Federal Safety Commission
Pushes CO Detectors

Odorless and colorless carbon monoxide gas (CO) is the number one cause of death by poisoning in the United States and accounts for more than one in five of all unintentional deaths by poisoning. Media accounts generally cite government data showing 5,000 CO deaths annually. This number comes from the U.S. Centers for Disease Control, which estimates there were 56,133 deaths from CO poisoning in the United States between 1979 and 1988. Of the total, 41,622 deaths--nearly 80%--resulted from suicide, fires, or homicide. Some 2,964 CO deaths resulted from "unidentifiable" causes while 11,547 deaths were classified as "unintentional."

According to the U.S. Consumer Product Safety Commission (CPSC), in 1989, the most recent year for which statistics are available, there were about 220 deaths from CO poisoning associated with gas-fired appliances, about 30 CO deaths associated with solid-fueled appliances (like charcoal grills), and about 45 CO deaths associated with liquid-fueled heaters.

These numbers probably underestimate the number of people exposed to non-fatal CO levels, because symptoms of poisoning--including headaches, nausea, fatigue, dizziness--are sometimes mistaken for the flu and go untreated.

Reflecting a growing awareness of the dangers of poisoning, the CPSC--which recommends that CO detectors meeting Underwriters Laboratory (UL) standard 2034 be installed in all existing residential buildings--has been moving toward requiring the detectors in newly built homes. The CPSC is also working with state and local code jurisdictions to incorporate CO detector requirements into state and local legislation, and is working with the National Fire Protection Association to develop a national installation standard.

Detectors meeting UL standard 2034 currently cost between $35 and $80. (for a list of manufacturers meeting the standard, contact UL, 333 Pfingsten Road, Northbrook, Illinois, 60062-2096, Tel: [708]272-8800). Because the toxic effect of CO depends on both concentrations and length of exposure, long-term exposure to a low concentration can produce effects similar to short-term exposure to a high concentration. Detectors meeting the UL standard, therefore, measure both high concentrations over short periods, and low concentrations over long periods. It is estimated that between one and two million such detectors are now in the marketplace, but demand for them is expected to grow rapidly (especially if codes require them).

Following several deaths by CO poisoning, and one particular case where an improperly installed furnace killed a family of ten, Chicago last year became the first city to adopt an ordinance requiring UL-approved CO detectors in all new single-family homes and in existing single-family residences equipped with new combustion furnaces.

The ordinance, however, has been somewhat controversial. When a detector alarm sounds, residents open doors and windows to ventilate their homes, and then call the fire department, gas utility, poison control center, or another response agency. When inspectors appear at the home, they may not be able to detect CO because the windows were opened. This leads to questions about whether the alarm sounded at an appropriate level, whether it was accurate, and so on.

With a burgeoning demand for CO detectors, proper protocols and measurement methods for responding to alarms are needed so that questions about the extent to which alarms represent actual situations of elevated CO levels can be answered. Who, for example, should respond to an alarm and under what conditions--the fire department, a heating technician, or the local utility?

In a space of two months, Chicago's fire department last year responded to nearly 6,000 calls, but according to one department spokesman, all but 33 of them were "unfounded." The fire department reportedly received 244 "unfounded" calls on Thanksgiving day alone.

Some city officials have called for a repeal of the ordinance, but other officials credit the ordinance with having saved 40 lives thus far.

Underwriters Laboratories, meanwhile, is considering changes to its CO detector standard. One proposal would change the "stability test" from 15 ppm for eight hours to 15 ppm for 30 days so that detectors would ignore lower CO concentrations. The standard was written in the context of the existing technology for detectors (in 1992) and new technology may soon dictate a need for an updated standard.