gcarrasco-publications

PUBLICATIONS

Abstract
The rising humid heat is regarded as a severe threat to human survivability, but the proper integration of humid heat into heat-health alerts is still being explored. Using state-of-the-art epidemiological and climatological datasets, we examined the association between multiple heat stress indicators (HSIs) and daily human mortality in 739 cities worldwide. Notable differences were observed in the long-term trends and timing of heat events detected by HSIs. Air temperature (Tair) predicts heat-related mortality well in cities with a robust negative Tair-relative humidity correlation (CT-RH). However, in cities with near-zero or weak positive CT-RH, HSIs considering humidity provide enhanced predictive power compared to Tair. Furthermore, the magnitude and timing of heat-related mortality measured by HSIs could differ largely from those associated with Tair in many cities. Our findings provide important insights into specific regions where humans are vulnerable to humid heat and can facilitate the further enhancement of heat-health alert systems.

Regional variation in the role of humidity on city-level heat-related mortality

Background
Heat can vary spatially within an urban area. Individual-level heat exposure may thus depend on an individual’s day-to-day travel patterns (also called mobility patterns or activity space), yet heat exposure is commonly measured based on place of residence.

Objective
In this study, we compared measures assessing exposure to two heat indicators using place of residence with those defined considering participants’ day-to-day mobility patterns.

Methods
Participants (n = 599; aged 35-80 years old [mean =59 years]) from San Diego County, California wore a GPS device to measure their day-to-day travel over 14-day intervals between 2014-10-17 and 2017-10-06. We measured exposure to two heat indicators (land-surface temperature [LST] and air temperature) using an approach considering their mobility patterns and an approach considering only their place of residence. We compared participant mean and maximum exposure values from each method for each indicator.

Results
The overall mobility-based mean LST exposure (34.7 °C) was almost equivalent to the corresponding residence-based mean (34.8 °C; mean difference in means = −0.09 °C). Similarly, the mean difference between the overall mobility-based mean air temperature exposure (19.2 °C) and the corresponding residence-based mean (19.2 °C) was negligible (−0.02 °C). Meaningful differences emerged, however, when comparing maximums, particularly for LST. The mean mobility-based maximum LST was 40.3 °C compared with a mean residence-based maximum of 35.8 °C, a difference of 4.51 °C. The difference in maximums was considerably smaller for air temperature (mean = 0.40 °C; SD = 1.41 °C) but nevertheless greater than the corresponding difference in means.

Impact
As the climate warms, assessment of heat exposure both at and away from home is important for understanding its health impacts. We compared two approaches to estimate exposure to two heat measures (land surface temperature and air temperature). The first approach only considered exposure at home, and the second considered day-to-day travel. Considering the average exposure estimated by each approach, the results were almost identical. Considering the maximum exposure experienced (specific definition in text), the differences between the two approaches were more considerable, especially for land surface temperature.

Is home where the heat is? comparing residence-based with mobility-based measures of heat exposure in San Diego, California

Objective To examine the associations between characteristics of daily rainfall (intensity, duration, and frequency) and all cause, cardiovascular, and respiratory mortality.

Design Two stage time series analysis.

Setting 645 locations across 34 countries or regions.

Population Daily mortality data, comprising a total of 109 954 744 all cause, 31 164 161 cardiovascular, and 11 817 278 respiratory deaths from 1980 to 2020.

Main outcome measure Association between daily mortality and rainfall events with return periods (the expected average time between occurrences of an extreme event of a certain magnitude) of one year, two years, and five years, with a 14 day lag period. A continuous relative intensity index was used to generate intensity-response curves to estimate mortality risks at a global scale.

Results During the study period, a total of 50 913 rainfall events with a one year return period, 8362 events with a two year return period, and 3301 events with a five year return period were identified. A day of extreme rainfall with a five year return period was significantly associated with increased daily all cause, cardiovascular, and respiratory mortality, with cumulative relative risks across 0-14 lag days of 1.08 (95% confidence interval 1.05 to 1.11), 1.05 (1.02 to 1.08), and 1.29 (1.19 to 1.39), respectively. Rainfall events with a two year return period were associated with respiratory mortality only, whereas no significant associations were found for events with a one year return period. Non-linear analysis revealed protective effects (relative risk <1) with moderate-heavy rainfall events, shifting to adverse effects (relative risk >1) with extreme intensities. Additionally, mortality risks from extreme rainfall events appeared to be modified by climate type, baseline variability in rainfall, and vegetation coverage, whereas the moderating effects of population density and income level were not significant. Locations with lower variability of baseline rainfall or scarce vegetation coverage showed higher risks.

Conclusion Daily rainfall intensity is associated with varying health effects, with extreme events linked to an increasing relative risk for all cause, cardiovascular, and respiratory mortality. The observed associations varied with local climate and urban infrastructure.

Rainfall events and daily mortality across 645 global locations: two stage time series analysis

Background:
We quantify the mortality burden and economic loss attributable to nonoptimal temperatures for cold and heat in the Central and South American countries in the Multi-City Multi-Country (MCC) Collaborative Research Network.

