Biomarkers

Environmental factors and health have often been studied by geographical region; few studies have focused on indoor environmental quality. The Artificial Intelligence Ready and Equitable Atlas for Diabetes Insights (AI-READI) project collects multimodal data from a diverse group of participants representing a range of Type 2 Diabetes Mellitus (T2DM) disease states (normal to insulin-dependent). We used AI-READI data to analyze associations between indoor environmental measures and early indicators of cognitive impairment.

Participants aged 40 and over with and without Type 2 diabetes are being recruited to the 4-year AI-READI project; data from the first 1067 individuals were released in year 2. Participants with Montreal Cognitive Assessment (MoCA) scores and environmental data were included. MoCA scores were binarized: 0 for scores > 25 (normal) and 1 for scores <=25 (impaired). Self-reported T2DM status was binarized: 0 for no diabetes or lifestyle-controlled diabetes, and 1 for insulin- or oral medication-controlled diabetes. Environmental data were collected for 10 days in participants' homes using a device measuring volatile organic compounds (VOC), nitrogen oxides (NOX), relative light intensity, and particulate matter (PM).

Out of 1037 participants analyzed, 488 (47%) had impaired cognition. Higher PM concentrations (1, 2.5, 4, and 10 um or smaller) were associated with lower MoCA scores. For example, participants with cognitive impairment had a mean PM2.5 concentration of 21.53 [18.07, 24.99, 95% CI] compared to 11.53 [10.24, 12.82 95% CI] for normal participants. (Figure 1) Lower light levels were associated with lower MoCA scores: median light intensity was 0.03 [0.027, 0.040 95% CI] for normal participants and 0.02 [0.015, 0.026 95% CI] for those with impaired cognition. (Figure 2) We found significant associations between MoCA (binarized) and T2DM (Chi-square, p-value= 4.65e-5). (Figure 3) There were no significant associations between cognitive function and VOC or NOX measures.

Preliminary results from the AI-READI cohort show an association between cognitive function (MoCA) and measures of ambient light, PM2.5, and T2DM status. Analysis of other factors (VOC, NOX) did not show significant associations.