The Planetary Health Alliance
Background: β-N-Methylamino-L-alanine (BMAA) is an inducer of neurodegenerative disorders. It has been suggested that chronic ingestion of BMAA through the food chain could cause amyotrophic lateral sclerosis (ALS) and other related pathologies including Parkinson syndrome. BMAA is produced by cyanobacteria in different environments, including the Ocean, and BMAA production has been shown to increase under nitrogen depletion in pure cultures of non-nitrogen-fixing cyanobacteria. In the Ocean, a major fraction of primary production is performed by cyanobacteria, mostly in regions connected to highly productive eastern boundary upwelling systems. Those regions are often characterized by strong oxygen depletion, and enhanced nitrogen loss, but provide an essential fraction of fish production for the local and global nutrition. Thus, those areas are of particular interest regarding BMAA production transport to humans through the food chain. The predicted expansion of those oxygen and nitrogen depleted waters may thus have a severe impact on BMAA production and consequently on human health development.
Methods: A coupled approach of satellite monitoring, chlorophyll, nitrogen and oxygen sensing via autonomous Ocean profilers (gliders) and ship-board measurements, and metagenomics mining to identify possible hotspots of BMAA production was applied. The obtained environmental datasets were complemented with studies of cyanobacterial pure cultures and enrichments to quantify potential rates of BMAA production.
Importance: In the upwelling systems of Mauritania and Peru, possibly important spots of BMAA production were identified by satellite analysis and direct pigment measurements. High abundances of cyanobacteria with the genetic capability for BMAA production have been measured. Culture experiments on related organisms, as well as on enrichment cultures from those regions confirmed enhanced BMAA production under nitrogen depletion. A statistical approach was used to estimate the BMAA production in both upwelling systems under different nitrogen availabilities, and the amount of BMAA ending up in anchovies, which form a major part of the fish catch in those areas. Follow-up analysis will target a potential increase in BMAA production and its transport through the food chains in both upwelling systems in the context of climate change and expansion of nitrogen depleted waters along productive upwelling regions.
Funding has been provided by the EU H2020 program, grant#704272 (NITROX), and by D-IAS. There are no competing interests.
Background: Brazil is home to a large portion of the world’s biodiversity, but, paradoxically, the country’s agriculture and food security are very much reliant on exotic or introduced crops or species. Native species are being neglected and forgotten, with natural landscapes and traditional foods being replaced by monocultures of commodity crops and pasture for livestock, causing habitat and biodiversity loss. At the same time, Brazil faces simplification of diets and high levels of malnutrition. Much of the native neglected biodiversity is highly nutritious and exploring these food sources could help provide sustainable solutions to diversifying diets, tackling malnutrition problems, promoting agricultural development through local food procurement, while also promoting biodiversity conservation, climate change adaptation and resilience. However, significant knowledge and evidence gaps still need to be addressed in order to better integrate biodiversity for enhancing food and nutrition security, including scientific data on the nutritional value and composition of native underutilized species. The “Biodiversity for Food and Nutrition” (BFN) project is working to generate food composition data, develop culinary recipes and increase appreciation and awareness about the value of underutilized, nutrient-rich biodiversity through strategic research partnerships with universities and research institutes in Brazil.
Methods: Partnership was established with 9 universities and research institutes for the generation of food composition data (macronutrients, vitamins and minerals) and development of culinary recipes for 70 native underutilized species with economic potential. Composition data was generated by compilation from available literature and laboratory analysis. Data will be made available at a food composition and recipes database being developed with the Information System on Brazilian Biodiversity (SiBBr).
Findings and interpretations: Food composition data revealed high nutrient contents of many of the studied species, such as peach palm (Bactris gasipaes), buriti (Mauritia flexuosa), camu-camu (Myrciaria dubia) and Mangaba (Hancornia speciosa). The data can be used as an advocacy tool for promotion of native species on public initiatives and allocation of incentives for improving their production and market chains. By working through regional partners, capacities were developed in different regions, engaging directly more than 100 students and professors. These groups act as multipliers, building additional human capacity and also operating as opinion leaders and policy advisors, including the provision of research and technical backstopping for municipal managers, school managers, nutritionists and cooks responsible for implementing the National School Feeding Programme (PNAE), thus the partnership is likely to favor the inclusion of biodiversity in school meals.
The goal of improving urban health and sustainability raises complex questions best addressed through inter- and even transdisciplinary approaches, in which scientific research and analysis and stakeholder engagement have important roles. We report pilot work in Nairobi and London that uses innovative methods to integrate qualitative and quantitative modelling to provide evidence to support policy development for health and sustainability in these cities.
