By Diana King

group of people sitting in a circle in a community center in Brazil
Marcia Castro leading a discussion between Harvard students and a community-based mental health project in Fortaleza, Brazil.

Marcia Castro, Andelot Professor of Demography, and chair of the Department of Global Health and Population at the Harvard T.H. Chan School of Public Health and CID faculty affiliate, was studying statistics at Rio de Janeiro State University when Brazil transitioned to a democracy in 1985, ending a 21-year military dictatorship.

She recalls the era as charged with possibility: “There were movements for direct elections, discussions about social welfare and healthcare,” the drafting of a new Constitution, and the building of vast new programs — including the current national health system, one of the largest universal healthcare systems in the world.

A first-generation college graduate whose family did not have regular access to healthcare, Castro has from the start intertwined her work with policy. “It was always important to me to figure out how my research can effect change,” she says.

Her first job after college was in the Brazil's Social Security Ministry, where she served as a statistician, analyzing the sustainability of the system, the aging of the population, and the impact of disability in the workforce. This on-the-job demographic training led her to pursue a master’s in demography at Cedeplar, a research institute affiliated with the Federal University of Minas Gerais, and a PhD in demography at Princeton.

Castro’s research applies statistics, demography, and fieldwork to better understand and control the spread of vector-borne diseases, such as malaria, dengue, and Zika – all of which are on the rise globally, in part due to climate change. The multi-disciplinary approach enables a broad and detailed view of how diverse risk factors (socioeconomic, behavioral, biological, environmental) influence disease spread, capturing regional differences, and providing policy insights.

A leading authority on malaria control in Africa and Brazil, Castro also studies child health and development, including, most recently, the impact of social distancing during COVID on maternal and child health in urban Brazil. Last summer, Castro co-led a training program in early childhood development for a diverse group of Brazilian leaders. Co-sponsored by the David Rockefeller Center for Latin American Studies (where Castro chairs the Brazil Studies program), the initiative aims to improve access, equity, quality and demand for early childhood services across Brazil.

Castro presenting to a group of people seated at a conference table
Marcia Castro leading a training on early childhood development in Brazil.

At the core of Castro’s projects is a commitment to rendering the faultlines, the existing inequities, of the public health system visible in order to address them. “The first step is to quantify,” she says. “But one number in a big, diverse country like Brazil doesn’t mean much.” Instead, Castro focuses on a spread, capturing social and geographical differences through spatial analysis methods she first pioneered during her doctoral research.

At the time, her dissertation advisor could not find a pattern in years of data collected from an agricultural settlement in Rondônia. After taking a NASA-sponsored course, Castro used satellite images and spatial analysis techniques to isolate subsections of the settlement, identifying an important spatial pattern, and uncovering new variables – ecological details such as altitude, vegetation, and type of agriculture turned out to matter a great deal.

She has applied spatial analysis on a larger scale, to all of Brazil, in a recent study on the impact of deforestation on malaria transmission between 2004 and 2022. Her method accounted for a crucial dynamic of transmission often overlooked in previous studies: seasonality.

“The time-scale is extremely important,” she says, because both malaria transmission and periods of deforestation are seasonal (typically disease transmission picks up in rainy months, while forests are cut only during the dry season). The results show a notable increase in transmission in the month following deforestation – an effect missed by past studies that relied solely on annual data. Castro hopes the new evidence will clear away doubts sowed by previous research and guide new policies.

To tackle a disease as complex as malaria, Castro insists, quantitative methods must be paired with fieldwork. “Data is crude; there’s no story, no context behind it,” she says. “It’s when you go into the field, talk to people, get your car stuck in the mud, and so forth, that you learn things not documented anywhere.”

left is car driving on muddy road, right is abandoned house on stilts in rural area
Scenes from Castro's fieldwork in the Brazilian Amazon.

Fieldwork taught Castro that “malaria is not just one disease.” It was remarkable “how little my knowledge of malaria in the Amazon applied in [a new] context,” she says. “The vectors were different, the local setting was different,” she says of her time in Dar es Salaam, where she helped the local government build a malaria control program that reduced infection using larvicide by over 20 percent.

Today, Castro is focused primarily on Brazil, where she “knows the culture, and can understand all the idiosyncrasies of the local setting.” A blunt critic of Brazil’s high levels of inequality, which were worsened by the pandemic, Castro is also frank about how much she benefited from the nation’s public education system.

She has seen governments come and go, and knows on a deep level that who is in charge matters. Good policies can transform lives, and as seen during the previous administration’s Covid response, poor ones cost lives.

“We have a unique window of opportunity with the current government’s focus on social issues to create legislation that saves lives, protects the environment, and leaves things a bit better,” says Castro, who already has her eyes set on the next threat.

A malaria vector native to Asia, highly adapted to cities, has invaded Africa and progressively moves westward – if it lands in Brazil, the cost in lives, left unchecked, would be catastrophic.

CID Faculty Affiliates
CID’s faculty affiliates embody the breadth and depth of international development research at Harvard. Over 125 affiliates hail from across the University and work in every region of the world, on every topic in development.
Read Next Post
View All Blog Posts