Methods:
We collected data for 66 locations from 13 countries in Central and South America to estimate location-specific temperature–mortality associations using time-series regression with distributed lag nonlinear models. We calculated the attributable deaths for cold and heat as the 2.5th and 97.5th temperature percentiles, above and below the minimum mortality temperature, and used the value of a life year to estimate the economic loss of delayed deaths.

Results:
The mortality impact of cold varied widely by country, from 9.64% in Uruguay to 0.22% in Costa Rica. The heat-attributable fraction for mortality ranged from 1.41% in Paraguay to 0.01% in Ecuador. Locations in arid and temperate climatic zones showed higher cold-related mortality (5.10% and 5.29%, respectively) than those in tropical climates (1.71%). Arid and temperate climatic zones saw lower heat-attributable fractions (0.69% and 0.58%) than arid climatic zones (0.92%). Exposure to cold led to an annual economic loss of $0.6 million in Costa Rica to $472.2 million in Argentina. In comparison, heat resulted in economic losses of $0.05 million in Ecuador to $90.6 million in Brazil.

Conclusion:
Most of the mortality burden for Central and South American countries is caused by cold compared to heat, generating annual economic losses of $2.1 billion and $290.7 million, respectively. Public health policies and adaptation measures in the region should account for the health effects associated with nonoptimal temperatures.

Mortality burden and economic loss attributable to cold and heat in Central and South America

Background:
Land-use and land-cover change (LULCC) can substantially affect climate through biogeochemical and biogeophysical effects. Here, we examine the future temperature–mortality impact for two contrasting LULCC scenarios in a background climate of low greenhouse gas concentrations. The first LULCC scenario implies a globally sustainable land use and socioeconomic development (sustainability). In the second LULCC scenario, sustainability is implemented only in the Organisation for Economic Cooperation and Development countries (inequality).

Methods:
Using the Multi-Country Multi-City (MCC) dataset on mortality from 823 locations in 52 countries and territories, we estimated the temperature–mortality exposure–response functions (ERFs). The LULCC and noLULCC scenarios were implemented in three fully coupled Earth system models (ESMs): Community Earth System Model, Max Planck Institute Earth System Model, and European Consortium Earth System Model. Next, using temperature from the ESMs’ simulations and the estimated location-specific ERFs, we assessed the temperature-related impact on mortality for the LULCC and noLULCC scenarios around the mid and end century.

Results:
Under sustainability, the multimodel mean changes in excess mortality range from −1.1 to +0.6 percentage points by 2050–2059 across all locations and from −1.4 to +0.5 percentage points by 2090–2099. Under inequality, these vary from −0.7 to +0.9 percentage points by 2050–2059 and from −1.3 to +2 percentage points by 2090–2099.

Conclusions:
While an unequal socioeconomic development and unsustainable land use could increase the burden of heat-related mortality in most regions, globally sustainable land use has the potential to reduce it in some locations. However, the total (cold and heat) impact on mortality is very location specific and strongly depends on the underlying climate change scenario due to nonlinearity in the temperature–mortality relationship.

Impacts of land-use and land-cover changes on temperature-related mortality

Abstract
Network analysis may improve the understanding of malaria epidemiology in rural areas of the Amazon region by explicitly representing the relationships between villages as a proxy for human population mobility. This study tests a comprehensive set of connectivity metrics and their relationship with malaria incidence across villages with contrasting PAMAFRO (a malaria control initiative) coverage levels in the Loreto department of Peru using data from the passive case detection reports from the Peruvian Ministry of Health between 2011 and 2018 at the village level. A total of 24 centrality metrics were computed and tested on 1608 nodes (i.e., villages/cities). Based on its consistency and stability, the betweenness centrality type outperformed other metrics. No appreciable differences in the distributions of malaria incidence were found when using different weights, including population, deforested area, Euclidian distance, or travel time. Overall, villages in the top quintile of centrality have a higher malaria incidence in comparison with villages in the bottom quintile of centrality (Mean Difference in cases per 1000 population; P. vivax = 165.78 and P. falciparum = 76.14). The mean difference between villages at the top and bottom centrality quintiles increases as PAMAFRO coverage increases for both P. vivax (Tier 1 = 155.36; Tier 2 = 176.22; Tier 3 = 326.08) and P. falciparum (Tier 1 = 48.11; Tier 2 = 95.16; Tier 3 = 139.07). The findings of this study support the shift in current malaria control strategies from targeting specific locations based on malaria metrics to strategies based on connectivity neighborhoods that include influential connected villages.

