a nationwide community-based database, to specifically identify the genetic liability for cognition in the Taiwan population and evaluate the main and modified effect of exposure to PM2.5 on the population. We expect the study findings provide new insights into intervention strategies to maintain human cognition and alleviate subsequent disease burdens.

Materials and methods

Study population

The study population included participants in the Taiwan Biobank [24, 25]. The Taiwan Biobank was designed to be a nationwide community-based study, recruiting cancer-free individuals aged 30–70 years. It is the largest government-supported biobank, including more than 120,000 participants. Detailed information on the participants’ socio-demographic characteristics, lifestyle and behavior, personal health status, family history of disease, physical and biological status, and genetic features were included. This study was approved by the Central Regional Research Ethics Committee of China Medical University, Taichung, Taiwan (CRREC-108-30).

Fine particulate matter (PM2.5)

PM2.5 was assessed using high-resolution, specified air pollution datasets maintained by the Atmospheric Composition Analysis Group of Washington University in St. Louis. These were global estimates of ground-level ambient PM2.5, gridded at approximately 1 km × 1 km [26]. We used zonal statistic function with the township boundary of Taiwan by ArcGIS (ArcMap, version 10.3; ESRI Inc., Redlands, CA, USA) to compute the annual concentration of PM2.5 in each township. Participants’ exposure to PM2.5 was specified by their residential address and was defined by the five-year average before the year of baseline determination. We further categorized the exposure to PM2.5 in quartiles (< 18.7, 18.7‒22.2, 22.2‒29.1, 29.1+) to elucidate the dose-response association between the level of environmental exposure and cognitive performance.

Genotyping and PRS

Genome-wide genotyping was performed for 131,048 TWB samples using the customized Taiwan Biobank chip and through the Axiom Genome-Wide Array Plate System (Affymetrix, Santa Clara, CA, USA). Details of quality control and SNP imputation have been previously described [27]. Genetic variants with low call rate (< 5%), low minor allele frequency (< 0.001), or significant deviation from the Hardy-Weinberg equilibrium (p < 1E-6) were excluded to reduce genotyping error. Genotype imputation was performed based on the reference panel of 504 East Asian (EAS) (Genomes Project Consortium, 2015) and 973 TWB panel, resulting in a total of 12,601,684 variants with imputation info > 0.7 available. Individuals (i) with genotyping missing rate > 2% in all chromosomes, (ii) EAS descent, and (iii) that are outliers in the heterozygosity/homozygosity test were filtered out. For two participants with the identity by descent sharing coefficients PI-HAT > 0.1875, only one of them was kept for further analysis to prevent cryptic relatedness. To account for population stratification bias, we conducted a principal component analysis and used the top 20 principal components (PCs) as covariates for subsequent analyses [4]. Finally, 106,806 unrelated participants were included in the PRS calculation.

PRS for cognitive performance (CP), AD, schizophrenia (SCZ), and MDD were computed. CP PRS was derived from COGENT consortium (N = 35,298) [13] and UK Biobank (N = 222,543) [28] of European descent using PRS-CS [29]. AD PRS was derived from a discovery sample of 39,918 cases and 358,140 controls of European descent (excluding 23andMe and proxy cases/controls from UK Biobank) [30] using PRS-CS. Genetic variations for APOE were excluded for the calculation of AD PRS to examine the non-APOE genetic predisposition to AD [19]. The APOE genotypes were classified as E3/E3, E3/E4, and E4/E4 according to rs7412 and rs429358. The Psychiatric Genomics Consortium meta-analysis was used as a discovery sample to identify the risk variants for SCZ and MDD. To optimize PRS prediction, we derived SCZ PRS using PRS-CSx [31] to integrate GWAS in 22,778 cases and 35,362 controls of EAS descent [32] and GWAS in 53,386 cases and 77,258 controls of European descent [33]. We derived MDD PRS using PRS-CSx to integrate GWAS in 246,363 cases and 561,190 controls of European descent [34] and GWAS in 13,042 cases and 88,467 controls of EAS descent [35]. All the PRSs were standardized as a z-distribution. The association between each PRS and their corresponding phenotype among TWB population was presented in Supplementary Table 1, showing satisfactory validity of the derived PRS.

