Author(s): Casey Grant. Published on May 1, 2017.

Survey Says

Why we need to learn more about the use of residential smoke and carbon monoxide alarms

One of the most important technologies in the history of modern fire protection has been residential smoke and carbon monoxide (CO) alarms, but we have very little reliable data today on how they are being used, or even what percentage of homes have working alarms. That’s because it’s been a quarter century since researchers have surveyed U.S. homes to test smoke alarms and conduct interviews.

This information is critical, because understanding how smoke and CO sensing technologies are being used is crucial for developing and implementing the most effective policies, codes and standards, education programs, and research initiatives. As detection and alarm technologies continue to advance, understanding how current devices are used will help answer questions about how to best implement future ones.


The importance and logistics of collecting this information was the topic of a meeting I attended in February at the Consumer Product Safety Commission (CPSC), the same organization that last conducted a smoke alarm survey of U.S. households in 1992. This new survey is scheduled to start late this year and will involve visits to a statistically balanced sampling of 1,200 residences. This is a big effort, and to do it right requires proper planning. Attendees, who included fire service members, enforcers, public educators, consumer representatives, technology providers, and others, offered important feedback for refining the plan. As I listened, my profound appreciation for the importance of this effort was fully confirmed.

Home fires are far and away our biggest threat in terms of how people die in fires each year. According to NFPA’s 2016 “Home Structure Fires” report, 74 percent of U.S. structure fires between 2010 and 2014 occurred in homes, killing an average of 2,520 people and injuring 12,720 annually. Roughly three of every five home fire deaths happened in homes without working smoke alarms, including 39 percent in homes with no smoke alarms at all and 19 percent in homes with smoke alarms that did not operate. In recent years, CO—known as a “silent killer” because it is a colorless, odorless, poisonous gas—has gained more notoriety. According to the Centers for Disease Control and Prevention, unintentional non-fire carbon monoxide poisoning killed an average of 430 people per year from 1999 to 2010. Even so, no comprehensive survey, including the 1992 CPSC study, has ever been conducted on CO detection and alarm use.

The results of the CPSC’s landmark smoke alarm survey, published in 1994, suggested that we still have much work to do, making this year’s update that much more important. In the study, which was conducted in 1992, researchers visited 1,000 homes to hold in-person interviews with residents about how many and what kind of smoke alarms were installed in their homes, the ways alarms can fail, factors leading to nonworking alarms, and the types of households most likely to have nonworking smoke alarms. They found that only 66 percent of all households had at least one smoke alarm that worked when tested. One-quarter of households with smoke alarms did not have alarms on every level. Roughly 20 percent of smoke alarms in homes did not have a functioning power source, while 12 percent of households that reported that all their smoke alarms worked had at least one nonworking alarm. We are overdue to find out what the landscape looks like today to better inform our efforts to promote these lifesaving devices.

Please make sure to support the upcoming alarm survey if it happens to include your jurisdiction. I believe this information is imperative as we continue to try and move the needle in reducing deaths and injuries due to fire and CO.

CASEY GRANT is executive director of the Fire Protection Research Foundation.