The researchers compared various time windows of exposure, finding that the previous decade of water arsenic exposure up to the time of a cardiovascular disease event contributed the greatest risk. The findings are published in the journal Environmental Health Perspectives.

“Our findings shed light on critical time windows of arsenic exposure that contribute to heart disease and inform the ongoing arsenic risk assessment by the EPA. It further reinforces the importance of considering non-cancer outcomes, and specifically cardiovascular disease, which is the number one cause of death in the U.S. and globally,” said Danielle Medgyesi, a doctoral Fellow in the Department of Environmental Health Sciences at Columbia Mailman School. “This study offers resounding proof of the need for regulatory standards in protecting health and provides evidence in support of reducing the current limit to further eliminate significant risk.”

According to the American Heart Association and other leading health agencies, there is substantial evidence that arsenic exposure increases the risk of cardiovascular disease. This includes evidence of risk at high arsenic levels (>100µg/L) in drinking water. The U.S. Environmental Protection Agency reduced the maximum contaminant level (MCL) for arsenic in community water supplies (CWS) from 50µg/L to 10µg/L beginning in 2006. Even so, drinking water remains an important source of arsenic exposure among CWS users. The natural occurrence of arsenic in groundwater is commonly observed in regions of New England, the upper Midwest, and the West, including California.

To evaluate the relationship between long-term arsenic exposure from CWS and cardiovascular disease, the researchers used statewide healthcare administrative and mortality records collected for the California Teachers Study cohort from enrollment through follow-up (1995-2018), identifying fatal and nonfatal cases of ischemic heart disease and cardiovascular disease. Working closely with collaborators at the California Office of Environmental Health Hazard Assessment (OEHHA), the team gathered water arsenic data from CWS for three decades (1990-2020).

The analysis included 98,250 participants, 6,119 ischemic heart disease cases and 9,936 CVD cases. Excluded were those 85 years of age or older and those with a history of cardiovascular disease at enrollment. Similar to the proportion of California’s population that relies on CWS (over 90 percent), most participants resided in areas served by a CWS (92 percent). Leveraging the extensive years of arsenic data available, the team compared time windows of relatively short-term (3-years) to long-term (10-years to cumulative) average arsenic exposure. The study found decade-long arsenic exposure up to the time of a cardiovascular disease event was associated with the greatest risk, consistent with a study in Chile finding peak mortality of acute myocardial infarction around a decade after a period of very high arsenic exposure. This provides new insights into relevant exposure windows that are critical to the development of ischemic heart disease.

Nearly half (48 percent) of participants were exposed to an average arsenic concentration below California’s non-cancer public health goal

These results highlight the serious health consequences not only when community water systems do not meet the current EPA standard but also at levels below the current standard. The study found a substantial 20 percent risk at arsenic exposures ranging from 5 to

“Our results are novel and encourage a renewed discussion of current policy and regulatory standards,” said Columbia Mailman’s Tiffany Sanchez, senior author. “However, this also implies that much more research is needed to understand the risks associated with arsenic levels that CWS users currently experience. We believe that the data and methods developed in this study can be used to bolster and inform future studies and can be extended to evaluate other drinking water exposures and health outcomes.”

Co-authors are Komal Bangia, Office of Environmental Health Hazard Assessment, Oakland, California; James V. Lacey Jr and Emma S. Spielfogel,California Teacher Study, Beckman Research Institute, City of Hope, Duarte, California; and Jared A Fisher^, Jessica M. Madrigal, Rena R. Jones, and Mary H. Ward^, Division of Cancer Epidemiology and Genetics, National Cancer Institute.

The study was supported by the National Cancer Institute, grants U01-CA199277, P30-CA033572, P30-CA023100, UM1-CA164917, and R01-CA077398; and also funded by the Superfund Hazardous Substance Research and Training Program P42ES033719; NIH National Institute of Environmental Health Sciences P30 Center for Environmental Health and Justice P30ES9089, NIH Kirschstein National Research Service Award Institutional Research Training grant T32ES007322, NIH Predoctoral Individual Fellowship F31ES035306, and the Intramural Research Program of the NCI Z-CP010125-28.