Socioeconomic covariates

Participants’ education level and the community-level socioeconomic covariates were evaluated here. Individual educational level was categorized into five groups: illiteracy, self-study, or elementary school; junior high school; senior high school or vocational school; university or college; and master’s degree or above. Individual socioeconomic characteristics were defined by the community-level socioeconomic statistics of their residential district, including unemployment rate, median household income, coefficient of variation (CV) of household income, and single-parent rate, from the 2015 Monthly Bulletin of Social Welfare Statistics and the 2015 Monthly Bulletin of Interior Statistics. The administrative district is classified as an urban, suburban, or rural area to reflect the degree of urbanization [36].

Assessment of cognitive performance

Cognitive performance was evaluated using the Mini-Mental State Examination (MMSE). MMSE is a validated cognitive screening tool for dementia [37]. It quantitatively assesses orientation, memory attention, calculation, and language function, with a 0–30 point score system. A higher MMSE score indicates better cognitive performance. In the Taiwan Biobank study, MMSE was conducted among participants ≥ 60 years.

Statistical analyses

The distribution of continuous variables is presented using means and standard deviations (SDs), and the distribution of categorical variables is presented using numbers and percentages. The association between personal characteristics, PRS, APOE genotype, PM2.5, community-level socioeconomic covariates, and cognitive performance was examined using multilevel linear regression to obtain unbiased estimates for individual- and group-level variables. The residential districts were incorporated in multilevel model as the random effect to account for the heterogeneous correlation across groups. Age, sex, batch effect, and the top 20 PCs were included in the multilevel linear model to control for population stratification bias and potential confounding. PM2.5, PRS or APOE genotype, and variables significantly associated with the MMSE score in the univariate analysis were used for multivariate modeling. The individual and the product term of PM2.5 and PRS or APOE genotype were included in the multivariate model to examine the gene-environment synergism. The simple slope analysis was further applied for the visualization of PM2.5–PRS and PM2.5–APOE genotype interaction. The stratified analyses for the association between PRS or APOE genotype and MMSE score using PM2.5 quartiles were conducted to examine whether the effect of PRS or APOE genotype on cognitive performance changed with the PM2.5 exposure level.

We restricted our analyses to participants whose residential areas were in the administrative districts with sample sizes > 20 to obtain unbiased and precise estimates in slope variances [38, 39]. The significance level was set at 0.05. All statistical analyses were performed using the SAS statistical software package (version 9.4 for Windows; SAS Institute Inc., Cary, NC, USA).

Results

Study participants

Among the TWB participants, 26,674 individuals aged > 60 years took examination of cognition via MMSE. After excluding 1,081 participants residing in 157 townships (each had a sample size of ≤ 20), a total of 25,593 participants from 164 townships were included in the analyses. The mean MMSE score was 27.52 (Table 1). Table 1 presents the distribution of participants’ baseline characteristics. In the older adult population, the mean age was 63.01 years, 61.18% were female, and most (75.42%) had higher educational levels than senior (vocational) high school. The mean PM2.5 concentration was 23.48 µg/m3, with the quartile cut-off points 18.7, 22.2, and 29.1 µg/m3. The mean community-level socioeconomic covariates were 3.78% for the unemployment rate, 641.16 thousand New Taiwan Dollars (NTD) for median household income, 211.61 for CV household income, and 7.35% for the single-parent rate. More than 68% of participants resided in the urban area. Most (81.66%) of TWB participants carried homozygous APOE E3 alleles and only a small portion (0.77%) of them carried homozygous APOE E4 alleles.

Table 1 The distribution of baseline characteristics and the univariate associations with MMSE score
Full size table

PM2.5, genetic liability, and cognitive performance

The univariate analyses of baseline characteristics and MMSE score are presented in Table 1. Older age (Beta=-0.0702, 95% CI, -0.0784 to -0.0620, p-value < 0.0001) and female sex (Beta=-0.2450, 95% CI, -0.3033 to -0.1867, p-value < 0.0001) were significantly associated with decreased MMSE scores. Individuals with educational levels of senior (vocational) high school or higher had better cognitive performance than those who were illiterate, self-studying, or educated in elementary school. For community-level socioeconomic variables, the higher the median income (Beta = 0.0051, 95% CI, 0.0040 to 0.0063, p-value < 0.0001) or CV household income (Beta = 0.0006, 95% CI, 0.0002 to 0.0010, p-value = 0.0040), the better the cognitive performance. The unemployment and single-parent rates are not significantly associated with MMSE score. A significantly lower cognitive performance was presented among individuals living in sub-urban (Beta=-0.7342, 95% CI, -0.9346 to -0.5377, p-value < 0.0001) and rural area( Beta=-0.7450, 95% CI, -1.1316 to -0.3584, p-value = 0.0002).