The role of connectivity on malaria dynamics across areas with contrasting control coverage in the Peruvian Amazon

Meteorological factors, population immunity, and COVID-19 incidence: A global multi-city analysis

Background:
Precipitation could affect the transmission of diarrheal diseases. The diverse precipitation patterns across different climates might influence the degree of diarrheal risk from precipitation. This study determined the associations between precipitation and diarrheal mortality in tropical, temperate, and arid climate regions.
Methods:
Daily counts of diarrheal mortality and 28-day cumulative precipitation from 1997 to 2019 were analyzed across 29 locations in eight middle-income countries (Argentina, Brazil, Costa Rica, India, Peru, the Philippines, South Africa, and Thailand). A two-stage approach was employed: the first stage is conditional Poisson regression models for each location, and the second stage is meta-analysis for pooling location-specific coefficients by climate zone.
Results:
In tropical climates, higher precipitation increases the risk of diarrheal mortality. Under extremely wet conditions (95th percentile of 28-day cumulative precipitation), diarrheal mortality increased by 17.8% (95% confidence interval [CI] = 10.4%, 25.7%) compared with minimum-risk precipitation. For temperate and arid climates, diarrheal mortality increases in both dry and wet conditions. In extremely dry conditions (fifth percentile of 28-day cumulative precipitation), diarrheal mortality risk increases by 3.8% (95% CI = 1.2%, 6.5%) for temperate and 5.5% (95% CI = 1.0%, 10.2%) for arid climates. Similarly, under extremely wet conditions, diarrheal mortality risk increases by 2.5% (95% CI = −0.1%, 5.1%) for temperate and 4.1% (95% CI = 1.1%, 7.3%) for arid climates.
Conclusions:
Associations between precipitation and diarrheal mortality exhibit variations across different climate zones. It is crucial to consider climate-specific variations when generating global projections of future precipitation-related diarrheal mortality.

Association between precipitation and mortality due to diarrheal diseases by climate zone: A multi-country modeling study

Background
Wildfire activity is an important source of tropospheric ozone (O3) pollution. However, no study to date has systematically examined the associations of wildfire-related O3 exposure with mortality globally.

Methods
We did a multicountry two-stage time series analysis. From the Multi-City Multi-Country (MCC) Collaborative Research Network, data on daily all-cause, cardiovascular, and respiratory deaths were obtained from 749 locations in 43 countries or areas, representing overlapping periods from Jan 1, 2000, to Dec 31, 2016. We estimated the daily concentration of wildfire-related O3 in study locations using a chemical transport model, and then calibrated and downscaled O3 estimates to a resolution of 0·25° × 0·25° (approximately 28 km2 at the equator). Using a random-effects meta-analysis, we examined the associations of short-term wildfire-related O3 exposure (lag period of 0–2 days) with daily mortality, first at the location level and then pooled at the country, regional, and global levels. Annual excess mortality fraction in each location attributable to wildfire-related O3 was calculated with pooled effect estimates and used to obtain excess mortality fractions at country, regional, and global levels.

Findings
Between 2000 and 2016, the highest maximum daily wildfire-related O3 concentrations (≥30 μg/m3) were observed in locations in South America, central America, and southeastern Asia, and the country of South Africa. Across all locations, an increase of 1 μg/m3 in the mean daily concentration of wildfire-related O3 during lag 0–2 days was associated with increases of 0·55% (95% CI 0·29 to 0·80) in daily all-cause mortality, 0·44% (–0·10 to 0·99) in daily cardiovascular mortality, and 0·82% (0·18 to 1·47) in daily respiratory mortality. The associations of daily mortality rates with wildfire-related O3 exposure showed substantial geographical heterogeneity at the country and regional levels. Across all locations, estimated annual excess mortality fractions of 0·58% (95% CI 0·31 to 0·85; 31 606 deaths [95% CI 17 038 to 46 027]) for all-cause mortality, 0·41% (–0·10 to 0·91; 5249 [–1244 to 11 620]) for cardiovascular mortality, and 0·86% (0·18 to 1·51; 4657 [999 to 8206]) for respiratory mortality were attributable to short-term exposure to wildfire-related O3.

Interpretation
In this study, we observed an increase in all-cause and respiratory mortality associated with short-term wildfire-related O3 exposure. Effective risk and smoke management strategies should be implemented to protect the public from the impacts of wildfires.

All-cause, cardiovascular, and respiratory mortality and wildfire-related ozone: a multicountry two-stage time series analysis

Background
The association between nonoptimal temperatures and cardiovascular mortality risk is recognized. However, a comprehensive global assessment of this burden is lacking.

Objectives
The goal of this study was to assess global cardiovascular mortality burden attributable to nonoptimal temperatures and investigate spatiotemporal trends.

Methods
Using daily cardiovascular deaths and temperature data from 32 countries, a 3-stage analytical approach was applied. First, location-specific temperature–mortality associations were estimated, considering nonlinearity and delayed effects. Second, a multivariate meta-regression model was developed between location-specific effect estimates and 5 meta-predictors. Third, cardiovascular deaths associated with nonoptimal, cold, and hot temperatures for each global grid (55 km × 55 km resolution) were estimated, and temporal trends from 2000 to 2019 were explored.

Results
Globally, 1,801,513 (95% empirical CI: 1,526,632-2,202,831) annual cardiovascular deaths were associated with nonoptimal temperatures, constituting 8.86% (95% empirical CI: 7.51%-12.32%) of total cardiovascular mortality corresponding to 26 deaths per 100,000 population. Cold-related deaths accounted for 8.20% (95% empirical CI: 6.74%-11.57%), whereas heat-related deaths accounted for 0.66% (95% empirical CI: 0.49%-0.98%). The mortality burden varied significantly across regions, with the highest excess mortality rates observed in Central Asia and Eastern Europe. From 2000 to 2019, cold-related excess death ratios decreased, while heat-related ratios increased, resulting in an overall decline in temperature-related deaths. Southeastern Asia, Sub-Saharan Africa, and Oceania observed the greatest reduction, while Southern Asia experienced an increase. The Americas and several regions in Asia and Europe displayed fluctuating temporal patterns.