We utilised two primary modelling methods, system dynamics and microsimulation, and sought to understand the value of these tools in combination to support policy decisions. System dynamics was used to derive an aggregated and nonlinear causal map of the interconnections between parameters of influence, and thus to gain insight into the policies and specific processes that need to be examined in further depth. It was a key tool for city-level stakeholder engagement. In part informed by the outcome of the system dynamics process, microsimulation was then used to quantify local impacts on health of selected policy options. The results were mapped using geographic information systems methods.
The combination of system dynamics and microsimulation models provided a framework that enhanced collective knowledge about the interrelationships of policy decisions, funding, public awareness and environmental and health effects. Our initial participatory system dynamics work on air pollution in Nairobi found that investments in monitoring and health impact assessment have the potential to trigger reinforcing mechanisms that create synergies among existing policies and elevate the return on investment. Preliminary microsimulation runs of the effects of policies to reduce fine particle pollution (PM2.5) in London revealed novel insights into the potential benefits and costs for health and the health care system. The two methods appeared to have valuable complementarity in their focus on aggregated dynamics at the policy level vs. local policy effects. The full paper will report on their combination in more detail.
The use of system dynamics can produce a quantitative model of the policy implementation process including the organisational barriers and opportunities for change. This can be extended to include aggregate outputs from other models to quantify a more holistic and high-level quantitative model of the dynamics of selected policy questions. Together, these methods can estimate regional environmental and local health effects of selected policies, but also inform about overcoming the barriers to these policies.
Projections suggest that by 2050 climate change will reduce global fish catch by 3-13%, with fish catch falling by as much as 30% in some tropical regions. Freshwater fisheries are particularly susceptible to the warming effects of climate change as shallower, hydrologically distinct water bodies are readily shaped by atmospheric temperatures and may not accommodate fish migrations, and poor and undernourished populations are especially dependent on freshwater fisheries. Despite the severity of projected climate change effects on fish catch and the risk to human health, little empirical work examines how fish catch is already responding to climate change, the ways fishers are adapting to these changes, and how it impacts their consumption of nutritious fish. We aim here to account for behavioral responses among fishermen to identify the ecological effect of climate change on fish catch, patterns of fish consumption, and nutrition in Cambodian rice field fisheries.
We use a panel dataset of 400 households over 3 years (19 distinct time points) to examine how changing temperatures alter households’ fish catch and whether fishing families respond by shifting their a) fishing participation, b) effort (i.e., hours, time of day, family members), or c) fish consumption. Our analytic approach involves three analyses: 1) the net effect of biophysical changes on household fish catch, 2) the effect of biophysical changes on household fish catch and fish consumption with the addition of controls for fishing effort and fish consumption, a key way that fishers may adapt to ecological changes, and 3) the direct effect of biophysical changes (temperature) on fishing effort and fish consumption.
Preliminary results suggest effects of climate change on fish catch and that fishing families are moderating their behavior to adapt to changing temperatures. Updated data and analyses will allow us to disentangle these effects by using an identification strategy that allows us to separate the ecological effects of temperature change on fish catch from the ways households respond to changes in fish availability.
Our results have broad implications for understanding the ways that climate change may impact the 10% of the global population that relies on small scale fisheries in developing countries. Analyses of fish catch will demonstrate the effects of temperature change on fish catch and how households respond to these changes. Further, analyses of the effects on consumption of fish will demonstrate how temperature further alters access to nutritious food. Our findings suggest the ways that changing climatic temperatures may impact the health and well-being of natural resource-dependent communities.
Background: Animal pollinators support biodiversity and increase crop yields. Anthropogenic environmental changes (including habitat loss, climate change, and the global use of agrochemicals) threaten the health of pollinator populations, posing a direct risk to agricultural systems and human health outcomes via impacts on nutrition. However, our tools for studying and predicting the impacts of these stressors (and their interactions) on animal pollinators such as bees are limited. Studying the effects of neonicotinoid pesticides (and their interactions with other stressors such as climate change) is particularly challenging, since these compounds affect bees by impacting subtle aspects of behavior at levels well below acute toxicity.
Methods: Here, we first describe the development of an automated, high-throughput system for studying the combined effects of neonicotinoid pesticides and temperature on bumble bees, the most important native crop pollinators in North America. Combining computer vision and thermal imaging with environmental control in a robotic tracking system, we quantify worker behavior and performance across multiple bee colonies in parallel exposed chronically to sub-lethal levels of imidacloprid (the most common neonicotinoid used globally). Second, we validate the findings of this laboratory-based approach with studies under field conditions.