PM2.5 concentration was significantly associated with poor cognitive performance (Table 1). Participants exposed to higher PM2.5 concentrations had a lower MMSE score (Beta=-0.0830 corresponding to a 1 µg/m3 increase in PM2.5 concentration, 95% CI, -0.0973 to -0.0688, p-value < 0.0001). PM2.5 concentrations between 22.2 and 29.1 µg/m3 (Beta=-0.1764, 95% CI, -0.3334 to -0.0193, p-value = 0.0277) and higher than 29.1 µg/m3 (Beta=-0.7839, 95% CI, -0.9611 to -0.6066, p-value < 0.0001) were associated with lower MMSE scores compared to PM2.5 concentration < 18.7 µg/m3. The dose-response relationship had statistical significance (p-value for trend test < 0.0001).

Genetic liability was crucial in participants’ cognitive performance (Table 1). CP PRS (Beta = 0.2268, 95% CI, 0.1985 to 0.2551, p-value < 0.0001) was positively correlated with the MMSE score, while AD PRS (Beta=-0.0475, 95% CI, -0.0760 to -0.0190, p-value = 0.0011) and SCZ PRS (Beta=-0.0520, 95% CI, -0.0805 to -0.0236, p-value = 0.0003) were negatively correlated with the MMSE score. MDD PRS (Beta = 0.0260, 95% CI, -0.0025 to 0.0544, p-value = 0.0735) showed a non-significant association with cognitive performance. APOE genotype also showed no significant relevance to cognition, with Beta=-0.0182 (95% CI, -0.0990 to 0.0627, p-value = 0.6598) for E3/E4 genotype and Beta=-0.2145 (95% CI, -0.5657 to 0.1366, p-value = 0.2311) for E4/E4 genotype compared to E3/E3 genotype.

We performed multivariable analyses to examine the association between PM2.5, PRS, and MMSE scores (Table 2). After adjusting for population stratification (top 20 PCs), batch effect, and confounding factors, CP PRS and PM2.5 were significantly associated with cognitive performance. A 0.0285 (adjusted Beta=-0.0285, 95% CI, -0.0392 to -0.0178, p-value < 0.0001) decrease in MMSE score corresponded with a 1 µg/m3 increase in PM2.5 concentration, and a 0.1729 (adjusted Beta = 0.1729, 95% CI, 0.1470 to 0.1988, p-value < 0.0001) increase in MMSE score corresponded with a one SD increase in CP PRS. SCZ PRS and PM2.5 exhibited a negative association with cognitive performance, presenting a lower MMSE score for higher PM2.5 concentration (adjusted Beta=-0.0281, 95% CI, -0.0389 to -0.0174, p-value < 0.0001) and polygenic loading (adjusted Beta=-0.0632, 95% CI, -0.0891 to -0.0374, p-value < 0.0001). The negative effect of AD PRS (adjusted Beta=-0.0321, 95% CI, -0.0580 to -0.0062, p-value = 0.0153) was significant for MMSE score as PM2.5 was included in the multivariate model. The effect of MDD PRS remained non-significant when the multiple-adjusted model was applied. The multivariate results of APOE genotype on cognition was presented in Table 3, showing non-significant effect for MMSE (E3/E4 vs. E3/E3: adjusted Beta=-0.0174, 95% CI, -0.0906 to 0.0558, p-value = 0.6421; E4/E4 vs. E3/E3: adjusted Beta=-0.2827, 95% CI, -0.6004 to 0.0350, p-value = 0.0811).

Table 2 Multivariate associations of PM2.5, PRS and MMSE score
Full size table
Table 3 Multivariate associations of PM2.5, APOE genotype and MMSE score
Full size table

Independent effect of genetic liability and PM2.5

The interaction between PM2.5 and PRS or APOE genotype was presented in Figure and Supplementary Tables 2, 3. The p-value for the product term of PM2.5-CP PRS (Fig. 1A), PM2.5-SCZ PRS (Fig. 1B), PM2.5-MDD PRS (Fig. 1C), PM2.5-AD PRS (Fig. 1D), and PM2.5-APOE genotype (Fig. 1E) was 0.7760, 0.3527, 0.7836, 0.3413, and 0.8452, respectively. It indicated the independent effect of PM2.5 and genetic liability associated with cognitive performance. Similar results were obtained when the categorized PM2.5 level was instead included in the multivariate model (Supplementary Tables 2–3) and stratified analyses (Supplementary Tables 4, 5).