Conclusions
Nonoptimal temperatures substantially contribute to cardiovascular mortality, with heterogeneous spatiotemporal patterns. Effective mitigation and adaptation strategies are crucial, especially given the increasing heat-related cardiovascular deaths amid climate change.

Global and Regional Cardiovascular Mortality Attributable to Nonoptimal Temperatures Over Time

Objectives: This study aims to estimate the short-term preventable mortality and associated economic costs of complying with the World Health Organization (WHO) air quality guidelines (AQGs) limit values for PM10 and PM2.5 in nine major Latin American cities.

Methods: We estimated city-specific PM-mortality associations using time-series regression models and calculated the attributable mortality fraction. Next, we used the value of statistical life to calculate the economic benefits of complying with the WHO AQGs limit values.

Results: In most cities, PM concentrations exceeded the WHO AQGs limit values more than 90% of the days. PM10 was found to be associated with an average excess mortality of 1.88% with concentrations above WHO AQGs limit values, while for PM2.5 it was 1.05%. The associated annual economic costs varied widely, between US$ 19.5 million to 3,386.9 million for PM10, and US$ 196.3 million to 2,209.6 million for PM2.5.

Conclusion: Our findings suggest that there is an urgent need for policymakers to develop interventions to achieve sustainable air quality improvements in Latin America. Complying with the WHO AQGs limit values for PM10 and PM2.5 in Latin American cities would substantially benefits for urban populations.

Health and Economic Benefits of Complying With the World Health Organization Air Quality Guidelines for Particulate Matter in Nine Major Latin American Cities

Background
In the near future, the incidence of mosquito-borne diseases may expand to new sites due to changes in temperature and rainfall patterns caused by climate change. Therefore, there is a need to use recent technological advances to improve vector surveillance methodologies. Unoccupied Aerial Vehicles (UAVs), often called drones, have been used to collect high-resolution imagery to map detailed information on mosquito habitats and direct control measures to specific areas. Supervised classification approaches have been largely used to automatically detect vector habitats. However, manual data labelling for model training limits their use for rapid responses. Open-source foundation models such as the Meta AI Segment Anything Model (SAM) can facilitate the manual digitalization of high-resolution images. This pre-trained model can assist in extracting features of interest in a diverse range of images. Here, we evaluated the performance of SAM through the Samgeo package, a Python-based wrapper for geospatial data, as it has not been applied to analyse remote sensing images for epidemiological studies.

Results
We tested the identification of two land cover classes of interest: water bodies and human settlements, using different UAV acquired imagery across five malaria-endemic areas in Africa, South America, and Southeast Asia. We employed manually placed point prompts and text prompts associated with specific classes of interest to guide the image segmentation and assessed the performance in the different geographic contexts. An average Dice coefficient value of 0.67 was obtained for buildings segmentation and 0.73 for water bodies using point prompts. Regarding the use of text prompts, the highest Dice coefficient value reached 0.72 for buildings and 0.70 for water bodies. Nevertheless, the performance was closely dependent on each object, landscape characteristics and selected words, resulting in varying performance.

Conclusions
Recent models such as SAM can potentially assist manual digitalization of imagery by vector control programs, quickly identifying key features when surveying an area of interest. However, accurate segmentation still requires user-provided manual prompts and corrections to obtain precise segmentation. Further evaluations are necessary, especially for applications in rural areas.

Using image segmentation models to analyse high-resolution earth observation data: new tools to monitor disease risks in changing environments

Background
The regional disparity of heatwave-related mortality over a long period has not been sufficiently assessed across the globe, impeding the localisation of adaptation planning and risk management towards climate change. We quantified the global mortality burden associated with heatwaves at a spatial resolution of 0.5°×0.5° and the temporal change from 1990 to 2019.

Methods and findings
We collected data on daily deaths and temperature from 750 locations of 43 countries or regions, and 5 meta-predictors in 0.5°×0.5° resolution across the world. Heatwaves were defined as location-specific daily mean temperature ≥95th percentiles of year-round temperature range with duration ≥2 days. We first estimated the location-specific heatwave-mortality association. Secondly, a multivariate meta-regression was fitted between location-specific associations and 5 meta-predictors, which was in the third stage used with grid cell-specific meta-predictors to predict grid cell-specific association. Heatwave-related excess deaths were calculated for each grid and aggregated. During 1990 to 2019, 0.94% (95% CI: 0.68–1.19) of deaths [i.e., 153,078 cases (95% eCI: 109,950–194,227)] per warm season were estimated to be from heatwaves, accounting for 236 (95% eCI: 170–300) deaths per 10 million residents. The ratio between heatwave-related excess deaths and all premature deaths per warm season remained relatively unchanged over the 30 years, while the number of heatwave-related excess deaths per 10 million residents per warm season declined by 7.2% per decade in comparison to the 30-year average. Locations with the highest heatwave-related death ratio and rate were in Southern and Eastern Europe or areas had polar and alpine climates, and/or their residents had high incomes. The temporal change of heatwave-related mortality burden showed geographic disparities, such that locations with tropical climate or low incomes were observed with the greatest decline. The main limitation of this study was the lack of data from certain regions, e.g., Arabian Peninsula and South Asia.