Findings: We find that chronic exposure to imidacloprid (at concentrations bees are likely to encounter in agricultural settings) significantly impacts critical components of nest behavior in bumble colonies (Bombus impatiens), including nest care and social interactions. Imidacloprid also reduces the colony’s ability to socially thermoregulate (i.e. maintain stable temperatures of developing young within the nest).
Interpretation: Our results suggest that neonicotinoid exposure interacts with climatic variation by exacerbating the impacts of temperatures fluctuations on bee colonies, likely resulting in impaired colony growth. We are currently extending this work to explore whether similar effects occur in honey bees (Apis mellifera), as well as using a modeling approach to predict the combined effects of temperature variation and neonicotinoid use on wild bumble bee populations and agricultural productivity in the continental USA under different climate scenarios. More broadly, our results highlight the potential of automated, ethomic approaches for screening agrochemicals and predicting their impacts on agricultural systems and human health outcomes.
Background: The food system is a major driver of climate change, land-use change and biodiversity loss, and pollution of aquatic and terrestrial ecosystems. Many of those domains might have already crossed planetary boundaries that describe the safe operating space for humanity. At the same time is the food system far away from providing healthy diets for a growing population. Current diets do not meet dietary recommendations, and imbalanced diets contribute substantially to the global burden of disease. Here we analyse the option space for the food system to stay within the planetary boundaries of climate change, land and water use, and biogeochemical flows of nitrogen and phosphorus. We differentiate between technological changes, changes in food loss and waste, and dietary changes, all of which can reduce the environmental pressures of the food system, and, in the case of dietary change, can have important health implications and potential co-benefits.
Methods: To analyse pathways towards sustainable food systems, we constructed a food systems model that connects food consumption in one region to food production and the associated environmental impacts in other regions. We used region-specific data and projections to derive current and future environmental footprints at the country and food group-level for land use, water use, greenhouse-gas emissions, nitrogen surplus, and phosphorous application. For each domain, we constructed scenarios of food-system change of standard and high ambition, including improvements in yields, fertilizer application, feed efficiency, and management practices; reductions in food loss and waste; and consumption changes towards dietary guidelines and alternative dietary patterns in line nutritional requirements. For each scenario, we considered different socio-economic pathways of population and income growth.
Results: In the absence of interventions, the pressures of the food system on climate change, land and water use, and biogeochemical flows of nitrogen and phosphorus are set to increase by 60-90% between 2010 and 2050. Reduction in food loss and waste reduced environmental pressures by 11-20%, changes in technology and management by 12-55%, and dietary changes by 4-82%, with high variability across domains and scenarios. Combining scenarios of standard ambition was not enough to stay below the suggested planetary boundaries. Instead, a combination of ambitious dietary change towards more flexitarian and plant-based diets, ambitious changes in technology and management, and reductions in food loss and waste were needed to simultaneously meet the food-related planetary boundaries for climate change, land and water use, and biogeochemical flows of nitrogen and phosphorus.
Change in how land is used is accelerating in many places globally and can, by altering the ecology of microhabitats suitable for the Anopheles mosquito vector, modify the potential for malaria infection in a landscape. This can in part explain why malaria transmission is spatio-temporally variable and why localized foci of a significantly higher burden are often observed. However, numerous other factors also contribute to the nonrandom distribution of malaria infections, such as human demography and behavior, immune history and current health status, and access to adequate health services. Therefore, integration of ecological, demographic, and socio-economic data is needed to fully understand the forces driving malaria transmission hotspots. Such research can be applied to develop land use practices that mitigate malaria transmission.
Large, rich, interdisciplinary data sets derived from two prospective cohort studies (n = 701 and n = 856; both sexes and all ages) and one cross-sectional study (n = 5,598; both sexes and all ages) from a total of 31 localities in Madagascar were analyzed. Clinical (including malaria infection outcome) and health and socio-economic survey data for each individual participant were paired with mosquito vector habitat transects and remote sensing geo-spatial data in hierarchical models. To better examine transmission patterns in the spatiotemporal distribution of malaria infections, genetic analyses using a panel of polymorphic genetic markers were performed to examine parasite genetic diversity.
We observed a nonrandom distribution of malaria infections at multiple spatial scales and provide evidence of hotspots in malaria transmission at the individual, household, and community levels. Malaria prevalence in many of the rural communities in Madagascar studied here was significantly higher than expected from published national or regional estimates. Genetic data, especially in the Northeast, were consistent with the presence of stable pockets of high transmission in remote areas where the landscape is dominated by recent and ongoing conversion of forest to agricultural land, primarily for rice cultivation.
