Scientists from Columbia University, the Columbia Mailman School of Public Health and New York University led the analysis, which addresses a widespread health concern. Naturally occurring arsenic in groundwater remains a significant challenge across the world. In the United States, more than 100 million people depend on groundwater that can contain arsenic, particularly those using private wells. Arsenic continues to be one of the most common chemical contaminants in drinking water.

"We show what happens when people who are chronically exposed to arsenic are no longer exposed," said co-lead author Lex van Geen of the Lamont-Doherty Earth Observatory, part of the Columbia Climate School. "You're not just preventing deaths from future exposure, but also from past exposure."

Two Decades of Data Strengthen the Evidence

Co-lead author Fen Wu of NYU Grossman School of Medicine said the findings offer the clearest proof yet of the connection between lowering arsenic exposure and reduced mortality risk. Over the course of two decades, the researchers closely tracked participants' health and repeatedly measured arsenic through urine samples, which strengthened the precision of their analysis.

"Seeing that our work helped sharply reduce deaths from cancer and heart disease, I realized the impact reaches far beyond our study to millions in Bangladesh and beyond now drinking water low in arsenic," said Joseph Graziano, Professor Emeritus at Columbia Mailman School of Public Health and principal investigator of the NIH-funded program. "A 1998 New York Times story first brought us to Bangladesh. More than two decades later, this finding is deeply rewarding. Public health is often the ultimate delayed gratification."

Clear Drop in Risk When Arsenic Exposure Falls

People whose urinary arsenic levels fell from high to low had mortality rates that matched those who had consistently low exposure for the entire study. The size of the drop in arsenic was closely tied to how much mortality risk declined. Those who continued drinking high-arsenic water did not show any reduction in chronic disease deaths.

Arsenic naturally accumulates in groundwater and has no taste or smell, meaning people can drink contaminated water for years without knowing it. In Bangladesh, an estimated 50 million people have consumed water exceeding the World Health Organization's guideline of 10 micrograms per liter. The WHO has described this as the largest mass poisoning in history.

From 2000 to 2022, the Health Effects of Arsenic Longitudinal Study (HEALS) monitored thousands of adults in Araihazar, Bangladesh. The project tested more than 10,000 wells in a region where many families rely on shallow tube wells with arsenic levels ranging from extremely low to dangerously high.

Researchers periodically measured arsenic in participants' urine, a direct marker of internal exposure, and recorded causes of death. These detailed data allowed the team to compare long-term health outcomes for people who reduced their exposure with those who remained highly exposed.

Community Efforts Created a Natural Comparison Group

Throughout the study period, national and local programs labeled wells as safe or unsafe based on arsenic levels. Many households switched to safer wells or installed new ones, while others continued using contaminated water. This created a natural contrast that helped researchers understand the effects of reducing exposure.

Arsenic exposure decreased substantially in Araihazar during the study. The concentration in commonly used wells fell by about 70 percent as many families sought cleaner water sources. Urine tests confirmed a corresponding decline in internal exposure, averaging a 50 percent reduction that persisted through 2022.

Reduced Exposure Brings Lasting Health Benefits

These trends held true even after researchers accounted for differences in age, smoking and socioeconomic factors. Participants who remained highly exposed, or whose exposure rose over time, continued to face significantly higher risks of death from chronic diseases.

The researchers compared the health benefits of lowering arsenic to quitting smoking. The risks do not disappear immediately but drop gradually as exposure decreases.

In Bangladesh, well testing, labeling unsafe sources, drilling private wells and installing deeper government wells have already improved water safety for many communities.

"Our findings can now help persuade policymakers in Bangladesh and other countries to take emergency action in arsenic 'hot spots'," said co-author Kazi Matin Ahmed of the University of Dhaka.

To reach more households, the research team is collaborating with the Bangladeshi government to make well data easier to access. They are piloting NOLKUP ("tubewell" in Bangla), a free mobile app created from more than six million well tests. Users can look up individual wells, review arsenic levels and depths, and locate nearby safer options. The tool also helps officials identify communities that need new or deeper wells.

Clean Water Investments Can Save Lives

The study shows that health risks can fall even for people who were exposed to arsenic for years. This highlights an important opportunity: investing in clean water solutions can save lives within a single generation.

"Sustainable funding to support the collection, storage and maintenance of precious samples and data over more than 20 years have made this critically important work possible," said Ana Navas-Acien, MD, PhD, Professor and Chair of Environmental Health Sciences at Columbia Mailman School of Public Health. "Science is difficult and there were challenges and setbacks along the way, but we were able to maintain the integrity of the samples and the data even when funding was interrupted, which has allowed us to reveal that preventing arsenic exposure can prevent disease."

The study team included researchers from Columbia University's Mailman School of Public Health, the New York University Grossman School of Medicine, Lamont-Doherty Earth Observatory, Boston University School of Public Health, the Department of Geology at the University of Dhaka and the Institute for Population and Precision Health at the University of Chicago.

The HEALS project was launched by Columbia University through the National Institute of Environmental Health Sciences' Superfund Research Program, with most U.S. collaborators based at Columbia when the study began.