Conclusions
Heatwaves were associated with substantial mortality burden that varied spatiotemporally over the globe in the past 30 years. The findings indicate the potential benefit of governmental actions to enhance health sector adaptation and resilience, accounting for inequalities across communities.

Global, regional, and national burden of heatwave-related mortality from 1990 to 2019: A three-stage modelling study

Background
Temperature variability (TV) is associated with increased mortality risk. However, it is still unknown whether intra-day or inter-day TV has different effects.

Objectives
We aimed to assess the association of intra-day TV and inter-day TV with all-cause, cardiovascular, and respiratory mortality.

Methods
We collected data on total, cardiovascular, and respiratory mortality and meteorology from 758 locations in 47 countries or regions from 1972 to 2020. We defined inter-day TV as the standard deviation (SD) of daily mean temperatures across the lag interval, and intra-day TV as the average SD of minimum and maximum temperatures on each day. In the first stage, inter-day and intra-day TVs were modelled simultaneously in the quasi-Poisson time-series model for each location. In the second stage, a multi-level analysis was used to pool the location-specific estimates.

Results
Overall, the mortality risk due to each interquartile range [IQR] increase was higher for intra-day TV than for inter-day TV. The risk increased by 0.59% (95% confidence interval [CI]: 0.53, 0.65) for all-cause mortality, 0.64% (95% CI: 0.56, 0.73) for cardiovascular mortality, and 0.65% (95% CI: 0.49, 0.80) for respiratory mortality per IQR increase in intra-day TV0–7 (0.9 °C). An IQR increase in inter-day TV0–7 (1.6 °C) was associated with 0.22% (95% CI: 0.18, 0.26) increase in all-cause mortality, 0.44% (95% CI: 0.37, 0.50) increase in cardiovascular mortality, and 0.31% (95% CI: 0.21, 0.41) increase in respiratory mortality. The proportion of all-cause deaths attributable to intra-day TV0–7 and inter-day TV0–7 was 1.45% and 0.35%, respectively. The mortality risks varied by lag interval, climate area, season, and climate type.

Conclusions
Our results indicated that intra-day TV may explain the main part of the mortality risk related to TV and suggested that comprehensive evaluations should be proposed in more countries to help protect human health.

Comparison for the effects of different components of temperature variability on mortality: A multi-country time-series study

Assessing the association between temperature frequency and mortality can provide insights into human adaptation to local ambient temperatures. We collected daily time-series data on mortality and temperature from 757 locations in 47 countries/regions during 1979–2020. We used a two-stage time series design to assess the association between temperature frequency and all-cause mortality. The results were pooled at the national, regional, and global levels. We observed a consistent decrease in the risk of mortality as the normalized frequency of temperature increases across the globe. The average increase in mortality risk comparing the 10th to 100th percentile of normalized frequency was 13.03% (95% CI: 12.17–13.91), with substantial regional differences (from 4.56% in Australia and New Zealand to 33.06% in South Europe). The highest increase in mortality was observed for high-income countries (13.58%, 95% CI: 12.56–14.61), followed by lower-middle-income countries (12.34%, 95% CI: 9.27–15.51). This study observed a declining risk of mortality associated with higher temperature frequency. Our findings suggest that populations can adapt to their local climate with frequent exposure, with the adapting ability varying geographically due to differences in climatic and socioeconomic characteristics.

Temperature frequency and mortality: Assessing adaptation to local temperature

Older adults are generally amongst the most vulnerable to heat and cold. While temperature-related health impacts are projected to increase with global warming, the influence of population aging on these trends remains unclear. Here we show that at 1.5 °C, 2 °C, and 3 °C of global warming, heat-related mortality in 800 locations across 50 countries/areas will increase by 0.5%, 1.0%, and 2.5%, respectively; among which 1 in 5 to 1 in 4 heat-related deaths can be attributed to population aging. Despite a projected decrease in cold-related mortality due to progressive warming alone, population aging will mostly counteract this trend, leading to a net increase in cold-related mortality by 0.1%–0.4% at 1.5–3 °C global warming. Our findings indicate that population aging constitutes a crucial driver for future heat- and cold-related deaths, with increasing mortality burden for both heat and cold due to the aging population.

Impact of population aging on future temperature-related mortality at different global warming levels

Background
Exposure to cold spells is associated with mortality. However, little is known about the global mortality burden of cold spells.

Methods
A three-stage meta-analytical method was used to estimate the global mortality burden associated with cold spells by means of a time series dataset of 1960 locations across 59 countries (or regions). First, we fitted the location-specific, cold spell-related mortality associations using a quasi-Poisson regression with a distributed lag non-linear model with a lag period of up to 21 days. Second, we built a multivariate meta-regression model between location-specific associations and seven predictors. Finally, we predicted the global grid-specific cold spell-related mortality associations during 2000–19 using the fitted meta-regression model and the yearly grid-specific meta-predictors. We calculated the annual excess deaths, excess death ratio (excess deaths per 1000 deaths), and excess death rate (excess deaths per 100 000 population) due to cold spells for each grid across the world.