Fig. 1
figure 1

Interaction between PM2.5 and PRS or APOE genotype on MMSE score. (A) PM2.5× CP PRS. (B) PM2.5× SCZ PRS. (C) PM2.5× MDD PRS. (D) PM2.5× AD PRS. (E) PM2.5×APOE genotype

Full size image

Discussion

This large-scale, population-based study was conducted to unravel the role of genetic liability, air pollution, and their interplay on cognition. Increased PM2.5 concentration, lower CP PRS, higher SCZ PRS and AD PRS were associated factors to cognitive impairment. No gene-environment interactions were identified in the study, indicating the distinct mechanisms of polygenic loading or PM2.5 in cognition causation.

The large-scale meta-analysis, including 86 studies from Europe, North America, Asia and Australia, indicated a PM2.5-cognition association of -0.02 (95% CI: -0.03, -0.00, p-value < 0.05) in middle-to-older populations [1]. PM2.5 was related to poor cognitive function in Chinese population [2]. In the present study, the hazard of PM2.5 on cognition was replicated in the Taiwanese population, with a 0.0285 (adjusted Beta=-0.0285, 95% CI, -0.0392 to -0.0178, p-value < 0.0001) decrease in MMSE score for the increase of PM2.5 concentration.

This study identified the independent effect of CP and SCZ liability on cognitive performance and it agrees with several Western studies [3, 11, 12]. The bidirectional relationship between schizophrenia and cognition was also identified in a genome-wide association meta-analysis [17]. Several studies have suggested a genetic correlation between cognition and AD through the involvement of APOE or genetic liability to AD [13, 19,20,21, 40, 41]. However, we observed a significant effect of AD PRS but not APOE genotype among Taiwnese population. This may be due to the limited sample size of APOE homozygous E4 genotype in our study. Evidence for the association between MDD PRS and cognition remains controversial. The genome-wide interaction study revealed a significant but distinct effect of MDD liability on different sub-domains of cognition, including reaction time, immediate memory, and global cognition [15], but a null effect was presented in the study.

Our study did not detect an interaction between polygenic effect and PM2.5, implying that a distinct mechanism of genetic and environmental factors may be involved. It was postulated that PM2.5 affects cognition via the degeneration of neurons or neuro-inflammation. Neurons in the olfactory bulb (OB) and hippocampus were considered critical for cognitive ability, including learning, memory, and sensory performance [42]. Calderón-Garcidueñas et al. reported ultrafine particles in human OB periglomerular neurons via the inhalation of air pollutants [43], causing direct contact with exogenous chemicals, which impacts neurogenesis and develops AD [44]. Meanwhile, PM2.5 induces immune response for the activation of macrophages and inflammatory cytokines, leading to the accumulation of reactive oxygen species (ROS) and inflammation. Inflammatory compounds and ROS in circulation may spread to blood-brain barrier [2], leading to the induction of neuro-inflammation and the impairment of cognitive function. In-vitro studies have reported poor microglial cell survival for high-dose PM2.5 due to neuro-inflammation and ROS production [45, 46]. Apart from the modulation of neurogenesis from PM2.5, the potential pathway of genetic architecture was through the physical change of brain structure, such as gray matter atrophy [12]. The cognition-associated genes dominate innate diversity between people and influence individual susceptibility, responding to alterations in brain structure over time, eventually impacting human cognition [12].

Our study had several strengths. First, this is the first community-based study to elucidate the complex roles of polygenic characteristics and PM2.5 on general cognition rather than clinical features, such as dementia. It may be more helpful to generate public health practice early during cognitive dysfunction and postpone the development of psychological disorders. Second, considering the diversity of genetic profiles and environmental factors worldwide, our study was advantageous for demonstrating their importance in Asian population and complementing the knowledge gap between countries. Third, we applied PRS to comprehensively summarize the subtle effect of genetic variants on cognition. The trans-ancestry PRS was computed for SCZ PRS and MDD PRS to integrate different genetic contributions across ethnic populations [31] to enhance the accuracy of effect size estimation.