Remote communities in Madagascar, communities at the frontier of the rapid ecological transformation being seen globally, have a disproportionately higher malaria burden. Integrating diverse data streams allows a more complete understanding of the ecological and socio-demographic contributors to this pattern and highlights the need for greater targeting of these vulnerable communities. We demonstrate the utility of a planetary health perspective in both identifying areas at greater risk and providing a framework for future policies and practice that jointly promote environmental and human health.
Evidence suggests that human-induced land-use such as agriculture and deforestation may be drivers or contributing factors of infectious disease outbreaks. Many studies describe mechanisms by which land use may impact infectious disease but fail to quantify the association. The aim of this study is to conduct a systematic review and meta-analysis investigating the risk of infectious diseases associated with land use in South East Asia.
A systematic review of the literature addressing land use or land use change and infectious disease prevalence or incidence in humans was conducted using a robust search strategy. Outcome data as a function of exposure to different land uses were extracted and crude odd’s ratios were calculated. A meta-analysis was conducted using a random effects model. Heterogeneity of effects was analysed using subgroup analysis, the I2 test for heterogeneity and the Cochrane Q Test. Sub-groups were defined a priori based on epidemiological groupings / hypothesised mechanisms to avoid post hoc data dredging. Publication bias was assessed using funnel plots, linear regression tests and the trim and fill method.
37 studies were deemed eligible for meta-analysis. 35 (95%) studies assessed land use exposures related to agricultural activities. The remaining studies assessed tin mining and reservoir construction. Overall, people exposed to agricultural land use were 60% more likely to be infected than controls (OR 1.60, CI 1.38–1.85, p < 0.001). The strongest effects were observed for mosquito borne diseases (OR 2.09, CI 1.39 – 3.16, I2 = 37.2%, p = 0.0004) followed by rodent, flea or tick-borne diseases (OR 1.76, CI 1.00 – 3.10, I2 = 87.5%, P = 0.0495) in forest monoculture such as palm oil and rubber. Nevertheless, effect sizes varied considerably across land-use types and disease categories. No evidence of significant publication bias was found.
Our study provides robust and generalisable evidence that agricultural land use significantly increases the risk of numerous infectious diseases with a range of epidemiology and ecologies in SE Asia. Given the large amount of environmental change that is being undergone globally, there is a need to assess causality of such mechanisms through longitudinal studies.
Urbanisation is a major demographic trend globally. Informal settlements account for much urban growth, exacerbating the inextricably linked challenges of sanitation, water provision, and public health. The Wellcome Trust-funded Revitalising Informal Settlements and their Environments (RISE) program is empirically testing an alternative, water-sensitive approach to this challenge that integrates sustainable design with the management of the water-cycle, benefiting human health and urban ecosystems.
The RISE program is empirically testing whether: (1) Water-sensitive revitalisation leads to improved physical characteristics, altering microbial and biological diversity, bioacoustics, greenness, and flood hazards; and decreasing potential sites for vector breeding and pathogen contamination; (2) The interruption of faecal–oral transmission reduces infection, leading to reduced enteric inflammation and carriage of drug-resistant gene markers, increased diversity of the gastrointestinal microbiome, and increased responsiveness to oral vaccines. The RISE hypothesis is that the changed physical environment and improved water-servicing will lead to enhanced psychological, social, and economic outcomes, resulting in further benefits to health and wellbeing. Collectively, improvement in these factors could have major implications for health at individual and community levels.
In close consultation with government stakeholders and communities, RISE is revitalising 12 settlements in Makassar and 12 in Suva, each site comprising an average of 50 dwellings assuming 5–6 people per dwelling with a total of 6,000–7,200 people. This intervention scale and sample size will ensure statistical power for primary health and environmental outcomes. The research program utilizes a parallel-cluster randomised controlled trial (RCT) design with stratified randomisation by tidal/non-tidal sites. Six settlements in each of Suva and Makassar will be randomly assigned to the intervention group, and 6 to controls. Environment and human health monitoring will be conducted for all settlements for 24 months post-construction to monitor and assess the effects on primary outcomes.
The presentation will report on progress to date in implementing the RISE research program. Site diagnostic and baseline environmental and health data will be presented for all 24 settlements under study.