Findings
Globally, 205 932 (95% empirical CI [eCI] 162 692–250 337) excess deaths, representing 3·81 (95% eCI 2·93–4·71) excess deaths per 1000 deaths (excess death ratio), and 3·03 (2·33–3·75) excess deaths per 100 000 population (excess death rate) were associated with cold spells per year between 2000 and 2019. The annual average global excess death ratio in 2016–19 increased by 0·12 percentage points and the excess death rate in 2016–19 increased by 0·18 percentage points, compared with those in 2000–03. The mortality burden varied geographically. The excess death ratio and rate were highest in Europe, whereas these indicators were lowest in Africa. Temperate climates had higher excess death ratio and rate associated with cold spells than other climate zones.

Interpretation
Cold spells are associated with substantial mortality burden around the world with geographically varying patterns. Although the number of cold spells has on average been decreasing since year 2000, the public health threat of cold spells remains substantial. The findings indicate an urgency of taking local and regional measures to protect the public from the mortality burdens of cold spells.

Global, regional, and national burden of mortality associated with cold spells during 2000–19: a three-stage modelling study

Background
Climate change can directly impact temperature-related excess deaths and might subsequently change the seasonal variation in mortality. In this study, we aimed to provide a systematic and comprehensive assessment of potential future changes in the seasonal variation, or seasonality, of mortality across different climate zones.

Methods
In this modelling study, we collected daily time series of mean temperature and mortality (all causes or non-external causes only) via the Multi-Country Multi-City Collaborative (MCC) Research Network. These data were collected during overlapping periods, spanning from Jan 1, 1969 to Dec 31, 2020. We projected daily mortality from Jan 1, 2000 to Dec 31, 2099, under four climate change scenarios corresponding to increasing emissions (Shared Socioeconomic Pathways [SSP] scenarios SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5). We compared the seasonality in projected mortality between decades by its shape, timings (the day-of-year) of minimum (trough) and maximum (peak) mortality, and sizes (peak-to-trough ratio and attributable fraction). Attributable fraction was used to measure the burden of seasonality of mortality. The results were summarised by climate zones.

Findings
The MCC dataset included 126 809 537 deaths from 707 locations within 43 countries or areas. After excluding the only two polar locations (both high-altitude locations in Peru) from climatic zone assessments, we analysed 126 766 164 deaths in 705 locations aggregated in four climate zones (tropical, arid, temperate, and continental). From the 2000s to the 2090s, our projections showed an increase in mortality during the warm seasons and a decrease in mortality during the cold seasons, albeit with mortality remaining high during the cold seasons, under all four SSP scenarios in the arid, temperate, and continental zones. The magnitude of this changing pattern was more pronounced under the high-emission scenarios (SSP3-7.0 and SSP5-8.5), substantially altering the shape of seasonality of mortality and, under the highest emission scenario (SSP5-8.5), shifting the mortality peak from cold seasons to warm seasons in arid, temperate, and continental zones, and increasing the size of seasonality in all zones except the arid zone by the end of the century. In the 2090s compared with the 2000s, the change in peak-to-trough ratio (relative scale) ranged from 0·96 to 1·11, and the change in attributable fraction ranged from 0·002% to 0·06% under the SSP5-8.5 (highest emission) scenario.

Interpretation
A warming climate can substantially change the seasonality of mortality in the future. Our projections suggest that health-care systems should consider preparing for a potentially increased demand during warm seasons and sustained high demand during cold seasons, particularly in regions characterised by arid, temperate, and continental climates.

Seasonality of mortality under climate change: a multicountry projection study

Despite progress towards malaria reduction in Peru, measuring exposure in low transmission areas is crucial for achieving elimination. This study focuses on two very low transmission areas in Loreto (Peruvian Amazon) and aims to determine the relationship between malaria exposure and proximity to health facilities. Individual data was collected from 38 villages in Indiana and Belen, including geo-referenced households and blood samples for microscopy, PCR and serological analysis. A segmented linear regression model identified significant changes in seropositivity trends among different age groups. Local Getis-Ord Gi* statistic revealed clusters of households with high (hotspots) or low (coldspots) seropositivity rates. Findings from 4000 individuals showed a seropositivity level of 2.5% (95%CI: 2.0%-3.0%) for P. falciparum and 7.8% (95%CI: 7.0%-8.7%) for P. vivax, indicating recent or historical exposure. The segmented regression showed exposure reductions in the 40–50 age group (β1 = 0.043, p = 0.003) for P. vivax and the 50–60 age group (β1 = 0.005, p = 0.010) for P. falciparum. Long and extreme distance villages from Regional Hospital of Loreto exhibited higher malaria exposure compared to proximate and medium distance villages (p < 0.001). This study showed the seropositivity of malaria in two very low transmission areas and confirmed the spatial pattern of hotspots as villages become more distant.