However, our study had some limitations. First, it was a cross-sectional observation. Each participant’s residence was only investigated at baseline, resulting in possible misclassification and underestimation of PM2.5 exposure. Second, only PM2.5 was considered in the study. A mixed exposure model with the inclusion of exercise and other pollutants such as nitrogen oxide [1], sulfur [47], road traffic noise [48], or second-hand smoking [2] was suggested to be more representative of real-life situations. Third, we included participants aged more than 60 for analyses because MMSE was only measured in a subsample of Taiwan Biobank. The generalizability was limited and further studies targeting on the association of PM2.5 and genetic liability with cognition among younger aged population may be required. Fourth, the MMSE score was not in a normal distribution (Kolmogorov-Smirnov p-value < 0.01) and the normality assumption was violated for regression modeling. However, normality examination has been reported invalid with large sample sizes [49]. Therefore, according to the Central Limit Theorem [50, 51] and the simulation study [52], even without adherence to normality assumption, parameter estimates would be in normal distribution and unbiased estimates would be still obtained in our large-scale study. Fifth, the common variants-based PRS did not capture the total genetic liability. Using cross-ancestry GWAS results to derive PRS may lead to a low prediction [53] because genetic architecture may differ across populations. Most large-scale GWAS have been performed in individuals of European ancestry, with only a few reported in individuals of East Asian ancestry, e.g., SCZ. Although a moderate-scale GWAS for MDD has been performed in East Asian ancestry, the limited sample size restricted the improvement of PRS prediction in East Asian populations [54] and subsequently limited the statistical power for cross-trait PRS association test. Further large-scale genetic research in individuals of diverse ancestries is needed.

Conclusion

In conclusion, independent associations were revealed between human polygenic loading, PM2.5, and cognitive performance. A prevention strategy targeting air pollution reduction may effectively improve cognition. Effects of multiple exposures and their influnces on the long-term change of cognition were await discovery in future research.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

AD:

Alzheimer’s disease

APOE:

Apolipoprotein E

CV:

Coefficient of variation

CP:

Cognition performance

EAS:

East Asian

GWAS:

Genome-wide association study

MDD:

Major depressive disorder

MMSE:

Mini-Mental State Examination

NTD:

New Taiwan Dollars

OB:

Olfactory bulb

PM2.5:

Fine particulate matter

PRS:

Polygenic risk score

PCs:

Principal components

ROS:

Reactive oxygen species

SCZ:

Schizophrenia

SDs:

Standard deviations

TWB:

Taiwan Biobank

References

  1. Thompson R, Smith RB, Karim YB, Shen C, Drummond K, Teng C, et al. Air pollution and human cognition: A systematic review and meta-analysis. Sci Total Environ. 2023;859(Pt 2):160234.

    Article  CAS  PubMed  Google Scholar 

  2. Wang J, Li T, Lv Y, Kraus VB, Zhang Y, Mao C, et al. Fine particulate matter and poor cognitive function among Chinese older adults: evidence from a Community-Based, 12-Year prospective cohort study. Environ Health Perspect. 2020;128(6):67013.

    Article  PubMed  Google Scholar 

  3. Liebers DT, Pirooznia M, Seiffudin F, Musliner KL, Zandi PP, Goes FS. Polygenic risk of schizophrenia and cognition in a Population-Based survey of older adults. Schizophr Bull. 2016;42(4):984–91.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Braithwaite I, Zhang S, Kirkbride JB, Osborn DPJ, Hayes JF. Air pollution (Particulate Matter) exposure and associations with depression, anxiety, bipolar, psychosis and suicide risk: A systematic review and Meta-Analysis. Environ Health Perspect. 2019;127(12):126002.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Liu R, Li D, Ma Y, Tang L, Chen R, Tian Y. Air pollutants, genetic susceptibility and the risk of schizophrenia: large prospective study. Br J Psychiatry. 2024;225:427–35.

    Article  PubMed  Google Scholar 

  7. Khan A, Plana-Ripoll O, Antonsen S, Brandt J, Geels C, Landecker H, et al. Environmental pollution is associated with increased risk of psychiatric disorders in the US and Denmark. PLoS Biol. 2019;17:e3000353.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Zhang Y, Lu Z, Sun Y, Guo L, Zhang X, Liao Y, et al. Interactive effect of air pollution and genetic risk of depression on processing speed by resting-state functional connectivity of occipitoparietal network. BMC Med. 2024;22(1):392.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Bouchard TJ Jr., McGue M. Genetic and environmental influences on human psychological differences. J Neurobiol. 2003;54(1):4–45.

