Association between area-level walkability and glycated haemoglobin: a Portuguese population-based study.

Diabetes poses a substantial disease burden, prompting preventive interventions. Physical inactivity, a major risk factor for type 2 diabetes, can potentially be mitigated by enhancing area-level walkability. Despite this, limited population-based studies have investigated the link between walkability and objective diabetes measures. Our study aims to estimate the association between area-level walkability and individual glycated haemoglobin levels in the Portuguese adult population without the diagnosis of diabetes. Data from the 2011 census and an updated street map were obtained to construct a walkability index based on residential density, land-use mix, and street connectivity. Individual health data were sourced from The National Health Examination Survey (INSEF) 2015, a representative survey of the Portuguese adult population. Gamma regression was employed for estimation of the main associations, revealing that residing in moderately walkable areas significantly reduced average glycated haemoglobin levels (Exp(ß) = 0.906; 95% CI: 0.821, 0.999) compared to the least walkable areas. The association was less pronounced and not statistically significant for the third tertile of walkability (Exp(ß) = 0.919; 95% CI: 0.822, 1.028). Our findings highlight a nonlinear protective association between walkability and glycated haemoglobin, emphasizing the potential policy implications for urban planning, diabetes prevention, and health promotion.

Unraveling the hurdles of a large COVID-19 epidemiological investigation by viral genomics.

COVID-19 local outbreak response relies on subjective information to reconstruct transmission chains. We assessed the concordance between epidemiologically linked cases and viral genetic profiles, in the Baixo Vouga Region (Portugal), from March to June 2020. A total of 1925 COVID-19 cases were identified, with 1143 being assigned to 154 epiclusters. Viral genomic data was available for 128 cases. Public health authorities identified two large epiclusters (280 and 101 cases each) with a central role on the spread of the disease. Still, the genomic data revealed that each epicluster included two distinct SARS-CoV-2 genetic profiles and thus more than one transmission network. We were able to show that the initial transmission dynamics reconstruction was most likely accurate, but the increasing dimension of the epiclusters and its extension to densely populated settings (healthcare and nursing home settings) triggered the misidentification of links. Genomics was also key to resolve some sporadic cases and misidentified direction of transmission. The epidemiological investigation showed a sensitivity of 70%-86% to detect transmission chains. This study contributes to the understanding of the hurdles and caveats associated with the epidemiological investigation of hundreds of community cases in the context of a massive outbreak caused by a highly transmissible and new respiratory virus.