Malaria seroepidemiology in very low transmission settings in the Peruvian Amazon

Introduction
The use of drones in environment and health research is a relatively new phenomenon. A principal research activity drones are used for is environmental monitoring, which can raise concerns in local communities. Existing ethical guidance for researchers is often not specific to drone technology and practices vary between research settings. Therefore, this scoping review aims to gather the evidence available on ethical considerations surrounding drone use as perceived by local communities, ethical considerations reported on by researchers implementing drone research, and published ethical guidance related to drone deployment.

Methods and analysis
This scoping review will follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for scoping reviews (PRISMA-ScR) and the Joanna Briggs Institute (JBI) guidelines. The literature search will be conducted using academic databases and grey literature sources. After pilot testing the inclusion criteria and data extraction tool, two researchers will double-screen and then chart available evidence independently. A content analysis will be carried out to identify patterns of categories or terms used to describe ethical considerations related to drone usage for environmental monitoring in the literature using the R Package RQDA. Discrepancies in any phase of the project will be solved through consensus between the two reviewers. If consensus cannot be reached, a third arbitrator will be consulted.

Ethics and dissemination
Ethical approval is not required; only secondary data will be used. This protocol is registered on the Open Science Framework (osf.io/a78et). The results will be disseminated through publication in a scientific journal and will be used to inform drone field campaigns in the Wellcome Trust funded HARMONIZE project. HARMONIZE aims to develop cost-effective and reproducible digital infrastructure for stakeholders in climate change hotspots in Latin America & the Caribbean and will use drone technology to collect data on fine scale landscape changes.

Ethical considerations related to drone use for environment and health research: A scoping review protocol

Vector-borne disease transmission is driven by environmental factors determining the distribution and spread of pathogens, vectors and human and animal populations. Earth observation (EO) data, such as satellite and drone imagery, can be used to characterise these factors and identify high risk geographical areas and populations. While the use of EO data to understand vector-borne disease epidemiology has a long history, the rapid expansion of satellite and aerial data, analysis methods and computing power offer new opportunities to integrate EO data into disease surveillance. We review sources and characteristics of EO data and analysis methods, identify commonly used EO-derived metrics for a range of diseases and present case studies on specific applications of EO data for disease surveillance. We additionally describe key considerations for disease control programmes considering the use of EO data, highlighting the applicability of different data types and analysis methods for different ecological contexts and use-cases.

Leveraging Earth observation data for surveillance of vector-borne diseases in changing environments

Objectives We aim to explore spatial variations in socioeconomic inequalities in HIV testing uptake in sub-Saharan Africa (SSA) at different geographical scales to identify potential geographical hotspots of inequalities. Additionally, to evaluate the potential benefits of HIV testing programmes, we assess whether local levels of HIV testing match the local levels of HIV prevalence.

Design A multi-country analysis of population-based cross-sectional surveys in SSA.

Setting We analysed data from 25 SSA countries with Demographic and Health Surveys between 2011 and 2019.

Participants Country-level analysis included 473 775 participants (312 104 women and 161 671 men) and cluster-level analysis included 328 283 individuals (241 084 women and 87 199 men). Women aged 15–49 years and men aged 15–54/59 years in selected households who were tested for HIV in the last 12 months were eligible. We quantified inequalities in self-reported recent HIV testing with the Slope Index of Inequality (SII) and the Relative Index of Inequality (RII) across geographical scales to capture sex-specific within-country spatial variations. We also conducted local Getis-Ord Gi* statistics to consider the autocorrelation in fine-scale SII and RII across countries. To assess the efficiency of HIV testing programmes, we measured the correlation between recent HIV testing and HIV prevalence through Spearman correlation across geographical scales.

Results We observed varying inequalities in recent HIV testing in magnitude and spatial distribution on both absolute and relative scales in many countries for both sexes at national and subnational levels. Hotspots of absolute and relative inequalities were mostly observed in Western and Central Africa with a few regions in Eastern and Southern Africa. Despite significant sex-specific correlations between testing and prevalence in all countries when assessed at the national level, we report an absence of such a correlation at fine scale in 17 of 50 sex-country combinations.

Conclusions We highlight the importance of investigating the spatial variability of various HIV indicators and related inequalities across different geographical levels. Results may help inform an equitable distribution of HIV testing services.

Investigating inequalities in HIV testing in sub-Saharan Africa: spatial analysis of cross-sectional population-based surveys in 25 countries

Background
The epidemiological evidence on the interaction between heat and ambient air pollution on mortality is still inconsistent.

Objectives
To investigate the interaction between heat and ambient air pollution on daily mortality in a large dataset of 620 cities from 36 countries.

Methods
We used daily data on all-cause mortality, air temperature, particulate matter ≤ 10 μm (PM10), PM ≤ 2.5 μm (PM2.5), nitrogen dioxide (NO2), and ozone (O3) from 620 cities in 36 countries in the period 1995–2020. We restricted the analysis to the six consecutive warmest months in each city. City-specific data were analysed with over-dispersed Poisson regression models, followed by a multilevel random-effects meta-analysis. The joint association between air temperature and air pollutants was modelled with product terms between non-linear functions for air temperature and linear functions for air pollutants.