    Article  PubMed  Google Scholar 

  10. Davies G, Tenesa A, Payton A, Yang J, Harris SE, Liewald D, et al. Genome-wide association studies Establish that human intelligence is highly heritable and polygenic. Mol Psychiatry. 2011;16(10):996–1005.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Joo YY, Cha J, Freese J, Hayes MG. Cognitive capacity Genome-Wide polygenic scores identify individuals with slower cognitive decline in aging. Genes (Basel). 2022;13(8).

  12. Tsapanou A, Mourtzi N, Gu Y, Habeck C, Belsky D, Stern Y. Polygenic indices for cognition in healthy aging; the role of brain measures. Neuroimage Rep. 2023;3(1).

  13. Trampush JW, Yang ML, Yu J, Knowles E, Davies G, Liewald DC, et al. GWAS meta-analysis reveals novel loci and genetic correlates for general cognitive function: a report from the COGENT consortium. Mol Psychiatry. 2017;22(3):336–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Toulopoulou T, Zhang X, Cherny S, Dickinson D, Berman KF, Straub RE, et al. Polygenic risk score increases schizophrenia liability through cognition-relevant pathways. Brain. 2019;142(2):471–85.

    Article  PubMed  Google Scholar 

  15. Thalamuthu A, Mills NT, Berger K, Minnerup H, Grotegerd D, Dannlowski U, et al. Genome-wide interaction study with major depression identifies novel variants associated with cognitive function. Mol Psychiatry. 2022;27(2):1111–9.

    Article  CAS  PubMed  Google Scholar 

  16. Calafato MS, Bramon E. The interplay between genetics, cognition and schizophrenia. Brain. 2019;142(2):236–8.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Savage JE, Jansen PR, Stringer S, Watanabe K, Bryois J, de Leeuw CA, et al. Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links to intelligence. Nat Genet. 2018;50(7):912–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Ranlund S, Calafato S, Thygesen JH, Lin K, Cahn W, Crespo-Facorro B, et al. A polygenic risk score analysis of psychosis endophenotypes across brain functional, structural, and cognitive domains. Am J Med Genet B Neuropsychiatr Genet. 2018;177(1):21–34.

    Article  PubMed  Google Scholar 

  19. Essers E, Binter AC, Neumann A, White T, Alemany S, Guxens M. Air pollution exposure during pregnancy and childhood, APOE epsilon4 status and alzheimer polygenic risk score, and brain structural morphology in preadolescents. Environ Res. 2023;216(Pt 2):114595.

    Article  CAS  PubMed  Google Scholar 

  20. Mormino EC, Sperling RA, Holmes AJ, Buckner RL, De Jager PL, Smoller JW, et al. Polygenic risk of alzheimer disease is associated with early- and late-life processes. Neurology. 2016;87:481–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Lupton MK, Strike L, Hansell NK, Wen W, Mather KA, Armstrong NJ, et al. The effect of increased genetic risk for Alzheimer’s disease on hippocampal and amygdala volume. Neurobiol Aging. 2016;40:68–77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Li Z, Yan H, Zhang X, Shah S, Yang G, Chen Q, et al. Air pollution interacts with genetic risk to influence cortical networks implicated in depression. Proc Natl Acad Sci U S A. 2021;118:46.

    Article  Google Scholar 

  23. Rutter M, Silberg J. Gene-environment interplay in relation to emotional and behavioral disturbance. Annu Rev Psychol. 2002;53:463–90.

    Article  PubMed  Google Scholar 

  24. Chen CH, Yang JH, Chiang CWK, Hsiung CN, Wu PE, Chang LC, et al. Population structure of Han Chinese in the modern Taiwanese population based on 10,000 participants in the Taiwan biobank project. Hum Mol Genet. 2016;25(24):5321–31.

    CAS  PubMed  PubMed Central  Google Scholar 

  25. Fan CT, Lin JC, Lee CH. Taiwan Biobank: a project aiming to aid Taiwan’s transition into a biomedical island. 2008.

  26. van Donkelaar A, Hammer MS, Bindle L, Brauer M, Brook JR, Garay MJ, et al. Monthly global estimates of fine particulate matter and their uncertainty. Environ Sci Technol. 2021;55(22):15287–300.

    Article  PubMed  Google Scholar 

  27. Liao SF, Su CY, Su MH, Chen CY, Chen CY, Lin YF, et al. Associations of polygenic risks, depression, and obesity-related traits in Taiwan biobank. J Affect Disord. 2023;320:397–403.