Background: Asthma is the most prevalent chronic respiratory disease worldwide, affecting 358 million people in 2015. Ambient air pollution exacerbates asthma among populations around the world and may also contribute to new-onset asthma. However, the number of air pollution-related cases of exacerbated asthma and new onset asthma globally remains unquantified. We quantitatively estimate asthma ERVs and, as a secondary analysis, new asthma cases that are attributable to ambient ozone, fine particulate matter (PM2.5), and nitrogen dioxide (NO2) globally.
Methods: We first combine national asthma incidence and prevalence rates and a newly constructed dataset of asthma emergency room visit rates from survey data in 54 countries and Hong Kong with globally gridded population counts to estimate the baseline number of new asthma cases and asthma ERVs at 0.1°x0.1° resolution. We then estimate the pollution-attributable fraction of asthma emergency room visits and new asthma cases using globally gridded pollution concentrations, derived from satellite remote sensing and chemical transport modeling (0.1°x0.1°), with concentration-response factors, drawn from several meta-analyses of epidemiological studies in 27 countries spanning North America, Europe, Asia, and Latin America.
Findings: We show that 9-23 million and 5-10 million annual asthma emergency room visits globally in 2015 could be attributable to ozone and fine PM2.5, corresponding to 8-20% and 4-9% of the global total. The range reflects the application of central risk estimates from different epidemiological meta-analyses. Anthropogenic emissions are responsible for approximately 37% of ozone impacts and 73% of PM2.5 impacts. The remaining impacts are attributable to naturally-occurring ozone precursor emissions (e.g. from vegetation and lightning) and PM2.5 (e.g. dust, sea salt), though several of these sources are also subject to human influence from land use changes, anthropogenic climate changes, and other interactions between humans and the environment. India and China experience 23% and 10% of all asthma emergency room visits attributable to ozone globally, and 30% and 12% of those attributable to PM2.5.
Interpretation: Our results show that mitigating ambient air pollution could substantially reduce the burden of asthma globally.
Although WHO discourages drinking demineralized water, rainwater is currently being promoted in salinity affected coastal Bangladesh as a low-salinity drinking water source. Rainwater has low sodium levels but can be associated with adverse health consequences due to lack of beneficial minerals such as calcium, magnesium and potassium. We compared the mineral intake and blood pressure (BP) among rainwater and groundwater drinkers.
Methods: We analyzed data from a stepped-wedge trial that followed-up 1,191 participants over five successive months between December, 2016 and April, 2017. At each visit, households reported their primary drinking water source; we measured participants’ BP, 24-hour urinary sodium, potassium, calcium and magnesium. We used mixed models with random intercepts at participants-, household- and community-level to determine the association between drinking water sources with mineral intake and BP. We reported cluster robust standard error and adjusted for age, sex, BMI, physical activities, smoking, marital and socio-economic status.
Result: The median 24-hour urinary concentrations among rainwater and groundwater drinkers were 140.3 & 162.5 mmol/day for sodium, 23.7 & 32.1 mmol/day for potassium, 26.8 & 38.2 mmol/day for calcium and 2.8 & 3.6 mmol/day for magnesium. In full adjusted models, compared to groundwater drinkers, drinking rainwater was associated with 13.12 (95% CI: 4.75, 21.56) mmol/day lower mean 24-hour urinary sodium, 1.71 (95% CI: -0.09, -3.5, 0.09) mmol/day lower potassium, 14.32 (95% CI: 9.57, 19.089.57) mmol/day lower urinary calcium, and 0.82 (95% CI: 0.33, 1.31) mmol/day lower urinary magnesium. The mean systolic and diastolic BP of the rainwater drinkers was 112.3 & 68.0 mmHg, compared to 108.3 & 64.7 mmHg for groundwater drinkers. In fully adjusted models, drinking rainwater was associated with +3.00 (95% CI: 1.47, 4.53) mmHg systolic and +2.20 (95% CI: 1.44, 2.96) mmHg diastolic BP compared to groundwater drinkers. We observed 0.80 (95% CI: 0.40, 1.20) mmHg increase in systolic BP associated with each 50 mmol/day increase in urinary sodium, 0.50 (95% CI: -0.80, -0.20) mmHg decrease associated with each 10 mmol/day increase in potassium, 0.15 (95% CI: -0.30, -0.01) mmHg decrease associated with each 10 mmol/day increase in calcium, and 0.26 (95% CI:- 0.39, -0.14) mmHg decrease associated with each 1 mmol/day increase in magnesium.
Inference: Drinking rainwater was associated with significantly lower intake of calcium and magnesium, and higher blood pressure compared to groundwater drinkers. In settings without other dietary sources of these minerals, rainwater should be supplemented by minerals to avoid potential adverse health consequences.