Results
We analyzed 22,630,598 deaths. An increase in mean temperature from the 75th to the 99th percentile of city-specific distributions was associated with an average 8.9 % (95 % confidence interval: 7.1 %, 10.7 %) mortality increment, ranging between 5.3 % (3.8 %, 6.9 %) and 12.8 % (8.7 %, 17.0 %), when daily PM10 was equal to 10 or 90 μg/m3, respectively. Corresponding estimates when daily O3 concentrations were 40 or 160 μg/m3 were 2.9 % (1.1 %, 4.7 %) and 12.5 % (6.9 %, 18.5 %), respectively. Similarly, a 10 μg/m3 increment in PM10 was associated with a 0.54 % (0.10 %, 0.98 %) and 1.21 % (0.69 %, 1.72 %) increase in mortality when daily air temperature was set to the 1st and 99th city-specific percentiles, respectively. Corresponding mortality estimate for O3 across these temperature percentiles were 0.00 % (-0.44 %, 0.44 %) and 0.53 % (0.38 %, 0.68 %). Similar effect modification results, although slightly weaker, were found for PM2.5 and NO2.

Conclusions
Suggestive evidence of effect modification between air temperature and air pollutants on mortality during the warm period was found in a global dataset of 620 cities.

Joint effect of heat and air pollution on mortality in 620 cities of 36 countries

Objective To evaluate lag-response associations and effect modifications of exposure to floods with risks of all cause, cardiovascular, and respiratory mortality on a global scale.

Design Time series study.

Setting 761 communities in 35 countries or territories with at least one flood event during the study period.

Participants Multi-Country Multi-City Collaborative Research Network database, Australian Cause of Death Unit Record File, New Zealand Integrated Data Infrastructure, and the International Network for the Demographic Evaluation of Populations and their Health Network database.

Main outcome measures The main outcome was daily counts of deaths. An estimation for the lag-response association between flood and daily mortality risk was modelled, and the relative risks over the lag period were cumulated to calculate overall effects. Attributable fractions of mortality due to floods were further calculated. A quasi-Poisson model with a distributed lag non-linear function was used to examine how daily death risk was associated with flooded days in each community, and then the community specific associations were pooled using random effects multivariate meta-analyses. Flooded days were defined as days from the start date to the end date of flood events.

Results A total of 47.6 million all cause deaths, 11.1 million cardiovascular deaths, and 4.9 million respiratory deaths were analysed. Over the 761 communities, mortality risks increased and persisted for up to 60 days (50 days for cardiovascular mortality) after a flooded day. The cumulative relative risks for all cause, cardiovascular, and respiratory mortality were 1.021 (95% confidence interval 1.006 to 1.036), 1.026 (1.005 to 1.047), and 1.049 (1.008 to 1.092), respectively. The associations varied across countries or territories and regions. The flood-mortality associations appeared to be modified by climate type and were stronger in low income countries and in populations with a low human development index or high proportion of older people. In communities impacted by flood, up to 0.10% of all cause deaths, 0.18% of cardiovascular deaths, and 0.41% of respiratory deaths were attributed to floods.

Conclusions This study found that the risks of all cause, cardiovascular, and respiratory mortality increased for up to 60 days after exposure to flood and the associations could vary by local climate type, socioeconomic status, and older age.

Mortality risks associated with floods in 761 communities worldwide: time series study

Evidence linking traffic noise to insulin resistance and diabetes is limited and unanswered questions remain regarding the potential effect modification by neighborhood socioeconomic status (nSES). We aimed to assess socioeconomic inequalities in noise exposure, whether road and aircraft noise exposures were associated with insulin resistance or diabetes, and whether nSES modified these relationships. Among the Community of Mine Study in San Diego County, road and aircraft noise exposure at enrollment was calculated based on the static (participant's administrative boundary, and circular buffer around participant homes), and dynamic (mobility data by global positioning system, GPS) spatio-temporal aggregation methods. Associations of noise with insulin resistance (HOMA-IR) or type 2 diabetes (T2DM) were quantified using generalized estimating equation models adjusted for sex, age, ethnicity, individual income, and air pollution (nitrogen dioxide) exposure. Additive interaction between noise and nSES was assessed. Among 573 participants (mean age 58.7 y), participants living in low nSES were exposed to higher levels of aircraft and road noise using noise level at the census tract, circular buffer, or Kernel Density Estimation (KDE) of GPS data. Participants exposed to road noise greater or equal to the median (53 dB(A)) at the census tract and living in low nSES had an increased level of insulin resistance (β = 0.15, 95%CI: −0.04, 0.34) and higher odds of T2DM (Odds Ratio = 2.34, 95%CI: 1.12, 4.90). A positive additive interaction was found as participants living in low nSES had higher odds of T2DM. The impact of noise exposure on insulin resistance and T2DM differs substantially by nSES. Public health benefits of reducing exposure to road or aircraft noise would be larger in individuals living in low nSES.

Aircraft and road traffic noise, insulin resistance, and diabetes: The role of neighborhood socioeconomic status in San Diego County