    Article  PubMed  Google Scholar 

  28. Lee JJ, Wedow R, Okbay A, Kong E, Maghzian O, Zacher M, et al. Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat Genet. 2018;50(8):1112–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Ge T, Chen CY, Ni Y, Feng YA, Smoller JW. Polygenic prediction via bayesian regression and continuous shrinkage priors. Nat Commun. 2019;10(1):1776.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Wightman DP, Jansen IE, Savage JE, Shadrin AA, Bahrami S, Holland D, et al. A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nat Genet. 2021;53(9):1276–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Ruan Y, Lin YF, Feng YA, Chen CY, Lam M, Guo Z, et al. Improving polygenic prediction in ancestrally diverse populations. Nat Genet. 2022;54(5):573–80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Lam M, Chen CY, Li Z, Martin AR, Bryois J, Ma X, et al. Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet. 2019;51(12):1670–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Trubetskoy V, Pardinas AF, Qi T, Panagiotaropoulou G, Awasthi S, Bigdeli TB, et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature. 2022;604(7906):502–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Howard DM, Adams MJ, Clarke TK, Hafferty JD, Gibson J, Shirali M, et al. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci. 2019;22(3):343–52.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Giannakopoulou O, Lin K, Meng X, Su MH, Kuo PH, Peterson RE, et al. The genetic architecture of depression in individuals of East Asian ancestry: A Genome-Wide association study. JAMA Psychiatry. 2021;78(11):1258–69.

    Article  PubMed  Google Scholar 

  36. Liu CY, Hung YT, Chuang YL, Chen YJ, Weng WS, Liu JS, et al. Incorporating development stratification of Taiwan townships into sampling design of large scale health interview survey. J Health Manag. 2006;4:1–22.

    CAS  Google Scholar 

  37. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive review. J Am Geriatr Soc. 1992;40(9):922–35.

    Article  CAS  PubMed  Google Scholar 

  38. Hox J. Multilevel modeling: When and why Classification, data analysis, and data highways 1998. pp. 147–54.

  39. Clarke P, Wheaton B. Addressing data sparseness in contextual population research using cluster analysis to create synthetic neighborhoods. Sociol Methods Res. 2007;35:311–51.

    Article  Google Scholar 

  40. Bertram L, McQueen MB, Mullin K, Blacker D, Tanzi RE. Systematic meta-analyses of alzheimer disease genetic association studies: the alzgene database. Nat Genet. 2007;39(1):17–23.

    Article  CAS  PubMed  Google Scholar 

  41. Zhang C, Pierce BL. Genetic susceptibility to accelerated cognitive decline in the US health and retirement study. Neurobiol Aging. 2014;35(6):e151211–8.

    Article  Google Scholar 

  42. Lledo PM, Valley M. Adult olfactory bulb neurogenesis. Cold Spring Harb Perspect Biol. 2016;8(8).

  43. Calderon-Garciduenas L, Reed W, Maronpot RR, Henriquez-Roldan C, Delgado-Chavez R, Calderon-Garciduenas A, et al. Brain inflammation and Alzheimer’s-like pathology in individuals exposed to severe air pollution. Toxicol Pathol. 2004;32(6):650–8.

    Article  PubMed  Google Scholar 

  44. Doty RL. The olfactory vector hypothesis of neurodegenerative disease: is it viable? Ann Neurol. 2008;63(1):7–15.

    Article  PubMed  Google Scholar 

  45. Kim RE, Shin CY, Han SH, Kwon KJ, Astaxanthin Suppresses. PM2.5-Induced neuroinflammation by regulating Akt phosphorylation in BV-2 microglial cells. Int J Mol Sci. 2020;21:19.

    Google Scholar 

  46. Wang BR, Shi JQ, Ge NN, Ou Z, Tian YY, Jiang T, et al. PM2.5 exposure aggravates oligomeric amyloid beta-induced neuronal injury and promotes NLRP3 inflammasome activation in an in vitro model of Alzheimer’s disease. J Neuroinflammation. 2018;15(1):132.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Wurth R, Kioumourtzoglou MA, Tucker KL, Griffith J, Manjourides J, Suh H. Fine particle sources and cognitive function in an older Puerto Rican cohort in greater Boston. Environ Epidemiol. 2018;2(3):e022.

