Background Restrictions on mobility have been implemented by many countries to limit the spread of novel coronavirus infection (COVID-19) yet have important social and economic consequences. Their impact on reducing transmission is, however, inadequately understood. Methods We examined the association of COVID-19 incidence rates with mobility changes, defined as changes in categories of domestic location, against a pre-pandemic baseline, using country-specific daily incidence data on newly confirmed COVID-19 cases and mobility data collected from mobile devices in all 34 OECD countries plus Singapore and Taiwan. The study period was from the day of the 100th case in each country to August 31, 2020. Daily incidence rates were lagged by 14 days and regressed to mobility changes using LOESS regression and logit regression. Findings In two thirds of examined countries, reductions of up to 40% in commuting mobility (to workplaces, transit stations, retailers, and recreation) were associated with decreased COVID-19 incidence, more so early in the pandemic. However, these decreases plateaued as mobility remained low or decreased further. We found smaller or negligible associations between mobility restriction and incidence rates in the late phase in most countries. Interpretation Mild to moderate degrees of mobility restriction in most countries were associated with reduced incidence rates of COVID-19 that appear to attenuate over time, while some countries exhibited no effect of such restrictions. More detailed research is needed to precisely understand the benefits and limitations of mobility restrictions as part of the public health response to the COVID-19 pandemic. Funding none This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=3706748 How well does societal mobility restriction help control the COVID-19 pandemic? Evidence from real-time evaluation

Optimal growth and development in childhood and adolescence is crucial for lifelong health and well-being1–6. Here we used data from 2,325 population-based studies, with measurements of height and weight from 71 million participants, to report the height and body-mass index (BMI) of children and adolescents aged 5–19 years on the basis of rural and urban place of residence in 200 countries and territories from 1990 to 2020. In 1990, children and adolescents residing in cities were taller than their rural counterparts in all but a few high-income countries. By 2020, the urban height advantage became smaller in most countries, and in many high-income western countries it reversed into a small urban-based disadvantage. The exception was for boys in most countries in sub-Saharan Africa and in some countries in Oceania, south Asia and the region of central Asia, Middle East and north Africa. In these countries, successive cohorts of boys from rural places either did not gain height or possibly became shorter, and hence fell further behind their urban peers. The difference between the age-standardized mean BMI of children in urban and rural areas was <1.1 kg m–2 in the vast majority of countries. Within this small range, BMI increased slightly more in cities than in rural areas, except in south Asia, sub-Saharan Africa and some countries in central and eastern Europe. Our results show that in much of the world, the growth and developmental advantages of living in cities have diminished in the twenty-first century, whereas in much of sub-Saharan Africa they have amplified.

Objectives To determine the impact of restrictions on mobility on reducing transmission of COVID-19. Design Daily incidence rates lagged by 14 days were regressed on mobility changes using LOESS regression and logit regression between the day of the 100th case in each country to August 31, 2020. Setting 34 OECD countries plus Singapore and Taiwan. Participants Google mobility data were obtained from people who turned on mobile device-based global positioning system (GPS) and agreed to share their anonymized position information with Google. Interventions We examined the association of COVID-19 incidence rates with mobility changes, defined as changes in categories of domestic location, against a pre-pandemic baseline, using country-specific daily incidence data on newly confirmed COVID-19 cases and mobility data. Results In two thirds of examined countries, reductions of up to 40% in commuting mobility (to workplaces, transit stations, retailers, and recreation) were associated with decreased COVID-19 incidence, more so early in the pandemic. However, these decreases plateaued as mobility remained low or decreased further. We found smaller or negligible associations between mobility restriction and incidence rates in the late phase in most countries. Conclusion Mild to moderate degrees of mobility restriction in most countries were associated with reduced incidence rates of COVID-19 that appear to attenuate over time, while some countries exhibited no effect of such restrictions. More detailed research is needed to precisely understand the benefits and limitations of mobility restrictions as part of the public health response to the COVID-19 pandemic.

The methods used in low- and middle-income countries’ (LMICs) household surveys have not changed in four decades; however, LMIC societies have changed substantially and now face unprecedented rates of urbanization and urbanization of poverty. This mismatch may result in unintentional exclusion of vulnerable and mobile urban populations. We compare three survey method innovations with standard survey methods in Kathmandu, Dhaka, and Hanoi and summarize feasibility of our innovative methods in terms of time, cost, skill requirements, and experiences. We used descriptive statistics and regression techniques to compare respondent characteristics in samples drawn with innovative versus standard survey designs and household definitions, adjusting for sample probability weights and clustering. Feasibility of innovative methods was evaluated using a thematic framework analysis of focus group discussions with survey field staff, and via survey planner budgets. We found that a common household definition excluded single adults (46.9%) and migrant-headed households (6.7%), as well as non-married (8.5%), unemployed (10.5%), disabled (9.3%), and studying adults (14.3%). Further, standard two-stage sampling resulted in fewer single adult and non-family households than an innovative area-microcensus design; however, two-stage sampling resulted in more tent and shack dwellers. Our survey innovations provided good value for money, and field staff experiences were neutral or positive. Staff recommended streamlining field tools and pairing technical and survey content experts during fieldwork. This evidence of exclusion of vulnerable and mobile urban populations in LMIC household surveys is deeply concerning and underscores the need to modernize survey methods and practices.

Background: Artificial intelligence (AI)-based computational models that analyze breast cancer have been developed for decades. The present study was implemented to investigate the accuracy and efficiency of combined mammography images and clinical records for breast cancer detection using machine learning and deep learning classifiers. Methods: This study was verified using 731 images from 357 women who underwent at least one mammogram and had clinical records for at least six months before mammography. The model was trained on mammograms and clinical variables to discriminate benign and malignant lesions. Multiple pre-trained deep CNN models to detect cancer in mammograms, including X-ception, VGG16, ResNet-v2, ResNet50, and CNN3 were employed. Machine learning models were constructed using k-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), Artificial Neural Network (ANN), and gradient boosting machine (GBM) in the clinical dataset. Results: The detection performance obtained an accuracy of 84.5% with a specificity of 78.1% at a sensitivity of 89.7% and an AUC of 0.88. When trained on mammography image data alone, the result achieved a slightly lower score than the combined model (accuracy, 72.5% vs. 84.5%, respectively). Conclusions: A breast cancer-detection model combining machine learning and deep learning models was performed in this study with a satisfactory result, and this model has potential clinical applications.