    Article  PubMed  Google Scholar 

  48. Tzivian L, Jokisch M, Winkler A, Weimar C, Hennig F, Sugiri D, et al. Associations of long-term exposure to air pollution and road traffic noise with cognitive function-An analysis of effect measure modification. Environ Int. 2017;103:30–8.

    Article  CAS  PubMed  Google Scholar 

  49. Ghasemi A, Zahediasl S. Normality tests for statistical analysis: a guide for non-statisticians. Int J Endocrinol Metab. 2012;10(2):486–9.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Ali MM, Sharma SC. Robustness to nonnormality of regression F tests. J Econom. 1996;71:175–205.

    Article  Google Scholar 

  51. Lumley T, Diehr P, Emerson S, Chen L. The importance of the normality assumption in large public health data sets. Annu Rev Public Health. 2002;23:151–69.

    Article  PubMed  Google Scholar 

  52. Knief U, Forstmeier W. Violating the normality assumption May be the lesser of two evils. Behav Res Methods. 2021;53:2576–90.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Martin AR, Gignoux CR, Walters RK, Wojcik GL, Neale BM, Gravel S, et al. Human demographic history impacts genetic risk prediction across diverse populations. Am J Hum Genet. 2017;100(4):635–49.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Su MH, Shih YH, Lin YF, Chen PC, Chen CY, Hsiao PC, et al. Familial aggregation and shared genetic loading for major psychiatric disorders and type 2 diabetes. Diabetologia. 2022;65(5):800–10.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This work was supported by the National Health Research Institutes (SH Wang, grant numbers CG-112-GP-10, CG-113-GP-10, CG-114-GP-10, NHRI-EX109-10931PI, NHRI-EX110-10931PI, NHRI-EX111-10931PI), Wan Fang Hospital (SF Liao, grant numbers 112-wf-eva-16, 113-wf-eva-07), Taipei Medical University (SF Liao, grant number TMU112-AE1-B14), Taiwan.

Author information

Authors and Affiliations

  1. Department of Medical Research, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan

    Shu-Fen Liao

  2. School of Public Health, College of Public Health, Taipei Medical University, Taipei, Taiwan

    Shu-Fen Liao

  3. Research Center for Humanities and Social Sciences, Academia Sinica, Taipei, Taiwan

    Ta-Chien Chan

  4. Institute of Public Health, School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan

    Ta-Chien Chan

  5. Department of Public Health, College of Public Health, China Medical University, Taichung, Taiwan

    Mei-Hsin Su

  6. Department of Psychiatry, Virginia Institute for Psychiatric Behavioral Genetics, Virginia Commonwealth University, Richmond, VA, USA

    Mei-Hsin Su

  7. National Center for Geriatrics and Welfare Research, National Health Research Institutes, 35, Keyan Road, Zhunan Town, Miaoli County 350, Miaoli, Taiwan

    Mei-Chen Lin, Chi-Shin Wu & Shi-Heng Wang

  8. Department of Psychiatry, National Taiwan University Hospital, Yunlin branch, Douliu, Taiwan

    Chi-Shin Wu

  9. Center for Population Neuroscience and Genetics, Laureate Institute for Brain Research, Tulsa, OK, USA

    Chun-Chieh Fan

  10. Department of Radiology, School of Medicine, University of California San Diego, La Jolla, CA, USA

    Chun-Chieh Fan

  11. Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan

    Shi-Heng Wang

Authors
  1. Shu-Fen Liao
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Ta-Chien Chan
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mei-Hsin Su
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Mei-Chen Lin
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Chi-Shin Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Chun-Chieh Fan
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Shi-Heng Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Conceptualization: SFL, SHW. Data curation and Formal analysis: SFL, TCC, SHW. Writing-original draft: SFL, TCC, SHW. Methodology and writing-review and editing: SFL, TCC, MHS, MCL, CSW, CCF, SHW. Approved the final article: all.

Corresponding author

Correspondence to Shi-Heng Wang.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the Central Regional Research Ethics Committee of China Medical University, Taichung, Taiwan (CRREC-108-30).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Conflict of interest

All authors declare no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liao, SF., Chan, TC., Su, MH. et al. The independent role of fine particulate matter and genetic liability on cognition in older adults. Ann Gen Psychiatry 24, 20 (2025). https://doi.org/10.1186/s12991-025-00559-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12991-025-00559-9

Keywords

Annals of General Psychiatry

ISSN: 1744-859X